Class D vs electrostatics

0 Members and 1 Guest are viewing this topic. Read 6094 times.

JLM

  • Full Member
  • Posts: 10654
  • The elephant normally IS the room
Class D vs electrostatics
« on: 26 Jul 2015, 11:27 pm »
Just read this from a manufacturer of electrostatic speakers and was curious if CIA amps (past or present) exhibit any problems with this type of driver:

"Some electrostatic speakers have very high capacitance. This raises the potential for three types of undesirable amplifier interaction:

1) All else being equal, the amplifier load from an ESL increases with frequency. High capacitance ESL's will have a very low impedance at high frequencies. When an amplifier that can not drive a low impedance is asked to source a lot of current at high frequencies, the amplifier will limit its output at all frequencies, creating compression, clipping, or some other overload effect.

2) The feedback loops in some amplifiers will not remain stable when driving a large capacitive load. If such an amplifier happens to begin oscillating only when called on to deliver high power, this will cause problems only at high loudness levels. The oscillations may be ultrasonic and thus inaudible, but the indirect, audible effects can include compression, hissing, intermodulation, and chirping, among others. It is also possible to throw an amplifier into an oscillating state called "motorboating", which sounds much like its namesake outboard engine. An amplifier may be harmed while oscillating, or just run warmer than usual.

3) Some switching (e.g., class D) amplifier designs are incompatible with ESL's. If a switching amplifier is designed without enough high frequency feedback, and no special considerations are made in the amplifier for driving ESL's or in the speaker for being driven by switching amplifiers, then the panel capacitance forms a tank with the amplifier's output filter inductance. This will resonate when excited by high frequency signal content, normally at frequencies somewhat but not far above the audio range. It can cause various sorts of distortion, and even arcing of the panels or damage to the amplifier, at sound levels well below the usual limit."

steve f

  • Full Member
  • Posts: 682
Re: Class D vs electrostatics
« Reply #1 on: 27 Jul 2015, 12:00 am »
Hi JLM,

We have a discussion on this very topic on the Solid State forum. Please join in.
I've been told by a well known ESL builder that Class D amps are best avoided for the reasons you just posted.
I wonder which Class D amps if any work well with ESL speakers. I've asked one manufacturer for his view, but no reply as of yet.
ESLs are such strange beasts that a lot of common audiophile perceptions simply don't apply.

Steve

Freo-1

Re: Class D vs electrostatics
« Reply #2 on: 27 Jul 2015, 12:19 am »
Some of the high power McIntosh amps (with autoformers) can handle ESL speakers just fine.  They have output taps down to one ohm impedance. 

The Pass Labs and Thershold amps can also handle ESL speakers rather well.

JLM

  • Full Member
  • Posts: 10654
  • The elephant normally IS the room
Re: Class D vs electrostatics
« Reply #3 on: 28 Jul 2015, 11:45 pm »
The quote I posted was from Janszen speaker website.  They are a long time builder of electrostatics (small non-dipole ones with built-in woofers).  According to the site their "properly designed" speakers aren't a problem for amps.
« Last Edit: 29 Jul 2015, 07:36 pm by JLM »

steve f

  • Full Member
  • Posts: 682
Re: Class D vs electrostatics
« Reply #4 on: 29 Jul 2015, 01:21 am »
I'm going to be using a different brand of ESL. The only Class D amps I have around are a couple of Crown XLS 1500s. They were fine with Quads. I'm afraid that item (3) may be an issue.

CIAudio

Re: Class D vs electrostatics
« Reply #5 on: 29 Jul 2015, 01:04 pm »
Our amplifiers work fine with capacitive loads due to the design of the feedback loop/output filter.
We have many customers using ESL's with great results.

steve f

  • Full Member
  • Posts: 682
Re: Class D vs electrostatics
« Reply #6 on: 29 Jul 2015, 05:14 pm »
Thank you. Would you care to elaborate about it a bit, unless its proprietary.

steve

CIAudio

Re: Class D vs electrostatics
« Reply #7 on: 29 Jul 2015, 05:29 pm »
Most class D designs take their feedback before the output filter, the Hypex UcD and Ncore take the feedback after the output filter (at the speaker terminals). This keeps the capacitance of the output filter from interacting with the speaker load capacitance.

jaywills

  • Full Member
  • Posts: 330
Re: Class D vs electrostatics
« Reply #8 on: 29 Jul 2015, 05:53 pm »
I've run my Quad 63's with both a home-brew UcD 400 (analog input) and a TacT S2150 ("equibit") (all digital signal) without difficulty or apparent problems.  No experience with other 'stats, though.

steve f

  • Full Member
  • Posts: 682
Re: Class D vs electrostatics
« Reply #9 on: 29 Jul 2015, 06:48 pm »
CIA, thank you for the explanation.

steve

steve f

  • Full Member
  • Posts: 682
Re: Class D vs electrostatics
« Reply #10 on: 29 Jul 2015, 06:59 pm »
Jaywills, I've experienced no problems with my Quad 989s either, and I got very good results with a pro amp, Crown xls1500. I'm upgrading to Sanders Sound and they are known to sound great but are one of the most difficult panels to drive. I couldn't afford a Magtech so I'm forced to consider more economical sources. I'm open to suggestions. The TL bass section will be driven by the Crowns.. I've rethought my amp philosophy and i really appreciate powerful non clipping amps. Thanks.

steve

JLM

  • Full Member
  • Posts: 10654
  • The elephant normally IS the room
Re: Class D vs electrostatics
« Reply #11 on: 29 Jul 2015, 07:37 pm »
Thanks Dusty.  Another reason to be a CIA customer.