OT's 'widget' schematics for the ESL57

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
What I could make out from OTAs description is that the widget makes an easy load for the amplifier. Some amplifiers do not tolerate the low impedance of the ESL exhibiting at high frequency, and could become instable. How is this impedance correction achived, I have no idea. But I can live without it :)
 
I have no idea, where is it placed in the circuit, that might give a clue as to what's inside of it.

It puts between the amplifier and the transformer's input
 

Attachments

  • image.jpeg
    image.jpeg
    49 KB · Views: 422
What I could make out from OTAs description is that the widget makes an easy load for the amplifier. Some amplifiers do not tolerate the low impedance of the ESL exhibiting at high frequency, and could become instable. How is this impedance correction achived, I have no idea. But I can live without it :)


My plan is to experiment with other amp than the Dynaco ST70
 
I have no idea, where is it placed in the circuit, that might give a clue as to what's inside of it.
From pics and descriptions posted on their website: One Thing Audio
  • The ESL-63 widget appears to be intended as a replacement for the RC input components.
  • The ESL-57 widget is likely designed to be placed at the input as well, but component values and PCB traces are not definitively discernible in the pic.
    With carefully selected R & C values some equalization(mentioned in the description) of the slightly sloping response of the ESL-57 should be possible.
 

Attachments

  • ESL63_RCinput.png
    ESL63_RCinput.png
    651.4 KB · Views: 847
  • Widget_ESL63.png
    Widget_ESL63.png
    295.4 KB · Views: 424
  • Widget_ESL57.png
    Widget_ESL57.png
    232.1 KB · Views: 539
Thank you for your time!


From pics and descriptions posted on their website: One Thing Audio
  • The ESL-63 widget appears to be intended as a replacement for the RC input components.
  • The ESL-57 widget is likely designed to be placed at the input as well, but component values and PCB traces are not definitively discernible in the pic.
    With carefully selected R & C values some equalization(mentioned in the description) of the slightly sloping response of the ESL-57 should be possible.
 
That input provides phase shift and a little response compensation. The impedance of the RC parallel pair is about 1.4 ohms at 20Hz and falls to essentially a short at 20KHz. That is is series with the transformer and panels and such, which also have a falling impedance (more or less). That's also a trick in cone and dome speaker crossovers.

In my humble opinion, that circuit has no place on the original ESL.

As an aside, adding about 10uF of film capacitor in parallel with that 220uF electrolytic really helps the sound of the 63's. for the 988's and later there is room under the hood to replace that electrolytic entirely with an enormous solen film cap. But the 63's are a bit cramped for that, and placement is nasty.

Sheldon
quadesl.com
 
0D6EEC65-5219-4FDE-B1F0-73376C2A4A0E.jpeg
(Quad 988)
Yes very tight fit. I purchased 5 - 200 uf Non polars to match them along with the wonder cap matching and got rid of the upstream fuse FS2

The stock 220 uf NP cap were quite a bit higher in value when pullled and measured.

I noticed Bennic makes a 220 uf PP film cap also along with the rest. (Solen Etc )
Very pricy and huge in size and can see why back to back electrolytics were used.
I might make up a board to see if there is a sonic difference worth pursuing


Regards
David
 
Many thanks for the informed responses.

Does the 10uF mod still have effect if you change to a 220 uF film cap ?

What effect does removing the posistor FS2 have ?

Have you heard that Quad are to release a UK made 8 panel ESL, which was unveiled at Munich.

I believe that someone has already made such a beast by combining 2x 63's.

I have 989's which I run naked, they are due a strip down to clean the panels, so increasing frame rigidity and maybe adding a couple of extra panels might be added to the to do list...
 
Many thanks for the informed responses.

Does the 10uF mod still have effect if you change to a 220 uF film cap ?

What effect does removing the posistor FS2 have ?

Have you heard that Quad are to release a UK made 8 panel ESL, which was unveiled at Munich.

I believe that someone has already made such a beast by combining 2x 63's.

I have 989's which I run naked, they are due a strip down to clean the panels, so increasing frame rigidity and maybe adding a couple of extra panels might be added to the to do list...

No, the 10uF is the rational and sane thing to do rather than change out the entire 220uF bipolar electrolytic to a film cap. Changing the whole thing out is the sort of craziness that causes a person to have quads in the first place.

The thermal fuse has very low but not zero resistance. removing it (or wiring across it is an easier solution) won't have much of an effect on the speaker response. I haven't noticed a big difference in sound quality when I shorted across mine. I currently have them in place, but I can respect people removing them as well, as not having them there is bound to be an improvement. But it is part of the protection of the panels and electronics, so remove them at your own peril.

I've heard about the new quads, I don't have any details. I have a pair of the 2912's and they finally fixed the glue problem, but the diaphragms still have a "rot" problem that started when manufacture shifted to china. my pair is 5 years old and has had 5 panels go bad. Very little has changed over the years electrically on these speakers, it's like they don't have anybody who fully understands them anymore. So I'm not expecting any real changes in the design, just a repackaging.

I'm skeptical of stacked or paired modern quads, because the concentric rings don't lend themselves to more than two panels. Where stacking the original quads make sense since it's a vertical line source that you are just extending.

Honestly, I find the 6 panel quads to not be as neutral and balanced as the 4 panel ones, but that could be a function of my room too.

The 989's have the most floppy frames of the bunch. The panels are attached to two vertical thin aluminum angles. These are held at the top and bottom of the frames. Because they are much longer than the 988's (or 63's), the are really floppy. They are so floppy that the panels can move out so far (in shipping) that they can hit the dust covers.
I use a 1/2" long standoff and screw the panel mounting rails to the speaker sides halfway up, and then the panels are really rigidly coupled to the frames. It makes a significant difference, but you have to remove the middle panels to do it, so it's labor intensive if you don't already have the panels out for repairs. If people are interested in this mod, I'll write it up and put it on my blog.

Sheldon
quadesl.com
 
Last edited:
View attachment 688743
(Quad 988)
Yes very tight fit. I purchased 5 - 200 uf Non polars to match them along with the wonder cap matching and got rid of the upstream fuse FS2

The stock 220 uf NP cap were quite a bit higher in value when pullled and measured.

I noticed Bennic makes a 220 uf PP film cap also along with the rest. (Solen Etc )
Very pricy and huge in size and can see why back to back electrolytics were used.
I might make up a board to see if there is a sonic difference worth pursuing


Regards
David


Dave, looking at your picture reminded me of another thing. The 1000uF cap that is the filter cap for the arcing clamp circuit is only rated at 16v. If you measure the DC of that supply it will be right at 16V with ripples over that. This causes those caps to fail over time. I replace them with 25V units. If the cap fails, the clamp circuit becomes hyper sensitive. It's a good thing to swap out next time you are in there.

Sheldon
quadesl.com
 
Yes, in that pic the 1000uf cap is a Panasonic FC 25 v replacement.

Those aluminum “ L “ mounting legs are mounted in rubber ends and isolate the panels from the really “ erector set “ cheap frame construction. One might want to approach coupling the panels too rigidly to it with respect as to not excite a resonance and make it vibrate.
The L mounting leg should be replaced with a thicker one easily sourced.
All the panels should be double sided taped together and clamped tightly and then screwed to the end piece.
I replaced all the perimeter foam with closed cell dense replacements and sandwiched the ends with long sticky versions and stuck and compressed the side frame when screwing the outside frame together ( visions of aluminum window frame construction , sshhsh)
The 57 is actually a more rigidly constructed speaker in comparison ,...

The 988 and even worse with the 989, the added length that Sheldon so rightly points out makes for increased vibrations and the additional 2 panels of the 989 now couple bass frequencys to the floor that muddy the midrange as many have noted.
You do get the point source near ear level, but as mentioned clarity suffers.
Quads will benefit with longer panel life when all the low end vibes are keep to a subwoofer ( 20 hz and Mylar are Acoustat territory and not Quads )
This will lessen the frame vibrations and the speaker will produce less distortion and sound cleaner and more transparent 422AE8F1-4072-4967-B360-32F3518F33BD.jpeg

34C087D3-EEBB-499A-80CE-EAFA0A80C82C.jpeg overall.
I did machine and add a carbon fiber sand filled pole like the 28/9 series use as the flexible nature of the plastic base interface is not rigid enough by itself.
Some pics included

Regards
David
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.