This article does not claim to be a bestseller of popular science literature, but rather a guide for beginners. The article describes the winding process itself, and not its calculation.

Sooner or later, in the practice of every radio amateur, the question arises of what to power this or that device. The most common ULF powers are 2*100 or 2*200. That's why the best option There is a “donut” with 150 watts of overall power, in the first case you need one for 2 channels, in the other a couple for dual mono. The toroidal transformer has the best size-power ratio, high efficiency, and minimal interference. This is why audiophiles love them so much. Let us consider the winding process of this type of transformers in more detail.

The main thing that a person who winds a transformer should know and understand is:

• wire length (number of turns) is voltage;
• the cross-section of the conductor is the current with which it can be loaded;
• if the number of turns in the primary circuit is small, then this is excess heating of the wire;
• if the overall power is insufficient (more than possible is consumed), this is again heat;
• overheating of the transformer leads to reduced reliability.

So, what is needed for winding:

1. Transformer iron in the shape of a torus (later I will write where to get it);
2. Lacquer wire (a winding wire is needed for the transformer winding);
3. Masking tape (paper);
4. PVA glue;
5. Fabric tape or wicker tape;
6. Pieces of wire in insulation;
7. And lastly, but most importantly, is desire.

TRANSFORMER HARDWARE

I won’t talk about how to calculate the power of iron; there are already a lot of articles for this... Calculating power is difficult from a practical point of view, since the grade of steel and the quality of its production are not known. Therefore, two cores with the same overall mass have different parameters. Let's consider an example of winding a core on an already “spent” core. One of the easiest to obtain cores, the quality of which is worthy of attention. The core is from the Soviet stabilizer "Ukraine-2" (SN-315). At one time, a lot of them burned down, and you can get such a device on the market for 20 UAH... We are interested in the torus. This donut is wound with aluminum varnish, we mercilessly wind it up (or bite it), we need a core (carefully so as not to damage the core). Aluminum wire can be used for other purposes (twisting brooms or wires), or in my case, I melt it down for other purposes (making radiators). After winding, you get a beautiful core with dimensions of 96-54-32 mm, respectively, outer, inner diameter and height. Below is an example of such a core ( Fig.1 ). The overall power of such a core is at least 120 watts (tested in practice).

Before winding, it is necessary to prepare the iron for winding. If you look at the corners of the transformer, you will see that they are at an angle of 90 degrees, at these points the wire will bend and the varnish will peel off, so that this would not be necessary, it is necessary to process the corners with a file, rounding them as much as possible (I understand that it is lazy, but it is necessary). The minimum radius of the circle is 3mm. In Fig. 1 you can see that the corners have already been processed, and the torus is ready for winding. A little trick: when processing corners with a file, you must avoid licking the steel so that the layers remain open to each other! To do this, move the file along the direction of the transformer tape. After processing, I recommend checking the angles for closing the layers and finishing them with a fine file.

In order to isolate the core from the winding, it is necessary to insulate it with FABRIC insulating tape (or a boiler impregnated with paraffin wax). It is better to use electrical tape with a width of about 25mm (Fig. 2), then there will be maximum coverage of the metal in one layer, which saves space in the window. We do not seal the end of the winding (read further).

After these operations, the core is ready for winding and we move on to the next step.

VARNISH PIPE

I call it a varnish pipeline electrical conductor the insulation of which is made of varnish (culturally, winding or winding wire). Happens different brands PEV, PEV-2, PET-155 and others. I recommend using PEV-2, rich orange color. Also, a very dark-looking wire (PEL), the color of rotten cherry, performed very well, this has thick layer insulation, which allows it to be used for high-voltage transformers (more than 500V). For example, a PEV-2 wire with a diameter of 1.6 mm has an insulation thickness of about 0.06-0.07 mm, and a “black” wire is 0.1-0.11 mm.

Calculating the wire cross-section is a very interesting process. There is a lot of literature on this topic on the Internet, and I will not write about all sorts of calculations and subtleties (Google to help). Depending on the current density you choose, the wire cross-section will be different. The main thing that is required is the correct power ratio. It is necessary that the power secondary winding did not exceed the capabilities of the primary one anymore. As you know, the efficiency of transformers in the form of a torus is very high and equals about 97%, therefore, when winding a torus with a power of 200 watts, 6 watts of losses is a trifle that can be neglected. We believe that the power primary winding greater than or equal to the power of the sum of all secondary windings.

Calculation example. You need to wind the transformer. The primary winding is designed for 220V. There are two secondary windings of 28V each. The diameter of the primary winding wire is 0.6 mm in varnish. The thickness of the varnish is about 0.06mm, and the “clean” diameter of the primary winding wire is about 0.54mm. We substitute the area of ​​the circle into the formula and get a cross section of 0.228 mm 2 (if you don’t know how I calculated this, then buy an amplifier and don’t bother). And so, based on the proportion, we get 220V/28V*2=3.92, which means that the secondary winding should have a cross-section 3.92 times thicker than the primary winding. As you can see, I did not use power and, accordingly, current density. Everyone takes the current density that they consider correct (for myself I take 4A/mm 2, and my thoughts are confirmed by a real trance test, which I will describe further).

For the core described above, it is better to use a primary wire of at least 0.6 mm in diameter. A wire of this cross-section and the required length can be found in old tube TVs, in the form of demagnetization loops. There are always people on the market who buy old TVs (“junk dealers”), and you can find the necessary wire from them. We have two types of loops on the market: small and large, smaller for 20 UAH, large for 50.

Small in diameter, 2 of these are used in TVs. The diameter of such a half-demagnetization loop is about 40-50 cm, the cross-section of the conductor is somewhere around 0.6 mm. At high-quality installation This loop is enough to wind the primary winding of one torus with a margin of a couple of meters.

If you use a large loop, then the length of the wire is literally one and a half times longer than the small one, so it is more profitable to buy small loops. Sometimes you come across a loop from a tube, color TV, the length of the wire in such a loop is similar, but the cross-section of the wire can reach 0.7 mm. If you get one, you're lucky.

And so you found a demagnetization loop, usually it is wrapped in keeper fabric (rag strip), and on top with transparent tape or electrical tape. Near the wire terminals there is a joint where you can catch and carefully unwind the loop. There is no need to cut, saw off, or tear off the insulation; you can damage the wire; in addition, we will still need this insulation. After winding, we are left with a beautiful wire that can be used. Some people rewind the wire onto a “shuttle”, I personally don’t do that, why bend the wire once again if it’s already the right shape, besides, if you wind small tori, the shuttle will take up more space and may not fit through the window, and also damage the varnish . Before you start winding it, you need to make twists so that the wire does not move apart. In order to make twists, you need to take pieces of single-core wire (preferably in PVC insulation) 5-7 cm long. We wind the loop in a circle with a slightly tight step, then during winding, in order to add (unwind the wires) you will just need to twist this spring and the wire will separate (see photo Fig. 3).

Now our loop has one end on the outside, and the other somewhere inside, we need the outside one. Next, let's return to the iron, which we have already processed and wrapped with electrical tape or wicker. Remember, we didn’t seal the edge, that’s why (look at Fig. 4). On the side where the top of the trance will be (the terminals go up), at the corner of the torus we make a cut in the center of the insulating tape and thread the varnish pipe there already in the insulation; this will be the outlet for the beginning of the winding. Some recommend soldering a piece of flexible stranded wire in insulation and making such a tap. I’m not happy with this option because in this way I don’t know which wire is in the primary, and even ten years later I measured it with a micrometer and you know what you can reap from it, but with a tap, who knows what the cross-section is there. It's up to you though.

Let's make leads for the wire. The winding terminals must be “strengthened” with additional insulation. PVC insulation (Soviet white) is very suitable for these things, but insulation made from wire of the required cross-section is even better. You can use heat shrink, but it is better to use PVC or insulation because the former tends to bend in one place, which we really don’t need; we are trying to protect ourselves from this so that the wire does not break off. In order to tighten the insulation, I recommend taking a wire that has additional insulation in the form of a thread wrapped around the conductor. In this case, the thread does not create a strong bond between the PVC and copper and allows the insulation to be pulled together. To make it easier to tighten the wire, you need to bend it a little (at 45 degrees). I recommend “stretching” the insulation at a time and using it. ( Fig.2).

Domestic winding wires

The most widespread are winding wires in enamel insulation based on high-strength synthetic varnishes with a temperature index (TI) in the range of 105...200. TI refers to the temperature of the wire at which it useful resource not less than 20,000 hours

Copper enameled wires with insulation based on oil varnishes(PEL) are available with a core diameter of 0.002...2.5 mm. Such wires have high electrical insulating characteristics, which are practically independent of the external influence of elevated temperatures and humidity.

Wires of the PEL type are characterized by a greater dependence on the external influence of solvents, compared to wires with insulation based on synthetic varnishes. PEL winding wire can be distinguished from others even by external sign - enamel coating The color is close to black.

Copper wires of types PEV-1 and PEV-2 (available with a core diameter of 0.02...2.5 mm) have polyvinyl acetate insulation and are golden in color. Copper wires of types PEM-1 and PEM-2 (with the same diameter as PEV) and rectangular copper conductors PEMP (section 1.4...20 mm2) have varnished insulation on polyvinyl-formal varnish. The index “2” in the corresponding designation of PEV and PEM wires characterizes two-layer insulation (increased thickness).

PEVT-1 and PEVT-2 are enameled wires with a temperature index of 120 (diameter 0.05...1.6 mm), they have insulation based on polyurethane varnish. Such wires are convenient to install. When soldering, it is not necessary to strip the varnished insulation and use fluxes. Regular POS-61 solder (or similar) and rosin are sufficient.

Enameled wires with insulation based on polyesteramide PET-155 have a TI equal to 155. They are produced with cores not only round section(diameter), but also rectangular (PETP) type with a conductor diameter of 1.6-1 1.2 mm2. In terms of their parameters, PET wires are close to the PEVT type wires discussed above, but have higher resistance to heat and thermal shock. Therefore, winding wires of the types PEVT and PET, PETP can be especially often found in powerful transformers, including transformers for welding.

WINDING PROCESS

To wind the trance you will need 4-5 evenings and 2 hours of time, why at least 4 days you will understand later.

We have already run one end of the wire and pressed it. Then the most dreary winding begins. I recommend winding it this way. We take a trance (for now it’s iron), put on a glove or take some rags from natural fabric. We sit down on the sofa or bed, turn on a movie we’ve already seen or music (so as not to be too distracted), and start watching. We thread each turn into an iron ring. You need to wind turn to turn from inside(some manage to do it externally, I can’t imagine how).

To make it easier to count the turns, it is better to group them into 5 or 10 turns. It is necessary to pull the wire not strictly perpendicular (dotted red line) to the tangent (pure red), but slightly inclined towards the winding (yellow), as if the inner part of the winding goes in front of the outer one (Fig. 5). In this way, the winding wire, when stretched, will itself be pressed against other already laid turns. If your wire is bent, it will not fit perfectly, so it must be as straight as possible; to do this, while winding it, you need to pull it strongly, thereby straightening it. That's why you need gloves or rags; if you don't use gloves, your fingers and palms get tired and sore very quickly. If you wind a wire with a cross-section of more than 1.5 mm (very difficult), then I recommend slightly bending the wire under tension for ease of straightening.

(My friend’s father winds 50 hertz welders, a secondary shredder lays 35 squares of copper with his hands perfectly evenly, so he bends 5 kopecks of Ukraine into a dumpling with his fingers).

During winding, the wire is inspected for flaws, especially in places of bending; if the varnish is broken, then carefully cover it with insulating capon varnish or paint (on extreme case regular nail polish).

When the layer has been wound to the end. Between layers it is necessary to make interlayer insulation. I'm lucky and I have some stash of varnished fabric, and the fabric is stretchy and impregnated with something sticky. If these stick to each other (formed), then it is very difficult to separate them. It makes my fingers stick together. This varnished fabric is an ideal insulator; in addition, the winding does not rattle even when overloaded. But very few people have this. The same functions of an insulator can be very well implemented using masking tape.

After we have wound the layer, we take it and insulate it with masking tape. We make strips about 15mm wide. And we initially wrap the trans with these strips to insulate it inner part winding the wire (from inside the donut). Then we isolate the gaps from the outside of the donut. As a result of insulation with adhesive tape, it turns out that the insulation from the inside, by applying layers, will become twice as thick, and from the outside it will be single. After wrapping, it is necessary to generously lubricate the torus with PVA glue, this is done in order some scotch tape did not unwind, and it will also become stronger and seem solid. In addition, the glue will hold the windings so that they do not “buzz”. There is no need to spare glue, lubricate it with your finger and rub it in lightly. After which the torus needs to dry. I usually wind the torus in the evening, soak the layer with glue, and place the torus itself on a needle radiator for good air circulation. Overnight the torus dries and can be wound further. That's why a minimum of 4 pm is required for winding (4 pm - 4 layers). If necessary, you can speed up the drying process with a hairdryer. Let's rewind next layer... the winding process itself is similar and no different. At the end of the winding, we place the end of the winding in the same insulation as at the beginning of the winding. Then we fix the end of the winding masking tape, insulate the winding with masking tape and soak it with glue.

There is another good option for insulating between layers. It will be very good if during winding you use baking paper (parchment) cut into the same strips and then wrapped. As a result, the trance will need to be soaked, but actually cooked on steam bath mixtures 50:50 respectively paraffin:wax. Steam bath Take some water into a saucepan and set it to boil (we need steam). We place a container on top in which the transformer and paraffin wax are placed. We tie the transformer to a wire in advance, leaving the end (when the mixture flows behind this thread, you need to soak the transformer like a tea bag in a cup). When you dip the transformer, you need to be careful so that drops of wax do not fall on the flame, it is very flammable!!! Previously, output transformers for tube ULFs were impregnated with exactly this “dissolution,” although other high-quality trances were also impregnated. When the mixture is heated, it has a very high fluidity, almost like water, as a result of which the paper becomes literally saturated with paraffin and wax. However, this option will not be initially effective if the trans is heated (warm) at a temperature of 50 degrees, the wax is already quite soft and will not restrain the wire from vibration of 50 Hz, although it will act as a dielectric. (Truth is precisely because of vibration and vibrationThe wires are frayed and a closed turn is obtained, which leads to damage already during operation).

For pulse transformers, I recommend using paper + BF-2 glue as impregnation rather than tape. This adhesive is primarily used in the manufacture of speaker coils. But in pulse transformer also performed very well. With repeated overload, not the slightest squeak at the conversion frequency of 15 KHz.Unwinding the windings from the frame, they were removed with a cable w Irina has 8 veins.

During winding, we periodically measure the no-load current; to do this, you need to connect the tester in series with the primary winding in ammeter mode (read the instructions for the tester). Measure the current x.x. You need to be very careful because it works from the network! To avoid any emergency situations, I recommend turning on a 220V light bulb in series with the primary, with a power of about 40W. The light will light up if the number of turns is very small; if the trans is wound correctly, then it should only be with pink tint, which indicates a low current flowing through it. The transformer has large starting currents; at the moment the transformer starts, overloads can reach 160 times. Therefore, starting the transformer must be done not directly through the tester, but using a “jumper”, which you then open and the current begins to flow through the tester. The jumper can be implemented by simply shorting the tester probes, which are then opened. I will write below what the no-load current should be.

For transformers with low current consumption, it is recommended to use a 10 or 100 Ohm resistor (2-5W) which is connected in series with the primary winding. After measuring the voltage drop across the resistor, use Ohm's law to resample the current. This method is more preferable than the first, but at the same time more dangerous at high current consumption - the resistor turns into coal in a fraction of seconds!!!

About how to measure current x.x. I briefly told you what I wrote, now about the meanings. Current norm x.x. Everyone determines it individually for each trance, but usually the norm is up to 50 mA at 230V, although some say that 0.5A is normal. The lower the current, the better! The lower the quiescent current, the more the current shape is x.x. looks like a sine. If you have current x.x. from 20-50 it’s tolerable, let’s say a C, from 10-20 it’s four, less than 10mA it’s clearly five. For small torics, the current will be small due to the high resistance of the primary winding, this must be taken into account! Although how can I wind a toroid by hand with less than a hundred watts, this is an atrocity! The number of turns of the primary winding in them reaches a couple of thousand.

The transformer I wound according to my method has a current of x.x. equal to 11mA (with 4 layers of primary).

If you do everything sequentially, you will get something similar:

TESTING AND MEASUREMENT PROCESS

About how to measure current x.x. I briefly told you what I wrote, now about the meanings. Current norm x.x. each is determined individually for each trance, but usually the norm is up to 50 mA at 230V, although some say that 0.5A is normal. The lower the current, the better! The lower the quiescent current, the more the current shape is x.x. looks like a sine. If you have current x.x. from 20-50 it’s tolerable, let’s say a C, from 10-20 it’s four, less than 10mA it’s clearly five. For small torics, the current will be small due to the high resistance of the primary winding, this must be taken into account! Although how can I wind a toroid by hand with less than a hundred watts, this is an atrocity! The number of turns of the primary winding in them reaches a couple of thousand.

It will be very useful to look at the shape of the no-load current in the primary winding using an oscilloscope. BUT!! this needs to be done very quickly special conditions! For this, an isolating transformer (220/220V) is required, while the induction must be very low whatever do not cause additional distortions of the “sine” shape. And also latr. I recommend doing this test item only very experienced specialists, the consequences are fraught with burnout of the oscilloscope!!!

When using my winding parameters, I “removed” 150 watts from such a trance for several hours (there was no time longer).

We isolate the primary winding from the secondary.

After winding the required number of layers of the primary winding, we come to the moment of winding the secondary. It is necessary to isolate the primary winding from the secondary very carefully.

If the secondary winding suddenly burns out, the worst consequences are failure of the ULF. But if at this moment the secondary winding somehow “shorts” to the primary, then this is already a danger to life! Because the secondary winding of the transformer at the middle point is connected to the wuxia body, imagine that when you turn the volume control knob you get an electric shock?! It’s unpleasant, therefore, grounding in the socket is not a desirable norm, it’s a necessity, if you value your health, I recommend that you devote it to this Special attention... (This was a slight digression).

Based on the fact that sockets VERY rarely have REAL grounding, you need to isolate the primary winding from the secondary as much as possible. For this operation, you can use the already established method and use masking tape. BUT the thickness of the layer needs to be at least doubled, or better yet tripled. Moreover, impregnation with glue is necessary; the glue will add elasticity and an additional layer. More the best option There will be the use of special electrical varnishes such as TsAPON (color is not important). In this case, we literally soak the torus in varnish, you can even soak it! The varnish will be more fluid if it is heated; when heated, the capon melts like water, and thereby saturates the windings well, insulating and securing them. Regarding the primary winding, these are some of the best measures, as for me even better than paraffin. If you are going to use impregnations, then it is logical that using any “yellow transformer” tape is contraindicated; a layer of tape simply will not allow it to leak deeper, unlike paper or varnished cloth. Regarding “fixing” and insulating the secondary winding with the help of varnishes, I am categorically against it (what if you need to rewind the secondary, this will not be possible, besides, the wound wire will only be used for scrap metal.)

If there is no varnish, and masking tape is not impressive. It would be a very good idea to insulate the windings with fluoroplastic, this material is a super insulator! In appearance it looks like a white, slightly transparent film (photo below).

The main feature is that it is thermally resistant to heat (from minus -268 to +260 degrees).When I need to increase the temperature of the soldering iron tip, I simply wrap it with fluoroplastic, preventing the “body” of the soldering iron from cooling).Such highlights can only be found in specialty stores, although there will also be J varnished fabric nearby, which is also very good. Not everyone has access to such assortments, but if you want... In this case, I recommend rummaging through the bins. Fluoroplastic of the shape we need can be obtained in FT type capacitors. If carefully dismantled aluminium case capacitor, then we will get a core (the capacitor itself) from tightly wound fluoroplastic so we need it. From a 0.022 microfarad capacitor you can wind two pieces of one meter each. To insulate the primary we need about 5-6 meters. That is, we are looking for at least 3 capacitors. Fluoroplastic capacitors have very good sound, so think first before spoiling them.

Keep in mind that fluoroplastic will not allow the trance winding to become saturated like tape, so if you want to soak it with paraffin, do it before insulating the windings with fluoroplastic.

I’ll describe shielding the primary winding from the secondary a little later; this is more likely to be included in the section on higher matters.

Final finishing of the trans and its fastening.

I skip the moment of winding the secondary, because it is absolutely similar to the process of winding the primary. As for the final finishing, there are some points you need to understand.

A toroidal transformer is a closed magnetic circuit, the core tape is wound in a dense roll after being annealed in an oven under vacuum. Winding it is complicated by the need to thread the wire through the window. Its advantage is that the core itself is located inside without emitting unnecessary interference, because at the moment they are picked up by the trance secondary. Thus, the core of the trance - a rough piece of iron - is inside, and a soft copper wire exposed with a fragile varnish (a piece of iron) bravely protects it. The toroid body is very susceptible to external damage. A torus falling from a decent height can “kill” it with the help of short-circuit windings. Whereas trances such as PL or Sh-shaped iron, on the contrary, protect the secondary winding. In this way, it is much easier to secure the TS-nick because it can and should be compressed very strongly with metal ties in order to reduce the gap in the core, and thereby minimize losses and hum-vibrations of the plates. It is much more difficult to secure a toroid, or rather there are minimal options. Before making the final finishing of the trans, you need to clearly understand how the trans will be attached to the body.

And yet, what are the options for insulation and finishing:

As an option, you can use transparent tape in which the demagnetization loop was packaged (by the way, some loops were wrapped in fluoroplastic, check if you are lucky). The result is very beautiful donuts (you can see the winding and a beautiful wire). But the increased temperature of the transformer will soften the insulation, thereby reducing its strength level. But that's not the main thing! When you insulate the transformer with a “film”, the level of heat transfer drops significantly, and the torus can heat up more. I think everyone is trying to buy things from natural materials, trying to avoid synthetics, because the body “does not breathe” in it and a person sweats... so why should the torus endure. For these things, it is better to use keeper tape (sheet cut into J strips). For whatever it was even stronger; before winding, I soaked it in that PVA glue. Then I wrap the torus; during winding the excess is squeezed out. After drying, a nice, stiff rag frame is formed... If you suddenly need to unwind it, just soak it for a while. Treatment options (on an already wrapped transformer) with both alkyd and water-based paint, or special varnishes are also allowed.

What are the mounting options:

One of the obvious ways to fasten a torus is to fasten it with a bolt threaded through the center of the torus. When fastening in this way, keep in mind that through the bolt, and then the bottom of the case, then along the walls of the case, the top cover, a coil of cross-section may form that is simply crazy (depending on the diameter of the fastening bolt). Under no circumstances should you attach the torus to the bottom and top cover, you will form a closed loop and burn the torus!

In addition, in the gap between the fasteners and top cover interference will be generated since the bolt is iron (magnetic). The smaller the gap, the higher the level. It’s not uncommon to say that without a lid the ULF plays everything fine, there’s no background, I cover it with a lid and a crazy background appears. Interference is induced; to avoid such interference, it is necessary to use a fastening bolt made of diamagnetic materials, for example, brass has shown itself to work well... (but do not forget about the possibility of a coil forming through the body).

Now you need to somehow rest against the torus winding, while the contact area should be maximum to minimize the pressure on the wire. For these purposes, I use the rear washer and the core from the magnetic system of the speakers, all you need is to drill a hole in the core and cut the thread, and then you get a very good fastener (photo below).

You can also cut out a piece of PCB or gitinax with a thickness of 3 mm and shape it for maximum contact of the “washer” with the surface of the torus. You need to use a gasket between the “washer” and the body of the torus; for this, use rubber, the thickness of which should be at least twice as thick as the diameter of the secondary winding (guess why), with a bed both below and above. When making this washer, it is possible to provide for the installation of copper rivets in order to fix the terminals on the “terminal block”. If anyone is unclear, there is a photo of such a design.

The diameter of a pin or bolt threaded through the center of the torus is unlikely to correspond to the diameter of the window. In order for the donut not to fly on this bolt like a hoop on a ballerina, you must either wrap it with electrical tape (to the required diameter) or you can use thick cone-shaped rubber. Motorists can easily find this kind of rubber band, for example, a rubber band from a VAZ2107 jet stabilizer or shock absorber, it has the right shape and costs a penny.

It is not uncommon in factory versions to fill the window with a compound by inserting a bushing into which the torus is attached. In practice, this is not used by radio amateurs (usually) because, again, it is not possible to disassemble the torus without damaging the wire. At home, such a plug can be created using epoxy.

Another version of the “spider” fastener. In essence, the same washer cover is made, only in larger sizes. Its form is usually square cover made of iron or textolite, the edges protrude beyond the boundaries of the outer part of the transformer. Holes are drilled in these corners and bolted to the body, so you don't have to thread a bolt through the center and create an unfinished loop through the ULF body.

It would be VERY good to make an iron “pot with a lid” from thick steel (min. 2mm) for the toroid, in which to place the torus and fill it with a compound such as paraffin or wax (or the same epoxy resin), although after epoxy it will not be disassembled. This solves not only the problem of fastening, but also shielding from interference. (I have a photo of a similar design lying around on my computer; I don’t remember the author, but I think he won’t be offended).

A little about shielding.

It would be very good to place a shielding winding between the primary and secondary windings. Ideally, this winding should practically cover all visible parts of the toroid, blocking the magnetic fluxes on the way from the core (primary winding) to the secondary. One end of the shielding winding should be “in the air”, and the other should be connected to the amplifier housing (sometimes through a resistor of up to 10 ohms). The first end can be well insulated and left inside the torus. The second one, the one that is connected to the ground of the housing, is brought out using a multi-core flexible wire.

Ideally, winding should be done with copper tape about 15-20mm wide, which is insulated on both sides with varnished fabric, electrical tape or fluoroplastic, or you can use masking tape, but very carefully so as not to tear or make microcracks (both in the tape and in the insulator), which will break through with voltage. Shielding in this way takes up a lot of space and creates a lot of voids, which impair heat transfer, adding hum and “in vain” moving the secondary away from the core. It will be “more” economical if you wind the screen with a wire with a diameter of about 0.6 mm. But if the core is visible, then be sure that the interference will pass through these “windows”, that is, either we wind it as needed very tightly in several layers, or we don’t do empty work! If possible, you can make such a screen, it will definitely be worse!

It is much better to shield the transformer after winding, that is, when the transformer is completely wound (Although, to be honest, you need to divide interference by class and type, and separately consider methods of dealing with them). Ideally, in this case, it would be to use permalloy rather than copper tape. Although if they look at you with brick eyes when you hear the word fluoroplastic, then you can dream about permalloy ;). It is very good to wrap the transformer in several layers of transformer iron; iron from any transformer is suitable for these purposes. (I use steel from an old 2-amp latra core).

Here is a torus shielded using a transformer tape, placed in metal cover and boiled in paraffin, current x.x. 1.5 mA, primary coils over 2500 turns, interlayer fluoroplastic, with sequential paraffin welding. I made it in a mug + transformer steel, it turned out very well (see above)! This toroid was used to work in a preamplifier.

It’s not worth making a pot out of aluminum; it won’t protect you from anything. It needs to be made of thick steel (at least 2mm), and it’s also very good to additionally shield the inside with copper (sheet about 1mm thick). Although I didn’t do such things myself (from copper), authoritative people advised me.

In conclusion about interference from tori, I will say that toroids very rarely generate interference on equipment, while a peculiarity has been noticed that the toroids that generate noise are those that are not home-wired and have a high current. or increased induction... Therefore, if you are not greedy and wind the toroid with a low magnetic induction (increase the number of turns per volt), then you are unlikely to encounter the problem of interference from the transformer.

It is planned to supplement the article with such “highlights”... very briefly for now...

Internal resistance.

All transformers and energy sources (power supplies) have such an abstract parameter as internal resistance. What does it mean?! In the case of a transformer, this resistance will be equal to active resistance windings When you connect a load to the trans, the flowing current and the resistance of the windings create a voltage drop. In order for the voltage drop to be minimal, it is necessary to increase the cross-section of the conductor (reducing its resistance). But at the same time, it is necessary to take into account this fact during operation, that the overall power of the windings will be higher than the overall power of the core, carefully so as not to overload the primary.

Sectional winding.

Low induction.

Implicit coil.

Shielding and types of interference.

P.S. My first article and not yet finished, please don’t throw tomatoes... There’s no time to finish, I’m posting what I’ve already dashed off a long time ago... Now this bagel is successfully working in Natalie 2012EA, you can look for a photo in the corresponding thread, and here it is

There were several transformers lying idle, and one of them (Soviet TCA-30-1, 30 W) I decided to use it for a universal power supply.

Since its original windings did not suit me (mainly due to permissible current), then I decided to remove all its secondary windings and wind my own. The process was accompanied by many “discoveries” and puzzling questions, in the process of solving which many useful details were collected that I wanted to share with newcomers to this business, like me.

The article contains a video with details of some of the stages.

Where I was unfairly lucky here:

1. Was free time and no one interfered.
2. There were a lot of different old stocks, incl. copper wire required length.
3. There is a lot of information on the Internet (especially regarding theory).

Zarathustra forgave me...

Video of transformer rewinding

Time different stages this video:

26 min 28 sec- foil screen between primary and secondary

27 min 52 sec- how to correctly connect the windings in series

36 min 43 sec- how to find out the direction of turns using a battery and a multimeter

44 min 14 sec- calculation and winding of a new secondary winding

1 hour 24 minutes 20 seconds- mains voltage drop and other losses

1 h 30 min 01 sec- no-load current

1 hour 32 minutes 14 seconds- aluminum soldering

1 hour 33 minutes 42 seconds- result

Study of a modified transformer

Transformer TCA-30-1 turned out to be wound with aluminum wire (the letter “A” just means aluminum).

Fortunately, there was enough information about him on the Internet, although the reality did not coincide with the passport found for him. According to the passport, one of the windings was supposed to be copper (the PEV-1 wire does not have the letter “A” in the name like the others - PEVA), and I planned not to touch it, but during the work it turned out that this winding was also aluminum . That's why I deleted it too. Those. Only the primary winding remained intact.

Aluminum foil screen

During the disassembly process, out of curiosity, I unrolled some paraffin paper over the primary winding and wanted to look at it, and came across one turn of foil that was present between the primary winding and the secondary. This roll of foil overlapped with the paper, i.e. he didn't lock himself, and only one the ends were connected by spot welding to the body with a piece of copper wire. This separation is used as a shield against interference, although there is debate about its effectiveness. The transformer is Soviet and the screen was installed at the manufacturer's factory - I did not touch it.

Direction of turns

The turns on the transformer were wound on different coils(left and right) absolutely the same (not mirrored, but exactly the same). Later it became clear that such winding was done purely for convenience at subsequent serial connection windings from different coils. Apparently, for the same reason, the direction of different secondary windings alternates. In this case, the jumpers between the windings in a series connection are simply convenient put on one side.

Metal terminals

The terminals of this transformer are very difficult to solder and tin, since they do not appear to be made of copper. Copper, the better you warm it up, the better it is soldered, and for steel (?) terminals, heating leads to the solder rolling into a ball and flowing from the terminal to the soldering iron tip. It is necessary to catch one of the initial moments of warming up so that the solder remains on the terminal in an acceptable form.

In the transformer under study it was doubly difficult, because aluminum was soldered to the metal terminals. Had to use it for soldering phosphoric acid followed by washing with water and drying on a radiator.

Primary winding

This transformer has two coils, and each winding is divided into two equal parts, which are wound on each of the two coils, with a series connection. It is believed that this way the efficiency is higher - the load is more uniform.

The primary winding consists of two 110v on each coil, connected in series with a jumper. In addition, a small additional winding is connected in series to each of the windings, which I disconnected and used for my own purposes (thus turning it into a secondary winding). The voltage of this additional pair is about 36v (at 230v in the network).

Calculation of the secondary winding of the transformer

The main mistake I made was calculating the secondary winding based on the voltage in the 220v network. Meanwhile, The network voltage during peak loads can drop to 185v, - this is almost 20% lower than expected! Therefore, when calculating the secondary winding, we must proceed from this indicator - not 220, but for example 180. Otherwise, you can seriously miscalculate.

When calculating the voltage in the power supply transformer, the following should be taken into account:

• Minimum network voltage ~180 V
• Voltage drop across the diode bridge - more than 2 V
• Voltage drop across the stabilizer - for example 3 V
• Voltage drop on the secondary windings with increasing load current (multiply by an average of 1.02 - 1.06, depending on the maximum current)

The figure below shows the voltage on one element of the diode bridge KBU801 at a current of 8 A it reaches 1.08 V. I.e. across the entire bridge the voltage drop will be more than 2 V (click to enlarge).

To clarify the number of turns per volt in the secondary winding, you can do temporary control winding(for example, 10 turns) and measure the voltage it produces ( Be sure to check the network voltage!). Then divide these 10 (turns) by the resulting voltage. This gives us the number of turns per volt.

IMPORTANT! It is necessary to divide the turns of the control winding by its voltage, and not vice versa!

Example.

A supply voltage of 20 V with a maximum constant current of 2 A is required.

A rough estimate looks something like this:

20 + 3 = 23 V (voltage drop across the stabilizer)

23 + 2.2 = 25.2 V (voltage drop across the diode bridge)

25.2 / 1.41 = ~17.3 V (we convert the DC voltage after the diode bridge with a capacitor into the required secondary variable)

17.3 * 1.06 = ~18.4 V (we take into account the voltage drop in the winding at maximum load current)

If we have, for example, 4.4 turns per volt at an ideal ~220 V, then at a voltage of ~180 V in the network, we will need

18.4 * 4.4 = 81 turns (for ideal voltage ~220 V)

81 * (220/180) = 99 turns (for peak voltage drop up to ~180 V)

Those. at ~220 V in the network, the secondary winding containing 99 turns will produce about ~22.5 V
(and when the network draws down to ~180 V, the required ~18.4 V)

Winding

I was reeling at the same time four parallel wires. As a result, I got four windings on each coil in each row. This number of windings makes it possible, by connecting them in series (or parallel), to combine the required voltage (and current).

For laboratory block power supply used as a tool during work, this is the most convenient option.

IMPORTANT! For an "O" core transformer with two coils on the right and left (such as the one discussed in this article), it is best to every The winding is divided into two (identical), wound on different coils and connected in series. In this case, the efficiency will be higher.

BY THE WAY when laying on the frame, it is advisable to slightly bend the wire outward before each bend at the corners, so that the turns do not then move away from the frame, forming a gap in which the winding density deteriorates. I also pressed down the wire pine with a block after each bend on the frame.

Calculation of wire length.
Before winding, it is necessary to measure the width of the frame and the width of the window between the coil frames (or frame and core).
After this, you need to calculate the length of the wire and take into account its diameter (with varnish insulation!). If winding occurs without disassembling the core, by threading the wire through the window, then a piece/pieces of wire of the required length will need to be “bitten off” in advance, so it is important not to make a mistake. If the wire is thin enough (for example, less than 0.5 mm) and long, then it makes sense to make a thin shuttle on which to wind the wire of the required length - this will make it easier to drag it through the window.

Here, for example, the internal length of the frame was 54 mm, and expecting to lay 52 turns of wire with a diameter of 1 mm, I did not guess - I had to partially overlap the last half turn (apparently I did not take into account the thickness of the varnish insulation).
See picture (click to enlarge):

When calculating the capabilities of a window, you need to take into account the total thickness of the insulating pads made of paper or varnished fabric between the windings.

To accurately calculate the required length, you need to make a control turn and measure its length. At the same time, in every next row the coil will be slightly longer (the thickness of the bottom row and the thickness of the inter-row insulating spacer will affect). You need to understand that, for example, with 50 turns, a length error of one millimeter per turn will give an error of 5 cm at 50 turns. You also need to take into account the margin for conclusions (I added to total length pieces of 10 cm on each side, i.e. only 20 cm - this was enough for both conclusions and a possible error).

Direction of turns

I had difficulty finding information about the direction of the winding turns - for this I had to refresh school course physics (gimlet rule, etc.). Although this question inevitably arises for a beginner.

The main rule is The direction of the winding turns does not matter...until the need arises connect windings with each other (series or parallel), or in the case of using a transformer in some devices where it is important signal phase.

It doesn’t matter in which direction you wind the turns - what matters is how the windings are then connected

Series connection of windings

When connecting the windings of a transformer in series, you need to mentally imagine that one winding is a continuation of the other, and the point of their connection is break of a single winding, wherein direction rotation turns around the core remains unchanged(and of course cannot turn in the opposite direction!).

In this case, any terminal of the winding can be the beginning or the end, and the direction of rotation itself can be any. The main thing is that this direction remains the same connected windings

In this case, the movement of the connected windings from top to bottom of the coil or from bottom to top does not matter (see the figure - enlarged by clicking the mouse).

In transformers in which the core is shaped like the letter "O" and the coils are wound on two frames on the right and left, the same rules apply. But for ease of understanding, you can mentally “tear” the core (from above or below), and imagine that it is straightened into one rod - this will make it easier to understand how one winding passes into another while maintaining the direction of rotation of the turns (clockwise or counterclockwise) . See the picture below (the picture can be enlarged by clicking the mouse).

Parallel connection of windings

When connecting in parallel, the length of the wire in the windings is important.

Even with same number of turns, different windings may have different lengths wires(the winding that is closer to the middle will be shorter, and the one further away will be longer). As a result, there may be flows.

If parallel connection of the windings is assumed, then it is better to wind them simultaneously in two (three, four...) wires. Then they will be the same length, which will eliminate cross-flows as much as possible during their further parallel connection.

Winding several wires is also used when there is no wire the required section(they are dialing large section several smaller wires).

Checking the direction of turns using a battery and a multimeter

If there is a transformer in which you need to connect two windings in series, but the direction of the turns is not visible or known, you can apply a pulse direct current from the battery to one of the windings, observing the voltage surge on the other winding.

When the voltage surge at the moment of connecting the battery on the multimeter (on the second winding) is in “+”, then the connection points of the windings will be any “+” and “-” of different windings (for example, “+” of the multimeter and “-” of the battery, or vice versa) . The other two ends will be the outputs of these windings after connection (see figure - click to enlarge).

Direction of turns on different coils

I repeat - the direction of winding is not important, it is the connection of the windings that is important.

There is one "but" though. If we talk about convenience, then on this type of transformer (with a core in the shape of the letter “O” and two coils), it is more convenient to wind the right and left coils the same(not mirrored, but identical). In this case, it will be more convenient to install jumpers when connecting two windings in series on different coils - the jumpers will be on one side, and not across the entire frame from top to bottom.

See the picture (to enlarge, click on the picture):

No-load current

If everything is done correctly and the transformer core was assembled (at the factory) with high quality, then the no-load current (current of the primary winding, with the secondary winding completely disconnected from the load) should be within acceptable limits.

In my case, this current was 27 mA, which is just an excellent indicator.

The ammeter must be connected to the gap network cable connected to the primary winding and, preferably connecting the multimeter probes, connect the transformer to the network. Then disconnect the probes and observe the readings. It is necessary to connect the probes before connecting them to the network to avoid failure of the multimeter, because The transformer may have a large starting current (tens of times higher than the rated current).

if you have power transformer with suitable(in this case S = 10.4 cm²) in terms of power cross-section of the core, but its secondary winding is designed for a different voltage, you can rewind the transformer.

In this case, you can not carry out such labor-intensive work as winding a multi-turn primary winding, but use a ready-made, old primary winding.

We determine the location of the primary and secondary windings on the frame. The primary winding is usually located on the frame closer to the core and is wound thin wire With big amount turns.
Next you need to determine the number of turns per volt w for this steel core. You cannot use the value of the number of turns per volt previously calculated for the previous article.
Let's connect the transformer to a 220 volt network. Let's measure the voltage on all secondary windings. Let's choose the winding with the lowest voltage. For example, it will be equal to U = 30 volts. Let's mark its location on the frame.
Next, you need to disassemble the transformer, removing the core plates and freeing the frame. You need to rewind the transformer, wind the old secondary (or secondary, if there are several) windings and count the number of turns in the selected winding.
We leave only the primary winding and inter-winding insulation.
Let's say the number of turns in the selected winding will be n = 140.

Then the number of turns per volt w for this transformer will be:

w = n: U = 140: 30 = 4.67 turns.

If there is no secondary winding at all, or there is no way to calculate it, we will proceed in a different way.
Wind 100 turns over the primary winding insulated wire any diameter is a “measuring” winding.
Let's assemble the transformer again, connect it to a 220-volt network and measure the voltage on the “measuring” winding with a voltmeter. Let's say it will be 21.5 volts.

Let's calculate the number of turns per 1 volt for this transformer:
w = n: U = 100: 21.5 = 4.65 turns.
Then the number of turns in the new 36 volt secondary winding will be:

U_2 = 36 4.65 = 167.8 turns. Let's round up to 170 turns.
The “measuring” winding should be removed and wound with a wire of the appropriate diameter.

How to wind a transformer on an W-shaped core?

This article is a continuation of the articles:

Winding the windings of the transformer frame on an W-shaped core must be done on a winding machine equipped with a revolution counter and special device for attaching the frame and bobbin with wire. But, as a rule, there is no such machine at hand.

We use a regular one for winding hand drill. Before winding, you need to remove and put the frame on the mandrel several times so that the frame sits more freely on the mandrel. Next, we put the frame back on the mandrel, reinforce it with two plywood planks (the planks are needed so that the cheeks of the frame do not spread out to the sides when winding the wire), tighten it with a bolt or pin and secure it in the chuck of a hand drill.The drill must be secured in a table vise.

It is necessary to calculate the gear ratio of the chuck and the drill handle. To do this, let's count the number of revolutions of the drill chuck per revolution of the handle. Or, if possible, count the number of teeth on both gears. The ratio of their number will give the conversion factor n.

For example: the number of teeth on the handle gear is 35 pcs., the number of teeth on the chuck is 7 pcs., then the coefficient n = 35 / 7 = 5. With one revolution of the drill handle, 5 turns of wire are wound onto the frame.

When winding the transformer frame on an W-shaped core, you need to count not the number of revolutions of the chuck, but the number of revolutions of the drill handle, which is much simpler and more convenient. Let's determine the number of turns of the handle for the network primary winding.
K = 1050/5 = 210 rpm.
To wind the primary winding you need to make 210 revolutions of the drill handle.

One practical advice: in order not to lose count of the number of revolutions when winding the coil, after every 10 revolutions of the drill handle, somewhere on the paper you need to make a mark - a tick.
I counted the number of ticks equal to 21 - that’s when the primary winding is ready.

It is necessary to make a hole in the cheek of the frame for the wire to exit. The hole is made with an awl in the cheek, which goes outside the transformer.
The enameled winding wire is connected to the stranded wire. The junction is covered with a piece thick paper as in the picture...

When winding transformer coils on an W-shaped core, it is best (I highly recommend) to wind it turn to turn, laying capacitor paper between the layers for insulation between the layers.

The width of the condenser paper should be 4-5 mm wider than the distance between the cheeks of the frame and have cuts along the entire length, as in the figure….
The reason for increasing the width of the paper is this: when winding, the turns of the wire press the paper, it becomes deformed and narrows in size. The turns of the lower layer are exposed, and interturn breakdown between the layers is possible.

Having wound the primary winding and brought out the end with a stranded wire, lay 2-3 layers of paper or varnished cloth (inter-winding insulation) to protect the wires of the network winding from accidental contact with the wires of the output winding.

Winding the secondary winding using a drill is not convenient, because the secondary winding wire is thick - 1 mm in diameter... It is best to wind the secondary winding manually by removing the workpiece with the frame from the drill chuck.

The secondary winding is also wound turn to turn with a strip of paper (the same as the primary winding) placed between the layers. Number of turns of the secondary winding at 36 volts there will be 180 turns.

The ends of the secondary winding are removed from the frame by the wire itself, without soldering to a stranded wire. You can only, for strength, put a thin vinyl chloride tube on the wire.

After winding the secondary winding, 2-3 layers of thick paper are laid again to protect the wire from external damage. Then the finished frame with windings is carefully removed from the mandrel, being careful not to damage it.

Then we assemble the transformer completely, insert the magnetic circuit plates across the roof, from different sides of the frame. First, we assemble without plates - jumpers, it’s more convenient. After all the W-shaped plates are inserted, we insert the jumper plates.

By lightly tapping the ends with a hammer, we trim the plates on a level surface. Then the entire magnetic circuit must be tightened with stud bolts or crimped with corners with mounting holes.

Finally, we got to an interesting moment - the launch of our creation - a transformer on an W-shaped core into the electrical network.

To test the transformer, we connect the power cable with a plug (through a 1 ampere fuse) to the primary winding of the transformer.

Voltmeter alternating current you need to check the presence of voltage on the secondary winding of the transformer. It should be 35 - 37 volts.

If all the work is done correctly, then after 5-10 minutes of operation, the transformer should not heat up. After connecting a 36 volt light bulb, the voltage may drop to 33-35 volts, this is normal.

Winding a transformer with your own hands is not so much a complicated process as it is a lengthy process that requires constant concentration.

For those who begin such work for the first time, it can be difficult to figure out what material to use and how to check the finished device. Step-by-step instruction, presented below, will give beginners answers to all questions.

Before you begin directly winding, you need to stock up on all the necessary devices and tools to complete the work:

Types and methods, directions of winding of transformer windings are presented in the photo:

Insulation of winding layers

In some cases, it is necessary to insert spacers between the wires for insulation. Most often, capacitor or cable paper is used for this.

The middle of adjacent transformer windings should be more insulated. For insulating and leveling the surface under the next layer of winding you will need a special varnished cloth, which must be wrapped on both sides with paper. If there is no varnished fabric, then you can solve the problem using the same paper folded in several layers.

Paper strips for insulation should be 2-4 mm wider than the winding.

To check, first of all you need to determine the conclusions of all its windings. Useful tips For information on how to test a transformer for functionality with a multimeter, read the following article.

Algorithm of actions

1. Fix the wire with the coil in the winding device, and the transformer frame is in the winding device. Make rotations soft, moderate, without disruption.
2. Lower the wire from the reel onto the frame.
3. Leave between the table and the wire minimum 20 cm so that you can place your hand on the table and fix the wire. Also on the table should be all related materials: sandpaper, scissors, insulation paper, included soldering tool, pencil or pen.
4. With one hand, smoothly rotate the winding device, and with the other, fix the wire. It is necessary that the wire lies evenly, turn to turn.
5. Transformer insulate the frame, and pass the removed end of the wire through the frame hole and briefly fix it on the axis of the winding device.
6. Winding should begin without haste: you need to “get your hands on it” so that you can lay the turns next to each other.
7. It is necessary to ensure that the wire angle and tension are constant. You should not wind each subsequent layer “all the way”, because the wires can slip and fall into the frame “cheeks”.
8. Set the counting device (if any) to zero or count the turns carefully orally.
9. Glue the insulating material together or press it with a soft rubber ring.
10. Make each subsequent turn 1-2 turns thinner than the previous one.

To learn how to wind transformer coils with your own hands, watch this video:

Wire connection

If a break occurs during winding, then:

• thin wires (thinner than 0.1 mm) twist and brew;
• medium wire ends (less than 0.3 mm) should be freed from insulating material 1-1.5 cm, twist and solder;
• ends of thick wires (thicker than 0.3 mm) you need to clean it up a little and solder without twisting;
• Insulate the place of soldering (welding).

Important points

If a thin wire is used for winding, then the number of turns must exceed several thousand. The top of the winding must be protected with insulation paper or leatherette.

If the transformer is wrapped with a thick wire, then external protection is not required.

Trial

After winding is completed, it is necessary to test the transformer in action, to do this, connect its primary winding to the network.

To check the device for occurrence short circuits, the primary winding and the lamp should be connected in series to the power source.

Insulation reliability degree checked by alternate touching the lead-out end of the wire of each lead-out end of the network winding.

The transformer test should be carried out very carefully and carefully so as not to come under voltage from the step-up winding.

If strictly follow the instructions provided and do not neglect any of the points, then winding the transformer manually will not present any difficulties, and even a beginner can cope with it.

If you have a power transformer with a suitable(in this case S = 10.4 cm²) in terms of power cross-section of the core, but its secondary winding is designed for a different voltage, you can rewind the transformer.

In this case, you can not carry out such labor-intensive work as winding a multi-turn primary winding, but use a ready-made, old primary winding.

We determine the location of the primary and secondary windings on the frame. The primary winding is usually located on the frame closer to the core and is wound with a thin wire with a large number of turns.
Next you need to determine the number of turns per volt w for this steel core. You cannot use the value of the number of turns per volt previously calculated for the previous article.
Let's connect the transformer to a 220 volt network. Let's measure the voltage on all secondary windings. Let's choose the winding with the lowest voltage. For example, it will be equal to U = 30 volts. Let's mark its location on the frame.
Next, you need to disassemble the transformer, removing the core plates and freeing the frame. You need to rewind the transformer, wind the old secondary (or secondary, if there are several) windings and count the number of turns in the selected winding.
We leave only the primary winding and inter-winding insulation.
Let's say the number of turns in the selected winding will be n = 140.

Then the number of turns per volt w for this transformer will be:

w = n: U = 140: 30 = 4.67 turns.

If there is no secondary winding at all, or there is no way to calculate it, we will proceed in a different way.
We wind 100 turns of insulated wire of any diameter over the primary winding - this is the “measuring” winding.
Let's assemble the transformer again, connect it to a 220-volt network and measure the voltage on the “measuring” winding with a voltmeter. Let's say it will be 21.5 volts.

Let's calculate the number of turns per 1 volt for this transformer:
w = n: U = 100: 21.5 = 4.65 turns.
Then the number of turns in the new 36 volt secondary winding will be:

U_2 = 36 4.65 = 167.8 turns. Let's round up to 170 turns.
The “measuring” winding should be removed and wound with a wire of the appropriate diameter.