|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
108.171.131.176
I tried some 5.1V zener diodes off a common B+ node for better driver stage channel separation in my RH84 inspired test amp.
Two diodes were connected to C3 (the cap that feeds the plate resistors of the driver stage, one diode going to each resistor).
THD and THD+N seemed uneffected by adding the diodes.
Follow Ups:
I reconfigured my PS filter with a separate UF4007 diode + R dropper + decoupling cap w/bleeder resistor for each driver stage channel.Still having a THD+N noise issue that seems to remain constant even at lower power levels.
This does not manifest as AC ripple on the output/speaker terminals as it is about 4 - 5mVAC with the RCA input jacks shorted.
Nor does it appear to be 60 cycle as that is down over 70dB.
Edit: I pulled the CCS on the EL84 cathode and subbed in a bias resistor of 250 ohms - made the THD jump up by .5%, the noise issues were still there, so I put the CCS back in.
I looked at the driver stage again and there is a lot of noise coming from it.
I had noted this characteristic on another site, someone posted a few spectrum graphs of 12AT7 vs. a pentode driver stage.
Plan on isolating the final stage and injecting a test signal into the EL84 grid to see what happens.
One of my EL84 sparked and started to red plate, I might have a few spares to continue testing.
From what I have read, the RH design has a bit higher 3rd harmonic that SE pentode amps that use global feedback giving it a bit more subjective "punch" to the sound.
Edits: 09/23/16
FWIW, PN junction diodes can serve as "check valves" in decoupling networks to improve channel separation,
Eli D.
deathtube, do you have schemo of how you used this Zener? And why would you use a Zener and not a simple diode?
Thanks!
My PS is LCLCRC.
The driver plate supply for both channels is taken off C3.
I had about 100 pieces of 5.1V zener diodes and not so many UF type 1A diodes, so I used the zeners.
Each diode is conducting all the time. The effect is essentially no different than if you increased the value of each plate resistor by 200 ohms or so.
--------------------------
Buy Chinese. Bury freedom.
I would probably need to set something up similar to Eli's schematic.
Thanks for the comment.
At the risk of dropping the voltage below nominal value.
Did you notice any changes audible in separation?
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: