Planar Speaker Asylum

Welcome! Need support, you got it. Or share your ideas and experiences.

Return to Planar Speaker Asylum

Planar Speaker Asylum

Service manual/schematic for Quad ESL-988 + modding tips?

80.235.39.51

Zombie

Any of the above would be welcome. Especially how to lower the HT back to ESL-63 levels.

Follow Ups:

Cardas Rhodium binding post upgrade - mashley 13:43:52 09/16/12 (0)
All Quad owners. Replacing the binding posts to Cardas Rhodium is quite a substantial tweak in itself.

Kentaja suggested this to me some time back, quite frankly I was floored.

They are real buggers to solder though, post here first before attempting or you'll learn the hard way.
RE: Service manual/schematic for Quad ESL-988 + modding tips? - kentaja 06:49:09 09/16/12 (10)
There is no service manual for the Chinese made Quads, but the circuit is identical to the ESL-63. There are any number of on-line resources for the schematic for the ESL-63. Sorry I don't have a link handy.
The motherboard of the ESL98 has component numbers silk screened on the PCB. Changing the value of R13 will adjust the output of the bias supply. Quad raised the bias voltage to improve the efficiency of the speaker in an attempt to make them play louder. It certainly made them play louder for static testing but in practice it decreased actual output before charge migration. Lowering the efficiency slightly will actually increase speaker output before charge migration, reduce leakage rate of the panels and increase panel life.
This change was actually done during late ESL-63 production so we can't blame the Chinese for this mistake. They simply copied the last production ESL-63 bad ideas and all.
A R13 value of 82K is about right. Measuring the bias voltage first is a good idea since the Chinese power transformer secondary voltages are all over the place. I aim for a bias voltage of around 4.7K-4.8K. I have seen very early ESL-63 production as low as 4.5K.
Improving the quality of the passive audio components will improve the speaker. This would include the RC network components R15 & C25. Eliminate FS2. This was a thermistor added late in ESL-63 production to keep small wattage amp clipping from damaging input components. Another bad idea IMO. Thermistors sound like crap and I don't want such a thing in the audio path. The better idea is to simply use an appropriate size amp that won't be clipping all the time.
While you are in there and want to go all the way replace R1A+B, R2A+B, C26. Better binding posts will help things. Cardas Rhodium posts are my choice.
All these changes will make the speaker sound much better, make them play louder and will increase panel service life. Given the way the Chinese execute the electronics all this work is a pain but well worth the effort.

Edits: 09/16/12
- RE: Service manual/schematic for Quad ESL-988 + modding tips? - Zombie 09:03:41 09/16/12 (9)
  Hello and thanks for the input.
  Have I understood you right:
  According to the schematic the EHT is 5.25 kV and 4.7-4.8 kV would be more appropriate? R13 is 56 kohm, but should be changed to 82k or a value that gets close to 4.7-4.8 kV?
  
  With what do I replace the R1A+B and R2A+B and C26? (Can't find C26 on the schematic I have - dated 5-10-89).
  - RE: Service manual/schematic for Quad ESL-988 + modding tips? - kentaja 09:33:58 09/16/12 (8)
    Yes the schematic states 5.25kV but they never actually polarized the panels quite that high. Closer to 5.1kV. 4.7-4.8kV will drop the efficiency by aboutt 1 - 1.5dB. Provided the secondary voltage of the transformer is 670V 82K will get the bias voltage around 4.7-4.8kV. I have seen secondary voltages as high as 700V on 98 speakers. In that case I would make the R13 value around 100K-120K.
    Dropping the voltage as low as possible while still maintaining good output without clipping the amp is key. I have chosen 4.7-4.8kV as the target for amps with 50-100 watts output which is what the typical Quad owner will use. I personally use 300 watts and drop the voltage closer to 4.5kV.
    Later ESL-63 schematic and the 98 schematic have an error omitting C26 but it is in the speaker. C26 is right next to resistors R1A,B and R2A,B on the PCB.
    C26 is 1.5uF/50V. I typically replace with a Mundorf MKP of the appropriate value. For the resistors I have used many different types depending on budget. I personally use Caddock metal films with heat sinks but standard Dale/Vishay RS-5 work nicely as well. Resistors are 3.3ohm and should be a minimum of 5 watts.
    Like anything tastes will vary on passive components and other equipment in the chain. Just about anything will sound better than the stock components.
    
    Edits: 09/16/12 09/16/12
    - RE: Service manual/schematic for Quad ESL-988 + modding tips? - Zombie 11:54:48 09/17/12 (1)
      I use a Quad 909, so 4,7 kV would OK or should I go lower, to 4,5 kV, maybe?
      
      Edits: 09/17/12
      - RE: Service manual/schematic for Quad ESL-988 + modding tips? - kentaja 12:20:04 09/17/12 (0)
        
        In Reply to: RE: Service manual/schematic for Quad ESL-988 + modding tips? posted by Zombie on September 17, 2012 at 11:54:48
        
        4.7kV will be fine. You only need to go lower if you enter into the insanity zone like I have!
    - RE: Service manual/schematic for Quad ESL-988 + modding tips? - Zombie 10:52:12 09/16/12 (5)
      Do you mean 670 V between the secondaries before the resistor "bridge"?
      What do you think about this: http://www.keith-snook.info/Loud-Speakers/QUAD-ESL63/QUAD-ESL63.html
      Here the cap is referred to as C14. Any other thoughts about doubling the 3R3 and the other suggestions in the article?
      
      Edits: 09/16/12
      - RE: Service manual/schematic for Quad ESL-988 + modding tips? - kentaja 13:11:48 09/16/12 (4)
        
        In Reply to: RE: Service manual/schematic for Quad ESL-988 + modding tips? posted by Zombie on September 16, 2012 at 10:52:12
        
        Yes the voltage should ideally be 670V before the resistor bridge. Measure across the yellow secondary wires that attach to the PCB.
        There are two 3R3 resistors in parallel so the resistance is actually 1R65. Two resistors were used to increase power handling and keep the resistors small. A single 10watt 1R65 would be sufficient. I would recommend that the stock value be maintained. If one has an amp with variable output impedance the value can be changed or the resistors may be eliminated. Not many amps with variable output impedance. C14 is the coupling capacitor for the transformers. On the 98 PCB this cap is marked C26. The value is 1.5uF/100V.
        
        I have seen the article before.
        The problem he notes in the panel is not an issue. This is simply the conductive coating on the membrane. Sometimes they can be a bit sloppy when applying the conductive coating but this does not affect panel performance in any way. I have seen far worse!
        Removing the RC network filter is an old tweak that started in the ESL-63 days. Some like the sound this way I don't. It changes the entire balance of the speaker and moves it in the wrong direction IMO. If someone prefers the speaker that way that is OK but I like the performance of the speaker with the RC network in place even with just the stock components.
        As Rob Flain pointed out the 1.5 ohm resistor of the filter provides a load for the amplifier if the speaker should go into protect mode. If the speaker should clamp without the resistor in place the amplifier will see a dead short. This momentary short may not present a problem with a tube amp or solid state amp with an appropriate protection circuit but why take the chance?
        I do think removing FS2 is a good idea. It provides little practical protection and does degrade performance. It was implemented late in the production of the ESL-63 and carried over to the current speakers. I routinely eliminate FS2 with late production ESL-63 and all new production speakers. I don't use FS2 on the new ESL-63 boards that I produce. I have never had any reports of problems with FS2 out of the circuit.
        I have produced new delay line assemblies for the speaker including better caps, Teflon in this case. Testing is under way and several units are in the field for evaluation.
        As great as the Quad sounds stock careful tweaking makes them even better. PJW knew what he was doing, everything is there for a reason and I believe maintaining the overall circuit values, etc., produces the best results.
        
        Edits: 09/16/12 09/16/12 09/16/12
        
        RE: Service manual/schematic for Quad ESL-988 + modding tips? - Zombie 23:27:16 09/17/12 (2)
        
        In Reply to: RE: Service manual/schematic for Quad ESL-988 + modding tips? posted by kentaja on September 16, 2012 at 13:11:48
        
        Would be interesting to get a reminder when the tests are evaluated.
        
        RE: Service manual/schematic for Quad ESL-988 + modding tips? - kentaja 07:47:48 09/18/12 (1)
        
        In Reply to: RE: Service manual/schematic for Quad ESL-988 + modding tips? posted by Zombie on September 17, 2012 at 23:27:16
        
        Drop me a private email in a month and I will be happy to share the results.
        
        RE: Service manual/schematic for Quad ESL-988 + modding tips? - Zombie 01:21:31 08/23/19 (0)
        
        In Reply to: RE: Service manual/schematic for Quad ESL-988 + modding tips? posted by 4120 on September 18, 2012 at 07:47:48
        
        Hello! How did the test go? A long time has passed :)
        
        RE: Service manual/schematic for Quad ESL-988 + modding tips? - kentaja 19:26:13 09/16/12 (0)
        
        In Reply to: RE: Service manual/schematic for Quad ESL-988 + modding tips? posted by kentaja on September 16, 2012 at 13:11:48
        
        A follow up. I read the article again and the comments about the panels catching fire is rather silly. I suppose in theory it is possible but highly unlikely I would say virtually impossible.
        
        I have done experiments defeating the sniffer circuit and let the panels spark away. Thousands of cycles of sparking. Have they ever caught on fire? Nope. Of course the panel looks like Swiss cheese afterwards!
        
        Is there a scenario where the Mylar might ignite and catch on fire? Yes. I produced a panel that dropped the surface resistance well below what would be acceptable for stable performance. In this experiment the Mylar did catch on fire. Talk about cool! Took several seconds for the Mylar to completely burn away. There was no damage to the stator assembly it was easily rebuilt and the panel worked fine again.
        
        What can happen without FS2 in place is someone uses a very small wattage amp and just clips the snot out of thing. How can someone listen to it that way? Can't they hear the amplifier clipping? What will happen when abused this why is the R1, R2 resistors will fry burning the PCB and making quite a mess but this take perhaps 10-15 minutes of constant continuous clipping. The board is a charred mess and has to be replaced. But I seriously doubt the speaker is going to catch on fire.
        
        I have never seen another electrostat that the maker is concerned with the speaker catching on fire and adds protection circuits to stop such a thing. This is really grasping at straws.
        
        If someone clips an amp in this fashion and continues to pound on the speaker and abuse the thing in this way they deserve the damage that is caused and the repair bill that will follow. I see no reason to degrade speaker performance to stop this kind of abuse.

Post a Followup:

FAQ

Post a Message!

Forgot Password?

Moniker (Username):
Password (Optional):
	Remember my Moniker & Password (What's this?) Eat Me
E-Mail (Optional):
Subject:

Message:	(Posts are subject to Content Rules)

Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:	Automagically notify you when someone responds.

To view your new posting or follow-up, click on the RELOAD or REFRESH button on your browser.

This post is made possible by the generous support of people like you and our sponsors:

Atma-Sphere Music Systems, Inc.