Hi,
with large unsegmented panels -and the AM is of such a type- impedance values of <1Ohm towards the upper bandwidth limit are typical. Since the capacitance of the panel forms a resonant circuit with the stray inductance of the audio trannie a raise in frequency response (or peak) results. A low-ohmic-value resistor connected in series to the audio trannie serves two purposes. First it dampens the resonance peak (so you can ´taylor´ the fequency response) and second it eases the load the amplifier sees by raising the ohmic value and reducing the phase shift.
The low ohmic value is not too critical with most amps since the speaker impedance drops to such low values at the upper frequency extreme where music doesn´t contain high signal levels any more. It is a matter though when using measuring signals with constant and high signal levels.
If the tranny is too low in primary inductance there will be a low ohmic value at the lower bandwidth limit too. Because of the high signal- and hence current levels involved this can be a serious problem for the amp. FR-ESLs are of course more prone to this problem than hybrid-ESLs that feature a series capacitor as ´safeguard´ in the crossover in most cases.
What is more problematic with most amps is the complex nature of the impedance. A phase shift of 45° means already 1A of ´wasted´ complex current that has to be supplied by the amplifier for every 1A of ´usable´ real current into the load. The ideal capacitor shows phase shift values of 90° --> only complex current, no real current into the load. The ideal transformer would present the amplifier with a perfectly transformed capacitance, hence 90° of phase shift. A real world transformer reduces the phase shift the better the worse the transformer is.
That´s good for the designer of the ESL since the use of a worse and therefore considerably cheaper tranny eases the stress on the poor amplifier alot. But if a good tranny (which is sonically superior) is used You have to be prepared that the amplifier shoud be able to handle at least 5times (power- and heat-wise) the output power it delivers to the speaker. Very lowloss trannies like some toroids show peak values of ~85°(!!). This means >11A of complex current for each 1A of real current!! This means too, that the powersupply and the heatsinks have to be designed very beefy even if the amplifier specs only moderate power capabilities.
Additionally if the amplifier uses global feedback (which most amps nowadays do) the high values of phase shift can lead to oscillations and eventually breakdown. The series resistor at the trannie´s primary side lowers the phase shift value and easens the amps life.
That are the reasons why You find a series resistor in nearly every ESL.
jauu
Calvin
with large unsegmented panels -and the AM is of such a type- impedance values of <1Ohm towards the upper bandwidth limit are typical. Since the capacitance of the panel forms a resonant circuit with the stray inductance of the audio trannie a raise in frequency response (or peak) results. A low-ohmic-value resistor connected in series to the audio trannie serves two purposes. First it dampens the resonance peak (so you can ´taylor´ the fequency response) and second it eases the load the amplifier sees by raising the ohmic value and reducing the phase shift.
The low ohmic value is not too critical with most amps since the speaker impedance drops to such low values at the upper frequency extreme where music doesn´t contain high signal levels any more. It is a matter though when using measuring signals with constant and high signal levels.
If the tranny is too low in primary inductance there will be a low ohmic value at the lower bandwidth limit too. Because of the high signal- and hence current levels involved this can be a serious problem for the amp. FR-ESLs are of course more prone to this problem than hybrid-ESLs that feature a series capacitor as ´safeguard´ in the crossover in most cases.
What is more problematic with most amps is the complex nature of the impedance. A phase shift of 45° means already 1A of ´wasted´ complex current that has to be supplied by the amplifier for every 1A of ´usable´ real current into the load. The ideal capacitor shows phase shift values of 90° --> only complex current, no real current into the load. The ideal transformer would present the amplifier with a perfectly transformed capacitance, hence 90° of phase shift. A real world transformer reduces the phase shift the better the worse the transformer is.
That´s good for the designer of the ESL since the use of a worse and therefore considerably cheaper tranny eases the stress on the poor amplifier alot. But if a good tranny (which is sonically superior) is used You have to be prepared that the amplifier shoud be able to handle at least 5times (power- and heat-wise) the output power it delivers to the speaker. Very lowloss trannies like some toroids show peak values of ~85°(!!). This means >11A of complex current for each 1A of real current!! This means too, that the powersupply and the heatsinks have to be designed very beefy even if the amplifier specs only moderate power capabilities.
Additionally if the amplifier uses global feedback (which most amps nowadays do) the high values of phase shift can lead to oscillations and eventually breakdown. The series resistor at the trannie´s primary side lowers the phase shift value and easens the amps life.
That are the reasons why You find a series resistor in nearly every ESL.
jauu
Calvin
otto88 said:
Otto, the simplest solution is to place a 10 watt 1,5ohm or so resistor in series with the step-up transformer inside your ESL. This'll raise the resistance to above 2,0ohm. As mentioned my Aspen Lifeforce has no issues with that - or the phase shift for that matter -.
Back to the resistor: The trick is not to choose a value that is too high. A high value resistor together with your transformer's primary windings could start to form a high-pass filter. The down-side is that there's now a series resistor in the signal path. That said, when studying XO's on multi-way loudspeakers you'll often find much more than that in the way of the signal. Some choose to forego on the series resistor and choose for a notch filter instead.
By the way, in the spirit of "Walk the walk...." I utilize a 10 Watt Mundorf "Metal Oxide Film" resistor of 1,5ohm.
Calvin
Thanks that’s mostly very clear, and a great explanation except
"A real world transformer reduces the phase shift the better the worse the transformer is" ??
Ronmeister
Thanks again, thats clear
Whats the Lifeforce "rated minimum" ohms?
The Hypex is well regarded by many, what swung you to the Lifeforce (which iirc is dearer).
Anthony, which of your amps would you recommend?
Do the Lifeforce, Hypex or Aussie Amplifiers use global feedback?
If an amp were tried, which isn’t really up to driving the load, is their a warning sign(s), or might it just eg blow – and is it the output stages that go?
Thanks that’s mostly very clear, and a great explanation except
"A real world transformer reduces the phase shift the better the worse the transformer is" ??
Ronmeister
Thanks again, thats clear
Whats the Lifeforce "rated minimum" ohms?
The Hypex is well regarded by many, what swung you to the Lifeforce (which iirc is dearer).
Anthony, which of your amps would you recommend?
Do the Lifeforce, Hypex or Aussie Amplifiers use global feedback?
If an amp were tried, which isn’t really up to driving the load, is their a warning sign(s), or might it just eg blow – and is it the output stages that go?
Hi Otto
I would recommend either the HPA-nxV300 or nxV500 modules
Both of the modules use Global feedback and will easily handle ESLs
The only sign you will get is the rapid over heating of the output stage
if the amplifier does not like an ESL system
I would recommend either the HPA-nxV300 or nxV500 modules
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Both of the modules use Global feedback and will easily handle ESLs
The only sign you will get is the rapid over heating of the output stage
if the amplifier does not like an ESL system
otto88 said:
Hi Otto, good questions! My road to Lifeforce was paved with 1) personal taste 2) measurements 3) my experience 4) experience of others. The reasons I did not go for the less costly Hypex had to do with my lack of experience as well as the lack of experience of my circle of friends with this product. I mention it because others have booked good results in difficult circumstances.
Back to Lifeforce and why? In my DIY audio past I was a "Solid State - ESL guy" from the late 80's to the mid 90's. My tastes, musically and sonically, evolved in to me being a "tube guy". However, the loudspeaker designs associated with SET never delivered the kind of magic I believe I experienced with ESLs. I booked decent results with a lot of loudspeaker builds over the years, single driver, BLH, multi-unit etc. etc. At some stage though, a few years ago I decided I missed ESL-magic sufficiently, missed building ESL's, so I brushed up on my theory and built a few sets, both wire and plate stator.
Unfortunately my Tube Amps, great as they are, did not suit ESLs. Though decent results were achieved in my immediate circle of friends with PP Tube designs, it is my personal opinion that SS befits the challenges we're discussing better. But then... I do love the "tube sound"... along came Aspen Amplifiers. A few of my contacts, people who's opinion I trust, pointed me towards Ozz, towards Hugh. I contacted Hugh, was very, very impressed with his willingness to discuss details at length without securing a sale, my trust in him and his product grew... a leap of faith... and there they were; two Lifeforce Modules. An extremely stable SS design with, IMHO, sonic characteristics similar to my beloved Tube Amps. An ease of musical reproduction that is stunning, detailed yet lively, precise yet warm... etc. It exceeded some of the extremely expensive commercial amps I had the pleasure of experiencing. But I digress...
The measurements, I don't believe Hugh published any figures below 4 ohm. I've subjected my Lifeforce to values slightly lower than 1 ohm - I'll not say how, for fear of losing face 😀 - and it did not break a sweat. That said, the way in which I designed my latest ESLs, they don't drop below 2.4 ohm, a value with which my Lifeforce has no problems whatsoever.
Apparently you don't need to spend what I spent on my modules. The friends that directed me towards Hugh's amps utilized the considerably cheaper AKSA 100 and AKSA 55 (discontinued) with ESL designs I'm familiar with and/or built. So there's no need to go overboard as I did... (my annual bonus had just hit my account! 😎 ). These same friends backed Hugh's guarantee that his modules work with ESL with actual measurements. Which completes the circle of experience, taste and measurements.
For completion sake, I now "run" the Lifeforce Amp, together with a DIY CD player (combination of philips transport and AN DAC) as well as a DIY EL84 Amp. The loudspeakers are a pair of full range ESLs for the Lifeforce and a pair of "hybrid-full-range" for the EL84.
I hope I did not "ramble on" too long. I'm passionate about this hobby of ours... One more time, my "Lifeforce" embedded in Granite...
Attachments
a.wayne said:
A.Wayne, when I first returned to building ESLs I utilized perforated metal sheets - freely translated in to "plate stator" in Dutchlish
. I find them easier to build, and a good start for me to get back in to it. The ESLs I referred to in the last part of my ramble, the ones in my living room now, are copper wire stators. Though I find them more difficult to build I've found them to be more rewarding musically.
Thanks for the reply Ron ,
My preference is also a wired stator Ron where most here favor perforated metal , i asked because you also have experienced both and i wanted to hear your thoughts on the matter ..
My preference is also a wired stator Ron where most here favor perforated metal , i asked because you also have experienced both and i wanted to hear your thoughts on the matter ..
a.wayne, good to see there're more of us "out there". I've had quite a bit of experience with both the past 3-4 years. IMHO there's something more "open" and "musical" about wire stators. What kind, brand, diy-model etc. do you have?
Thanks Ronmeister for a *very interesting "ramble".
I also like both tubes and ESLs –
The AKSA 55/ 100 (which yes have a warm "tuby" but clean sound) works ok into *ESLs? I'd heard from Hugh that about 3 – 3.5 ohms was their minimum.
How low were friends' ohms driving the AKSA??
I also like both tubes and ESLs –
The AKSA 55/ 100 (which yes have a warm "tuby" but clean sound) works ok into *ESLs? I'd heard from Hugh that about 3 – 3.5 ohms was their minimum.
How low were friends' ohms driving the AKSA??
otto88 said:
After impedance correction 2.9 and 3.3 ohm. The 2.9 ohm is a full range ESL, the 3.3 ohm is a hybrid. Both were built by me a while ago, (very) loosely based on Fikier's designs. I'm sure Hugh likes to err on the side of caution. What's important to note is that the phase shift and complex current is often more problematic for amplifiers. I believe it is here that the AKSA amps have proven to be stable. The low impedance can easily be corrected as described previously. If you're a "tube guy" like I am, it'd be worth considering AKSA.
Ronmeister said:
Totally agree with your assessment Ron . It's been awhile since i have fooled with ESL , but our determination years ago ( 15 + ) was that wired stators exceeded metal perf stators in openness and detail.
back then wired stators were called open stators and perf as closed stators , well that is what i re- call us calling them . we built a few , full range and hybrid , if memory serves me right the first was based on an article that was printed in speaker builder , again 15 + yrs ago and then we rolled our own , based on desired performance , up graded power supply etc ..
My current project is a 3 way RIbbon hybrid , that is producing really good results .( well except for it's power hungriness ) I hope to do a wired stator ESL next as i move to my next beast dynamic/ESL /ribbon hybrid ...
Regards,
Would love to see what you come up with!
I'll be completing another full range wire-stator model early next year. They'll be 260cm2 - I'd expect a fairly smooth frequency response between 60hz - 20 Khz. I have the parts, including Vanderveen Transformers, but will not have the time this summer. The central European winters are perfect for building... stuff...
I'll be completing another full range wire-stator model early next year. They'll be 260cm2 - I'd expect a fairly smooth frequency response between 60hz - 20 Khz. I have the parts, including Vanderveen Transformers, but will not have the time this summer. The central European winters are perfect for building... stuff...
ESLs are used above say 150 Hz - Active subs
Below c 150 Hz, I'll be using the active servo controlled 2 * 30 cm a side dipole woofers I linked to earlier.
http://www.gr-research.com/index.as...PROD&ProdID=140
http://www.audiocircle.com/index.php?topic=54875.0
If eg a SE tube amp used were that way with normal cone mid-bass drivers, the demands on the often "weakest link" in the amp, the SE's output transformers, would be significantly reduced.
If ESLs are used above say 150 Hz, does that make any difference to the demands - requirements on a SS amp?
I created a new thread for this:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=148779
Thanks
Below c 150 Hz, I'll be using the active servo controlled 2 * 30 cm a side dipole woofers I linked to earlier.
http://www.gr-research.com/index.as...PROD&ProdID=140
http://www.audiocircle.com/index.php?topic=54875.0
If eg a SE tube amp used were that way with normal cone mid-bass drivers, the demands on the often "weakest link" in the amp, the SE's output transformers, would be significantly reduced.
If ESLs are used above say 150 Hz, does that make any difference to the demands - requirements on a SS amp?
I created a new thread for this:
http://www.diyaudio.com/forums/showthread.php?s=&threadid=148779
Thanks
- Status
- Not open for further replies.