The reason I tried these toroids in first place is that I wasn't happy with the the treble response with the 100:1 EI core trannys I was using (I think 100:1 is too high for good treble response).
I will speculate that toroids from different manufacturers may perform differently and maybe I just got lucky with these toroids but I can tell you the treble is better than with the EI-core trannys I was using.
I suppose I lied about having no measuring equipment: Actually, I did do a rudimentary comparison using my graphic EQ. I first wired up the toroids into one speaker only and then compared the two speakers using the mic and rudimentary RTA feature in my Audio Control C101 graphic EQ. The speaker with the EI core trannys (no EQ boost applied) was down about 4db from the baseline in the 7860 hz band and dropped completely off scale in the 15.5khz band. Whereas, the speaker with the toroids was ruler flat through the 7860hz band and about 4db down in the 15.5 khz band. My RTA/EQ is not lab-standard equipment by any means and my hearing isnt what it used to be but I do hear and measure a clear improvement in the treble ouput of my speakers with these toroids. Hope this helps!
Hi CharlieM,
I actually get the opposite of what you get. I can get much more treble from the EI core custom wound trannys that I'm using now. I guess it's a matter of different winding techniques that go into making good trannys. Here in Thailand, it is still possible to make custom wound transformers at reasonable prices.
By the way, your ESLs look beautiful. Your wood working skill is among the best that I have every seen.
Wachara C.
I actually get the opposite of what you get. I can get much more treble from the EI core custom wound trannys that I'm using now. I guess it's a matter of different winding techniques that go into making good trannys. Here in Thailand, it is still possible to make custom wound transformers at reasonable prices.
By the way, your ESLs look beautiful. Your wood working skill is among the best that I have every seen.
Wachara C.
Hi,
when a toroid sounds dull on a panel You should rather assume that the combination of both doesn´t fit and after revising if You did every connection right You should assume Your panel to be of lower quality.
EI-cores trannys show higher losses and increased leakage inductance.
This leakage inductance forms a resonant circuit with the panel´s capacitance which leads to a high frequency peak that can reach several dBs.
So if Your panel shows a falling amplitude response (reasons can be a membrane thicker than 6µm, or a ´heavy´ coating et al) You can beef up the response using a tranny that leaks. The larger the value of the inductance the lower the resonance frequency. Old ML Sequel´s resonance was around 12kHz with the frequency response dropping sharply (-12dB) above. It was a common practise to use rather crappy transormers so You needed a rather large resistor to kill the resonance and flatten amplitude response. The high losses of those trannies lead to less phase shift and the high resistor value increased high frequency impedance. So the electrical load the amplifier needs to drive becomes much less demanding and mora amplifiers were capable to drive the ESL.
A similar resonance effect show the output filters of switching amps! So watch out when You drive a panel through a switching amp. The peak may become so serious that even noise becomes audible. Don´t drive panels with a class-d without a series resistor!
Now a toroid isn´t free from the effect, but the resonance may be found in the ultrasonic range above audio. A resistor connected in series with the primary dampens the resonance and lowers the peak. So with that measurement You can tune the amplitude response too. A larger than needed resistor introduces a drop in frequency response. So there are several reasons why a ESL may sound dull. I´m quite sure the reason is not the the toroid, at least it is not the single reason.
jauu
Calvin
when a toroid sounds dull on a panel You should rather assume that the combination of both doesn´t fit and after revising if You did every connection right You should assume Your panel to be of lower quality.
EI-cores trannys show higher losses and increased leakage inductance.
This leakage inductance forms a resonant circuit with the panel´s capacitance which leads to a high frequency peak that can reach several dBs.
So if Your panel shows a falling amplitude response (reasons can be a membrane thicker than 6µm, or a ´heavy´ coating et al) You can beef up the response using a tranny that leaks. The larger the value of the inductance the lower the resonance frequency. Old ML Sequel´s resonance was around 12kHz with the frequency response dropping sharply (-12dB) above. It was a common practise to use rather crappy transormers so You needed a rather large resistor to kill the resonance and flatten amplitude response. The high losses of those trannies lead to less phase shift and the high resistor value increased high frequency impedance. So the electrical load the amplifier needs to drive becomes much less demanding and mora amplifiers were capable to drive the ESL.
A similar resonance effect show the output filters of switching amps! So watch out when You drive a panel through a switching amp. The peak may become so serious that even noise becomes audible. Don´t drive panels with a class-d without a series resistor!
Now a toroid isn´t free from the effect, but the resonance may be found in the ultrasonic range above audio. A resistor connected in series with the primary dampens the resonance and lowers the peak. So with that measurement You can tune the amplitude response too. A larger than needed resistor introduces a drop in frequency response. So there are several reasons why a ESL may sound dull. I´m quite sure the reason is not the the toroid, at least it is not the single reason.
jauu
Calvin
Hey CharlieM,
I see from your webpage that you used the pneumatic diaphragm stretching technique. I hope this doesn't deflect the thread too much, but can you tell me whether you had to use multiple inner tubes to accommodate your diaphragms, and if so, how you went about it? I've played with cutting inner tubes and then joining several together to make one large one but am interested in whether others have come up with better solutions.
Thanks,
Few
I see from your webpage that you used the pneumatic diaphragm stretching technique. I hope this doesn't deflect the thread too much, but can you tell me whether you had to use multiple inner tubes to accommodate your diaphragms, and if so, how you went about it? I've played with cutting inner tubes and then joining several together to make one large one but am interested in whether others have come up with better solutions.
Thanks,
Few
Hi Few,
I made the diapgram stretcher only a couple of inches longer and wider than my 12x48 panels, then stretched a single 1.25 x 26 bike tube around it with no splice. It took a lot of effort to stretch the tube that far but I got it on without breaking it and it worked fine--- takes very little air pressure to stretch a 6 micron film past it’s breaking point. I actually broke the film on my first attempt, just as it reached 1.5% elongation so I backed off a little and no further problems.
I want to take your advice and get some measuring software. Can you recommend something? What kind of microphone would I need to plug into my laptop?
It would be nice to have the capability to plot both the frequency response and the impedance. As it is, I’m using a 1 ohm resistor in front of those trannys, which I might not need but I’m hesitant to reduce it because I don’t know the speakers’ impedence and don’t want to risk lowering it.
My hearing isn’t good enough to hear the top octave anymore anyway and my speakers sound great to my old ears as they are now but it would be nice to post the actual frequency response plot for the benefit of others who may want to use these toroids.
Maybe Mavric will chime in (hint, hint), since he's younger with good ears and he dropped by and listened to my speakers for a couple of hours last weekend-- he left pretty stoked to build some for himself so I guess he liked their sound OK.
Good to hear from you again… your encouragement during my build was a big help.
Charlie
You are correct about there being more to the sound of a transformer than the core type. The choice of electrical steel grade and thickness can affect high frequency performance more than the core type assuming they are all well made. Just because it is a toroid doesn't mean that cheap, thick steel can't be used, which will drive up magnetizing and eddy losses. Something to be aware of though is coupling coefficient if close enough to one means that whatever appears across the primary will appear on the secondary. If you have a low output impedance amp you can Force a flatter high frequency response across a crappy transformer even though the loss is high, but if there are a bunch of resistors or other impedances in series with the primary you will get less flux in the core as loss increases with frequency. So what you will get through a transformer depends highly on the circuit that it's in.
Also, regarding other questions I saw on this thread, if you have a transformer that has been designed for a particular winding voltage at 60Hz, of course you can put the same voltage across two of those windings at 30Hz without overdriving the core. Or the same voltage at 20Hz, across three primaries.
I have a taste for sound. So I will comment on this one. Charlie has put time and effort on his projects, he keeps the other ones in the closet. I was told there was a sweet spot. I walked all over his house, his diffuser behind the ESL might be why when you close your eyes, it is very hard to find a source of where the music is coming from. Granted that his living room lookes at or about 30 ft wide and 20 ft deep, with a tower that large, I understand a sweet spot. I have to say, at medium or low volume, best sound I have heard, and I have been in car audio for 18 yrs, with some in home, all as professional installation. NO ONE sells ESL's for home audio anywhere near here, so custom is the way, however as mentioned, no one has the time. Charlie was kind and gave rather lended me a book. still reading, and reading. I have posted lining the stator with something else,, I will try it and test it to see how thin or thick I can spray bed liner to 10-20 mils. i myself think it can be done, insulation, duration, and maybe some bragging rights, just kidding. Never hurts to try. as my motto goes, i would be a fool not to.
Hi Charlie,
Thank you for the email and sharing your knowledge with your ESL build. Tons of stuff to learn, and many different ways to build. Here is a link to info regarding free measurement software and speaker measurement guides HTGuide Forum - Measurement, Design (sim), EQ, Software & Hardware links. Arta is excellent for windows, and Fuzzmeasure free trial is even better for Macs. I do have an extra Behringer mic you may borrow (it is not calibrated, but "close") as well as a Mobile pre and cables.
Looking for perforated steel locally in 52" x 16" sizes locally and will contact you fairly soon.
Craig
Hi Charlie,
Craigj beat me to the punch with a couple of measurement package suggestions. I don't have first hand experience with either of the packages he listed, but those are the same ones I was going to suggest. I have used a now out of date version Sound(Qu)easy. The price isn't too bad, it does lots of useful stuff, but I have to say I find it sort of a pain to use. I ultimately wrote my own measurement software using LabVIEW.
Come to think of it, you might also check http://www.diyaudio.com/forums/multi-way/145662-holmimpulse-measurements-practice.html here on the diyaudio site. In any case, you'll gain so much by using any sort of measurement of frequency response and impedance that the differences between packages are dwarfed by the advantages of having a quantitative measure of what your speakers are doing.
The Behringer mic is a good way to go--just be sure you have phantom power available with your mic preamp.
Post your results if you decide to make some measurements!
Few
Craigj beat me to the punch with a couple of measurement package suggestions. I don't have first hand experience with either of the packages he listed, but those are the same ones I was going to suggest. I have used a now out of date version Sound(Qu)easy. The price isn't too bad, it does lots of useful stuff, but I have to say I find it sort of a pain to use. I ultimately wrote my own measurement software using LabVIEW.
Come to think of it, you might also check http://www.diyaudio.com/forums/multi-way/145662-holmimpulse-measurements-practice.html here on the diyaudio site. In any case, you'll gain so much by using any sort of measurement of frequency response and impedance that the differences between packages are dwarfed by the advantages of having a quantitative measure of what your speakers are doing.
The Behringer mic is a good way to go--just be sure you have phantom power available with your mic preamp.
Post your results if you decide to make some measurements!
Few
GraigJ, Few,
Thanks for the info guys-- I will be checking out the measurement thing as time permits. I'm actually drawing up another ESL (simpler this time), even though I don't need any (this esl thing is a damn psychosis). Anyway, Mavric got me interested in it again and now I'm all about figuring how to build em simple and cheap and post whatever might be helpful. It would be cool if more people were able to build and hear ESL's and I think one of the bigger impediments to that has been finding suitable and affordable trannys. Thanks again, guys!
Thanks for the info guys-- I will be checking out the measurement thing as time permits. I'm actually drawing up another ESL (simpler this time), even though I don't need any (this esl thing is a damn psychosis). Anyway, Mavric got me interested in it again and now I'm all about figuring how to build em simple and cheap and post whatever might be helpful. It would be cool if more people were able to build and hear ESL's and I think one of the bigger impediments to that has been finding suitable and affordable trannys. Thanks again, guys!
Charlie,
any luck on the trannies, that will be the investment, and another, does the dual trans outperform single? If you feel confident on the single, I just might come down there and buy them, or if you have a better suggestion, i am open for lessons. By the way, had a customer, worked at the same place you do, no names, but said he knew you, I will jest say airplanes, either way, nice person, spoke highly of you.
any luck on the trannies, that will be the investment, and another, does the dual trans outperform single? If you feel confident on the single, I just might come down there and buy them, or if you have a better suggestion, i am open for lessons. By the way, had a customer, worked at the same place you do, no names, but said he knew you, I will jest say airplanes, either way, nice person, spoke highly of you.
Found some El-cheapo toroids that work for ESL's
Hello All,
Just updating my thread:
I just found some el-cheapo ($13 each), readily available, Antek AN-0506, 50VA 2x 115v/2x6v toroidal transformers that sound great driving my ESL's!
My fear about dual output windings is the possibility of shorting between the windings and damaging my amps. I hooked up a tandem of these with 6v windings wired parallel and 115v windings wired series and a 1- ohm/10 watt resistor in front, then cringed like hell while I gradually fed them more and more power. I am happy to report that they sound very nice and I cranked out some pretty serious volume too (not all that my amps can deliver but much louder than I ever care to listen again) and nothing bad happened. I have no equipment to measure anything except my ears and my ears say these $13 trannys will fill the bill for hybrid use (mine are crossed over at 275 Hz). Here's the web address for the Antek AN-506 trannys:
Antek - AN-0506
Hello All,
Just updating my thread:
I just found some el-cheapo ($13 each), readily available, Antek AN-0506, 50VA 2x 115v/2x6v toroidal transformers that sound great driving my ESL's!
My fear about dual output windings is the possibility of shorting between the windings and damaging my amps. I hooked up a tandem of these with 6v windings wired parallel and 115v windings wired series and a 1- ohm/10 watt resistor in front, then cringed like hell while I gradually fed them more and more power. I am happy to report that they sound very nice and I cranked out some pretty serious volume too (not all that my amps can deliver but much louder than I ever care to listen again) and nothing bad happened. I have no equipment to measure anything except my ears and my ears say these $13 trannys will fill the bill for hybrid use (mine are crossed over at 275 Hz). Here's the web address for the Antek AN-506 trannys:
Antek - AN-0506
Attachments
Charlie, what was that last jumper for? at the end of the step up? you are right, that was made in Italy, spagetti, lasagna, some noodles. JK, looks just like what we have have, only smaller gauge wire. I am curoius of course on how did yu find this and come to the conclussion that this will work,
Mavric
Mavric
Last edited:
That power supply isn't adjustable but the jumper does allow me to choose between two different output voltages by either using or bypassing he last two diodes and caps in the cirucuit. The schematic is shown on my blogpage-- just something I'm playing around with. Those trannys are equivalent to the 230v trannys that you and I are already using except they have (2) 115v windings instead of a single 230v winding. If you connect those windings in series, you get excactly the same step-up ratio as the trannys we have. The only question was if the insulation between the 115v windings would withstand the voltage without shorting out and they didn't so I'm OK with using them.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.