I'd be curious to see the higher frequency response of these, out into the ultrasonic so that one can tell what is really going on... I had a friend who got a pair of toroidal tube PP output iron and the HF response was abysmal. This was a highly touted brand too... so I am skeptical about any general purpose power iron being used at audio freqs.
Yes, I will check that as well, bear.
I know my own cores have an astronomical bandwidth with resonance bump at about 650Khz as tested here,
http://www.diyaudio.com/forums/planars-exotics/161485-step-up-transformer-design.html#post2093314
I see signals from the local AM radio stations when I have my scope connected to them.
Sometimes it makes getting precise measuring with the scope set to it's highest sensitivity a hassle but I manage.
jer
I know my own cores have an astronomical bandwidth with resonance bump at about 650Khz as tested here,
http://www.diyaudio.com/forums/planars-exotics/161485-step-up-transformer-design.html#post2093314
I see signals from the local AM radio stations when I have my scope connected to them.
Sometimes it makes getting precise measuring with the scope set to it's highest sensitivity a hassle but I manage.
jer
Hello CharlieM,
I would be happy to test/measure the AS-1206 and compare results with the AN-506.
Let me know if interested, and I'll IM you a shipping address.
Thanks for the offer but Jer replied first and I've already sent a 1206 transformer to him. It will be interesting to see the results.
Bear,
For that test all I had was my signal generator to use hooked up to a winding and my scope connected to another.
I don't remember off hand what windings I had used at the time but I think it was the two 120v windings so the signal was straight trough the transformer and isolated the to devices.
In the later tests were it should it to be starting to drop at about 18khz or so was actually later found out to be the response of the amplifier that I was using to drive it at the time.
Since then I have retrieved my bigger and better amps from Florida and was at the time I had not been on in a while.
They also have flatter low frequency response as well,
Although this is not as important as the peak saturation voltage as that is all that is needed to be known for that range.
jer
For that test all I had was my signal generator to use hooked up to a winding and my scope connected to another.
I don't remember off hand what windings I had used at the time but I think it was the two 120v windings so the signal was straight trough the transformer and isolated the to devices.
In the later tests were it should it to be starting to drop at about 18khz or so was actually later found out to be the response of the amplifier that I was using to drive it at the time.
Since then I have retrieved my bigger and better amps from Florida and was at the time I had not been on in a while.
They also have flatter low frequency response as well,
Although this is not as important as the peak saturation voltage as that is all that is needed to be known for that range.
jer
Last edited:
Here are the original scope picture of the frequency response of my cores.
It is very flat and again the roll off at either end was found to be that of the amp itself.
http://www.diyaudio.com/forums/planars-exotics/158115-material-esl-3.html#post2085791
I don't know if I had taken any pictures of the square wave tests but they were the cleanest square waves that I have ever seen coming out of a piece iron up to this point up past 50khz to 100Khz or so.
If I do find any I will repost the links and I will also include them in the next tests as this give an excellent example of its High Frequency capability's.
jer
It is very flat and again the roll off at either end was found to be that of the amp itself.
http://www.diyaudio.com/forums/planars-exotics/158115-material-esl-3.html#post2085791
I don't know if I had taken any pictures of the square wave tests but they were the cleanest square waves that I have ever seen coming out of a piece iron up to this point up past 50khz to 100Khz or so.
If I do find any I will repost the links and I will also include them in the next tests as this give an excellent example of its High Frequency capability's.
jer
See, that just doesn't make sense because it takes a lot of effort to make a standard audio transformer have a nominally flat response to 60-75kHz. To get one to go all the way to 100kHz or above is extraordinary and doesn't just randomly happen. Certainly not with your average toroidal power transformer. So, something is going on here that at least isn't exactly clear to me, i've tried to read through some of the linked posts but can't yet determine what was done that would account for the apparently extraordinary results.
Just to clarify, you are taking a step down stock power toroidal transformer and feeding the test signal into the 120vac primary and looking at the lower voltage output?? No load?
Just to clarify, you are taking a step down stock power toroidal transformer and feeding the test signal into the 120vac primary and looking at the lower voltage output?? No load?
Yes, this is correct.
At the time I had done this study I had little knowledge of transformers other than different core material types and transformation ratios.
The effects of Leakage inductance and self capacitance was new ground for me.
Also the importance of the load resistance I was aware of but didn't realize how much it meant under certain conditions.
This I learned more about as I dug deeper into the situation.
My findings are such that the core size and number of primary turns determines the lowest frequency of operation as per peak input voltage and is true for all transformers.
And the self capacitance along with the leakage inductance determines the resonate frequency of the transformer and you want to stay away from that range although added some resistance to dampen this resonance is common practice.
I have not gotten in to core losses of the materials at high frequency's yet.
But, I do know that the name of the game these days is to have better efficiency.
It is known that better core materials are used these days compared to that of yesteryear.
However amongest my extensive search's I find it difficult to find exactly what percentages of the materials are actually being used and for which product for any particular company.
This is where us as DIYer's are left at our mercy.
Unless one decides to buy a particular core material of known parameters at a higher cost (aside from the increased cost due to the bad economy and the fact that buying in volume is cheaper).
The B-H curves that I got from the typical cores that I have laying around where much better and straighter in the Toroid's than some of the E-I that I have from even back as far as the early 80's.
Although not the greatest compared to some more exotic materal's.
http://www.diyaudio.com/forums/planars-exotics/161485-step-up-transformer-design-5.html#post2194033
As for as using a load I did some tests with and without a load and I did use a load using the transformer in a 1:1 transformation ratio frequency response test using just the to 120v windings,If I remember correctly.
In that test I used a resistor and I also used a my little two way sony speaker.
In the tests showing the flat response I was using a resistor as a load for the scope and sweep test pictures.
When I hooked up the speaker I found very little difference if any in the quality of sound with or without the transformer in the circuit.
Being that the windings were for 120v at 60Hz I shouldn't have any core saturation problems getting to 20hz at 30v rms.
Listening to some music the only thing I noticed was a subtle increase of muddiness and distortions in the extreme low end.
But, It wasn't that bad to call it terrible it just sound loser and above about 200hz every thing sounded as clean and crisp as the tests without the transformer.
This was when I deemed that this transformer might work very well for my ESL's and it did.
I also did a quick test of ultrasonic's with the ESL setup and found it worked as well up to about 50KHZ or so and was enough to catch the barking dogs next door's attention!!!
The limit's of my microphone caused me not to get any real accurate measurements at the time.
But, Hey, I least I got them to stop barking every time I beamed the signal to them out my open window !!! He,he,he,he
So when you talk about the cost of audio transformer R&D certainly play's a big part as does the cost of the materials these days.
The bigger the core the more material.
The higher the primary voltages ( and vise versa depending on Step-up or step-down) the more turns are required.
This raises the self capacitance and more measures must be accounted for in order to reduce this effect.
Thus the time it takes to start over and test it again!
Therefore a even bigger core is need to use a thicker winding insulation between the layers to lower the self capacitance.
Another good thing about toroid cores is that they are more efficient and have a lower leakage inductance than the typical E-I.
Not to mention the better materials commonly used these days.
This also helps to reduce there size as well.
Just like the Acoustat's interfaces it takes a lot of iron to make a good fullrange transformer for ESL's
Alpha core has some very large cores that can be used to affordably make such a transformer.
They have some very large cores and also should have enough room to use extra extra thick winding insulation layer's in order to help cut down on the the transformer's self capacitance.
This is a great thing and would be perfect for some large panels but the gain is only small for the small panels that I have been working with. Although it would still be a large improvement.
More on that later.
The last and foremost reason the cost is so high is that the manufacture's have you right were they want you and know that their product is exactly what you are looking for and have very little options other than to by their product!.
I know inflation has caused things to go up considerably, But seriously $400 for one Plitron with only a 90 watt rating?
Even 4 Antek 0506's per panel would cost 1/8 that and give you just as good if not better performance and a higher power rating as well !!!
jer
At the time I had done this study I had little knowledge of transformers other than different core material types and transformation ratios.
The effects of Leakage inductance and self capacitance was new ground for me.
Also the importance of the load resistance I was aware of but didn't realize how much it meant under certain conditions.
This I learned more about as I dug deeper into the situation.
My findings are such that the core size and number of primary turns determines the lowest frequency of operation as per peak input voltage and is true for all transformers.
And the self capacitance along with the leakage inductance determines the resonate frequency of the transformer and you want to stay away from that range although added some resistance to dampen this resonance is common practice.
I have not gotten in to core losses of the materials at high frequency's yet.
But, I do know that the name of the game these days is to have better efficiency.
It is known that better core materials are used these days compared to that of yesteryear.
However amongest my extensive search's I find it difficult to find exactly what percentages of the materials are actually being used and for which product for any particular company.
This is where us as DIYer's are left at our mercy.
Unless one decides to buy a particular core material of known parameters at a higher cost (aside from the increased cost due to the bad economy and the fact that buying in volume is cheaper).
The B-H curves that I got from the typical cores that I have laying around where much better and straighter in the Toroid's than some of the E-I that I have from even back as far as the early 80's.
Although not the greatest compared to some more exotic materal's.
http://www.diyaudio.com/forums/planars-exotics/161485-step-up-transformer-design-5.html#post2194033
As for as using a load I did some tests with and without a load and I did use a load using the transformer in a 1:1 transformation ratio frequency response test using just the to 120v windings,If I remember correctly.
In that test I used a resistor and I also used a my little two way sony speaker.
In the tests showing the flat response I was using a resistor as a load for the scope and sweep test pictures.
When I hooked up the speaker I found very little difference if any in the quality of sound with or without the transformer in the circuit.
Being that the windings were for 120v at 60Hz I shouldn't have any core saturation problems getting to 20hz at 30v rms.
Listening to some music the only thing I noticed was a subtle increase of muddiness and distortions in the extreme low end.
But, It wasn't that bad to call it terrible it just sound loser and above about 200hz every thing sounded as clean and crisp as the tests without the transformer.
This was when I deemed that this transformer might work very well for my ESL's and it did.
I also did a quick test of ultrasonic's with the ESL setup and found it worked as well up to about 50KHZ or so and was enough to catch the barking dogs next door's attention!!!
The limit's of my microphone caused me not to get any real accurate measurements at the time.
But, Hey, I least I got them to stop barking every time I beamed the signal to them out my open window !!! He,he,he,he
So when you talk about the cost of audio transformer R&D certainly play's a big part as does the cost of the materials these days.
The bigger the core the more material.
The higher the primary voltages ( and vise versa depending on Step-up or step-down) the more turns are required.
This raises the self capacitance and more measures must be accounted for in order to reduce this effect.
Thus the time it takes to start over and test it again!
Therefore a even bigger core is need to use a thicker winding insulation between the layers to lower the self capacitance.
Another good thing about toroid cores is that they are more efficient and have a lower leakage inductance than the typical E-I.
Not to mention the better materials commonly used these days.
This also helps to reduce there size as well.
Just like the Acoustat's interfaces it takes a lot of iron to make a good fullrange transformer for ESL's
Alpha core has some very large cores that can be used to affordably make such a transformer.
They have some very large cores and also should have enough room to use extra extra thick winding insulation layer's in order to help cut down on the the transformer's self capacitance.
This is a great thing and would be perfect for some large panels but the gain is only small for the small panels that I have been working with. Although it would still be a large improvement.
More on that later.
The last and foremost reason the cost is so high is that the manufacture's have you right were they want you and know that their product is exactly what you are looking for and have very little options other than to by their product!.
I know inflation has caused things to go up considerably, But seriously $400 for one Plitron with only a 90 watt rating?
Even 4 Antek 0506's per panel would cost 1/8 that and give you just as good if not better performance and a higher power rating as well !!!
jer
Yes they are, DeltaStar.
I have not used them as of yet,But many have and use just two of them.
I would suggest to use 4 of them as I mentioned earlier as this would help to lower the saturation frequency point and increase the power handling only this is not really an issue at the higher frequency's.
I don't have a 0506 to test as Bolserst has already done that.
Since the 120v windings are bifillar wound using them in parallel and/or just one per core may reduce the the total transformer capacitance for the system compared to using them in series, on top of increasing the power handling to well over 200 watts for the 4 of them.
jer
I have not used them as of yet,But many have and use just two of them.
I would suggest to use 4 of them as I mentioned earlier as this would help to lower the saturation frequency point and increase the power handling only this is not really an issue at the higher frequency's.
I don't have a 0506 to test as Bolserst has already done that.
Since the 120v windings are bifillar wound using them in parallel and/or just one per core may reduce the the total transformer capacitance for the system compared to using them in series, on top of increasing the power handling to well over 200 watts for the 4 of them.
jer
Last edited:
Hello DeltaStar,
I use 230V/6V Farnell/Multi-Comp transformers, which are not sold in the US. However, I've also sound-tested a pair of those Antek AN-506's briefly, driving my hybrid ESL's, and they sound great. Another builder here on the forum, dochungwell, has been using the Antek 506's for over a year now I believe and he's quite happy with them.
On my blogpage (link below) there are schematics showing the Antek AN-506's wired up in both tandem and quad configurations. There is also a link on my blog page to Bolserst's test report on the Antek AN506's. In addition, I think Jer will be doing some testing and measurements on the larger Antek AS-1206 soon.
Here's a photo of those 50VA Antek 506's:
I use 230V/6V Farnell/Multi-Comp transformers, which are not sold in the US. However, I've also sound-tested a pair of those Antek AN-506's briefly, driving my hybrid ESL's, and they sound great. Another builder here on the forum, dochungwell, has been using the Antek 506's for over a year now I believe and he's quite happy with them.
On my blogpage (link below) there are schematics showing the Antek AN-506's wired up in both tandem and quad configurations. There is also a link on my blog page to Bolserst's test report on the Antek AN506's. In addition, I think Jer will be doing some testing and measurements on the larger Antek AS-1206 soon.
Here's a photo of those 50VA Antek 506's:
Last edited:
I’ll take a stab at describing why simplex wound power toroids provide extended HF response when driving hybrid ESLs, while much effort is needed to do the same in tube amplifiers. It basically boils down to two major differences:
1) ESL step up transformers are driven from low impedance sources while tube amp transformers are driven by relatively high impedances sources.
2) For hybrid ESL use, the required LF bandwidth is limited to 200-300Hz while transformers for tube amplifiers generally need to handle another 2-3 octaves.
These two major differences allow the use of significantly lower number of primary turns, because:
1) Low impedance drive allows the use of much lower primary inductance since the voltage applied to the primary will not be modulated by voltage drop across the source impedance due to current that varies with frequency and load. Low impedance drive results in flat low distortion voltage response applied to the primary of the transformer even with widely varying current from impedance variations and core related loads. For reference, the Antek AN-0506 has a primary inductance of only about 15mH.
2) Since the LF bandwidth is limited to 200-300Hz, lower number of primary turns can be used while retaining adequate power handling before core saturation.
Reducing primary turns is very beneficial as leakage inductance is proportional to number of primary turns squared. So reducing the primary turns results in significant gains in HF bandwidth. Leakage inductance is also inversely proportional to the length of the winding, so longer winding is better. In the case of the toroid, it’s core shape allows a given number of turns to be spread over 4-6 times the length as for an EI core with the same core area. This is why simplex wound toroidal power transformers are useable for ESL step-up but simplex wound EI power transformers are not.
Now I’m not sure about the 100Khz+ bandwidth that was mentioned, I believe that was a test with only a 10 turn primary which significantly reduced the power handling capability. For the Antek AN-0506, the 6V windings have 44 turns and provide enough power handling capability for hybrid ESLs crossed in the 200-300Hz range.
Attachment #1 shows the response using two of the AN-0506 connected as described below driving a 1200pF panel.
With no load, bandwidth was 47kHz.
See Attachment #2 borrowed from Jazzman’s website:
Jazzman's DIY Electrostatic Loudspeaker Page: The Electronics Package
For ESL step-up transformer use, the power toroids are driven in reverse. (ie the low voltage secondaries are used as the input and the 120VAC primaries are used as the output). Most setups use dual 120VAC primary and dual secondary(6VAC) transformers. The four 6VAC windings are connected in parallel and used as the input. The four 120VAC windings are connected in series and used as the output. The result will be step-up ratio of about 75:1 depending on the exact turn ratio used in the power toroids.
Attachments
Last edited:
Yes, I use them although I have had a few short on me before from current surges from arcs, but most all of them survived.
Use the ones with a higher current ratings, I think they range from .75 to 1.5 amps.
I have been using them in my HV supply and have yet to blow one out,
http://www.esldiy.com/esldiy/forum/index.php?topic=5.msg29#msg29
jer
Use the ones with a higher current ratings, I think they range from .75 to 1.5 amps.
I have been using them in my HV supply and have yet to blow one out,
http://www.esldiy.com/esldiy/forum/index.php?topic=5.msg29#msg29
jer
Last edited:
In this circuit they see at most 650V across them which is roughly 2/3 of their rating.
I have never had a problem with 1N4007 diodes working at this voltage level...my oldest ESL in regular use is 18 years old.
There A lot of repeat on this info.....thanks for geting the 1k diodes useg out there one more time.... thay work great an can be had at the rat-shak any time in the US...
I have seen the Trodel in the ML summit an it look just like one of the Antek AN-506's same size an wate...thay do use an crossover in the primary centap...this may give a edg
happy new year
I have seen the Trodel in the ML summit an it look just like one of the Antek AN-506's same size an wate...thay do use an crossover in the primary centap...this may give a edg
happy new year
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.