|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
24.140.192.106
Experimenting with plate to plate feedback and took some measurements with ARTA.
Low power SE 6V6 triode mode , 6SL7 with CCS , cap coupled stages.
Before feedback was applied:
Power out before clipping : 530mW
Voltage Gain In/Out = 11
Vin for full output before clipping: .187V
THD = 2.9%
After feedback was applied with 420K ohm resistor:
Power out before clipping: 550mW
Voltage Gain In/Out = 4
Vin for full output before clipping = .528V
THD = 1.1%
Limiting factors : undersized PT and small SE UL OPT's from Triode Electronics (reviewed on Tube LAB site).
DT 667
Follow Ups:
Hi DT, why not use pentode mode for the 6v6 you would have 5watts output? I built a 6l6/6v6 amp of the RH type which uses 6sl7 input/driver tube which is a plate to plate feedback circuit. I used a 200k plate to plate resistor. My operating voltages were higher than yours at 250v on the anode of the 6v6.
I need to get some bigger PT's to build anything better.
My motivation was seeing if I could use plate to plate feedback with a CCS load on the driver plate , so far it works OK.
I've taken this amp as far as it can go for now.
It's not too surprising that the numbers improved with the addition of NFB. What power level did you use for the THD measurements?
--------------------------
Buy Chinese. Bury freedom.
The power level for THD% measures in triode mode was just over 1/2W before onset of clipping.I get more power in UL mode, over 1W.
The PT in this case is a bit undersized, probably better for a preamp than a stereo SE amp, so the biasing is conservative.
Edit: in SE UL mode w/fbk, Max Power Out was 1.67W @ THD = 1.78%Vin at max power = .74V , Voltage Gain In/Out = 5
In SE UL without feedback @ 1W power out , THD = 3.88%
I'll have to see it improved THD% with feedback = a better listening experience or not.
Edits: 08/26/16
I measured my little 6V6 SE amp last year while hand-picking tubes. Don't remember the distortion numbers being nearly that high. I need to find my notes...
--------------------------
Buy Chinese. Bury freedom.
I am curious as to how you dertermine "max power out" for each variation? Also why are the distortion numbers for the feedback version different from the non-feedback version? I would think that the term "max power" would be referenced to some arbitrary distortion number and that referenced matched in the two cases so a better comparison can be made.
dave
For max power out, a scope probe is hooked up across a load resistor on the OPT secondary.
I run a 1KHz test signal from my soundcard into the amp and look at the point where the waveform clips, then back off until the sinewave looks undistorted.
I also have a true RMS voltmeter across the load resistor to record the AC signal voltage and use the typical ohm's law equation for power.
As for why the THD% is lower with local FBK than non-feedback version, that follows every SE pentode class circuit I have ever tried building.
I guess the conflict I see here is the visual clipping which is a very subjective thing and the harmonic distortion measurement which is purely objective are not agreeing with each other.
In order to make a "fair" comparison either the output power or the harmonic distortion (clipping) should be held constant and then the other result documented. Of course this doesn't take into account the distortion spectra but for a SE amp where 2nd order dominates the gross distortion figures things should line up fairly well.
dave
One problem is that the internal sine generator in ARTA is in 1dB steps.
I would need a way to fine tune the signal going into the amp so the output signal over the load resistor can be held with more precision.
Then I could get the same power output and compare THD% when in non-feedback vs. feedback modes.
Or I could use my external signal generator, but that tends to output more distortion than the soundcard.
Thanks for your comment.
Arta is limited to 1dB increments on the audio generator? Wow...
Even so that is in the voltage realm and the increase of 1dB of voltage is about 25% more power leaving the maximum variation between any two settings at 12.5% and statistically you will probably end up in the 3-10% range of variation.
dave
You should get about 2 W. from a Class "A", triode wired, 6V6 family tube.
Building a modest cost, high performance, B+ PSU for a stereoblock using SE "12" W. tubes is pretty easy. "Full wave" voltage double the O/P of a Triad N-68X isolation trafo. 2 A./600 PIV Schottky diodes will be switching noise free and more than tough enough.
Eli D.
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: