Printed scheme. PCB at home. PCB manufacturing

Many people say that making your first PCB is very difficult, but it is actually very simple.

Now I will tell you a couple of well-known ways to make a printed circuit board at home.

To begin with, a short plan of how a printed circuit board is made:

1. Preparation for manufacturing
2. Conductive paths are drawn
2.1 Draw with varnish
2.2 Draw with a marker or nitro paint
2.3Laser ironing
2.4Printing with film photoresist
3.Etching the board
3.1Etching with ferric chloride
3.2 Etching with copper sulphate with table salt
4. Tinning
5.Drilling

1. Preparation for PCB manufacturing

To begin with, we need a sheet of foil textolite, metal scissors or a hacksaw, a regular pencil grater and acetone.

Carefully cut out the necessary piece of foil textolite. Then it is necessary to carefully clean our textolite, on the copper side, with a pencil grater to a shine, then wipe our workpiece with acetone (this is done for degreasing).

Fig 1. Here is my workpiece

Everything is ready, now do not touch the shiny side, otherwise you will have to degrease again.

2. Draw conductive paths

These are the paths through which the current will be conducted.

2.1 We draw paths with varnish.

This method is the oldest and simplest for the Saami. We need the simplest nail polish.

Carefully draw conductive paths with nail polish. Be careful, because the varnish sometimes blurs and the tracks merge. Let the varnish dry. That's all.

Fig 2. Paths painted with varnish

2.2 Draw paths with nitro paint or marker

This method is no different from the previous one, only everything is drawn much easier and faster.

Fig 3. Paths drawn with nitro paint

2.3 Laser ironing

Laser ironing is one of the most common ways to make printed circuit boards. The method is not laborious and takes little time. I have not personally tried this method, but many people I know use it with great success.

First, we need to print a drawing of our printed circuit board on a laser printer. If you don't have a laser printer, you can print on an inkjet printer and then make copies on a copier. I use Sprint-Layout 4.0 to draw up the drawings. Only when printing, be careful using a mirror, many have killed boards in this way more than once.

We will print on some old unnecessary magazine with glossy paper. Before printing, set your printer to maximum toner consumption, this will save you a lot of problems.

Fig 4. Printing a drawing on glossy magazine paper

Now carefully cut out our drawing in the form of an envelope.

Fig 5. Envelope with a diagram

Now we put our blank in the envelope and carefully seal it at the back with tape. We glue it so that the textolite does not move in the envelope

Fig 6. Finished envelope

Now iron the envelope. We try not to miss a single millimeter. It depends on the quality of the board.

Fig 7. Ironing the board

When the ironing is finished, carefully place the envelope in a bowl of warm water.

Fig 8. Soaking the envelope

When the envelope is soaked, we roll up the paper without sudden movements, no matter what damage the toner tracks. If there are defects, take a cd or dvd marker and fix the tracks.

Fig 9. Almost finished board

2.4 PCB fabrication with film photoresist

As in the previous method, we make a drawing using the Sprint-Layout 4.0 program and press print. We will print on a special film for printing on inkjet printers. Therefore, we set up printing: We remove the sides f1, m1, m2; In the options, check the Negative and Frame checkboxes.

Fig 10. Printing setup

We set up the printer for black and white printing and set the maximum intensity in the color settings.

Figure 11. Printer setup

We print on the matte side. This side is working, you can determine it by sticking it to your fingers.

After printing, our template is laid to dry.

Fig 12. Drying our template

Now we cut off the piece of photoresist film we need

Figure 13. Photoresist film

Carefully remove the protective film (it is matte), glue it to our textolite blank

Fig 14. We glue photoresist to textolite

You need to glue it carefully, and remember, the better you press the photoresist, the better the tracks on the board will be. Here's roughly what should happen.

Fig 15. Photoresist on textolite

Now, from the film on which we printed, we cut out our drawing and apply it to our photoresist with textolite. Do not confuse the sides, otherwise you will get a mirror. And covered with glass

Fig 16. We apply a film with a drawing and cover it with glass

Now we take an ultraviolet lamp and illuminate our paths. For each lamp, its parameters for development. Therefore, choose the distance to the board and the glow time yourself

Fig 17. We illuminate the tracks with an ultraviolet lamp

When the tracks light up, we take a small plastic dish, make a solution of 250 grams of water, a spoonful of soda, and lower our board there already without our board template and the second transparent photoresist film.

Fig 18. We lay down the board in a soda solution

After 30 seconds, our print tracks appear. When the dissolution of the photoresist is over, our board will turn out, which we wanted. Rinse thoroughly under running water. All is ready

Fig 19. Finished board

3. Etching of the new PCB. Etching is a way to remove excess copper from PCB.

For etching, special solutions are used, which are made in plastic dishes.

After making the solution, a printed circuit board is lowered there and etched for a certain time. You can speed up the etching time by maintaining the temperature of the solution in the region of 50-60 degrees and constant stirring.

Remember to use rubber gloves when working and then wash your hands well with soap and water.

After etching the board, you need to rinse the board well under water and remove the remaining varnish (paint, photoresist) with ordinary acetone or nail polish remover.

Now a little about solutions

3.1 Etching with ferric chloride

One of the most famous etching methods. For etching, ferric chloride and water are used with a ratio of 1:4. Where 1 is ferric chloride, 4 is water.

Preparing is simple: the right amount of chlorinated iron is poured into the dishes and poured with warm water. The solution should turn green.

Etching time for a 3x4 cm board, around 15 minutes

You can get ferric chloride in the market or in radio electronics stores.

3.2 Etching with copper sulphate

This method is not as common as the previous one, but it is also common. I personally use this method. This method is much cheaper than the previous one, and getting the components is easier.

Pour 3 tablespoons of table salt, 1 tablespoon of copper sulfate into the dishes and pour 250 grams of water at a temperature of 70 degrees. If everything is correct, the solution should turn turquoise, and a little later green. To speed up the process, it is necessary to mix the solution.

Etching time for a 3x4 cm board, around one hour

You can get copper sulphate in agricultural products stores. Copper sulfate is a blue-colored fertilizer. It is in the form of a crystal powder. Battery protection device from full discharge

Hello dear visitor. I know why you are reading this article. Yes yes I know. No what are you? I'm not a telepath, I just know why you got to this particular page. Surely…….

And again, my friend Vyacheslav (SAXON_1996) wants to share his experience on the columns. Word to Vyacheslav I somehow got one 10MAS speaker with a filter and a tweeter. I haven't for a long time…….

DRC

Every PCB manufacturer has technology restrictions on minimum trace widths, trace spacing, hole diameters, and so on. If the board does not meet these restrictions, the manufacturer refuses to accept the board for production.

When creating a PCB file, the default technology limits are set from the default.dru file in the dru directory. As a rule, these limits do not correspond to the limits of real manufacturers, so they need to be changed. You can set the limits just before generating the Gerber files, but it's better to do it right after the board file is generated. To set restrictions, press the DRC button

gaps

Distances

Minimum dimensions

Belts

masks

Running a check

Gerber file generation

For a drill file, the output device ( device) should be EXCELLON, but not GERBER_RS274X

The article discusses the topology of high-frequency boards from a practical point of view. Its main purpose is to help beginners get a feel for the many points that must be taken into account when designing printed circuit boards (PCB) for high-frequency devices. It will also be useful for improving the skills of those specialists who had a break in the development of boards. The main attention is paid to ways to improve the characteristics of circuits, to speed up the time of their development and introduction of changes.

The issues discussed and the proposed methods are applicable to the topology of high-frequency circuits in general. When the operational amplifier (op-amp) is operating at high frequencies, the main characteristics of the circuit depend on the topology of the PCB. Even with good design, circuit performance can be mediocre due to a poorly designed or sloppy printed circuit board. It is possible to be sure that the circuit will show the calculated parameters only by thinking in advance and paying attention to the main points during the entire process of developing the PCB layout.

Scheme

A good circuit is a necessary but not sufficient condition for a good topology. When designing it, you should not skimp on additional information on the drawing, and carefully monitor the direction of the signal. The continuity of the signal from left to right is likely to have the same effect on the PCB. Maximum useful information in the scheme will provide optimal performance developers, technicians, engineers, who will be very grateful to you, and customers in case of any difficulties will not have to urgently look for a developer.

What information, in addition to the usual reference designations, power dissipation and tolerances, should be applied to the circuit? Here are some tips on how to make a super circuit out of a regular circuit: add waveforms, mechanical information about packages or dimensions, specify track lengths, areas where parts should not be placed, parts that should be on the top side of the PCB; add tuning instructions, element rating ranges, thermal information, matched impedance lines, short circuit operation definitions, and so on.

Trust no one

If you're not a topologist yourself, take the time to go through the circuit with the topologist. It is much easier and faster to pay attention to the topology at the beginning than to deal with endless improvements later. Don't count on the layout designer to be able to read your mind.Introduction and guidance are most important at the beginning of the board layout process. The more information and participation in the wiring process, the better the board will turn out. Indicate to the developer the milestones at which you want to familiarize yourself with the wiring process. These " checkpoints» protect the board from far-reaching errors and minimize topology corrections.

Instructions to the developer should include: a brief description of the functions of the circuit; sketch of the board, which shows the location of the inputs and outputs; board stack up (i.e. board thickness, number of layers, details of signal layers and solid layers - power, ground - analog, digital, high-frequency); signals that should be on each layer; placement of critical elements; accurate placement of decoupling elements; critical tracks; lines with matched impedance; tracks of the same length; element sizes; paths far (or near) from each other; chains closer (or farther) from each other; elements near (or far) from each other; elements on the top and bottom side of the board. No one will accuse you of an excess of information, if there is too little - they will complain, on the contrary - never.

Location, location and more location

When placing a circuit on a board, everything is important, from the layout of individual elements to the choice of which nets should be placed side by side.

Usually the location of the inputs, outputs and power is determined. Particular attention should be paid to topology: the location of critical elements - both individual circuits and the circuit as a whole. Determining the location of major components and signal paths from the start ensures that the circuit will perform as intended. This reduces costs, solves problems and reduces wiring times.

Power decoupling

Decoupling the power supply from the amplifier's power pins to minimize noise is a critical aspect of the PCB design process, both for high-speed op-amp circuits and other high-frequency circuits. Typically, one of two configurations is used for decoupling high-speed op-amps.

Between power rail and ground

This method works better in most cases and allows the use of capacitors connected in parallel from the op-amp's power pins directly to ground. Usually two are enough, but some circuits benefit from multiple capacitors connected in parallel.

Parallel connection of capacitors with different capacitances ensures that the power pins will have a low impedance across alternating current over a wide frequency range. This is especially important when the Power Supply Instability Ratio (PSR) drops - the capacitors compensate the amplifier for this drop. Providing a low impedance path to ground for many decades of frequency prevents unwanted noise from entering the op-amp. On fig. 1 shows the advantages of this method. At lower frequencies, capacitors with large capacity provide little resistance to ground. At the self-resonant frequency of the capacitor, the quality of the capacitor deteriorates, and it becomes an inductance. Therefore, it is important to use many capacitors: when frequency response one drops, the other becomes significant, providing low AC impedance over many decades of frequency.

Rice. 1. Capacitor impedance versus frequency

Directly near the power leads of the op-amp, a capacitor with a smaller capacitance and smaller geometric dimensions should be located on the same side as the op-amp - and as close as possible to the amplifier. The ground side of the capacitor must be connected to the ground plane with minimum lead and track lengths. The connection should be as close to the amplifier termination as possible to minimize interference between the power rails and ground. Rice. 2 illustrates this technique.

Rice. 2. Connecting power rails to ground with parallel capacitors

This process should be repeated with the next largest capacitor. A good rule of thumb is to start with the smallest 0.01uF capacitor and work your way up to a 2.2uF low ESR (equivalent series resistance) oxide capacitor. The first of those indicated in the 0508 case has a small series inductance and excellent high-frequency parameters.

Between one tire and another

An alternative configuration is to use one or more capacitors connected between the positive and negative power rails of the op-amp. This method is used when it is difficult to install all four capacitors in the circuit. The disadvantage is the increase in the size of the capacitors, since the voltage across them is doubled compared to blocking each source individually. In this case, a capacitor with a large breakdown voltage is required, which leads to an increase in its size. However, this option improves both PSR and distortion performance.

Since each circuit and its layout is different, the configuration, number and capacitance of capacitors will be determined by the specific requirements of the circuit.

Where C- capacity; A- lining area in cm²; k- relative permittivity of the board material; And d- distance between plates in cm.

Rice. 5. Capacitance of a plane-parallel capacitor

Consideration should also be given to strip inductance due to excessive track length and insufficient ground plane. Equation 2 gives the trace inductance formula (Figure 6):

Where W- track width; L- its length; And H- thickness. All dimensions are in millimeters.

Rice. 6. Track inductance

Rice. 7. Response to an impulse without a layer and with a layer of earth

Where T- board thickness and d is the diameter of the via in centimeters.

Rice. 8. Via Dimensions

earth layer

Here we will touch on some key points of this issue. List of links to this topic given at the end of the article.

Since the earth layer usually has a large area and cross section, its resistance is kept to a minimum. At low frequencies current flows along the path of least resistance, but at high frequencies it flows along the path of least resistance. However, there are exceptions, and sometimes a smaller ground layer works better. This also applies to high-speed op-amps, if you remove part of the ground under the input and output pads.

Analog and digital circuits, including their ground and substrates, should be separated whenever possible. Steep pulse edges create current peaks that flow through the ground layer and create noise, degrading the analog parameters of the circuit.

At high frequencies, attention should be paid to a phenomenon called the skin effect. It causes the current to flow along the outer surface of the conductor, as if making it narrower and increasing the resistance compared to the value of the conductor at direct current. Although the skin effect is beyond the scope of this article, here is an approximate expression for calculating the skin depth in copper (in cm):

To reduce the skin effect, a coating of metals that reduce the possibility of its occurrence can be useful.

Corps

Rice. 9. Differences in the topology of circuits with op amps: a) SOIC package; b) SOT-23 package; c) SOIC package with RF resistor on the underside of the board.

The topology of the board with the SOT-23 package is almost ideal: the minimum length of feedback tracks, the minimum use of vias; the load and decoupling capacitor are connected to ground in short paths to one point; positive voltage decoupling capacitor, not shown in fig. 9b is placed directly below the negative voltage capacitor on the underside of the board.

Low distortion amplifier pinout

The new distortion-reducing pinout used in some of Analog Devices' op amps (such as the AD8045) helps eliminate both of the problems mentioned above and improves performance in two other important areas. The low distortion LFCP pinout shown in fig. 10 is derived from a traditional op amp pinout by turning it counterclockwise by one pin and adding a second output pin dedicated to the feedback loop.

Rice. 10. Op-amp with pinout for low distortion

The low distortion pinout allows a short connection between the output (the feedback pin) and the inverting input, as shown in fig. 11. This greatly simplifies the topology and gives it a rational form.

Rice. 11. PCB Topology for AD8045 Low Distortion Op-Amp

The second advantage of the case is the attenuation of the second harmonic of the non-linear distortion. One of the reasons for its occurrence is the connection between the non-inverting input and the output of the negative supply voltage. The low-distortion pinout of the LFCP case eliminates this coupling and significantly attenuates the second harmonic; in some cases, its reduction can be up to 14 dB. On fig. Figure 12 shows the distortion difference between AD8099 op amps in SOIC and LFCSP packages.

Rice. 12. Comparison of AD8099 op-amp distortion in different packages - SOIC and LFCSP

This case has another advantage - in power dissipation. The package has an open chip substrate, which reduces its thermal resistance, improving θ JA by about 40%. In this case, the microcircuit operates at lower temperatures, which increases its reliability.

There are currently three Analog Devices high speed op amps available in the new low distortion packages: AD8045, AD8099, and AD8000.

Wiring and shielding

On printed circuit boards of electronic circuits, a wide variety of signals can be simultaneously present - analog and digital, with high and low voltage, high and low current - from direct current to gigahertz frequencies. Keeping them from interfering with each other is a difficult task.

It is important to think ahead of time about how the signals are handled on the board, note which ones are sensitive, and determine the steps to keep them intact. Layers of the earth, other than providing a reference potential for electrical signals, can also be used for shielding. When it is necessary to isolate signals, the first step is to ensure sufficient distance between signal traces. Let's look at a few practical steps:

• Minimizing the length of parallel lines and avoiding close proximity between signal traces on the same layer will reduce inductive coupling.
• Minimizing trace lengths on adjacent layers will prevent capacitive coupling.
• Signal tracks requiring special insulation should be run on different layers and, if they cannot be spaced apart, perpendicular to each other, a layer of earth should be laid between them. Perpendicular wiring minimizes capacitive coupling and ground forms an electrical shield. This technique is used when forming lines with a matched impedance (wave impedance).

High-frequency (HF) signals are usually carried on impedance-matched lines. That is, the impedance of the track is provided equal to, for example, 50 ohms (typical for RF circuits). Two widely used types of matched lines - microstrip and stripline - can give the same results, but have different implementations.

The microstrip matched line shown in fig. 13, can pass on either side of the board; it uses the ground layer immediately below it as its reference ground plane.

Rice. 13. Microstrip transmission line

To calculate the characteristic wave resistance lines on the FR4 board, you can use the following formula:

Where H- distance from the ground plane to the track; W- track width; T- track thickness; all dimensions are in mils (1 mil = 10 -3 inches). εr- relative permittivity of the board material.

The stripline matched line (Figure 14) uses two layers of the ground plane and a signal track between them. This method uses more traces, requires more layers, is sensitive to changes in insulator thickness, and is more expensive, so it is usually only used in more demanding applications.

Rice. 14. Stripline matched line

Equation for calculating the characteristic impedance of a strip line:

Rice. 15. Protective rings: a) inverting and non-inverting circuit; b) implementation of both options in the SOT-23-5 package

There are many other shielding and wiring options. For more information on these and other topics mentioned above, the reader is invited to consult the links below.

Conclusion

Reasonable PCB topology is essential for successful design of devices based on high-speed op-amps. It is based on a good circuit, and close cooperation between the circuit engineer and the PCB designer is also important, especially when placing elements and connecting them.

Literature

1. Ardizzoni J. Keep High-Speed ​​Circuit-Board Layout on Track // EE Times, May 23, 2005.
2. Brokaw P. An IC Amplifier User's Guide to Decoupling, Grounding, and Making Things Go Right for a Change // Analog Devices Application Note AN-202.
3. Brokaw P., Barrow J. Grounding for Low- and High-Frequency Circuits // Analog Devices Application Note AN-345.
4. Buxton J. Careful Design Tames High-Speed ​​Op Amps // Analog Devices Application Note AN-257.
5. DiSanto G. Proper PC-Board Layout Improves Dynamic Range // EDN, November 11, 2004.
6. Grant D., Wurcer S. Avoiding Passive-Component Pitfalls // Analog Devices Application Note AN-348.
7. Johnson H. W., Graham M. High-Speed ​​Digital Design, a Handbook of Black Magic. Prentice Hall, 1993.
8. Jung W., ed., Op Amp Applications Handbook // Elsevier-Newnes, 2005.

Now most electronic circuits are made using printed circuit boards. According to the printed circuit board manufacturing technologies, microelectronics assemblies are also made - hybrid modules that contain components of various functional purposes and degrees of integration. Multilayer printed circuit boards and highly integrated electronic components reduce the weight and size characteristics of electronic components and computer science. Now the printed circuit board is more than a hundred years old.

Printed circuit board

This (in English PCB - printed circuit board)- a plate made of electrically insulating material (getinaks, textolite, fiberglass and other similar dielectrics), on the surface of which thin electrically conductive strips (printed conductors) with contact pads for connecting mounted radio elements, including modules and integrated circuits, are somehow applied. This wording is literally taken from the polytechnic dictionary.

There is a more general formulation:

A printed circuit board is a structure of fixed electrical interconnections on an insulating base.

The main structural elements of a printed circuit board are a dielectric base (rigid or flexible) on the surface of which there are conductors. The dielectric base and conductors are the necessary and sufficient elements for a printed circuit board to be a printed circuit board. To install components and connect them to conductors, additional elements are used: contact pads, transitional metallized and mounting holes, connector lamellas, areas for heat removal, shielding and current-carrying surfaces, etc.

The transition to printed circuit boards marked a qualitative leap in the design of electronic equipment. The printed circuit board combines the functions of the carrier of radio elements and the electrical connection of such elements. The latter function is not feasible if a sufficient level of insulation resistance is not provided between the conductors and other conductive elements of the printed circuit board. Therefore, the PCB substrate must act as an insulator.

Historical reference

In the 1920s and 1930s, many patents were issued for printed circuit board designs and methods for making them. The first methods of manufacturing printed circuit boards remained predominantly additive (development of the ideas of Thomas Edison). But in its modern form, the printed circuit board appeared thanks to the use of technologies borrowed from the printing industry. Printed circuit board - a direct translation from the English printing term printing plate ("printing form", or "matrix"). Therefore, the Austrian engineer Paul Eisler is considered the true "father of printed circuit boards". He was the first to come to the conclusion that printing (subtractive) technologies can be used for mass production of printed circuit boards. In subtractive technologies, an image is formed by removing unnecessary fragments. Paul Eisler worked out the technology of galvanic deposition of copper foil and its etching with ferric chloride. The technologies for mass production of printed circuit boards were in demand already during the Second World War. And from the mid-1950s, the formation of printed circuit boards began as a constructive basis for radio equipment not only for military, but also for domestic purposes.

PCB materials

Basic dielectrics for printed circuit boards

Knowing the parameters of materials for printed circuit boards, both single-layer and multilayer, is important for everyone involved in their application, especially for printed circuit boards of devices with increased speed and microwave. When designing an MPP, developers are faced with such tasks as:
- calculation of wave resistance of conductors on the board;
- calculation of the value of interlayer high-voltage insulation;
- choice of the structure of blind and hidden holes.
Available options and thicknesses of various materials are shown in tables 2-6. It should be taken into account that the material thickness tolerance is usually up to ±10%, therefore the thickness tolerance of the finished multilayer board cannot be less than ±10%.

Tg - glass transition temperature (structural failure)

Dk - dielectric constant

Basic dielectrics for microwave printed circuit boards

Typical designs of printed circuit boards are based on the use of standard fiberglass type FR4, with an operating temperature of -50 to +110 °C, and a glass transition temperature Tg (softening) of approx. 135 °C.
With increased requirements for heat resistance or when mounting boards in a lead-free furnace (t up to 260 °C), a high-temperature FR4 High Tg or FR5.
When the requirements for continuous operation at high temperatures or with sudden temperature changes are used polyimide. In addition, polyimide is used for the manufacture of high-reliability circuit boards, for military applications, and also in cases where increased dielectric strength is required.
For boards with microwave circuits(over 2 GHz) separate layers apply microwave material, or the entire board is made of microwave material. The most famous suppliers of special materials are Rogers, Arlon, Taconic, Dupont. The cost of these materials is higher than FR4, and is conventionally shown in the penultimate column of the table relative to the cost of FR4.

* Dk - dielectric constant

Dk - permittivity

PCB pad coatings

Most often, pads are coated with tin-lead alloy, or PBC. The method of applying and leveling the surface of the solder is called HAL or HASL (from the English Hot Air Solder Leveling - solder leveling with hot air). This coating provides the best solderability of the pads. However, it is being replaced by more modern coatings, as a rule, compatible with the requirements of the international RoHS directive.

This directive requires the presence of harmful substances, including lead, to be banned from products. So far, RoHS does not apply to the territory of our country, but remembering its existence is useful.

Possible options for covering WFP sites are in Table 7.

HASL is universally applicable unless otherwise specified.

Immersion (chemical) gilding used to provide a smoother board surface (this is especially important for BGA pads), but has a slightly lower solderability. Oven soldering is performed in much the same way as HASL, but hand soldering requires the use of special fluxes. Organic Coating, or OSP, protects the copper surface from oxidation. Its disadvantage is a short solderability retention period (less than 6 months).

immersion tin provides a smooth surface and good solderability, although it also has a limited solder life. Lead-free HAL has the same properties as lead-containing, but the solder composition is approximately 99.8% tin and 0.2% additives.

Blade connector contacts, subjected to friction during the operation of the board, are electroplated with a thicker and more rigid layer of gold. Both types of gold plating use a nickel underlayer to prevent the gold from diffusing.

Note: All finishes except HASL are RoHS compliant and suitable for soldering with lead-free solders.

Protective and other types of printed circuit board coatings

Protective coatings are used to isolate the surfaces of conductors not intended for soldering.

To complete the picture, consider the functional purpose and materials of printed circuit board coatings.

1. Solder Mask - applied to the surface of the board to protect conductors from accidental short circuits and dirt, as well as to protect fiberglass from thermal shocks during soldering. The mask does not carry any other functional load and cannot serve as protection against moisture, mold, breakdown, etc. (with the exception of cases where special types of masks are used).
2. Marking - applied to the board with paint over the mask to simplify the identification of the board itself and the components located on it.
3. Peelable mask - is applied to specified areas of the board that need to be temporarily protected, for example, from soldering. In the future, it is easy to remove, since it is a rubber-like compound and simply peels off.
4. Carbon contact coating - applied to certain places on the board as contact fields for keyboards. The coating has good conductivity, does not oxidize and is wear resistant.
5. Graphite resistive elements - can be applied to the board surface to act as resistors. Unfortunately, the accuracy of the nominal values ​​is not high - not more precisely ± 20% (with laser adjustment - up to 5%).
6. Silver contact jumpers - can be applied as additional conductors, creating another conductive layer when there is not enough space for routing. They are mainly used for single-layer and double-sided printed circuit boards.

PCB design

The most distant predecessor of printed circuit boards is ordinary wire, most often insulated. He had a significant disadvantage. Under conditions of high vibrations, it required the use of additional mechanical elements to fix it inside the REA. For this, carriers were used on which radio elements were installed, the radio elements themselves and structural elements for intermediate connections, fixing wires. This is a massive montage.

Printed circuit boards are free from these shortcomings. Their conductors are fixed on the surface, their position is fixed, which makes it possible to calculate their mutual connections. In principle, printed circuit boards are now approaching flat designs.

At the initial stage of application, printed circuit boards had a single-sided or double-sided arrangement of conductive tracks.

Single sided PCB- this is a plate, on one side of which there are printed conductors. In double-sided printed circuit boards, the conductors also occupied the empty wrong side of the plate. And for their connection, various options were proposed, among which the most common are via metalized vias. Fragments of the design of the simplest single-sided and double-sided printed circuit boards are shown in fig. 1.

Double-sided PCB- their use instead of one-sided was the first step towards the transition from the plane to the volume. If we abstract (mentally discard the substrate of a double-sided printed circuit board), then we get a three-dimensional construction of conductors. By the way, this step was taken quite quickly. The application of Albert Hanson has already indicated the possibility of placing conductors on both sides of the substrate and connecting them using through holes.

Rice. Fig. 1. Fragments of the design of printed circuit boards a) one-sided and 6) two-sided: 1 - mounting hole, 2 - contact pad, 3 - conductor, 4 - dielectric substrate, 5 - via metallized hole

Further development of electronics - microelectronics led to the use of multi-pin components (chips can have more than 200 pins), the number of electronic components grew. In turn, the use of digital microcircuits and the growth of their performance led to an increase in the requirements for their shielding and power distribution to components, for which special shielding conductive layers were included in the multilayer boards of digital devices (for example, computers). All this led to the growth of interconnections and their complication, which resulted in an increase in the number of layers. In modern printed circuit boards, it can be much more than ten. In a sense, the multilayer printed circuit board has gained volume.

Construction of multilayer printed circuit boards

In the first, most common, version, the inner layers of the board are formed from double-sided copper-laminated fiberglass, which is called the "core". The outer layers are made of copper foil pressed to the inner layers with a binder - a resinous material called "prepreg". After pressing at high temperature, a “pie” of a multilayer printed circuit board is formed, in which holes are then drilled and metallized. Less common is the second option, when the outer layers are formed from "cores" held together by prepreg. This is a simplified description, there are many other designs based on these options. However, the basic principle is that the prepreg acts as a binder between the layers. Obviously, there cannot be a situation where two double-sided "cores" are adjacent without a prepreg pad, but the foil-prepreg-foil-prepreg...etc structure is possible, and is often used in boards with complex combinations of blind and hidden holes.

Multilayer printed circuit boards now account for two-thirds of the world's printed circuit board production in terms of value, although in quantitative terms they are inferior to single and double-sided boards.

Schematically (simplified) a fragment of the design of a modern multilayer printed circuit board is shown in fig. 2. Conductors in such printed circuit boards are placed not only on the surface, but also in the bulk of the substrate. At the same time, the layered arrangement of the conductors relative to each other was preserved (a consequence of the use of planar printing technologies). Layering is inevitably present in the names of printed circuit boards and their elements - one-sided, double-sided, multilayer, etc. Layering really reflects the construct and the technologies for manufacturing printed circuit boards corresponding to this construct.

Rice. Fig. 2. Fragment of a multilayer printed circuit board design: 1 - through metallized hole, 2 - blind microvia, 3 - hidden microvia, 4 - layers, 5 - hidden interlayer holes, 6 - pads

In reality, the design of multilayer printed circuit boards differs from those shown in Fig. 2.

In terms of its structure, MPP is much more complicated than double-sided boards, as is the technology for their production. And their structure itself differs significantly from that shown in Fig. 2. They include additional shield layers (ground and power), as well as several signal layers.

In reality they look like this:

a) Schematically	To ensure switching between the layers of the MFP, interlayer junctions (vias) and microvias (microvias) are used. 3.a. Interlayer transitions can be made in the form of through holes connecting the outer layers to each other and to the inner layers. Deaf and hidden transitions are also used. A blind junction is a connecting metallized channel, visible only from the top or bottom side of the board. Hidden vias are used to interconnect the internal layers of the board. Their use makes it possible to significantly simplify the layout of boards, for example, a 12-layer MPC design can be reduced to an equivalent 8-layer one. switching. Microvias have been developed specifically for surface mounting, connecting pads and signal layers.
c) for clarity in 3D form	For the manufacture of multilayer printed circuit boards, several dielectrics laminated with foil are connected to each other, for which gluing gaskets - prepregs are used. In Figure 3.c, the prepreg is shown in white. The prepreg glues the layers of a multilayer printed circuit board by thermal pressing. The overall thickness of multilayer printed circuit boards grows disproportionately fast as the number of signal layers increases. In this regard, it is necessary to take into account the large ratio of the board thickness to the diameter of through holes, which is a very strict parameter for the process of through hole plating. However, even given the difficulty of plating small diameter through-holes, multilayer printed circuit board manufacturers prefer to achieve high packing density with more relatively cheap layers than with fewer high-density but correspondingly more expensive layers.
With) Drawing 3	Figure 3.c shows an approximate layer structure of a multilayer printed circuit board with an indication of their thicknesses.

Vladimir Urazaev [L.12] believes that the development of structures and technologies in microelectronics is in accordance with the objectively existing law of the development of technical systems: tasks related to the placement or movement of objects are solved by moving from a point to a line, from a line to a plane, from a plane to three-dimensional space.

I think that printed circuit boards will have to obey this law. There is a potential possibility of implementing such multilevel (infinitely level) printed circuit boards. This is evidenced by the rich experience in using laser technologies in the production of printed circuit boards, the equally rich experience in using laser stereolithography to form three-dimensional objects from polymers, the tendency to increase the heat resistance of base materials, etc. Obviously, such products will have to be called something else. Since the term "printed circuit board" will no longer reflect either their internal content or manufacturing technology.

Perhaps it will.

But I think they already know volumetric structures in PCB design, these are multilayer printed circuit boards. And the volumetric mounting of electronic components with the location of contact pads on all surfaces of radio components reduces the manufacturability of their installation, the quality of interconnections and complicates their testing and maintenance.

Future will tell!

Flexible printed circuit boards

For most people, a printed circuit board is just a rigid plate with electrically conductive interconnects.

Rigid printed circuit boards are the most massive product used in radio electronics, which almost everyone knows about.

But there are also flexible printed circuit boards, which are increasingly expanding their range of applications. An example is the so-called flexible printed cables (loops). Such printed circuit boards perform a limited scope of functions (the function of the substrate for radioelements is excluded). They serve to connect conventional printed circuit boards, replacing bundles. Flexible printed circuit boards acquire elasticity due to the fact that their polymer "substrate" is in a highly elastic state. Flexible printed circuit boards have two degrees of freedom. They can even be folded into a Möbius strip.

Drawing 4

One or even two degrees of freedom, but very limited freedom, can also be given to ordinary rigid printed circuit boards, in which the polymer matrix of the substrate is in a rigid, glassy state. This is achieved by reducing the thickness of the substrate. One of the advantages of embossed printed circuit boards made from thin dielectrics is the ability to give them a "roundness". Thus, it becomes possible to coordinate their shape and the shape of objects (rockets, space objects, etc.) in which they can be placed. The result is a significant savings in the internal volume of products.

Their significant drawback is that with an increase in the number of layers, the flexibility of such printed circuit boards decreases. And the use of conventional inflexible components makes it necessary to fix their shape. Since the bends of such printed circuit boards with inflexible components lead to high mechanical stresses at the points of their connection with the flexible printed circuit board.

An intermediate position between rigid and flexible printed circuit boards is occupied by "ancient" printed circuit boards, consisting of rigid elements folded like an accordion. Such "accordions" probably led to the idea of ​​​​creating multilayer printed circuit boards. Modern flexible-rigid printed circuit boards are implemented in a different way. We are talking mainly about multilayer printed circuit boards. They can combine rigid and flexible layers. If the flexible layers are taken out of the rigid ones, it is possible to obtain a printed circuit board consisting of a rigid and flexible fragments. Another option is to connect two rigid fragments with a flexible one.

The classification of PCB designs based on the layering of their conductive pattern covers most but not all PCB designs. For example, for the manufacture of woven circuit boards or loops, not printing printing, but weaving equipment turned out to be suitable. Such "printed circuit boards" already have three degrees of freedom. Just like ordinary fabric, they can take on the most bizarre shapes and shapes.

Printed circuit boards with high thermal conductivity

Recently, there has been an increase in heat dissipation of electronic devices, which is associated with:

Increasing the performance of computing systems,

high power switching needs,

Increasing use of electronic components with increased heat dissipation.

The latter is most clearly manifested in LED lighting technology, where interest in the creation of light sources based on high-power ultra-bright LEDs has sharply increased. The luminous efficiency of semiconductor LEDs has already reached 100lm/W. Such ultra-bright LEDs replace conventional incandescent lamps and find their application in almost all areas of lighting technology: lamps street lighting, automotive lighting, emergency lighting, advertising signs, LED panels, indicators, running lines, traffic lights, etc. These LEDs have become indispensable in decorative lighting, in dynamic lighting systems due to their monochrome color and switching speed. It is also advantageous to use them where it is necessary to save energy drastically, where frequent maintenance is expensive and where electrical safety requirements are high.

Studies show that approximately 65-85% of the electricity during the operation of the LED is converted into heat. However, subject to the thermal regimes recommended by the LED manufacturer, the service life of the LED can reach 10 years. But, if the thermal regime is violated (usually this is operation with a junction temperature of more than 120 ... 125 ° C), the service life of the LED can drop by 10 times! And in case of gross non-compliance with the recommended thermal conditions, for example, when turning on LEDs of the emitter type without a radiator for more than 5-7 seconds, the LED may fail even during the first switch-on. Increasing the transition temperature, in addition, leads to a decrease in the brightness of the glow and a shift in the operating wavelength. Therefore, it is very important to correctly calculate the thermal regime and, if possible, dissipate the heat generated by the LED as much as possible.

Large manufacturers of high power LEDs such as Cree, Osram, Nichia, Luxeon, Seoul Semiconductor, Edison Opto, etc., have long manufactured LED modules or clusters on printed circuit boards with metal base (in the international classification IMPCB - Insulated Metal Printed Circuit Board, or AL PCB - printed circuit boards on an aluminum base).

Figure 5

These printed circuit boards on an aluminum base have a low and fixed thermal resistance, which allows, when installed on a radiator, to simply ensure heat removal from the p-n junction of the LED and ensure its operation throughout the entire service life.

Copper, aluminum, various types of ceramics are used as materials with high thermal conductivity for the bases of such printed circuit boards.

Problems of industrial production technology

The history of the development of printed circuit board technology is the history of improving quality and overcoming problems that arise in the course of development.

Here are some of her details.

Printed circuit boards manufactured by through-hole metallization, despite their widest application, have a very serious drawback. From a constructive point of view, the weakest link in such printed circuit boards is the junction of metallized posts in vias and conductive layers (pads). The connection of the metallized column and the conductive layer goes along the end face of the pad. The connection length is determined by the thickness of the copper foil and is typically 35 µm or less. Galvanic plating of the walls of vias is preceded by the stage of chemical plating. Chemical copper, unlike galvanic copper, is looser. Therefore, the connection of the metallized column with the end surface of the contact pad occurs through an intermediate sublayer of chemical copper, which is weaker in terms of strength characteristics. The coefficient of thermal expansion of fiberglass is much greater than that of copper. When passing through the glass transition temperature of epoxy resin, the difference increases sharply. During thermal shocks, which the printed circuit board experiences for a variety of reasons, the connection is subjected to very high mechanical loads and ... breaks. As a result, the electrical circuit is broken and the performance of the electrical circuit is disrupted.

Rice. 6. Interlayer transitions in multilayer printed circuit boards: a) without dielectric undercoating, 6) with dielectric undercoating 1 - dielectric, 2 - pad of the inner layer, 3 - chemical copper, 4 - galvanic copper

Rice. Fig. 7. A fragment of the construction of a multilayer printed circuit board made by layer-by-layer build-up: 1 - interlayer transition, 2 - conductor of the inner layer, 3 - mounting pad, 4 - conductor of the outer layer, 5 - dielectric layers

In multilayer printed circuit boards, an increase in the reliability of internal vias can be achieved by introducing an additional operation - underetching (partial removal) of the dielectric in vias before metallization. In this case, the connection of metallized posts with contact pads is carried out not only along the end, but also partially along the outer annular zones of these pads (Fig. 6).

A higher reliability of metallized transitions of multilayer printed circuit boards was achieved using the technology of manufacturing multilayer printed circuit boards by the layer-by-layer build-up method (Fig. 7). The connections between the conductive elements of the printed layers in this method are carried out by galvanic build-up of copper in the holes of the insulation layer. In contrast to the through-hole plating method, in this case, the vias are filled entirely with copper. The connection area between the conductive layers becomes much larger, and the geometry is different. Breaking such connections is not so easy. And yet this technology is also far from ideal. The transition "galvanic copper - chemical copper - galvanized copper" still remains.

Printed circuit boards made by through-hole metallization must withstand at least four (multilayer at least three) resoldering. Embossed printed circuit boards allow a much larger number of resoldering (up to 50). According to the developers, metalized vias in embossed printed circuit boards do not reduce, but increase their reliability. What caused such a sharp qualitative leap? The answer is simple. In the technology of manufacturing embossed printed circuit boards, the conductive layers and the metallized columns connecting them are implemented in a single technological cycle (simultaneously). Therefore, there is no transition "galvanic copper - chemical copper - galvanized copper". But such a high result was obtained as a result of the rejection of the most mass-produced technology for the manufacture of printed circuit boards, as a result of the transition to another construct. For many reasons, it is undesirable to abandon the method of metallization of through holes.

How to be?

Responsibility for the formation of a barrier layer at the junction of the ends of the contact pads and metallized caps lies mainly with the technologists. They were able to solve this problem. Revolutionary changes in the technology of manufacturing printed circuit boards have been introduced by methods of direct metallization of holes, which excludes the stage of chemical metallization, being limited only to pre-activation of the surface. Moreover, the processes of direct metallization are implemented in such a way that a conductive film appears only where it is needed - on the surface of the dielectric. As a result, there is simply no barrier layer in the plated vias of printed circuit boards made by the direct through-hole plating method. Isn't it a beautiful way to resolve a technical contradiction?

It was also possible to overcome the technical contradiction related to the plating of vias. Plated holes can become weak link printed circuit boards for another reason. Via wall thickness should ideally be uniform throughout their height. Otherwise, again there are problems with reliability. The physical chemistry of electroplating processes counteracts this. The ideal and real profile of the coating in plated vias are shown in fig. 5. The thickness of the coating at the depth of the hole is usually less than at the surface. The reasons are very different: uneven current density, cathodic polarization, insufficient electrolyte exchange rate, etc. In modern printed circuit boards, the diameter of metalized vias has already crossed the mark of 100 microns, and the ratio of the height to the diameter of the hole in some cases reaches 20:1. The situation became extremely complicated. Physical methods (using ultrasound, increasing the intensity of fluid exchange in the holes of printed circuit boards, etc.) have already exhausted their possibilities. Even the viscosity of the electrolyte begins to play a significant role.

Rice. 8. Cross-section of the via to be plated in the printed circuit board. 1 - dielectric, 2 - ideal metallization profile of the hole walls, 3 - real metallization profile of the hole walls,
4 - resist

Traditionally, this problem was solved by using electrolytes with equalizing additives, which are adsorbed in areas where the current density is higher. The sorption of such additives is proportional to the current density. Additives create a barrier layer to counteract excessive settling electroplating on sharp edges and areas adjacent to them (closer to the surface of the printed circuit board).

A different solution to this problem has been theoretically known for a long time, but practically it was possible to implement it quite recently - after the industrial production of switching power supplies was mastered high power. This method is based on the use of a pulsed (reverse) power supply mode for galvanic baths. Most of the time direct current is supplied. When this occurs, the deposition of the coating. Reverse current is supplied for a lesser part of the time. Simultaneously, the dissolution of the deposited coating occurs. Uneven current density (greater at sharp corners) in this case is only beneficial. For this reason, the dissolution of the coating occurs first and to a greater extent at the surface of the printed circuit board. In this technical solution, a whole “bouquet” of methods for resolving technical contradictions is used: use a partially redundant action, turning harm into favor, apply the transition from a continuous process to an impulse one, do the opposite, etc. And the result obtained corresponds to this “bouquet”. With a certain combination of the duration of forward and reverse pulses, it even becomes possible to obtain a coating thickness in the depth of the hole greater than at the surface of the printed circuit board. That's why this technology has proved indispensable for filling blind vias with metal (a property of modern printed circuit boards), due to which the density of interconnects in PCBs approximately doubles.

The problems associated with the reliability of metalized junctions in printed circuit boards are of a local nature. Consequently, the contradictions that arise in the process of their development, in relation to printed circuit boards as a whole, are also not universal. Although such printed circuit boards occupy the lion's share of the market of all printed circuit boards.

Also, in the process of development, other problems that technologists face, but consumers do not even think about them, are solved. We get multilayer printed circuit boards for our needs and apply them.

Microminiaturization

At the initial stage, the same components were installed on printed circuit boards that were used in the volumetric installation of REA, although with some refinement of the conclusions to reduce their size. But the most common components could be installed on printed circuit boards without rework.

With the advent of printed circuit boards, it became possible to reduce the size of the components used on printed circuit boards, which in turn led to a decrease in operating voltages and currents consumed by these elements. Since 1954, the Ministry of Power Plants and the Electrical Industry has mass-produced the Dorozhny tube portable radio receiver, which used a printed circuit board.

With the advent of miniature semiconductor amplifier devices - transistors, printed circuit boards began to dominate in household appliances, a little later in industry, and with the advent of fragments of electronic circuits combined on a single chip - functional modules and microcircuits, their design already provided for the installation of exclusively non-printed circuit boards.

With the continued reduction in the size of active and passive components, a new concept has appeared - “Microminiaturization”.

In electronic components, this resulted in the appearance of LSI and VLSI containing many millions of transistors. Their appearance made it necessary to increase the number of external connections (see the contact surface of the graphics processor in Figure 9.a), which in turn caused the complication of the wiring of conductive lines, which can be seen in Figure 9.b.

Such a GPU panel, and CPU also - nothing more than a small multilayer printed circuit board, on which the processor chip itself is placed, the wiring of the chip pins with the contact field and attached elements (usually filter capacitors of the power distribution system)

Figure 9

And let it not seem like a joke to you, the 2010 CPU from Intel or AMD is also a printed circuit board, and a multilayer one at that.

Figure 9a

The development of printed circuit boards, as well as electronic technology in general, is a line of reduction of its elements; their compaction on the printed surface, as well as the reduction of elements of electronic technology. Under the "elements" in this case, one should understand both the own property of printed circuit boards (conductors, vias, etc.), and elements from the supersystem (printed circuit assembly) - radio elements. The latest in speed of microminiaturization are ahead of printed circuit boards.

Microelectronics is engaged in the development of VLSI.

Increasing the location density element base requires the same from the conductors of the printed circuit board - the carrier of this element base. In this regard, there are many problems that need to be solved. We will talk about two such problems and how to solve them in more detail.

The first methods of manufacturing printed circuit boards were based on gluing copper foil conductors to the surface of a dielectric substrate.

It was assumed that the width of the conductors and the gaps between the conductors are measured in millimeters. In this version, this technology was quite efficient. The subsequent miniaturization of electronic technology required the creation of other methods for manufacturing printed circuit boards, the main variants of which (subtractive, additive, semi-additive, combined) are still used today. The use of such technologies made it possible to implement printed circuit boards with element sizes measured in tenths of a millimeter.

Achieving PCB resolution levels of approximately 0.1 mm (100 µm) has been a milestone. On the one hand, there was a transition "down" by one more order. On the other hand, a kind of qualitative leap. Why? The dielectric substrate of most modern printed circuit boards is fiberglass - laminated plastic with a polymer matrix reinforced with fiberglass. Reducing the gaps between the conductors of the printed circuit board has led to the fact that they have become commensurate with the thickness of the glass filaments or the thickness of the knots of the interlacing of these filaments in fiberglass. And the situation in which the conductors are "closed" with such knots has become quite real. As a result, the formation of peculiar capillaries in fiberglass, "closing" these conductors, has also become real. Under conditions of high humidity, capillaries eventually lead to a deterioration in the level of insulation between the conductors of printed circuit boards. And to be more precise, this happens even in conditions of normal humidity. Moisture condensation in the capillary structures of fiberglass is also noted under normal conditions. Moisture always reduces the level of insulation resistance.

Because in modern radio electronic equipment such printed circuit boards have become commonplace, we can conclude that the developers of basic materials for printed circuit boards still managed to solve this problem using traditional methods. But will they cope with the next significant event? Another qualitative leap has already taken place.

It is reported that Samsung specialists have mastered the technology for manufacturing printed circuit boards with a width of conductors and gaps between them of 8-10 microns. But this is not the thickness of a glass thread, but fiberglass!

The task of providing insulation in ultra-small gaps between the conductors of current and especially future printed circuit boards is difficult. What methods it will be solved - traditional or non-traditional - and whether it will be solved - time will tell.

Rice. Fig. 10. Etching profiles of copper foil: a - ideal profile, b - real profile; 1 - protective layer, 2 - conductor, 3 - dielectric

There were difficulties in obtaining ultra-small (ultra-narrow) conductors in printed circuit boards. For many reasons, subtractive methods have become widespread in PCB manufacturing technologies. In subtractive methods, an electrical circuit pattern is formed by removing unnecessary foil fragments. Even during the Second World War, Paul Eisler worked out the technology of etching copper foil with ferric chloride. Such unpretentious technology is still used by radio amateurs. Industrial technology is not far removed from this "kitchen" technology. Unless the composition of pickling solutions has changed and elements of process automation have appeared.

The fundamental drawback of absolutely all etching technologies is that etching proceeds not only in the desired direction (toward the surface of the dielectric), but also in an undesirable transverse direction. The lateral undercutting of the conductors is commensurate with the thickness of the copper foil (about 70%). Usually, instead of an ideal conductor profile, a mushroom-like profile is obtained (Fig. 10). When the width of the conductors is large, and in the simplest printed circuit boards it is even measured in millimeters, they simply turn a blind eye to the side undercut of the conductors. If the width of the conductors is commensurate with their height or even less than it (the realities of today), then "lateral aspirations" cast doubt on the feasibility of using such technologies.

In practice, the amount of lateral underetching of printed conductors can be reduced to some extent. This is achieved by increasing the etching rate; using jet pouring (etchant jets coincide with the desired direction - perpendicular to the plane of the sheet), as well as in other ways. But when the width of the conductor approaches its height, the effectiveness of such improvements becomes clearly insufficient.

But advances in photolithography, chemistry, and technology are now making it possible to solve all these problems. These solutions are taken from microelectronics technologies.

Radio amateur technologies for the production of printed circuit boards

The manufacture of printed circuit boards in amateur radio conditions has its own characteristics, and the development of technology is increasing these possibilities. But processes continue to be their basis

The question of how to make printed circuit boards at home cheaply has been of concern to all radio amateurs, probably since the 60s of the last century, when printed circuit boards were widely used in household appliances. And if then the choice of technologies was not so great, today, thanks to the development of modern technology, radio amateurs are able to quickly and efficiently produce printed circuit boards without the use of any expensive equipment. And these opportunities are constantly expanding, allowing them to bring the quality of their creations closer to industrial designs.

Actually, the whole process of manufacturing a printed circuit board can be divided into five main stages:

• preliminary preparation of the workpiece (surface cleaning, degreasing);
• application of a protective coating in one way or another;
• removal of excess copper from the board surface (etching);
• cleaning the workpiece from the protective coating;
• hole drilling, flux coating, tinning.

We consider only the most common "classical" technology, in which excess copper areas are removed from the board surface by chemical etching. In addition, it is possible, for example, to remove copper by milling or using an electric spark machine. However, these methods have not been widely used either in the amateur radio environment or in industry (although the manufacture of boards by milling is sometimes used in cases where it is necessary to produce simple printed circuit boards in single quantities very quickly).

And here we will talk about the first 4 points of the technological process, since drilling is performed by a radio amateur using the tool that he has.

At home, it is impossible to make a multilayer printed circuit board capable of competing with industrial designs, therefore, in amateur radio conditions, double-sided printed circuit boards are usually used, and in microwave device designs only double-sided ones.

Although home PCB fabrication should strive to use as many surface mount components as possible when designing a circuit, this in some cases allows almost the entire circuit to be routed on one side of the board. This is due to the fact that no technology of metallization of vias that is really feasible at home has been invented so far. Therefore, if the board cannot be wired on one side, you should wire on the second side using the leads of various components installed on the board as vias, which in this case will have to be soldered on both sides of the board. Of course, there are various ways to replace the plating of holes (using a thin conductor inserted into the hole and soldered to the tracks on both sides of the board; using special caps), but all of them have significant drawbacks and are inconvenient to use. Ideally, the board should only be routed on one side using a minimum number of jumpers.

Let us now dwell in more detail on each of the stages of manufacturing a printed circuit board.

Preliminary preparation of the workpiece

This stage is the initial one and consists in preparing the surface of the future printed circuit board for applying a protective coating on it. In general, over a long period of time, surface cleaning technology has not undergone any significant changes. The whole process is reduced to the removal of oxides and contaminants from the surface of the board using various abrasive products and subsequent degreasing.

To remove stubborn dirt, you can use fine-grained sandpaper (“zero”), fine abrasive powder, or any other tool that does not leave deep scratches on the surface of the board. Sometimes you can simply wash the surface of the printed circuit board with a hard washcloth for washing dishes with detergent or powder (for this purpose it is convenient to use an abrasive washcloth for washing dishes, which looks like felt with small inclusions of some substance; often such a washcloth is glued to a piece of foam rubber) . In addition, if the surface of the printed circuit board is sufficiently clean, you can skip the abrasive treatment altogether and go straight to degreasing.

If there is only a thick oxide film on the printed circuit board, it can be easily removed by treating the printed circuit board for 3-5 seconds with a solution of ferric chloride, followed by rinsing in cold running water. However, it should be noted that it is desirable to either carry out this operation immediately before applying the protective coating, or after it, store the workpiece in a dark place, since copper quickly oxidizes in the light.

The final step in surface preparation is degreasing. To do this, you can use a piece of soft cloth that does not leave fibers, moistened with alcohol, gasoline or acetone. Here you should pay attention to the cleanliness of the surface of the board after degreasing, since recently acetone and alcohol with a significant amount of impurities have begun to come across, which leave whitish stains on the board after drying. If so, then you should look for another degreaser. After degreasing, the board should be washed in running cold water. The quality of cleaning can be controlled by observing the degree of wetting of the copper surface with water. A surface that is completely wetted with water, without the formation of drops on it and breaks in the water film, is an indicator of a normal level of cleaning. Disturbances in this water film indicate that the surface has not been sufficiently cleaned.

Protective coating

The application of a protective coating is the most important stage in the PCB manufacturing process, and it is this that determines the quality of the manufactured board by 90%. Currently, there are three most popular methods of applying a protective coating in the amateur radio environment. We will consider them in ascending order of the quality of boards obtained by using them.

First of all, it should be clarified that the protective coating on the surface of the workpiece must form a homogeneous mass, without defects, with even clear boundaries and resistant to the chemical components of the pickling solution.

Manual application of protective coating

With this method, the drawing of the printed circuit board is transferred to the fiberglass manually using some kind of writing device. Recently, a lot of markers have appeared on sale, the dye of which is not washed off with water and gives a fairly strong protective layer. In addition, for manual drawing, you can use a drawing pen or some other device filled with dye. So, for example, it is convenient to use a syringe with a thin needle for drawing (insulin syringes with a needle diameter of 0.3-0.6 mm are best suited for this purpose), cut to a length of 5-8 mm. In this case, the rod should not be inserted into the syringe - the dye should flow freely under the action of the capillary effect. Also, instead of a syringe, you can use a thin glass or plastic tube stretched over the fire to achieve the desired diameter. Particular attention should be paid to the quality of the processing of the edge of the tube or needle: when drawing, they should not scratch the board, otherwise already painted areas can be damaged. As a dye when working with such devices, you can use bituminous or some other varnish diluted with a solvent, zaponlak, or even a solution of rosin in alcohol. In this case, it is necessary to choose the consistency of the dye in such a way that it flows freely when drawing, but at the same time does not flow out and does not form drops at the end of the needle or tube. It should be noted that the manual process of applying a protective coating is quite laborious and is suitable only in cases where it is necessary to make a small board very quickly. The minimum track width that can be achieved when drawing by hand is in the order of 0.5 mm.

Using "laser printer and iron technology"

This technology appeared relatively recently, but immediately became widespread due to its simplicity and High Quality payments received. The basis of the technology is the transfer of toner (powder used in printing in laser printers) from any substrate to a printed circuit board.

In this case, two options are possible: either the substrate used is separated from the board before etching, or, if the substrate is used aluminum foil, it is etched together with copper .

The first stage of using this technology is to print mirror image printed circuit board pattern on the substrate. The printer's print settings should be set to the highest print quality (because in this case the thickest toner layer is applied). As a substrate, you can use thin coated paper (covers from various magazines), fax paper, aluminum foil, laser printer film, Oracal self-adhesive film backing, or some other materials. If you use too thin paper or foil, you may need to glue them around the perimeter to a sheet of thick paper. Ideally, the printer should have a kink-free paper path that prevents such a sandwich from creasing inside the printer. This is also of great importance when printing on foil or base from Oracal film, since the toner on them is very weak, and if the paper is folded inside the printer, there is a high probability that you will have to spend several unpleasant minutes cleaning the printer's oven from adhering toner residues. It is best if the printer can feed the paper through itself horizontally while printing on the top side (like the HP LJ2100 is one of the best printers for PCB applications). I would like to immediately warn owners of printers such as HP LJ 5L, 6L, 1100, so that they do not try to print on foil or base from Oracal - usually such experiments end in failure. Also, in addition to the printer, you can also use a copier, the use of which sometimes gives even better results compared to printers due to the application of a thick layer of toner. The main requirement for the substrate is the ease of its separation from the toner. Also, if paper is used, it should not leave lint in the toner. In this case, two options are possible: either the substrate is simply removed after transferring the toner to the board (in the case of a film for laser printers or a base from Oracal), or it is pre-soaked in water and then gradually separated (coated paper).

The transfer of toner to the board consists in applying a substrate with toner to a pre-cleaned board, followed by heating to a temperature slightly above the melting point of the toner. There are a huge number of options for how to do this, but the simplest is to press the substrate to the board with a hot iron. At the same time, in order to evenly distribute the pressure of the iron on the substrate, it is recommended to lay several layers of thick paper between them. A very important issue is the temperature of the iron and the exposure time. These parameters vary on a case-by-case basis, so you may have to run more than one experiment before you get good results. There is only one criterion here: the toner must have time to melt enough to stick to the surface of the board, and at the same time it must not have time to reach a semi-liquid state so that the edges of the tracks do not flatten out. After "welding" the toner to the board, it is necessary to separate the substrate (except for the case of using aluminum foil as a substrate: it should not be separated, since it dissolves in almost all etching solutions). Oracal's laser printer film and backing simply peel off gently, while regular paper requires pre-soaking in hot water.

It is worth noting that due to the peculiarities of laser printers printing, the toner layer in the middle of large solid polygons is quite small, so you should avoid using such areas on the board as much as possible, or after removing the substrate, you will have to retouch the board manually. In general, the use of this technology, after some training, makes it possible to achieve the width of the tracks and the gaps between them up to 0.3 mm.

I have been using this technology for many years (since the laser printer became available to me).

Application of photoresists

A photoresist is a substance that is sensitive to light (usually in the near ultraviolet) and changes its properties when exposed to light.

Recently, several types of imported photoresists in aerosol packaging have appeared on the Russian market, which are especially convenient for home use. The essence of using a photoresist is as follows: a photomask () is applied to a board with a layer of photoresist applied to it and it is illuminated, after which the illuminated (or unexposed) areas of the photoresist are washed off with a special solvent, which is usually caustic soda (NaOH). All photoresists are divided into two categories: positive and negative. For positive photoresists, the track on the board corresponds to a black area on the photomask, and for negative ones, accordingly, it is transparent.

The most widespread are positive photoresists as the most convenient to use.

Let us dwell in more detail on the use of positive photoresists in aerosol packaging. The first step is to prepare the photomask. At home, it can be obtained by printing a board pattern on a laser printer on film. In this case, special attention should be paid to the density of black on the photomask, for which it is necessary to disable all modes of saving toner and improving print quality in the printer settings. In addition, some companies offer the output of a photomask on a photoplotter - while you are guaranteed a high-quality result.

At the second stage, a thin film of photoresist is applied to the pre-prepared and cleaned surface of the board. This is done by spraying it from a distance of about 20 cm. In this case, one should strive for maximum uniformity of the resulting coating. In addition, it is very important to ensure that there is no dust during the spraying process - each dust particle that gets into the photoresist will inevitably leave its mark on the board.

After applying a layer of photoresist, it is necessary to dry the resulting film. It is recommended to do this at a temperature of 70-80 degrees, and first you need to dry the surface at a low temperature and only then gradually bring the temperature to the desired value. Drying time at the specified temperature is about 20-30 minutes. In extreme cases, drying the board at room temperature for 24 hours is allowed. Boards with applied photoresist should be stored in a dark, cool place.

The next step after applying the photoresist is exposure. At the same time, a photomask is superimposed on the board (with the print side to the board, this helps to increase clarity during exposure), which is pressed against a thin glass or. With sufficiently small sizes of plates for pressing, a photographic plate washed from the emulsion can be used. Since the region of maximum spectral sensitivity of most modern photoresists is in the ultraviolet range, it is desirable to use a lamp with a large fraction of UV radiation in the spectrum (DRSH, DRT, etc.) for illumination. As a last resort, you can use a powerful xenon lamp. The exposure time depends on many factors (the type and power of the lamp, the distance from the lamp to the board, the thickness of the photoresist layer, etc.) and is selected experimentally. However, in general, the exposure time is usually no more than 10 minutes even when exposed to direct sunlight.

(Plastic, transparent in visible light, I do not recommend using plates for pressing, since they have a strong absorption of UV radiation)

The development of most photoresists is carried out with a solution of caustic soda (NaOH) - 7 grams per liter of water. It is best to use a freshly prepared solution having a temperature of 20-25 degrees. The development time depends on the thickness of the photoresist film and ranges from 30 seconds to 2 minutes. After development, the board can be etched in common solutions, since the photoresist is resistant to acids. When using high-quality photomasks, the use of photoresist makes it possible to obtain tracks with a width of up to 0.15-0.2 mm.

Etching

There are many compositions for chemical etching of copper. All of them differ in the rate of the reaction, the composition of the substances released as a result of the reaction, as well as the availability of the chemical reagents necessary for preparing the solution. Below is information about the most popular pickling solutions.

Ferric chloride (FeCl)

Perhaps the most famous and popular reagent. Dry ferric chloride dissolves in water until a saturated golden yellow solution is obtained (this will require about two tablespoons per glass of water). The etching process in this solution can take from 10 to 60 minutes. The time depends on the concentration of the solution, temperature and agitation. Stirring greatly speeds up the reaction. For this purpose, it is convenient to use an aquarium compressor, which provides mixing of the solution with air bubbles. The reaction is also accelerated when the solution is heated. After etching, the board must be washed with plenty of water, preferably with soap (to neutralize acid residues). The disadvantages of this solution include the formation of waste during the reaction, which are deposited on the board and prevent the normal course of the etching process, as well as a relatively low reaction rate.

ammonium persulfate

Light crystalline substance, soluble in water based on the ratio of 35 g of the substance to 65 g of water. The etching process in this solution takes about 10 minutes and depends on the area of ​​the copper coating being etched. To ensure optimal conditions for the reaction, the solution must have a temperature of about 40 degrees and be constantly mixed. After etching, the board must be washed in running water. The disadvantages of this solution include the need to maintain the required temperature and mixing.

Hydrochloric acid solution (HCl) and hydrogen peroxide (H 2 O 2)

- To prepare this solution, add 200 ml of 35% hydrochloric acid and 30 ml of 30% hydrogen peroxide to 770 ml of water. The finished solution should be stored in a dark bottle, not hermetically sealed, since gas is released during the decomposition of hydrogen peroxide. Caution: When using this solution, all precautions must be taken when working with caustic chemicals. All work must be done only in the fresh air or under a hood. If the solution comes into contact with the skin, it must be washed immediately with plenty of water. Etching time is highly dependent on the agitation and temperature of the solution and is in the order of 5-10 minutes for a well-stirred fresh solution at room temperature. Do not heat the solution above 50 degrees. After etching, the board must be rinsed with running water.

This solution after etching can be restored by adding H 2 O 2 . The assessment of the required amount of hydrogen peroxide is carried out visually: the copper board immersed in the solution should be repainted from red to dark brown. The formation of bubbles in the solution indicates an excess of hydrogen peroxide, which slows down the etching reaction. The disadvantage of this solution is the need for strict adherence to all precautions when working with it.

A solution of citric acid and hydrogen peroxide from Radiokot

In 100 ml of pharmacy 3% hydrogen peroxide, 30 g of citric acid and 5 g of salt are dissolved.

This solution should be enough to etch 100 cm2 of copper, 35 µm thick.

Salt in the preparation of the solution can not be spared. Since it plays the role of a catalyst, it is practically not consumed in the etching process. Peroxide 3% should not be further diluted. when other ingredients are added, its concentration decreases.

The more hydrogen peroxide (hydroperite) is added, the faster the process will go, but do not overdo it - the solution is not stored, i.e. is not reused, which means that hydroperite will simply be overused. An excess of peroxide is easily identified by abundant "bubble" during pickling.

However, the addition of citric acid and peroxide is quite acceptable, but it is more rational to prepare a fresh solution.

Workpiece cleaning

After etching and flushing of the board, it is necessary to clean its surface from the protective coating. This can be done with any organic solvent, for example, acetone.

Next, you need to drill all the holes. This should be done with a sharpened drill at maximum speed of the electric motor. If, when applying a protective coating, no empty space was left in the centers of the contact pads, it is necessary to first mark the holes (this can be done, for example, with a core). After that, the defects (fringe) on the reverse side of the board are removed by countersinking, and on a double-sided printed circuit board on copper - with a drill with a diameter of about 5 mm in a manual clamp for one turn of the drill without applying force.

The next step is to cover the board with flux, followed by tinning. You can use commercially available fluxes (best water washable or no rinse at all) or simply cover the board with a weak solution of rosin in alcohol.

Tinning can be done in two ways:

Solder immersion

The help of a soldering iron and a metal braid impregnated with solder.

In the first case, it is necessary to make an iron bath and fill it with a small amount of low-melting solder - an alloy of Rose or Wood. The melt must be completely covered with a layer of glycerin on top to avoid oxidation of the solder. To heat the bath, you can use an inverted iron or electric stove. The board is immersed in the melt, and then removed with the simultaneous removal of excess solder with a hard rubber squeegee.

Conclusion

I think this material will help readers get an idea about the design and manufacture of printed circuit boards. And for those who start to deal with electronics, get the basic skills of making them at home. For a more complete acquaintance with printed circuit boards, I recommend reading [L.2]. It can be downloaded from the Internet.

Literature

1. Polytechnic Dictionary. Editorial staff: Inglinsky A. Yu. et al. M.: Soviet Encyclopedia. 1989.
2. Medvedev A. M. Printed circuit boards. Structures and materials. Moscow: Technosphere. 2005.
3. From the history of printed circuit board technology // Electronics-NTB. 2004. No. 5.
4. Novelties of electronic technology. Intel is ushering in the era of 3D transistors. Alternative to traditional planar devices // Elektronika-NTB. 2002. No. 6.
5. Truly three-dimensional microcircuits - the first approximation // Components and technologies. 2004. No. 4.
6. Mokeev M. N., Lapin M. S. Technological processes and systems for the production of woven circuit boards and cables. L.: LDNTP 1988.
7. Volodarsky O. Does this computer suit me? Electronics woven into fabric becomes fashionable // Electronics-NTB. 2003. No. 8.
8. Medvedev AM Printed circuit board production technology. Moscow: Technosphere. 2005.
9. Medvedev A. M. Impulse metallization of printed circuit boards // Technologies in the electronic industry. 2005. No. 4
10. Printed circuit boards - development lines, Vladimir Urazaev,

Printed circuit board (English printed circuit board, PCB, or printed wiring board, PWB) is a dielectric plate, on the surface and / or in the volume of which electrically conductive circuits of an electronic circuit are formed. The printed circuit board is designed for electrical and mechanical connection of various electronic components. Electronic components on a printed circuit board are connected with their leads to the elements of the conductive pattern, usually by soldering.

In contrast to surface mounting, on a printed circuit board, the electrically conductive pattern is made of foil, entirely located on a solid insulating base. The printed circuit board contains mounting holes and pads for mounting pin or planar components. In addition, printed circuit boards have vias for electrical connection of foil sections located on different layers of the board. From the outside, the board is usually coated with a protective coating (“solder mask”) and markings (an auxiliary figure and text according to the design documentation).

Depending on the number of layers with an electrically conductive pattern, printed circuit boards are divided into:

single-sided (SPP): there is only one layer of foil glued to one side of the dielectric sheet.

double-sided (DPP): two layers of foil.

multilayer (MPP): foil not only on two sides of the board, but also in the inner layers of the dielectric. Multilayer printed circuit boards are obtained by gluing several single or double sided boards together.

As the complexity of the designed devices and the density of mounting increase, the number of layers on the boards increases.

The basis of the printed circuit board is a dielectric, the most commonly used materials are fiberglass, getinaks. Also, a metal base coated with a dielectric (for example, anodized aluminum) can serve as the basis for printed circuit boards; copper foil tracks are applied over the dielectric. Such printed circuit boards are used in power electronics for efficient heat removal from electronic components. In this case, the metal base of the board is attached to the radiator. As a material for printed circuit boards operating in the microwave range and at temperatures up to 260 ° C, fluoroplastic reinforced with glass fabric (for example, FAF-4D) and ceramics are used. Flexible boards are made from polyimide materials such as Kapton.

What material will we use for the manufacture of boards

The most common, affordable materials for the manufacture of circuit boards are Getinaks and Steklotekstolit. Getinax paper impregnated with bakelite varnish, fiberglass textolite with epoxy. We will definitely use fiberglass!

Foiled fiberglass is sheets made on the basis of glass fabrics impregnated with a binder based on epoxy resins and lined on both sides with copper electrolytic galvanic-resistant foil 35 microns thick. The maximum allowable temperature is from -60ºС to +105ºС. It has very high mechanical and electrical insulating properties, lends itself well to machining by cutting, drilling, stamping.

Fiberglass is mainly used one or two-sided with a thickness of 1.5mm and with copper foil with a thickness of 35μm or 18μm. We will use a 0.8mm thick single-sided fiberglass with a 35µm thick foil (why will be discussed in detail later).

Methods for making printed circuit boards at home

Boards can be manufactured chemically and mechanically.

With the chemical method, in those places where there should be tracks (drawing) on ​​the board, a protective composition (lacquer, toner, paint, etc.) is applied to the foil. Next, the board is immersed in a special solution (ferric chloride, hydrogen peroxide, and others), which "corrodes" the copper foil, but does not affect the protective composition. As a result, copper remains under the protective composition. The protective composition is subsequently removed with a solvent and the finished board remains.

The mechanical method uses a scalpel (for manual production) or a milling machine. A special cutter makes grooves on the foil, eventually leaving islands with foil - the necessary pattern.

Milling machines are quite expensive, as well as the cutters themselves are expensive and have a small resource. So, we will not use this method.

The simplest chemical method is manual. With a risograph varnish, tracks are drawn on the board and then we etch with a solution. This method does not allow making complex boards with very thin traces - so this is not our case either.

The next method for making boards is with a photoresist. This is a very common technology (boards are made by this method at the factory) and it is often used at home. There are a lot of articles and methods for manufacturing boards using this technology on the Internet. It gives very good and repeatable results. However, this is also not our option. The main reason is rather expensive materials (photoresist, which also deteriorates over time), as well as additional tools (UV lamp, laminator). Of course, if you have a bulk production of boards at home - then the photoresist is out of competition - we recommend mastering it. It is also worth noting that the equipment and technology of photoresist allows the production of silk-screen printing and protective masks on circuit boards.

With the advent of laser printers, radio amateurs began to actively use them for the manufacture of circuit boards. As you know, a laser printer uses "toner" to print. This is a special powder that sinters under temperature and sticks to paper - as a result, a pattern is obtained. The toner is resistant to various chemicals, which allows it to be used as a protective coating on the copper surface.

So, our method is to transfer the toner from the paper to the surface of the copper foil and then etch the board with a special solution to obtain a pattern.

Due to its ease of use, this method has earned a very wide distribution in amateur radio. If you type in Yandex or Google how to transfer the toner from paper to the board, you will immediately find such a term as "LUT" - laser ironing technology. Boards using this technology are made as follows: a pattern of tracks is printed in a mirror version, paper is applied to the board with a pattern to copper, we iron this paper on top, the toner softens and sticks to the board. The paper is further soaked in water and the board is ready.

There are "a million" articles on the Internet about how to make a board using this technology. But this technology has many disadvantages that require direct hands and a very long attachment to it. That is, you have to feel it. Payments do not come out the first time, they are obtained every other time. There are many improvements - to use a laminator (with alteration - in the usual one there is not enough temperature), which allow to achieve very good results. There are even methods for building special heat presses, but all this again requires special equipment. The main disadvantages of LUT technology:

overheating - the tracks spread out - become wider

underheating - tracks remain on paper

the paper is “cooked” to the board - even when it gets wet it is difficult to leave - as a result, the toner may be damaged. There is a lot of information on the Internet about which paper to choose.

Porous toner - after removing the paper, micropores remain in the toner - the board is also etched through them - corroded tracks are obtained

repeatability of the result - excellent today, bad tomorrow, then good - it is very difficult to achieve a stable result - you need a strictly constant toner warm-up temperature, you need a stable board pressure.

By the way, this method did not work for me to make a board. Tried to do both on magazines and on coated paper. As a result, he even spoiled the boards - copper swelled from overheating.

For some reason, there is undeservedly little information on the Internet about another method of toner transfer - the method of cold chemical transfer. It is based on the fact that toner does not dissolve with alcohol, but with acetone. As a result, if you choose such a mixture of acetone and alcohol, which will only soften the toner, then it can be “re-pasted” onto the board from paper. I really liked this method and immediately paid off - the first board was ready. However, as it turned out later, I could not find detailed information anywhere that would give a 100% result. We need a method by which even a child could make a payment. But for the second time, the payment did not work out, then again it took a long time to select the necessary ingredients.

As a result, after a long time, a sequence of actions was developed, all components were selected that give, if not 100% then 95% of a good result. And most importantly, the process is so simple that the child can make the payment completely on his own. This is the method we will use. (Of course, it can be further improved to the ideal - if it works out better for you, then write). The advantages of this method:

all reagents are inexpensive, available and safe

no additional tools are needed (irons, lamps, laminators - nothing, although not - you need a pan)

there is no way to spoil the board - the board does not heat up at all

paper moves away by itself - you can see the result of the transfer of toner - where the transfer did not come out

there are no pores in the toner (they are sealed with paper) - accordingly, there are no mordants

do 1-2-3-4-5 and always get the same result - almost 100% repeatability

Before we start, let's see what boards we need, and what we can do at home with this method.

Basic requirements for manufactured boards

We will make devices on microcontrollers, using modern sensors and microcircuits. Microcircuits are getting smaller and smaller. Accordingly, the following requirements must be met:

boards must be two-sided (as a rule, it is very difficult to separate a single-sided board, it is quite difficult to make four-layer boards at home, microcontrollers need a ground layer to protect against interference)

the tracks should be 0.2mm thick - this size is quite enough - 0.1mm would be even better - but there is a possibility of pickling, track departure during soldering

the gaps between the tracks - 0.2mm - this is enough for almost all circuits. Reducing the gap to 0.1mm is fraught with merging of tracks and difficulty in monitoring the board for short circuits.

We will not use protective masks, and also do silk-screening - this will complicate the production, and if you are making the board for yourself, then this is not necessary. Again, there is a lot of information on the Internet on this topic, and if you wish, you can make a “marafet” yourself.

We will not tinker with the boards, this is also not necessary (unless you are making a device for 100 years). For protection, we will use varnish. Our main goal is to quickly, efficiently, cheaply make a board for the device at home.

This is what the finished board looks like. made by our method - tracks 0.25 and 0.3, distances 0.2

How to make a double-sided board from 2 single-sided

One of the problems with making double-sided boards is aligning the sides so that the vias line up. Usually a "sandwich" is made for this. 2 sides are printed on a sheet of paper at once. The sheet is bent in half, the sides are precisely aligned with the help of special marks. Double-sided textolite is inserted inside. With the LUT method, such a sandwich is ironed and a double-sided board is obtained.

However, in the cold transfer toner method, the transfer itself is carried out with the help of a liquid. And therefore it is very difficult to organize the process of wetting one side simultaneously with the other side. Of course, this can also be done, but with the help of a special device - a mini press (vice). Thick sheets of paper are taken - which absorb the toner transfer fluid. The sheets are wetted so that the liquid does not drip and the sheet holds its shape. And then a “sandwich” is made - a wetted sheet, a sheet of toilet paper to absorb excess liquid, a sheet with a pattern, a double-sided board, a sheet with a pattern, a sheet of toilet paper, again a wetted sheet. All this is clamped vertically in a vise. But we will not do this, we will do it easier.

A very good idea slipped through the board manufacturing forums - what a problem it is to make a double-sided board - we take a knife and cut the textolite in half. Since fiberglass is a puff material, it is not difficult to do this with a certain skill:

As a result, from one double-sided board with a thickness of 1.5 mm, we get two one-sided halves.

Next, we make two boards, drill and that's it - they are perfectly aligned. It was not always possible to cut the textolite evenly, and as a result, the idea came up to immediately use a thin one-sided textolite with a thickness of 0.8 mm. Then you can not glue the two halves, they will be held by soldered jumpers in vias, buttons, connectors. But if necessary, you can glue it with epoxy glue without any problems.

The main advantages of this trip:

Textolite with a thickness of 0.8 mm is easily cut with scissors on paper! In any shape, that is, it is very easy to cut to fit the body.

Thin textolite - transparent - by shining a lantern from below, you can easily check the correctness of all tracks, short circuits, breaks.

Soldering one side is easier - the components on the other side do not interfere and you can easily control the soldering of microcircuit pins - you can connect the sides at the very end

You need to drill twice as many holes and the holes may slightly misalign.

The rigidity of the structure is slightly lost if you do not glue the boards, and gluing is not very convenient

One-sided fiberglass 0.8mm thick is difficult to buy, mostly 1.5mm is sold, but if you couldn’t get it, you can cut a thicker textolite with a knife.

Let's move on to the details.

Necessary tools and chemistry

We will need the following ingredients:

Now that all this is there, let's do it step by step.

1. Layout of board layers on a sheet of paper for printing using InkScape

Automatic collet set:

We recommend the first option - it is cheaper. Next, you need to solder wires and a switch to the motor (preferably a button). It is better to place the button on the body, so that it is more convenient to quickly turn the motor on and off. It remains to choose a power supply, you can take any power supply for 7-12V with a current of 1A (or less), if there is no such power supply, then charging via USB at 1-2A or a Kron battery may be suitable (you just need to try - not everyone likes to charge motors, the motor may not start).

The drill is ready, you can drill. But it is only necessary to drill strictly at an angle of 90 degrees. You can build a mini machine - there are various schemes on the Internet:

But there is an easier solution.

Drill jig

To drill exactly at 90 degrees, it is enough to make a drilling jig. We'll do something like this:

It is very easy to make it. We take a square of any plastic. We put our drill on a table or other flat surface. And we drill a hole in the plastic with the right drill. It is important to ensure a smooth horizontal displacement of the drill. You can lean the motor against a wall or rail and plastic too. Next, use a large drill to drill a hole for the collet. On the reverse side, drill or cut off a piece of plastic so that the drill can be seen. A non-slip surface can be glued to the bottom - paper or an elastic band. Such a conductor must be made for each drill. This will ensure perfectly accurate drilling!

This option is also suitable, cut off the top part of the plastic and cut off the corner from the bottom.

Here is how drilling is done with it:

We clamp the drill so that it sticks out 2-3 mm when the collet is fully immersed. We put the drill in the place where it is necessary to drill (when etching the board, we will have a mark where to drill in the form of a mini hole in copper - in Kicad we specially set a checkbox for this, so that the drill will get up there by itself), press the conductor and turn on the motor - the hole ready. For illumination, you can use a flashlight by placing it on the table.

As we wrote earlier, you can only drill holes on one side - where the tracks fit - the second half can be drilled without a jig along the first guide hole. This saves some power.

8. Tinning board

Why tin boards - mainly to protect copper from corrosion. The main disadvantage of tinning is overheating of the board, possible damage to the tracks. If you do not have a soldering station - definitely - do not tin the board! If it is, then the risk is minimal.

It is possible to tin the board with ROSE alloy in boiling water, but it is expensive and difficult to obtain. It is better to tin with ordinary solder. To do this qualitatively, a very thin layer must be made a simple device. We take a piece of braid for soldering parts and put it on the sting, fasten it with a wire to the sting so that it does not come off:

We cover the board with a flux - for example, LTI120 and a braid too. Now we collect tin into the braid and we drive it along the board (we paint it) - we get an excellent result. But with use, the braid falls apart and copper fibers begin to remain on the board - they must be removed, otherwise there will be a short circuit! It is very easy to see this by shining a flashlight on the back of the board. With this method, it is good to use either a powerful soldering iron (60 watts) or ROSE alloy.

As a result, it is better not to tin the boards, but to varnish at the very end - for example, PLASTIC 70, or a simple acrylic varnish bought in auto parts KU-9004:

Fine tuning of the toner transfer method

There are two points in the method that are amenable to tuning, and may not work right away. To set them up, you need to make a test board in Kicad, tracks in a square spiral of different thicknesses, from 0.3 to 0.1 mm and at different intervals, from 0.3 to 0.1 mm. It is better to immediately print several of these samples on one sheet and adjust.

Possible issues we will be fixing:

1) tracks can change geometry - spread, become wider, usually not very much, up to 0.1mm - but this is not good

2) the toner may not adhere well to the board, move away when removing the paper, it may not adhere well to the board

The first and second problems are interrelated. I solve the first, you come to the second. We must find a compromise.

The tracks can spread for two reasons - too much clamping weight, too much acetone in the composition of the resulting liquid. First of all, you need to try to reduce the load. The minimum load is about 800g, you should not reduce it below. Accordingly, we put the load without any pressure - we just put it on top and that's it. Be sure to have 2-3 layers of toilet paper for good absorption of excess solution. You must ensure that after removing the load, the paper should be white, without purple smudges. Such smudges indicate a strong melting of the toner. If it was not possible to adjust the load with the load, the tracks still blur, then we increase the proportion of nail polish remover in the solution. Can be increased to 3 parts liquid and 1 part acetone.

The second problem, if there is no geometry violation, indicates an insufficient weight of the cargo or a small amount of acetone. Again, it's worth starting with the load. More than 3 kg does not make sense. If the toner still does not adhere well to the board, then you need to increase the amount of acetone.

This problem mostly occurs when you change your nail polish remover. Unfortunately, this is not a permanent and not a pure component, but it was not possible to replace it with another one. I tried to replace it with alcohol, but apparently the mixture is not homogeneous and the toner sticks with some inclusions. Also, nail polish remover may contain acetone, then it will need less. In general, you will need to carry out such tuning once until the liquid runs out.

Board ready

If you do not immediately solder the board, then it must be protected. The easiest way to do this is to coat with alcohol rosin flux. Before soldering, this coating will need to be removed, for example, with isopropyl alcohol.

Alternatives

You can also make a payment:

Additionally, a custom board manufacturing service is now gaining popularity - for example, Easy EDA. If a more complex board is needed (for example, a 4-layer board), then this is the only way out.