Ground reference for ESLs

[bolserst]

Originally Posted by arend-jan

I don't know the answer to that. But I'm willing to take a guess.

Perhaps there were problems with the reliability of the spark detector? It seems preferable to have the spark detector referenced to the high voltage section in a predictable manner, not just by parasitics and whatever amplifier is hooked up.

Interesting thought...certainly the spark detector would be more sensitive to the RF energy emitted during arcing if the detector was referenced to part to the circuit through which the spark would conduct. However, several other ESL companies use the later arrangement where the primary and secondary are tied together and do not have a spark detector circuit to concern themselves with. Perhaps they just followed Quad's lead without much thought put into it? Wouldn't be the first time circuit arrangments were borrowed and used without fully understanding them.

In the later models the input section and the high voltage section are referenced to the same internal ground. Which means connecting it so safety earth would connect one of the speaker input terminals to safety earth. Not all amplifiers are going to handle that gracefully so that might be the reason to keep the whole speaker floating?

Interesting that you should mention this since I just recently read on the ML forum that some models(the Quest was mentioned) had this arrangement and would destroy the output stages of balanced power amplifiers that were connected to safety ground through their own power cord.

[tyu]

Bolserst
Thanks for the Quad info...looks like thay were using the {limiter} way back whin...What is that a NPN..PNP..no fet??...an how about that Cap..1.5uf
nothing is new....

[bolserst]

Originally Posted by tyu

Bolserst
Thanks for the Quad info...looks like thay were using the {limiter} way back whin...What is that a NPN..PNP..no fet??...an how about that Cap..1.5uf
nothing is new....

Power BJT technology was much more mature back in the 60s than power FET. I'm sure that weighed in to the decision to use a BJT device for the limiter rather than a MOSFET. Most likely the 1.5uF cap is there to filter out the voltage transients from when the NPN transistor attempts to turns OFF after limiting a voltage peak.

Yes it does seem sometimes that we keep re-discovering old ideas.
This seems to be the case when technological advances are considered proprietary and are not patented or published.

[john65b]

My Icepower amp, for instance, does not have the amp speaker ground referenced to ground, so forcing it to ground could destroy the amp. I have my ESL floating...

Wouldn't this also apply to any amp that is bridged?

[bolserst]

Originally Posted by john65b

My Icepower amp, for instance, does not have the amp speaker ground referenced to ground, so forcing it to ground could destroy the amp. I have my ESL floating...

Wouldn't this also apply to any amp that is bridged?

Absolutely.
Connecting safety ground to the (-) speaker input terminal on an ESL is just not a good idea.