Driving ESL transformer more efficient on SS-amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
..But funny, I have neither one of them. I drove to Zurich 2015 to 2nd hand dealer. I got only 2 normal connectors, and a power in. No external resistor at all.
Ha! That is funny…you have Type C ;)
Does your step-up transformer look the same? Can you post a pic?

I conclude lowering the resistor value will do the same as those versions, lift the highs few dB. But also the poor 10 watt resistor has to stand a lot. Gets hot, changes value, bä. I bought a bunch of 44 MOX with 47R each. Would give me 2.1 Ohm and 220 watts, wrapped in a cooler, with 0.45% eff tolerance. Not yet installed though. Nice idea?
That should work just fine. You could also try the RC circuit you asked about in post#2. With the values shown below the response should be very similar, but the impedance at frequencies < 100Hz may be a little easier for the amplifier.

well considering that the official manual from the company sais it can run more than 100dB.... I think ive been usually faaaar away from that. Still the fostex amp with 150 watts easily clipped.
Maybe not as far away as you think. Based on your measurement in post#9, 2.83Vrms @ 1m = about 80dB.
So 35Vrms(ie 150W/8ohm) would be about 102dB.
Reducing the resistance from 6.8ohm to 2.2ohm will mainly flatten out the HF response, but you will also gain a few dB in the midrange and bass.

Hopefully you can get together with Calvin to have the step-up transformers and bias supply checked out to set you mind at ease.
 

Attachments

  • ESL_RC_input.png
    ESL_RC_input.png
    33.8 KB · Views: 266
Guess that second circuit will drop from 4 to 2 Ohms to the higher freq. ??
You are correct; the impedance will drop down close to 2 ohm at 20kHz. Based on impedance measurement you show in post#1, it would likely be about 2.5 ohm. I am still curious about how that impedance measurement was made, and what voltage level was used. You will find that the impedance of the transformer at the lower frequency (20Hz – 200Hz) will increase with signal level by as much as a factor of 5 or more, because the permeability of the core increases with flux density…a very nice thing for amplifiers. This increase continues until signal level is such that the core is saturated and then the impedance will drop quickly with any additional increase in signal level. If the impedance measurement was made with a small voltage level < 1Vrms, the low frequency impedance may be giving a pessimistic view of what the amplifier load would be at normal listening levels. If possible, try measuring the impedance without the 6 ohm resistor in series(or at most a 2 ohm) and at voltage levels of 1Vrms and something a bit higher like 5Vrms. This will tell us a bit more about what the low frequency capability of the transformer is.

Thanks for the photos. Very interesting to see another variation on the Capaciti interface design. Looking at the first photo it struck me how similar the transformer is to the one used by Audio Exklusiv. I recall asking Calvin once if there was a connection between the two companies but he was not aware of one.
HIFISTATEMENT | netmagazine - Audio Exklusiv P 6.1
 

Attachments

  • audio_exklusiv_trans.png
    audio_exklusiv_trans.png
    868.3 KB · Views: 212
  • Audio Exklusiv P 6p1 translate.pdf
    786.5 KB · Views: 112
Last edited:
...That explains a lot. I measured with 0.1 volts.
Heres 0.1 and 1 and 3.5 volt. Cant go higher. Does that look good?
Yes that looks very good :up:
This confirms the primary inductance of the transformer is > 200mH not 5-7mH as was indicated by the first measurement. You can see now why using a 2.2 ohm resistor will mainly affect the impedance at higher frequency. At low frequencies the impedance is dominated by the transformer primary inductance, so adding or subtracting a few ohms of series resistance makes little difference. Based on this, the second circuit shown in post#21 is not really needed.
 
Disabled Account
Joined 2017
Id only need it for higher DCR to be able to run bridged mode, to have more amp power or.
Though u guys say, it wont need 1000 watts, it would at least give my amp 10dB Headroom and run way easier, considering the phase shift might be also a little difficulty for the amp, just my thinking.
In general lowering impedance would flatten the high frequencies and the strong rolloff, then maybe only increased voltage would bring the last 5dB overall and all as good as possible? :)

Of course, given the assumption nothing is broken which should be checked before.
Does the impedance tell you so far anything bout the health of the transformer???
Regarding voltage I’ll buy resistors for divider soon and check.
 
...Does the impedance tell you so far anything bout the health of the transformer???
To me, the impedance curve looks like a healthy ESL.
The next thing to do is to measure impedance of both speakers at the same voltage level and compare. Also measure frequency response of both speakers with mic in same relative position and same voltage level. If both of these measurements match well between the two speakers, it is highly likely that your ESLs are healthy. It is unusual for two speakers to be damaged in exactly the same way.
 
Disabled Account
Joined 2017
Hi bolsert

I was thinking again... actually wanted to use a sub below, crossing the capacity active anyway. So what about skipping the parallel resistor in second proposal completely?
Id have a serial capacitor and resistor only. Thinking about 100uF and 2 Ohms.
That should give infinite high DCR that makes the bridged amps very happy for their DC offset. And dropping still to 2 Ohms in the highs.

But... seeing how the impedance in lower frequency increases with the voltage so extremely...
Wouldn’t that react with the caps, changing the response depending on the voltage level.
Respectively create a lot of IM Distortion ???

Cheers
Josh
 
If you are crossing active, why are you needing a 100uF in series with the 2 ohm?
The output required from the amps at low frequencies and DC should be minimal if active crossover is used.

You are correct that the varying impedance would react with the capacitor resulting in frequency response variations with volume level and other distortion types. The usual way to handle this is with a shunt resistor upstream of the capacitor to swamp the variations. Here is a link to the Acoustat Spectra 11 high pass filter network.
Martin Logan CLS Power Supply
 
Disabled Account
Joined 2017
Ah thats a useful link thank you!

Well the thing is, i need some resistance right, not to blow the amp.
At least 2 Ohms. On my Parasound HCA2200 II .
As I intend to use them bridged, i need to double up to 4 ohms, not to risk overheating due to DC offset.
But the resistor burns approx that energy which im adding by bridging. So the bridge is for nothing.
(yes i dont need 1k Watt, but i intend to use an amp so strong that its bored on the ESL as I always feel they sound slightly stressed)

I use quiet steep linearphase FIR filter actively and would like to settle around 200Hz as the capacitis distortion raises exactly below 200 Hz.

So, using a first order highpass Cap would block all DCR and DCoffset from the bridge amp, without burning the precious amplifier effort in a bigger resistor, still having 2 Ohm resistor behind the cap.

Well, seems like that would be too nice to be true... looking at your other post.
Im thinking now, what specs this shunt resistor should have to be still audiophile.
Being parallel to the amp/speaker, it should not disturb much the speaker i guess, even if I take a cheap highly inductive ceramic resistor or ? Should just be high power version.

Other question is if this network of caps + serial resistor + parallel resistor isnt slowly disturbing the sound more than it profits from power capacity on amp side :)

But most crucial, will it burn power? I mean something is floating through the resistor, something that the amp worked for before. So the total efficiency of dB per Watt would still be better in this setup compared to normal 2 ohm serial on non-bridged amp with half the power????

cheers
Josh
 
To avoid any sensitivity loss from shunt resistor, place it upstream of the 2 ohm series resistor.

My recollection is that the Parasound HCA-2200 and 2200 II both had DC servos, so I don't think DC offset should be a problem. If you are measuring more than 5mV of offset, there may be a problem with the servo. With an active crossover rolling off below 200Hz, you should have no problem running with just the 2 ohm series resistor and no capacitor or shunt resistance.
 

Attachments

  • ESL_HP_input.png
    ESL_HP_input.png
    14.7 KB · Views: 208
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.