Tube Amps / Music Electronics
|For current discussions, please visit Music Electronics Forum.|
|10/5/1998 1:27 PM|
Hey guys. I'm still working on this 2 X el84 class A amp, and I've got a question about something I'm hearing. The distortion sounds great. It's a simple 2 stage preamp with most of the distortion in the output section. I'm getting alot of "swell" in the distortion as the notes sustain though. You know - that strange freq change that you hear in older tweed amps. What causes this? Is it from the output trans? I'd like to get it out of there if I can. Thanks
|And now, a word from our sponsors:
|10/5/1998 6:41 PM|
Kyle: is this amp cathode biased? Hows your filter caps holding up? [Richie]
|10/6/1998 6:19 AM|
I was working on a Princeton rebuild with EL84's. The preamp was basically a tweed Bassman circuit from the volume pot to the outputs, which were cathode biased (no feedback). I found a strange component to the output stage overdrive; mainly a huge amount of crossover distortion which looked like the output waveform stayed at zero volts for a short time midway between the waveform peaks (I hope this explanation makes sense?) The output stage distorted symmetrically (as it should). The amp sounded grainy or sizzly from this problem. Undistorted, there was no crossover distortion on the scope. I tried replacing the Princeton output with the trans from a Lafayette amp (designed with dual EL84's); no help. Strangely enough, the only helpful solution I found was to disconnect the EL84 bias resistor bypass cap, this helped somewhat but not quite enough. Everything else in the amp looked OK, the EL84's were biased to 10V (if memory serves), I tried several different batches of EL84's, no help.
I don't know if this is like your problem, but I hope the gurus out there may help us both.
|10/6/1998 7:25 AM|
Kyle&MKB: I hope this might help in the descrition, I worked on an ampeg that did this kind of thing, It sounded good, but would sound like it was losing power, and coming back,it was the cathode resistor and cap,that I had to change to fix the problem.I think I know what Kyle is talking about, but is hard to describe in words.All I can say is, the amp peaks out, and falls off a bit,[weaker sounding]then comes back.The sound is the same,[tone wise] but just has an up and down effect. [best I can describe] [Richie]
|10/6/1998 8:24 AM|
Hey guys - here's what I've got. A common 100 ohm cathode resistor with a 100uf cap used for both tubes. 318vdc plate to cathode and 44ma per tube. It's not really a volume change I'm hearing. It more of a tone change as the note sustains. The highs dip down after the attack and it gets dark and fuzzy sounding. It also only shows up when cranking the amp up.
|10/6/1998 9:36 AM|
What you describe sounds a little more like the output transformer crashing with high current peaks.
When the OT passes more signal current over its idle current, then it was designed to do, funny things happen to the frequency response.
The highs get much darker and the lows go away.
The frequency response at -3dB points of the OT comes together in the middle with no lows or no highs. It passes mids OK but the ends are gone.
A good clue might be if the amp is getting a darker squanky sound when playing very loud higher notes and at the same time the lows sound a little weaker and muddy with no sizzle on top.
But it does not do this when played at moderate volumes.
Since the amp sounds like it is self biased in class A, I would suspect the OT is not built for class A in this amp at this current level.
The other thing Kyle, to consider, as Richie alluded, as the positive going drive signal on the grids of the final tubes gets too high it can actually momentarily charge the coupling caps with a positive voltage.
This falsely turns the final tube on full blast with no audio signal until the grid load resistors drain off the capacitor's charge.
That sorta is like the tube has turned into a diode and it chops the signal but sounds funky...BAD.
The trick here is to reduce the drive voltage and come up with the right combination of coupling cap value, total grid load resistance and grid stopper value.
Try a smaller value cap to the final tubes.
Also keeping the bias voltage at a point where the tube is sitting with a little less current at idle might help.
In other words, maybe a bit more relative negative voltage on the grids when compared to the cathode would help.
Or the other way around when self biased through the cathodes of the final tubes....a wee bit more positive voltage on the cathodes when compared to the grids.
What kind of caps are you using to couple the driver section to the power tubes?
Maybe they are flakey too.
|10/6/1998 10:37 AM|
I think what Bruce described pretty much covers it.
What kind of transformer are you using? Is it reclaimed from a small hi-fi amp? If so, it was probably designed for much less idling current than your amp is set up for presently.
A more traditional common cathode resistor size for a P-P EL84 pair would be in the range of 150-200 ohms, usually closer to 200. Unless the output transformer is a big one, for a 6L6 amp or a class A EL84 amp, the primary current should be brought down. You'll probably have to bias this thing in class AB.
I've noticed that some EL84 amps that the designer expected to be overdriven have seemingly large signal grid stopper resistors fitted. This may be required in your amp to alleviate some of the grid blocking problems. I've seen 5.6k on marshall EL34 amps, and was somewhat startled to see an EL84 amp with ten times that size. What value grid stoppers are installed in your circuit?
|Page 1 of 2||Next>||Last Page>>|