The other one I read is probably controversial (for some), if sharp bends are placed in the wire, think.... unwind a transformer and notice any kinks in the wire.
Next... some suggest the sound of OPT's improve with use.
So was the sound degraded from the kinks, and after use the wire (through cycling through use and temp increase and cooling) slowly normalises the kinks. (or maybe only partly so)
.
Next... some suggest the sound of OPT's improve with use.
So was the sound degraded from the kinks, and after use the wire (through cycling through use and temp increase and cooling) slowly normalises the kinks. (or maybe only partly so)
.
The internet is a great source of information - it's also a great source of disinformation - while I don't believe there is any scientific evidence to support what you say regarding the kinks in the wire and the 'improvement' in sound after a period of use/burn-in, there are those that claim to hear a difference and you can't really refute those claims as you don't hear what they hear. For me kinks are not good - they disrupt even layers.
Some claim that paper insulation in vintage output transformers imparts a magical quality to the sound - there may be some scientific truth in that as the interwinding capacitances will be different, however - I wound 2 identical transformers - one had a plastic bobbin, nomex and mylar insulation and the other wound the old way, paper former and insulation, layer of paper between each copper layer etc. Both of them were auditioned by a guitarist in a Fender Deluxe amp build, he could not detect any difference in the sound between each transformer - he could though hear the difference between a Brimar, and Miniwatt preamp tube, I myself heard no difference with the transformers or with the preamp tubes.
Perhaps the best way to resolve for your self is to try it - wind 2 identical transformers - put some kinks in one if them and then with a stereo amp and mono signal AB the channels.
Some claim that paper insulation in vintage output transformers imparts a magical quality to the sound - there may be some scientific truth in that as the interwinding capacitances will be different, however - I wound 2 identical transformers - one had a plastic bobbin, nomex and mylar insulation and the other wound the old way, paper former and insulation, layer of paper between each copper layer etc. Both of them were auditioned by a guitarist in a Fender Deluxe amp build, he could not detect any difference in the sound between each transformer - he could though hear the difference between a Brimar, and Miniwatt preamp tube, I myself heard no difference with the transformers or with the preamp tubes.
Perhaps the best way to resolve for your self is to try it - wind 2 identical transformers - put some kinks in one if them and then with a stereo amp and mono signal AB the channels.
Sorry, I don’t have more pictures. When I’m working, I stay highly focused to avoid making mistakes and in any case, uglier than the ones Alexander makes.
Both have a similar construction double C-core 3100g and a single C-core 2200g. The primary is 12k, designed for a 211 tube. Well, I don’t like to talk about numbers, I prefer to talk about sound, which is the real target.
I wind with all the secondaries in parallel, but the question is if has anyone wound the secondary in series with a similar diameter as the parallel configuration, using Litz wire?
Both have a similar construction double C-core 3100g and a single C-core 2200g. The primary is 12k, designed for a 211 tube. Well, I don’t like to talk about numbers, I prefer to talk about sound, which is the real target.
I wind with all the secondaries in parallel, but the question is if has anyone wound the secondary in series with a similar diameter as the parallel configuration, using Litz wire?
It takes an ample amount of practice, years, to avoid basic mistakes when winding. It takes a lot of concentration. I still have no idea why, probably because of the complexity of some transformers.
You must be careful with series connected secondaries. I'm preparing tutorial projects to shed more light into this, unfortunately, we're living into very busy times.
Beautiful work!
The overall sound of transformers has to do mostly with the materials used to build it. And the weight. But the same goes for the amplifier materials as well. It can be measured, but acoustically, via microphones/accelerometers.
The audible differences from copper wire come mostly from it's hardness. Harder copper wire sounds more "in your face", while softer wire has a laid back soundstage. However, you cannot anneal the wire without screwing its enamel insulation. There is a theory that copper self-anneals to a little extent while resting, but this could take years / decades.
Recent pictures of a large 9.1cm2 core interstage 1:1 to be driven by a 845. And other stuff..
You must be careful with series connected secondaries. I'm preparing tutorial projects to shed more light into this, unfortunately, we're living into very busy times.
Beautiful work!
The overall sound of transformers has to do mostly with the materials used to build it. And the weight. But the same goes for the amplifier materials as well. It can be measured, but acoustically, via microphones/accelerometers.
The audible differences from copper wire come mostly from it's hardness. Harder copper wire sounds more "in your face", while softer wire has a laid back soundstage. However, you cannot anneal the wire without screwing its enamel insulation. There is a theory that copper self-anneals to a little extent while resting, but this could take years / decades.
Recent pictures of a large 9.1cm2 core interstage 1:1 to be driven by a 845. And other stuff..
Attachments
Last edited:
May be a bit off-topic, but anyway... Since its a transformer design thread in some degree. I opted to run this technical question on a DeepSeek AI.
I have transformer with sectioned winding to reduce leakage capacitance. Do reverse winding direction reduces stray capacitance between primary and secondary?
DeepSeek AI writes a quite lengthy lengthy argumentative essay, and then short summary. Thus, I will split this post into 2.
------------
Okay, so I have this transformer with sectioned windings to reduce leakage capacitance. The user is asking if reversing the winding direction of the primary or secondary would reduce the stray capacitance between them. Hmm, let me think.
First, I remember that in transformers, capacitance between windings is a big deal for high-frequency performance. Leakage or stray capacitance can cause issues like noise coupling or affect the frequency response. Sectioned windings, like interleaving or separating primary and secondary into sections, are methods to reduce this capacitance. But the question is about winding direction.
So, winding direction—how does that affect capacitance? Capacitance between two conductors depends on the voltage difference and the physical overlap. If the windings are in opposite directions, maybe the voltage gradient along the windings could influence the overall capacitance. Let me visualize a typical transformer winding. Suppose the primary is wound clockwise from start to finish, and the secondary is also wound clockwise. The voltage increases along the winding, so each turn of the primary has a certain potential relative to the secondary. If the secondary is wound counterclockwise, maybe the potentials at adjacent points are closer, reducing the voltage difference between nearby turns. Wait, but if the primary and secondary are wound in opposite directions, the start of the primary might be near the finish of the secondary. Depending on the phase relationship, this could either increase or decrease the average voltage difference between adjacent layers.
Alternatively, if the windings are in the same direction, the voltage gradients might be more aligned. For example, the highest voltage on the primary is next to the highest voltage on the secondary if they're both wound the same way. But if wound in opposite directions, the high voltage of the primary could be next to the low voltage of the secondary, leading to a larger voltage difference, which might actually increase the capacitance. Because capacitance is proportional to the voltage difference. Wait, but capacitance is also about the electric field between conductors. If the voltage difference is higher, the electric field is stronger, leading to more stored charge, hence higher capacitance. But if the voltage difference is lower, then the capacitance effect is less.
But in a transformer, the primary and secondary are magnetically coupled but electrically isolated. The stray capacitance is between the windings, which are separated by insulation. The physical distance and the dielectric material between them also affect capacitance. So winding direction might influence how the turns are positioned relative to each other. For example, if primary and secondary are wound in opposite directions, the physical overlap might have sections where the voltage difference is lower. Suppose the primary starts at one end and goes left to right, increasing voltage, while the secondary starts at the right and goes left, decreasing voltage. Then, adjacent layers might have a smaller voltage difference compared to if both windings go in the same direction. So maybe that reduces the overall capacitance.
But I'm not entirely sure. Let me check some references. From what I recall, in some transformer designs, reversing the winding direction can help balance the capacitive coupling. For instance, in a flyback transformer, interleaving primary and secondary with opposing windings might reduce the effective capacitance by canceling some of the displacement currents. Also, the concept of "non-inductive winding" where adjacent layers are wound in opposite directions to minimize inductance, but that's a different parameter.
Wait, but the user is talking about sectioned windings. So maybe the primary is divided into sections, and the secondary is also sectioned, with each section interleaved. If each section is wound in the opposite direction, perhaps the cumulative effect reduces the overall capacitance. Alternatively, if you reverse the direction of one winding relative to the other, the proximity of high and low voltage points could average out the voltage gradient, leading to lower effective capacitance.
But I should think about the basics. Capacitance between two conductors is C = εA/d, where A is the area of overlap, d the distance, and ε the permittivity. If reversing the winding direction changes the overlap area or the effective distance, then capacitance changes. If winding in opposite directions reduces the overlap area between high-voltage sections, then capacitance would decrease. For example, if the primary's high-voltage end is near the secondary's low-voltage end due to opposite winding directions, the voltage difference between adjacent sections is minimized. But the area of overlap might still be the same, just with different voltage gradients.
Alternatively, the average voltage difference across the overlapping sections might be lower if the windings are in opposite directions, leading to a lower overall capacitive coupling. Let's say the primary has a voltage that increases from 0 to V along its length, and the secondary also increases from 0 to V if wound in the same direction. Then adjacent layers have a voltage difference that's a fraction of V. But if the secondary is wound in reverse, starting at V and going to 0, then the voltage difference between primary and secondary at each point could be V - (V - x) = x, which might average out to the same. Wait, maybe that's not the case. If the primary is wound 0 to V left to right, and secondary is wound V to 0 left to right (opposite direction), then the voltage difference at each point is |V_primary - V_secondary|. At the left end, primary is 0, secondary is V, so difference V. At the right end, primary is V, secondary is 0, difference V. In the middle, maybe difference is |V/2 - V/2| = 0. So the average voltage difference is V/2. If they were wound in the same direction, then the primary and secondary both go 0 to V left to right, so at each point, the voltage difference is the same, say if the windings are isolated. Wait, but in a transformer, the secondary voltage is determined by the turns ratio. Let's assume it's a 1:1 transformer for simplicity. Then primary and secondary have the same voltage. If they're in phase, then the voltage difference between adjacent turns is zero. But in reality, the secondary voltage is induced and depends on the load. Wait, maybe this is getting too complicated.
Alternatively, considering that the capacitance is related to the integral of the electric field between the windings. If the windings are in opposite directions, the electric field variation along the length might lead to some cancellation. But I'm not sure. Maybe another approach: think about how the displacement currents flow due to the capacitance. If the windings are in the same direction, displacement currents from primary to secondary would add up, but in opposite directions, they might partially cancel. For example, if the primary starts at 0 and goes to V, and the secondary starts at V and goes to 0, the displacement current from the primary's start (0V) to secondary's start (V) is high, but from primary's end (V) to secondary's end (0V) is also high. Wait, but that might just be two high displacement currents at both ends, leading to higher overall capacitance. Hmm, conflicting thoughts.
Alternatively, maybe the key factor is the physical arrangement. Sectioned windings already reduce capacitance by increasing the distance between primary and secondary sections. Winding direction might influence how the electric fields interact. If the windings are in opposite directions, the potentials at adjacent points might alternate, reducing the effective capacitance. For example, in a layer wound transformer, if primary and secondary layers are alternated but wound in opposite directions, the voltage difference between adjacent layers could be minimized. But if they are wound in the same direction, each layer's voltage increases, leading to higher potential differences between adjacent layers.
Alternatively, maybe the winding direction doesn't matter as much as the physical separation and the insulation between windings. But the user is specifically asking if reversing the winding direction helps. From some sources I recall, in high-voltage transformers, sometimes the windings are arranged in a way that adjacent layers have opposing voltage gradients to minimize the capacitance. For example, if the primary is wound clockwise and the secondary counter-clockwise, the voltage gradients are in opposite directions, leading to a lower overall capacitance because the average voltage difference between adjacent conductors is reduced.
But I need to verify this. Suppose you have two parallel conductors, one with voltage increasing from left to right, and the other decreasing from left to right. The voltage difference between them at any point is the sum of their individual voltages at that position. If the primary goes 0 to V and the secondary goes V to 0 along the same axis, then at each point x, the primary voltage is V*(x/L) and the secondary is V*(1 - x/L), so the difference is V*(2x/L -1). The average absolute difference would be V/2, same as if both were going 0 to V. Wait, but that's if they're overlapping. Alternatively, if they are interleaved in sections, reversing the winding direction might cause adjacent sections to have lower voltage differences. For example, if the primary has sections P1, P2, P3, each wound in a direction, and secondary sections S1, S2, S3 wound in the opposite direction, then adjacent primary and secondary sections might have lower voltage differences. Alternatively, if the primary is wound in one direction and the secondary in the opposite, the start of the primary is near the end of the secondary, which might have a higher voltage difference, but maybe the overall integral over the entire winding averages out to a lower capacitance. Hmm. This is getting a bit unclear.
I think the key point is that reversing the winding direction can affect the distribution of voltage gradients between the primary and secondary, potentially reducing the average voltage difference across the windings, thereby reducing the stray capacitance. However, this depends on how the windings are physically arranged. If the windings are sectioned and interleaved, reversing the direction might lead to adjacent sections having complementary voltage gradients, which could lower the overall capacitance. But if the windings are not interleaved and are simply placed opposite, the effect might be minimal.
Another aspect is the proximity effect. If windings are in opposite directions, the current flows in opposite directions, which might influence the magnetic fields and proximity effects, but that's more about inductance and losses rather than capacitance.
In some transformer designs, such as those used in switching power supplies, minimizing interwinding capacitance is critical for reducing EMI and improving high-frequency performance. Techniques like using a Faraday shield between windings, increasing insulation thickness, or using sectional windings are common. Winding direction might be a secondary factor but could contribute if it leads to a more optimal voltage gradient distribution.
So, putting it all together, yes, reversing the winding direction can reduce stray capacitance between primary and secondary, especially if it results in adjacent sections having lower voltage differences. However, the effectiveness depends on the specific winding arrangement, such as how the sections are interleaved and the physical layout. It's not a universal solution but can be beneficial in certain designs when combined with other techniques like sectioning.
I have transformer with sectioned winding to reduce leakage capacitance. Do reverse winding direction reduces stray capacitance between primary and secondary?
DeepSeek AI writes a quite lengthy lengthy argumentative essay, and then short summary. Thus, I will split this post into 2.
------------
Okay, so I have this transformer with sectioned windings to reduce leakage capacitance. The user is asking if reversing the winding direction of the primary or secondary would reduce the stray capacitance between them. Hmm, let me think.
First, I remember that in transformers, capacitance between windings is a big deal for high-frequency performance. Leakage or stray capacitance can cause issues like noise coupling or affect the frequency response. Sectioned windings, like interleaving or separating primary and secondary into sections, are methods to reduce this capacitance. But the question is about winding direction.
So, winding direction—how does that affect capacitance? Capacitance between two conductors depends on the voltage difference and the physical overlap. If the windings are in opposite directions, maybe the voltage gradient along the windings could influence the overall capacitance. Let me visualize a typical transformer winding. Suppose the primary is wound clockwise from start to finish, and the secondary is also wound clockwise. The voltage increases along the winding, so each turn of the primary has a certain potential relative to the secondary. If the secondary is wound counterclockwise, maybe the potentials at adjacent points are closer, reducing the voltage difference between nearby turns. Wait, but if the primary and secondary are wound in opposite directions, the start of the primary might be near the finish of the secondary. Depending on the phase relationship, this could either increase or decrease the average voltage difference between adjacent layers.
Alternatively, if the windings are in the same direction, the voltage gradients might be more aligned. For example, the highest voltage on the primary is next to the highest voltage on the secondary if they're both wound the same way. But if wound in opposite directions, the high voltage of the primary could be next to the low voltage of the secondary, leading to a larger voltage difference, which might actually increase the capacitance. Because capacitance is proportional to the voltage difference. Wait, but capacitance is also about the electric field between conductors. If the voltage difference is higher, the electric field is stronger, leading to more stored charge, hence higher capacitance. But if the voltage difference is lower, then the capacitance effect is less.
But in a transformer, the primary and secondary are magnetically coupled but electrically isolated. The stray capacitance is between the windings, which are separated by insulation. The physical distance and the dielectric material between them also affect capacitance. So winding direction might influence how the turns are positioned relative to each other. For example, if primary and secondary are wound in opposite directions, the physical overlap might have sections where the voltage difference is lower. Suppose the primary starts at one end and goes left to right, increasing voltage, while the secondary starts at the right and goes left, decreasing voltage. Then, adjacent layers might have a smaller voltage difference compared to if both windings go in the same direction. So maybe that reduces the overall capacitance.
But I'm not entirely sure. Let me check some references. From what I recall, in some transformer designs, reversing the winding direction can help balance the capacitive coupling. For instance, in a flyback transformer, interleaving primary and secondary with opposing windings might reduce the effective capacitance by canceling some of the displacement currents. Also, the concept of "non-inductive winding" where adjacent layers are wound in opposite directions to minimize inductance, but that's a different parameter.
Wait, but the user is talking about sectioned windings. So maybe the primary is divided into sections, and the secondary is also sectioned, with each section interleaved. If each section is wound in the opposite direction, perhaps the cumulative effect reduces the overall capacitance. Alternatively, if you reverse the direction of one winding relative to the other, the proximity of high and low voltage points could average out the voltage gradient, leading to lower effective capacitance.
But I should think about the basics. Capacitance between two conductors is C = εA/d, where A is the area of overlap, d the distance, and ε the permittivity. If reversing the winding direction changes the overlap area or the effective distance, then capacitance changes. If winding in opposite directions reduces the overlap area between high-voltage sections, then capacitance would decrease. For example, if the primary's high-voltage end is near the secondary's low-voltage end due to opposite winding directions, the voltage difference between adjacent sections is minimized. But the area of overlap might still be the same, just with different voltage gradients.
Alternatively, the average voltage difference across the overlapping sections might be lower if the windings are in opposite directions, leading to a lower overall capacitive coupling. Let's say the primary has a voltage that increases from 0 to V along its length, and the secondary also increases from 0 to V if wound in the same direction. Then adjacent layers have a voltage difference that's a fraction of V. But if the secondary is wound in reverse, starting at V and going to 0, then the voltage difference between primary and secondary at each point could be V - (V - x) = x, which might average out to the same. Wait, maybe that's not the case. If the primary is wound 0 to V left to right, and secondary is wound V to 0 left to right (opposite direction), then the voltage difference at each point is |V_primary - V_secondary|. At the left end, primary is 0, secondary is V, so difference V. At the right end, primary is V, secondary is 0, difference V. In the middle, maybe difference is |V/2 - V/2| = 0. So the average voltage difference is V/2. If they were wound in the same direction, then the primary and secondary both go 0 to V left to right, so at each point, the voltage difference is the same, say if the windings are isolated. Wait, but in a transformer, the secondary voltage is determined by the turns ratio. Let's assume it's a 1:1 transformer for simplicity. Then primary and secondary have the same voltage. If they're in phase, then the voltage difference between adjacent turns is zero. But in reality, the secondary voltage is induced and depends on the load. Wait, maybe this is getting too complicated.
Alternatively, considering that the capacitance is related to the integral of the electric field between the windings. If the windings are in opposite directions, the electric field variation along the length might lead to some cancellation. But I'm not sure. Maybe another approach: think about how the displacement currents flow due to the capacitance. If the windings are in the same direction, displacement currents from primary to secondary would add up, but in opposite directions, they might partially cancel. For example, if the primary starts at 0 and goes to V, and the secondary starts at V and goes to 0, the displacement current from the primary's start (0V) to secondary's start (V) is high, but from primary's end (V) to secondary's end (0V) is also high. Wait, but that might just be two high displacement currents at both ends, leading to higher overall capacitance. Hmm, conflicting thoughts.
Alternatively, maybe the key factor is the physical arrangement. Sectioned windings already reduce capacitance by increasing the distance between primary and secondary sections. Winding direction might influence how the electric fields interact. If the windings are in opposite directions, the potentials at adjacent points might alternate, reducing the effective capacitance. For example, in a layer wound transformer, if primary and secondary layers are alternated but wound in opposite directions, the voltage difference between adjacent layers could be minimized. But if they are wound in the same direction, each layer's voltage increases, leading to higher potential differences between adjacent layers.
Alternatively, maybe the winding direction doesn't matter as much as the physical separation and the insulation between windings. But the user is specifically asking if reversing the winding direction helps. From some sources I recall, in high-voltage transformers, sometimes the windings are arranged in a way that adjacent layers have opposing voltage gradients to minimize the capacitance. For example, if the primary is wound clockwise and the secondary counter-clockwise, the voltage gradients are in opposite directions, leading to a lower overall capacitance because the average voltage difference between adjacent conductors is reduced.
But I need to verify this. Suppose you have two parallel conductors, one with voltage increasing from left to right, and the other decreasing from left to right. The voltage difference between them at any point is the sum of their individual voltages at that position. If the primary goes 0 to V and the secondary goes V to 0 along the same axis, then at each point x, the primary voltage is V*(x/L) and the secondary is V*(1 - x/L), so the difference is V*(2x/L -1). The average absolute difference would be V/2, same as if both were going 0 to V. Wait, but that's if they're overlapping. Alternatively, if they are interleaved in sections, reversing the winding direction might cause adjacent sections to have lower voltage differences. For example, if the primary has sections P1, P2, P3, each wound in a direction, and secondary sections S1, S2, S3 wound in the opposite direction, then adjacent primary and secondary sections might have lower voltage differences. Alternatively, if the primary is wound in one direction and the secondary in the opposite, the start of the primary is near the end of the secondary, which might have a higher voltage difference, but maybe the overall integral over the entire winding averages out to a lower capacitance. Hmm. This is getting a bit unclear.
I think the key point is that reversing the winding direction can affect the distribution of voltage gradients between the primary and secondary, potentially reducing the average voltage difference across the windings, thereby reducing the stray capacitance. However, this depends on how the windings are physically arranged. If the windings are sectioned and interleaved, reversing the direction might lead to adjacent sections having complementary voltage gradients, which could lower the overall capacitance. But if the windings are not interleaved and are simply placed opposite, the effect might be minimal.
Another aspect is the proximity effect. If windings are in opposite directions, the current flows in opposite directions, which might influence the magnetic fields and proximity effects, but that's more about inductance and losses rather than capacitance.
In some transformer designs, such as those used in switching power supplies, minimizing interwinding capacitance is critical for reducing EMI and improving high-frequency performance. Techniques like using a Faraday shield between windings, increasing insulation thickness, or using sectional windings are common. Winding direction might be a secondary factor but could contribute if it leads to a more optimal voltage gradient distribution.
So, putting it all together, yes, reversing the winding direction can reduce stray capacitance between primary and secondary, especially if it results in adjacent sections having lower voltage differences. However, the effectiveness depends on the specific winding arrangement, such as how the sections are interleaved and the physical layout. It's not a universal solution but can be beneficial in certain designs when combined with other techniques like sectioning.
Reversing the winding direction of the primary or secondary in a transformer with sectioned windings can indeed help reduce stray capacitance between the windings, depending on the design specifics. Here's a structured summary of the key points:
Reversing winding direction can reduce stray capacitance in sectioned windings by strategically aligning voltage gradients to minimize average voltage differences. However, its efficacy is maximized when paired with interleaving and proper physical design. It is not a standalone solution but a complementary technique in transformer optimization.
- Capacitance Fundamentals:
Stray capacitance (CC) between windings depends on the voltage difference (VV), overlapping area (AA), distance (dd), and dielectric properties (ϵϵ). Reducing the average voltage difference between adjacent sections or minimizing the effective overlap area lowers CC. - Winding Direction Impact:
- If windings are reversed (e.g., primary wound clockwise, secondary counterclockwise), the voltage gradients along the windings may oppose each other. This can reduce the average voltage difference between adjacent layers/sections, especially in interleaved designs.
- For example, in a 1:1 transformer, reversing the secondary winding direction could place the primary's low-voltage end near the secondary's high-voltage end, and vice versa. While local voltage differences at the ends remain high, the integrated voltage gradient across overlapping sections may average out to a lower effective CC.
- Sectioned/Interleaved Windings:
- Sectioning already reduces capacitance by increasing physical separation. Reversing winding direction in this context can further optimize the voltage distribution.
- Adjacent sections with opposing voltage gradients (e.g., primary ascending, secondary descending) may experience partial cancellation of displacement currents, reducing net capacitive coupling.
- Practical Considerations:
- Effectiveness depends on the physical layout (e.g., layer arrangement, insulation thickness) and whether sections are interleaved.
- In high-frequency transformers (e.g., flyback, RF), this technique is often combined with shielding, increased insulation, or fractional winding strategies for optimal results.
Reversing winding direction can reduce stray capacitance in sectioned windings by strategically aligning voltage gradients to minimize average voltage differences. However, its efficacy is maximized when paired with interleaving and proper physical design. It is not a standalone solution but a complementary technique in transformer optimization.
IMHO, switching from forth-back, back-forth winding to Z-winding changes the lumped capacitance multiplier, of, for example, a primary section or a choke, from a factor of 1,33 to 1,00.
Not really worth it, especially considering an OPT, where primary lumped capacitance is negligible compared to primary/secondary capacitance.
Not really worth it, especially considering an OPT, where primary lumped capacitance is negligible compared to primary/secondary capacitance.
In SE output XFRM primary wire that goes to anode is on the first layer to the core of the XFRM. That reduces unwanted ractive elements. And Scondary can be in same or opposite dirr.
.
Direction of winding if You want can be as water draining into the bath.
So on the north hemisphere is one direction on the south hemisphere is another...
One case making issue that is ecuador - water does not make circular shape but outgoing in the straight line... That is connection to the planet magenetic field...
Ha ha.
.
.
Direction of winding if You want can be as water draining into the bath.
So on the north hemisphere is one direction on the south hemisphere is another...
One case making issue that is ecuador - water does not make circular shape but outgoing in the straight line... That is connection to the planet magenetic field...
Ha ha.
.
Hi everyone
To put it simply
In the case of SE Tramy, it is good to wind the last outer section of the primary in the opposite direction to the other primary sections, then, you need to connect the HT at the beginning of this section.
In this way, the less active side layer of this section is next to the secondary =>less capacity
Generaly secondary section is winding in one layer (many in //) it is easy to get the good arrangement to recpect phase antiphase 2 stages or 3 stages for exemple
To put it simply
In the case of SE Tramy, it is good to wind the last outer section of the primary in the opposite direction to the other primary sections, then, you need to connect the HT at the beginning of this section.
In this way, the less active side layer of this section is next to the secondary =>less capacity
Generaly secondary section is winding in one layer (many in //) it is easy to get the good arrangement to recpect phase antiphase 2 stages or 3 stages for exemple
The same principle applies for PP trannies. Besides, compared to SE, they're easier in terms of capacitance. Not only you get capacitance reduction, but also a chance for symmetry between the capacitive values, which can be detrimental. SE transformers are harder, not only because of magnetizing DC currents, but require more advanced interleaving. SE trannies cannot have equal voltage gradients compared to PP. I'd say intuitively, SE xformers are 3x times harder to get right.
Frankly I've learned about output transformers here from you guys more than any book on this subject.
Here is some pictures from my latest build; 2.5K to 8 ohms 10W SE OPT:
Here is some pictures from my latest build; 2.5K to 8 ohms 10W SE OPT:
Recently rewound a Lux-kit OY15 -5 output transformer - wound on a tube/former rather than a bobbin, can't say it all went smoothly, a mistake on my part and I had to unwind and start again, but I managed in the end, was very hot hot here, over 36 deg C in the workshop, tape adhesive tended to let go - probably from the heat - this caused some problems securing taps and windings. This transformer has 5 primary and 4 parallel connected secondaries, each secondary consists of 2 layers, one layer is just straight forward full layer of 75 turns, the other layer has 4 taps bought out to the edge, this makes a total of 10 leads bought out to the edge - multiply that by 4 and one ends up with 40 leads to keep track of for the final connection.
Amazing work. Having wound a few tfmr's I know how hard it is, especially the ends of a layer. To stop the end turns of one layer on top of another layer from slipping off the insulation onto the one underneath is very tricky. I'm getting better at it but couldn't wind the OPT above in post #354.
I'm just about to wind my first proper OPT. It's taken me weeks of calculating having consulted several books and online sources but I've only skimmed the surface. I've just re-read most of this thread & am now worrying about my layering scheme & the inter connections of the various windings.
I'm after winding a PP 22w 5k pri, 8r secondary for 2C34 triodes in parallel, I followed Pat Turner for the most part. Proposed layering is 2p-s-4p-s-4p-s-2p. This is using E114 GOSS laminations and a core area of 1900mm sq. This is a bit big but what I have. Being bigger than needed though means I have a large window so I thought of using larger than needed gauge wire, we'll see. Any tips before I start winding?
Andy.
I'm just about to wind my first proper OPT. It's taken me weeks of calculating having consulted several books and online sources but I've only skimmed the surface. I've just re-read most of this thread & am now worrying about my layering scheme & the inter connections of the various windings.
I'm after winding a PP 22w 5k pri, 8r secondary for 2C34 triodes in parallel, I followed Pat Turner for the most part. Proposed layering is 2p-s-4p-s-4p-s-2p. This is using E114 GOSS laminations and a core area of 1900mm sq. This is a bit big but what I have. Being bigger than needed though means I have a large window so I thought of using larger than needed gauge wire, we'll see. Any tips before I start winding?
Andy.
I use margins for that purpose; margin is a space from the bobbin end which wire is not wound. I cut the insulation paper exactly to the window size, an then use a credit card with 1 mm diameter between the bobbin flange and first two or three turns and then start the motor for the rest of turns:
Another method for this is to wind the whole window space without any margin, but cut the insulation paper a little longer than the window. This method is described by Turner in the "Practical Transformer Winding" article. Some people also make little cuts at the paper ends to prevent paper from foldindg in this method.
If you don't have a herniated disc like me, using a bigger core is always better for output transformers; It lets you to have less turns so the leakage inductance would be lower and high frequency response should be better. Also as you mentioned it lets you to use thicker wire so insertion loss will be lower. You can use my spreadsheet for your design, it is based on Turner method and lets you to test your design with different wire sizes and different cores to get the best results in terms of high and low frequency response and insertion loss:
I don´t have any drawings on the descrption described, but some years ago I wound a couple of high power o/p push-pull transformers with multiple equal winding secondaries, that is combination for 1-2-4-8-15 ohms and including sectionalized UL primaries, a 19 sectioned design close to the design in the Radiotron handbook. Do the theory and one will find the integer for the 8 ohm turns out odd as it is impossible to get the winding integer to suit. So a twin parallel and two series arrangement is close. Many expert winding engineers fluent to the "winding autobahn" manage to create the internal links to other sectionalized windings, something I had to do externally. It is a job done without interruption.
I followed Turners ways, but in the end settled for Partridge/Williamson configuration. The live performance test with a P-P amp at low power didn´t quite follow expectations as those who are after a good squarewave, will have to do as in my case, re-arrange the 4 equal secondary windings for the best square performance, that is keeping the primaries unchanged. Clearly it is impossible to get the leakage parasitics equal throughout the winding process on the bobbin, but I never take for granted the manufacturers permutation for secondaries. All they do is a 50Hz throughput to check ratios and voltages and the rest theory and CAD to predict.
Sowter is one that provides sectionalized 4 secondary configurations; it is up to us to figure out the best squarewave performance. Only my view, sonically speaking, is it really worth 19 sectionalized windings ? (done twice only and no more done!)
I have heard through many 5 sectioned designs and surprised to their performance vs. cost. 3 sectioned designs tend towards MI quality.
Bench Baron
I followed Turners ways, but in the end settled for Partridge/Williamson configuration. The live performance test with a P-P amp at low power didn´t quite follow expectations as those who are after a good squarewave, will have to do as in my case, re-arrange the 4 equal secondary windings for the best square performance, that is keeping the primaries unchanged. Clearly it is impossible to get the leakage parasitics equal throughout the winding process on the bobbin, but I never take for granted the manufacturers permutation for secondaries. All they do is a 50Hz throughput to check ratios and voltages and the rest theory and CAD to predict.
Sowter is one that provides sectionalized 4 secondary configurations; it is up to us to figure out the best squarewave performance. Only my view, sonically speaking, is it really worth 19 sectionalized windings ? (done twice only and no more done!)
I have heard through many 5 sectioned designs and surprised to their performance vs. cost. 3 sectioned designs tend towards MI quality.
Bench Baron
A nice looking square wave response cannot be summarized by the amount of interleavings, many aspects are at play.
Thanks Sajad, I have played with your spreadsheet a few times & have found my calculations pretty near the results of your spreadsheet. That's a good trick with the credit card. where I've had a gap at the ends of layers I've also use flat polyester ribbon of approx 3mm by 2mm to fill the gap.
After re-reading this thread I'm a bit confused about connecting all the various primary's & secondary's. 50AE mentioned several times about not connecting the secs in series as this may increase inter winding C.
Andy.
After re-reading this thread I'm a bit confused about connecting all the various primary's & secondary's. 50AE mentioned several times about not connecting the secs in series as this may increase inter winding C.
Andy.
Hi @sajadho1988 I look at the Your spread sheet for SE OT
If You allowed please,
You starting calculations with min primary inductance as
Z primary / (2 Pi f)
it is wrong...
You completly neglecting the tube parameters?
Actually the equivalent resistance of the generator, which is tube with load,
is close to parallel value of
Ri = internal resistance of the tube
Rload = Z primary
in parallel
So more accurate and very different value is
[(Ri x Rload) / (Ri + Rload)] / (2 Pi f)
.
Other thing considering tube as factor is Voltage p-p to Load (primary OT)
It is generally maximum p-p voltage (2 x Ug) x A
where
A is amplification ratio of the tube in circuit (not amplification factor of the tube mju)
without these datas You simply can not determine max saturation for wanted Tesla
...
If You allowed please,
You starting calculations with min primary inductance as
Z primary / (2 Pi f)
it is wrong...
You completly neglecting the tube parameters?
Actually the equivalent resistance of the generator, which is tube with load,
is close to parallel value of
Ri = internal resistance of the tube
Rload = Z primary
in parallel
So more accurate and very different value is
[(Ri x Rload) / (Ri + Rload)] / (2 Pi f)
.
Other thing considering tube as factor is Voltage p-p to Load (primary OT)
It is generally maximum p-p voltage (2 x Ug) x A
where
A is amplification ratio of the tube in circuit (not amplification factor of the tube mju)
without these datas You simply can not determine max saturation for wanted Tesla
...