Ideal bias supply?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi,

I can recommend using a conventional Cockroft-Walton-kaskade with a neon oscillator at the output. The neonoscillator is made of a simple neonbulb + resistor with a capacitor parallel. It works like a filter and you can see the amount of charge running to the membrane by the frequency of the neon bulb going on and off.
No need for high power here.
 
A good ESL panel (IE low leakage) should be able to run for many minutes without a bias supply at all. My DIY panels, the Quads, and all the ESL's I've played with play perfectly fine for minutes/hours with the plug pulled after they charge up. There is a slow decrease in SPL as they discharge, but over a few minutes, we're talking fractional dB's.

So given that the bias supply charges the panel through a big-honkin (tm) resistor, and that charging and discharging frequency is closer to geological frequencies than audio frequencies, the bias supply plays a remarkably small roll in overal sound quality of the speaker.

I get a lot of folks asking about what power cables they should use for the original ESL's. I suggest that they pull the plugs and listen and see if they hear any difference. When they don't they should take that "power cord" money and go buy more music.

All that said, the bias supply should be noise free, which implies adequite filtering. I designed a switching supply that works up about 10KHz, which means that I can use slightly smaller caps than a standard 60 Hz (or 50Hz) supply and get way way way better noise filtering because my noise is many octaves above the 120 Hz noise from a sttandard linear supply run from that wall.


Sheldon
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.