You may be making a mountain out of a molehill. Remember, a 0.5 VRMS I/P will drive the amp to full power O/P and GE's curves are spaced at 5 V. intervals.
Sy, you're correct.
I just made some measurments on my PP EL34 amp. The driver is only a 12AT7 diff pair at the input, CCS in the tail. EL34's are triode, and cathode biased. With 441 v supply to the 12AT7, 68K plate R, 2.87ma each side, I get an output of 31.8v RMS with a 2.6v RMS input. A p-p voltage per side of the 12AT7 of 90v. That's in circuit, so any loading by the EL34 is in there as well. By the way, as you'd guess, there is no GNFB in this arrangement.
Oos, my mistake, I forgot there is some feedback. I took it out and measured gain of 18 (2v in gives me about 36v out).
Cool, I have been modeling this circuit in tubecad using a CCS. This matches the number the sim gives. Kind of validating.
Where are you setting the operating conditions of the output stage? I keep finding conflicting information. Most seem class AB with about 40mA current per tube. In triode, if I don't care about power much, it seems that I would be better with a bias current of 70 or so for less distortion and more class A, at least with my B + of 400ish.
The raw B+ is 460 (using Dynaco ST-70 replacement transformers and 5AR4 rectifier). B+ on the EL34 plates is 457. Bias is through shared 470 ohm R, 36v, I = 76ma total. I could boost that a fair amount, I guess.
Thanks smbrown for measuring. jrenkin, how are you setting up tubecad to sim those results? I keep getting a gain of 23 and with the ccs I get a gain of 43.
I'll have to get back to you on that. It's 1AM and we are at the hospital waiting for the baby to make her grand entrance! Will review my sim when I get home, but where are you setting the current? I think my foggy bewildered brain recalls setting the current low at 3-4 ma per triode. This won't look like the best place in the curves, but per wisdom gleaned from more knowledgeable sources, is a good place to run this setup.
- Status
- Not open for further replies.