[slowburn]

Hello All,

A few years ago I homebrewed Mr. Hag's fantastic Cornet using octals. I love the Cornet's sound (10,000 hours!) but I feel like trying a constant current source on the cathode followers.

The suggested CCS circuit ties to ground in two places: one on the small bypass cap and the other at on the resistor that's tied to the silicon device.

Should both CCS ground points be the signal ground point, just like I did with the big cathode load resistors?

Or, because I'm tapping the filament supply, do I tie one or both of the CCS ground points to the power supply ground, where the filament supply now grounds?

Thanks in advance.

[hagtech]

Hmmm, not sure how you wired it, but I think the heater ground might be good.  The compliance of the current sink should reject most power supply noise, keep it out of the signal path.

Depends on the isolation between heater and signal ground.  When using the circuit board, this is not an issue.

jh

[slowburn]

Thanks, Mr. H.

I replaced both cathode resistors with your CCS suggestion about a week-and-a-half ago. Tied both  cap and resistor to signal ground buss. They work fine, no ground noise.

Used generic components. Think there's benefit to using a fancy little cap--Nichicon Muse, Elna Silmic or the like?

Seems, too, like the resistor that ties the sand device to ground might like to be a flavor that I've found pleasureable in cathode resistors in the past. Wha'd'ya think about this? How is that resistor similar or different from the simple cathode resistor?

Thanks again. It's a great circuit!

[hagtech]

Hard to say.  You never know what component upgrade will have an effect - even in the most innocuous locations.  The emitter resistor would be important as is the transistor, but the 470uF capacitor may not be.  But then again, it is worth experimenting.

jh