• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

The Eros Clone Electrostatic Speaker

Joined
May 15, 2019
Messages
24
Likes
46
#1
Hi All,

The exquisitely crafted speakers shown below were built by my online collaborator, Mervyn Tims.

I have a website (Jazzman's DIY Electrostatic Loudspeaker Page) dedicated to building electrostatic speakers, and I often receive requests for drawings of my (3) ESL designs. Most people take one look at my drawings and cringe at the amount work involved, and that's as far as they get. But every now and then someone surprises me and actually builds one. Such was the case with Mervyn Tims, and his beautiful new DIY speakers, which I believe are the only pair of their kind in existence.

I've only built two of my designs, so I was pleased to finally see the Eros Clone design actually built. It's called the 'Eros Clone' because it's compact transmission line woofer system was patterned on Roger Sanders' Eros speaker and it's overall design is similar except for the wire stators.

Mervyn has given me permission to post the pics below, and I'm sure you will be as impressed as I am with his craftsmanship. In the second photo, note that the cover is removed from the pocket that houses the segmentation resistor network (Mervyn's modification, and very neatly done).

Enjoy!
merv 1.jpg
merv 2.jpg
 
Last edited:
Joined
Mar 25, 2019
Messages
13
Likes
7
#2
I have read your ESL page and I hope to try to make my own sometime. I have a question: is the diaphragm adhered at the top and bottom in addition to the sides and vertical supports?
 
Joined
May 15, 2019
Messages
24
Likes
46
#3
I have read your ESL page and I hope to try to make my own sometime. I have a question: is the diaphragm adhered at the top and bottom in addition to the sides and vertical supports?
Yes, the diaphragm is evenly tensioned in all directions and adhered to the stator frames at top/bottom, sides, and vertical supports.
 

Eurasian

Member
Patreon Donor
Joined
Nov 2, 2018
Messages
93
Likes
52
#4
I have a pair of original Eros! In the tiny sweet-spot/line they're excellent with great detail and fantastic imaging... everywhere else, not so much. These things are crazy directional!

I'm sure your segmented stats eliminate or reduce the centerline beaming of the originals -- I would love to hear a pair.
 
Joined
May 15, 2019
Messages
24
Likes
46
#5
I have a pair of original Eros! In the tiny sweet-spot/line they're excellent with great detail and fantastic imaging... everywhere else, not so much. These things are crazy directional!

I'm sure your segmented stats eliminate or reduce the centerline beaming of the originals -- I would love to hear a pair.
My first ESLs used unsegmented flat panels, and their imagining was superior to any curved or segmented panel. Beaming isn't a flaw-- but it is inconvenient when company drops in.

I tried to have it both ways once, by building a panel with switch- selectable wide and narrow dispersion modes. This was accomplished with a multi-pole rotary switch that could jump over (disable) or leave engaged the segmentation resistors. It was fun to play with for a while but the novelty wore off pretty quickly because I had to power down to make the switch (else the high voltage would fry the switch) and then re-EQ the panel for each mode.
 
Last edited:

Eurasian

Member
Patreon Donor
Joined
Nov 2, 2018
Messages
93
Likes
52
#6
My first ESLs used unsegmented flat panels, and their imagining was superior to any curved or segmented panel. Beaming isn't a flaw-- but it is inconvenient when company drops in.
True about company, but my issue with them is that I like to listen from several locations in my room separated by distance and large angles. For this reason, I am going to wide, but controlled, dispersion loudspeakers.
 
Joined
Mar 25, 2019
Messages
13
Likes
7
#7
Yes, the diaphragm is evenly tensioned in all directions and adhered to the stator frames at top/bottom, sides, and vertical supports.
Got it. If I was making a set I might consider scaling down a little bit as your design would be a little imposing in my small apartment. More specifically, I would likely scale the height by 60-75% and position them for listening on a couch. My concern with that plan is that if the vertical tension was kept the same but the height shortened then the diaphragm might not have the same freedom of movement and affect the frequency response. The options I could think of are:
1. Don't adhere the diaphragm at the top and bottom.
2. Adhere at the top and bottom, tension the horizontal dimension by 1%, but not in the vertical dimension.
3. Adhere at the top and bottom, tension in all directions, but by an amount less than 1%.

Options 1 & 2 seem like they would have similar results, but option 1 would produce a more cylindrical wavefront with little sound energy above and below the top and bottom of the speaker. Option 3 seems like a challenge to figure out the correct stretching amount without trial and error. Do you have any thoughts?
 
Joined
May 15, 2019
Messages
24
Likes
46
#8
Got it. If I was making a set I might consider scaling down a little bit as your design would be a little imposing in my small apartment. More specifically, I would likely scale the height by 60-75% and position them for listening on a couch. My concern with that plan is that if the vertical tension was kept the same but the height shortened then the diaphragm might not have the same freedom of movement and affect the frequency response. The options I could think of are:
1. Don't adhere the diaphragm at the top and bottom.
2. Adhere at the top and bottom, tension the horizontal dimension by 1%, but not in the vertical dimension.
3. Adhere at the top and bottom, tension in all directions, but by an amount less than 1%.

Options 1 & 2 seem like they would have similar results, but option 1 would produce a more cylindrical wavefront with little sound energy above and below the top and bottom of the speaker. Option 3 seems like a challenge to figure out the correct stretching amount without trial and error. Do you have any thoughts?
I can tell you that 1% elongation is not universal-- it's just the appropriate amount of tension for the specific spans between diaphragm supports used in my panels. It's the diaphragm's drum head resonance that is determined by the diaphragm tension and span between supports. For a given elongation/tension; the resonance frequency drops if the span between supports increases, and visa-versa. Also; the drum head resonance effectively sets a lower limit for the crossover frequency because you need to set the crossover frequency at least one octave above the drum head resonance using a steep crossover slope... so as not to excite the resonance (the drum head resonance is quite loud and we want to avoid exciting it)

I haven't given a lot of thought to tensioning the diaphragm in only one direction. With a curved panel you have to tension the diaphragm in one direction only, to avoid pulling the diaphragm into the rear stator, but a flat panel can be tensioned in all directions. Modifying the tension directions and/or anchoring edges would certainly affect the diaphragm's resonance modes. You could make a whole research project out of that alone.
 

LSB

Member
Joined
Jul 22, 2018
Messages
10
Likes
4
#9
My first ESLs used unsegmented flat panels, and their imagining was superior to any curved or segmented panel. Beaming isn't a flaw-- but it is inconvenient when company drops in.

I tried to have it both ways once, by building a panel with switch- selectable wide and narrow dispersion modes. This was accomplished with a multi-pole rotary switch that could jump over (disable) or leave engaged the segmentation resistors. It was fun to play with for a while but the novelty wore off pretty quickly because I had to power down to make the switch (else the high voltage would fry the switch) and then re-EQ the panel for each mode.
Those ESL look beautiful. Very nice design in so many technical details.

Please can you write a bit more about your expereince with the sound quality and imaging of the wide versus narrow dispersion modes? I like listening locked right in the sweet spot of any stereo system and do all my serious listening alone. I think you might be the only one with the same ESL panels that can switch from narrow to wide so the only one with hard won experience. If unsegmented panels actually sound better it will save me loads of work soldering. Thanks.
 
Joined
May 15, 2019
Messages
24
Likes
46
#10
Those ESL look beautiful. Very nice design in so many technical details.

Please can you write a bit more about your expereince with the sound quality and imaging of the wide versus narrow dispersion modes? I like listening locked right in the sweet spot of any stereo system and do all my serious listening alone. I think you might be the only one with the same ESL panels that can switch from narrow to wide so the only one with hard won experience. If unsegmented panels actually sound better it will save me loads of work soldering. Thanks.
The link below shows my old welding rod panels which had the switch-selectable wide and narrow dispersion modes:
http://jazzman-esl-page.blogspot.com/2008/03/a-segmented-wire-stator-esl-with.html?m=1

My website explains how segmentation works so I won’t go into it here, except to note that a panel’s dispersion pattern can be tailored from narrow to wide or by various degrees in between, depending on the segmentation scheme one chooses.

If you’ve never heard a pair of wide, flat, unsegmented panels, there is really no way to describe just how directional and intense they sound at their point of focus. You have to hear it to believe it. And the sonic character is very pure because you’re hearing predominantly direct sound from the panels and little or no secondary arrivals reflected from room surfaces. The imaging is fantastic and addictive—but there is no free lunch, as I will describe next.

Presumably; you will have tuned the frequency response into balance at the focus, so if you move your head only a few inches outside the tightly focused sweet spot, the highs literally fall off a cliff. It’s as if your head were locked in a vise (a.k.a. the ‘head-in-a-vise’ effect). While such beaming is great for solo listening at the focus, it’s not so good anywhere else in the room, and not good at all for entertaining guests.

Segmentation completely eliminates the head-in-a-vise effect and works extremely well, but the wider dispersion does take away some of that magical imaging and purity that a big beamy flat panel provides.

As regards adding the switch mode feature; you could use 2-position (on/off) multi-pole rotary switches, preferably the ‘make-before-break’ type, and wire it up so that “ON” jumps the resistors for narrow mode operation and “OFF” leaves the resistors engaged for wide mode. And if you really wanted to get creative, you could stack on a secondary switching scheme to merge wire groups and/or change resistor values to achieve some intermediate level of dispersion. The appropriate switches would be quite expensive—I lucked up finding a pair of surplus high quality Soviet military rotary switches on Ebay for cheap but deals like that are hard to find.

Of course; the switches would have to be on the high-voltage side of the transformer(s), which would make them susceptible to being destroyed by arcing if the switching occurred while playing music. So, you would have to power down to make the switch. Submerging the switches in linseed oil would probably restrain the arcing enough to allow switching modes with the panels playing.

Beaming gives unsegmented flat panels a characteristic rising frequency response which requires EQ’ing to restore the frequency balance. Whereas; a uniformly segmented flat panel has a naturally flat frequency response, all the way down to its dipole roll off frequency. That being the case; switching modes also requires re- EQ’ing the panel each time. This hassle could be reduced by using a digital EQ with recall memory pre-sets. Even so; for me, the novelty of the switch mode feature wore off pretty quickly and I found myself just leaving it in wide mode and adding some acoustic treatments in my listening room to tamp down the first reflections.

When you build a pair of ESLs you want others to hear them, and I think in the end most ESL builders who initially built un-segmented panels find themselves wishing for a wider sweet spot. Now that I have segmented wire panels (much more reliable than easier to build perf metal panels, BTW), I'm not going back.

Regards,
Charlie
 
Last edited:

LSB

Member
Joined
Jul 22, 2018
Messages
10
Likes
4
#11
Hi Charlie
Thanks for your generous advice on electrical segmenting of wire ESL with your last post and link to your website. That's really cool having a switch.
Im looking around for a high voltage switch but you got lucky. Noting on eBay today.
 
Joined
May 15, 2019
Messages
24
Likes
46
#12
Hi Charlie
Thanks for your generous advice on electrical segmenting of wire ESL with your last post and link to your website. That's really cool having a switch.
Im looking around for a high voltage switch but you got lucky. Noting on eBay today.
It's unlikely you would find a switch with the required number of poles and also a high enough voltage rating to switch modes while the panel is playing. If you're handy you could design and build your own multi-pole sliding switch -- or you could simply just jump the resistors with alligator-clip jumper wires (with the panel not playing).

As I explained earlier; the novelty of the switch mode feature wore off after a while-- so I opted to omit it in my newest segmented wire panels.
 
Last edited:

LSB

Member
Joined
Jul 22, 2018
Messages
10
Likes
4
#13
It's unlikely you would find a switch with the required number of poles and also a high enough voltage rating to switch modes while the panel is playing. If you're handy you could design and build your own multi-pole sliding switch -- or you could simply just jump the resistors with alligator-clip jumper wires (with the panel not playing).

As I explained earlier; the novelty of the switch mode feature wore off after a while-- so I opted to omit it in my newest segmented wire panels.
Thanks for the great advice. I do like simplicity so wont need a switch.

It would be great if the Eros Clone sound and vision could be put up on Youtube at some point. Do you think you will hear the Eros Clone sometime?
 
Joined
May 15, 2019
Messages
24
Likes
46
#14
Thanks for the great advice. I do like simplicity so wont need a switch.

It would be great if the Eros Clone sound and vision could be put up on Youtube at some point. Do you think you will hear the Eros Clone sometime?
I would love to hear them myself. And I will drop Mervyn a hint about a Youtube. Still; we would only be hearing a lap top speaker or ear buds or whatever device we were using to play the Youtube.

I suppose I could ask Mervyn to invite me over for a listen but I'n not even sure he lives in the US. We've never met; only exchanged a few emails in reply to his questions.
 
Top Bottom