Audiofanatic-
I for one am very interested in your transformer- would you share it's specs with us? Maybe photo? E160 isn't a bad price. I have some Plitron 50:1 toroids currently- They list for about $380 Canadian. Also thanks for your data on the ES-300 panel.
Also, the d/s spacing of your panels would be helpful. Thanks,
Jay
I for one am very interested in your transformer- would you share it's specs with us? Maybe photo? E160 isn't a bad price. I have some Plitron 50:1 toroids currently- They list for about $380 Canadian. Also thanks for your data on the ES-300 panel.
Also, the d/s spacing of your panels would be helpful. Thanks,
Jay
Jay
The distance is 2.5 mm to one of the stators.
The price is from 2006 and probably much higher now but with a GB it'll probably get to a reasonable price.
The transformer type is a Toroide type and the lowest freq I could get out of the panel with a 300B on extremely low volume was 25 Hz. Those were the most beautiful bass tones I ever heard
deep and clean! But I could not go loud enough with the 300B it was to much for the poor animal
Never tested it on the KRELL KSA250, I don't like so uch power in my transformers.
The top is (to my ears) sweet but not lazy It's unlike an ESL it's absolutely smooth and a bit warm. That's what my best friend says. I find it just good and well balanced .
I'll try to take a picture of it in the weekend when I'm ta the office.
It's with every single penny. I had tow Amplimo's on it but that didn't work, it was to small and not capable of driving the panels at all.
ESL's are nice speakers IF they are made well, if you think that it's easy and it'll work anyway than I got news for you, I'll play but not at it's best.
I know this for sure cause I do this for a living
All the best,
Audiofanatic
The distance is 2.5 mm to one of the stators.
The price is from 2006 and probably much higher now but with a GB it'll probably get to a reasonable price.
The transformer type is a Toroide type and the lowest freq I could get out of the panel with a 300B on extremely low volume was 25 Hz. Those were the most beautiful bass tones I ever heard
deep and clean! But I could not go loud enough with the 300B it was to much for the poor animal Never tested it on the KRELL KSA250, I don't like so uch power in my transformers.
The top is (to my ears) sweet but not lazy It's unlike an ESL it's absolutely smooth and a bit warm. That's what my best friend says. I find it just good and well balanced .
I'll try to take a picture of it in the weekend when I'm ta the office.
It's with every single penny. I had tow Amplimo's on it but that didn't work, it was to small and not capable of driving the panels at all.
ESL's are nice speakers IF they are made well, if you think that it's easy and it'll work anyway than I got news for you, I'll play but not at it's best.
I know this for sure cause I do this for a living
All the best,
Audiofanatic
I am also rethinking my entire build. I currently have 16X48 panels with perforated steel stators. I was hoping to augment the LF response with panels of similar construction. My main panels sound fine, but they are beamy as hell in the high end, and there is only one spot I can sit that lets me hear everything. I am crossing at 200 Hz using 24 dB per octave Linkwitz-Reily filters. I would liked to have gotten another 2 octaves (50 Hz) with ESL panels.
SDStokes mentioned in his first reply (#21) using wire stators and "mechanically" EQing the panels using resistors. The more I think about this, the more I think I want to try it.
When you think about it, a wire stator panel is a big array of small value capacitors. Putting a resistance on an individual wire will result in a low-pass response within that area of capacitance. It strikes me that at the very least, we can taper the HF response over the width of the panel, and thus control the HF dispersion.
If we make the panel big enough, we can get pretty decent response in the low end as well. If I could get response to 50 Hz before I cross into my subs, I will be a happy camper.
Years ago, Peter Walker of Quad fame published an analysis of elctrostatic loudspeakers in a now (possibly) out of print book entitled "Loudspeakers". This book was part of a 5 book set which I believe was titled "Audio Encyclopedia", published in the 50-60's era. I had a copy, which may be packed or donated to a library - don't know. I think this analysis could be very helpful, so keep your eye open.
SDStokes mentioned in his first reply (#21) using wire stators and "mechanically" EQing the panels using resistors. The more I think about this, the more I think I want to try it.
When you think about it, a wire stator panel is a big array of small value capacitors. Putting a resistance on an individual wire will result in a low-pass response within that area of capacitance. It strikes me that at the very least, we can taper the HF response over the width of the panel, and thus control the HF dispersion.
If we make the panel big enough, we can get pretty decent response in the low end as well. If I could get response to 50 Hz before I cross into my subs, I will be a happy camper.
Years ago, Peter Walker of Quad fame published an analysis of elctrostatic loudspeakers in a now (possibly) out of print book entitled "Loudspeakers". This book was part of a 5 book set which I believe was titled "Audio Encyclopedia", published in the 50-60's era. I had a copy, which may be packed or donated to a library - don't know. I think this analysis could be very helpful, so keep your eye open.
Check Out the cores from Alpha core !
It would be very cost effective to wind a custom transformer.
Start with a very large core in the 2000 to 3000 watt range this would get a good low frequency handling.
Also a core this size will have enough room for thick insulating between the layers in order to keep the stray capacitance to a minimum.
A core this size won't require as many turns either on the primary possibly keeping the secondary to only one layer or two depending on your chosen turns ratio.
This would be much cheaper than the custom wound ones from over sea's.
Yes,it would be tedious,But it is DIY and a larger core will also be easier to handle and wind, except for the weight.
This is something I plan on doing sometime if I decide to build some larger panels just for the low frequency's.
Cores that size are in the $25 to $65 range from Alpha Core.
They do have custom winding available as well but I have not inquired about their services yet.
Right now I have plenty of 200watt cores to mess with and I am planning on stripping one or two stacked and giving it a try myself.
A very large core and a 1000' spool of some kynar covered 30 Ga. (wire wrap ) wire seems like a very viable alternative of around $100 for two complete transformers.
Cheers!
jer
It would be very cost effective to wind a custom transformer.
Start with a very large core in the 2000 to 3000 watt range this would get a good low frequency handling.
Also a core this size will have enough room for thick insulating between the layers in order to keep the stray capacitance to a minimum.
A core this size won't require as many turns either on the primary possibly keeping the secondary to only one layer or two depending on your chosen turns ratio.
This would be much cheaper than the custom wound ones from over sea's.
Yes,it would be tedious,But it is DIY and a larger core will also be easier to handle and wind, except for the weight.
This is something I plan on doing sometime if I decide to build some larger panels just for the low frequency's.
Cores that size are in the $25 to $65 range from Alpha Core.
They do have custom winding available as well but I have not inquired about their services yet.
Right now I have plenty of 200watt cores to mess with and I am planning on stripping one or two stacked and giving it a try myself.
A very large core and a 1000' spool of some kynar covered 30 Ga. (wire wrap ) wire seems like a very viable alternative of around $100 for two complete transformers.
Cheers!
jer
Last edited:
Check out some of these links as well on, driver size vs frequency dispersion,
http://www.diyaudio.com/forums/plan...se-cancellation-esl-speakers.html#post2857081
I have found them to be quite informative and useful.
They have shed quite a bit of light on the subject of why larger panels are more directive at the higher frequency's,for me.
enjoy !
jer
http://www.diyaudio.com/forums/plan...se-cancellation-esl-speakers.html#post2857081
I have found them to be quite informative and useful.
They have shed quite a bit of light on the subject of why larger panels are more directive at the higher frequency's,for me.
enjoy !
jer
If you haven't read it already, you would probably be interested in this AES paper.
Audiostatic, Acoustat(in their Spectra series), and STAX all used ladder resistor segmenations in their ESLs.
So, the technique of segmentation is not new, but this paper does a very good job of providing theory behind the technique.
AES E-Library: Wide-Range Electrostatic Loudspeaker with a Zero-Free Polar Response
I posted an Excel spreadsheet a while back that incorporates concepts from the paper.
It allows you to quickly get an idea what different panel width and segmentation sizes are capable of in terms of bandwidth & SPL.
http://www.diyaudio.com/forums/planars-exotics/48120-experiences-esl-directivity-9.html#post2218526
Actually, pretty good discussion throughout the thread, starting from post # 52 where kavermei summarized the AES paper.
http://www.diyaudio.com/forums/planars-exotics/48120-experiences-esl-directivity-6.html#post2202327
Last edited:
...
A very large core and a 1000' spool of some kynar covered 30 Ga. (wire wrap ) wire seems like a very viable alternative of around $100 for two complete transformers.
Cheers!
jer
If you are going to make a secondary from Kynar wire, think twice:
Kynar has high permittivity, dissipation factor as high as 0.09...0.25, even at 100 cycles, not mentioning piezoelectric abilities
Thanks,alexburg,I wasn't completely aware of all of Kynar's characteristics yet as I had planned on looking more into this.
But I do know that it has a very high Dielectric Strength.
I was a little worried if it might have some piezoeletric abilities as the transformer I am using now does sing a bit and is quite irritating but won't be as noticeable once it is mounted in a suitable enclosure.
Maybe some triple coated enamel magnet wire would be better?
The search is on!
That's to bad since I had found a good source for some Kynar wire.
jer
But I do know that it has a very high Dielectric Strength.
I was a little worried if it might have some piezoeletric abilities as the transformer I am using now does sing a bit and is quite irritating but won't be as noticeable once it is mounted in a suitable enclosure.
Maybe some triple coated enamel magnet wire would be better?
The search is on!
That's to bad since I had found a good source for some Kynar wire.
jer
No need too.
I just need to find a suitable wire to use.
In some quick searches magnet wire is about the same cost as the kynar and actually cheaper in most cases per 1000', depending on the gauge.
I am researching some resources and spec's as we speak.
I just realized that each one of my cores has at least 500' of wire on each one for 1155 turns with about 12 ohms DCR.
I only need to have 2 to 4 times this to reach my current goals for now as I am not going below about 200hz with this particular project.
With a larger core lower frequency's are much more capable with the same amount of turns that I am now using.
My only concern is the voltage breakdown between windings when the bias is pushed up.
Right now I am using about 8kv to 10kv.
My supply can do 13.8kv but I am experiencing leakage some where in the panel or the transformer as the most I am getting with the supply full on is slightly above 10kv and there seems to be a considerable amount of current draw but no arcing.
This is of importance because a large bass panel with a large D/S will require these types of operating voltages.
jer
I just need to find a suitable wire to use.
In some quick searches magnet wire is about the same cost as the kynar and actually cheaper in most cases per 1000', depending on the gauge.
I am researching some resources and spec's as we speak.
I just realized that each one of my cores has at least 500' of wire on each one for 1155 turns with about 12 ohms DCR.
I only need to have 2 to 4 times this to reach my current goals for now as I am not going below about 200hz with this particular project.
With a larger core lower frequency's are much more capable with the same amount of turns that I am now using.
My only concern is the voltage breakdown between windings when the bias is pushed up.
Right now I am using about 8kv to 10kv.
My supply can do 13.8kv but I am experiencing leakage some where in the panel or the transformer as the most I am getting with the supply full on is slightly above 10kv and there seems to be a considerable amount of current draw but no arcing.
This is of importance because a large bass panel with a large D/S will require these types of operating voltages.
jer
Hi,
my enquieries about triple coated magnet wire -as it seems very promising for a wire stator- were a bit disappointing. Essex wire and a second major company both asked for minimum quotes of 250kg and would only manufacture on demand. Besides the sheer amount of wire, the raw cost for the copper alone is considerable. Maybe things have changed to the positive -it´d be THE single great exception from the rule- and one maybe lucky to get such wire under acceptable conditions.
jauu
Calvin
my enquieries about triple coated magnet wire -as it seems very promising for a wire stator- were a bit disappointing. Essex wire and a second major company both asked for minimum quotes of 250kg and would only manufacture on demand. Besides the sheer amount of wire, the raw cost for the copper alone is considerable. Maybe things have changed to the positive -it´d be THE single great exception from the rule- and one maybe lucky to get such wire under acceptable conditions.
jauu
Calvin
Hi Jer
Wen you try the 2000 to 3000 watt range, make sure you use it only for the lover freq. I did that with a few transformers and the result is what I ended up with.
I always say and keep repeating myself: it's not what you do, but the way that you do it!
You have to wind the transformer in a different way to get it right.
I've unwind power transformers in the past and it did work but not full range and not quite the way I want an ESL transformer to work with a full range panel.
But hey, this is DIYAudio and I urge everybody to try whatever is cheap enough to do and if you are happy with the result thatn who am I or you or anyone to judge anyone
Mine is from 2006 and it was never my intention to make any profit selling these transformers. But if others want to try them out I'm more than willing to order a few of them for my fellow members on DIYAudio.
I will not tell anybody how I did it though, I will just make it possible for them to get a good sounding and working ESL transformer for full range ESL's. Nothing more.
All the best and keep us posted with your findings regarding your transformers.
Regards,
Audiofanatic
Check Out the cores from Alpha core !
It would be very cost effective to wind a custom transformer.
Start with a very large core in the 2000 to 3000 watt range this would get a good low frequency handling.
Also a core this size will have enough room for thick insulating between the layers in order to keep the stray capacitance to a minimum.
A core this size won't require as many turns either on the primary possibly keeping the secondary to only one layer or two depending on your chosen turns ratio.
This would be much cheaper than the custom wound ones from over sea's.
Yes,it would be tedious,But it is DIY and a larger core will also be easier to handle and wind, except for the weight.
This is something I plan on doing sometime if I decide to build some larger panels just for the low frequency's.
Cores that size are in the $25 to $65 range from Alpha Core.
They do have custom winding available as well but I have not inquired about their services yet.
Right now I have plenty of 200watt cores to mess with and I am planning on stripping one or two stacked and giving it a try myself.
A very large core and a 1000' spool of some kynar covered 30 Ga. (wire wrap ) wire seems like a very viable alternative of around $100 for two complete transformers.
Cheers!
jer
Wen you try the 2000 to 3000 watt range, make sure you use it only for the lover freq. I did that with a few transformers and the result is what I ended up with.
I always say and keep repeating myself: it's not what you do, but the way that you do it!
You have to wind the transformer in a different way to get it right.
I've unwind power transformers in the past and it did work but not full range and not quite the way I want an ESL transformer to work with a full range panel.
But hey, this is DIYAudio and I urge everybody to try whatever is cheap enough to do and if you are happy with the result thatn who am I or you or anyone to judge anyone
Mine is from 2006 and it was never my intention to make any profit selling these transformers. But if others want to try them out I'm more than willing to order a few of them for my fellow members on DIYAudio.
I will not tell anybody how I did it though, I will just make it possible for them to get a good sounding and working ESL transformer for full range ESL's. Nothing more.
All the best and keep us posted with your findings regarding your transformers.
Regards,
Audiofanatic
Why would not you make some using raw wire and heatshrink tubing... of any kind isolation you like... done this to make 50 kV rated magnet wire
Combining ESL with Maggie bass
Combining Martin Logan Summits with Magneplanar Tympani IV bass
panels is what I put together. Provides great bass to 25 hz with
outstanding transient response. Below 25 hz is handled by a rotary
subwoofer and all three together are coherent and give me all the output
that I need.
Combining Martin Logan Summits with Magneplanar Tympani IV bass
panels is what I put together. Provides great bass to 25 hz with
outstanding transient response. Below 25 hz is handled by a rotary
subwoofer and all three together are coherent and give me all the output
that I need.
Attachments
I am quite suprised learning that Kynar has those drawbacks. I use 28 Ga WW wire for my interconnects and 3 pairs of the same wire braided over teflon tubing for my amp to speaker hook-ups.
So, in the meantime, the search is on for a suitable transformer wire.
Question- if the disturbance in Kynar wire is induced by vibration, can we not dampen or eliminate this by potting or encapsulating the thing? I'm not sure about your use of the term Piezoelectric. Are stray voltages developed when the wire is moving (vibrating), or are you just hearing the buzz of loose windings?
Maybe we need to look into the insulation used in neon transformers. 15KV is a lot.
Jay
So, in the meantime, the search is on for a suitable transformer wire.
Question- if the disturbance in Kynar wire is induced by vibration, can we not dampen or eliminate this by potting or encapsulating the thing? I'm not sure about your use of the term Piezoelectric. Are stray voltages developed when the wire is moving (vibrating), or are you just hearing the buzz of loose windings?
Maybe we need to look into the insulation used in neon transformers. 15KV is a lot.
Jay
Neon transformers are usually potted.
I was hoping to get away from not having to do this using the Kynar covered wire.
At least until I got a working unit as it might need to be unwound and rewound a few times.
And even if it is potted it may not be as bad.
But may not justify the added cost of the Kynar, as good quality magnet wire is a bit cheaper.
You get 50% more length for 32 Ga. and nearly double for 34 Ga. for roughly the same price of 30 Ga. magnet wire.
I wouldn't go much thinner than these gauges for fear of any arc overload opening up the winding.
Or at least 32Ga. for that matter.
I am not sure what gauge is used in a neon transformer.
But I have a few of them, and, I made a Jacobs ladder out of one of them once and the thing didn't burn out as I messed with it for quite along time.
One of the ones I have is rated at 450VA at 15KV and 30ma. and I have a couple of smaller ones as well.
This one (450VA) is potted in black tar and has a broken insulator and all I can see is the lead wire to the terminal as it is fairly heavy, but I doubt that it is the secondary wire.
jer
I was hoping to get away from not having to do this using the Kynar covered wire.
At least until I got a working unit as it might need to be unwound and rewound a few times.
And even if it is potted it may not be as bad.
But may not justify the added cost of the Kynar, as good quality magnet wire is a bit cheaper.
You get 50% more length for 32 Ga. and nearly double for 34 Ga. for roughly the same price of 30 Ga. magnet wire.
I wouldn't go much thinner than these gauges for fear of any arc overload opening up the winding.
Or at least 32Ga. for that matter.
I am not sure what gauge is used in a neon transformer.
But I have a few of them, and, I made a Jacobs ladder out of one of them once and the thing didn't burn out as I messed with it for quite along time.
One of the ones I have is rated at 450VA at 15KV and 30ma. and I have a couple of smaller ones as well.
This one (450VA) is potted in black tar and has a broken insulator and all I can see is the lead wire to the terminal as it is fairly heavy, but I doubt that it is the secondary wire.
jer
Last edited:
But why use Kynar for winding transformers, when magnet wire has been specifically developed for winding transformers and such? You can make it work, but it will be something oddball. The formulas for transformer design will not work quite right for Kynar due to the insulation thickness being a little weird.
Sheldon
Sheldon
If you are winding the transformer by hand, you probably don't want to go thinner than 32 ga anyway because it gets really easy to break the wire. I wound a tube power transformer years ago, and from my experience, particularly with toroids, I wouldn't want to use anything thinner.
Here's the power transformer docs with calculations. Not really applicable, but might be interesting for some:
http://quadesl.com/pdf/trans_inst.pdf
Sheldon
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.