Sub for Electrostats

[CharlieM]

These work for me:

Jazzman's DIY Electrostatic Loudspeaker Page

 

[bentoronto]

Beautiful. You've even done magic with that wild wood grain.

B.

 

[CharlieM]

Thanks... it's my special stain mix for oak: Natural oil stain tinted with powdered turmeric root 🙂

 

[61051]

I am currently using the Gradient SW63 with the Quad ESL. The woofers are tired and the XO unit is getting noisy. I began looking at re-driver'ing the Gradient but then started considering other possibilities. Surely in the past 30 years there are better approaches that I can put four new drivers into.

The Gradient kept up well with the Quad. I have another set of subs, the Dayton RSS315, which put out plenty, but just don't sync with the speed of the Quad. And they start to drop fast after 40HzView attachment 1099889View attachment 1099890.

I like the concept of the dipole and have been looking at a number of H, W, Z, V etc. configurations. The Linkwitz design for his LX521 seems to have some interesting advantages but I'd really prefer to avoid the active equalization.

In the end, I'm open to whatever the best solution is- I'm not married to any particular approach.

Thanks in advance for the suggestions and guidance.

FWIW, I changed all the drivers in my Gradients for Peerless SLS - 830669, as these were cost effective and I couldn't find anyone to re foam the originals. Their mounting hole footprint was different. There were also slightly deeper than the originals and fouled the grilles, so I removed these - see photo. Official replacements for the originals are also obsolete, The Contrapunt (?) Dipole site has some other replacements listed. There was also a guy (on here, I think) who did quite a bit of work sourcing suitable replacements - I got these from his suggestions.
Best
Joe

 

[61051]

These are the ones in service....

 

[jan.didden]

I was lucky to find a local outfit who refurbished my woofers, replaced the wheathered foam surrounds.
But I still haven't finished rebuilding the Gradients ...
Too many projects.

Jan

 

[61051]

Yes indeed, I still have the originals stored in the Peerless boxes, just in case I find someone here.

 

[jan.didden]

Trying to find the original circuit for those Gradient subs, it was a line-level filter.
I have seen it here, I believe posted by Patrick Turner (r.i.p.), but can't find it back.
Anybody has it?

thanks,

Jan

 

[mkc]

Hi Jan,

I have made this. It's not complete. Let me know if you need more. I have pictures of the PCB which I reversed.

Gradient filter

Mogens

 

[jan.didden]

Excellent - thank you Mogens! Is there a more readable version of the graphs?
I plan to implement it in DSP.
Which part is not complete?

Jan

 

[mkc]

Hi Jan,

I only drew one channel and so on. I have the Kicad project if you can read that. I made the curves using LTSpice so I should have that asc-file somewhere. Let me have a look and I will get back to you.

Mogens

 

[jan.didden]

I'm not using KiCad, but that's OK, I can read the schematic.
The LTspice .asc would be very welcome.
I can then match those curves with the DSP curves.

Jan

 

[mkc]

Hi Jan,

I can unfortunately not locate the LTspice file. I have zipped the Kicad project and the jpeg-files of the actual board. I hope this is of some use to you.

Kicad project

Btw. Looks forward to your write-up of your DD-amp for the ESL-63

Cheers,
Mogens

 

[jan.didden]

No worries, I'll just enter it into LTspice.

I'm doing a presentation and demo of my DD amp on Nov 12 at the European Triode Festival.
It'll be on the street then ;-)

Jan

 

[jan.didden]

Mogens, if I can bother you one more time, I entered the circuit in LTspice but I get a weird response.
Unfortunately I cannot read the graphs on your schematic.
Could you please take a look, did I f*ck up?

Jan

 

[mkc]

Hi Jan,

I had a quick look at the schematic and it looks fine to me. I found the screen shoot of the graphs I used in the schematic. I will try out your LTspice file,

Graphs

Mogens

 

[jan.didden]

Thanks. It looks like my time constants are off ...
Back to LTspice!

Jan

 

[mkc]

Well, it could just as well be me who has made an error in the LTspice sim or even the schematic. So please be aware.

Mogens

 

[mkc]

Hi Jan,

It seems I have made an mistake in my schematic. C11 in my schematic, C5 in your LTspice file should be 15n.

Mogens

 

[jan.didden]

OK, thanks, I'll check it out.

Jan