I used a 25K 24 step attenuator off ebay and I have plenty of gain. I think 6L6's build used a Goldpoint, I think his was 50K. Hikari1 used the same pot I did, many suggested 10k, but I didnt think this affected gain? I'm wrong all the time though! 😀😀😀
Russellc
Last edited:
OK here's another thing I just discovered. When I turn it on (cold) and with the attenuator all the way up, putting 1V in I get 10V out. As it warms up (maybe 10 minutes) the wave begins to clip and voltage slowly drops to 6V. I tried messing with P3, no change.
I think I have two separate problems:
1) I don't have enough useable steps in my attenuator.
2) The problem with the gain and it varying.
I have plenty of gain to drive an F5 in my system, but wouldn't have enough to drive an F4, which is my goal.
Last edited:
OK, I tested VGS on Mosfets and Jfets:
Right
Q1=56mV
Q2=46mV
Q3=2.0V
Q4=2.3V
Left
Q1=54mV
Q2=48mv
Q3=1.9V
Q4=2.1V
Right
Q1=56mV
Q2=46mV
Q3=2.0V
Q4=2.3V
Left
Q1=54mV
Q2=48mv
Q3=1.9V
Q4=2.1V
That's the warm up drift and its fine. We were looking for volts scale loss on both rails or even on one polarity because its just another way to cut on the preamp's voltage swing envelope spec. Nothing there though.
Kevin:
What's your DC offset (before the output cap) when it's warm and the output goes down?
It may be that the offset is zero when cold, zero offset let's it swing more volts and the offset is drifting as it warms up..which reduces the output swing.
It wire some probes up so you can watch the bias and offset as it warms up. Mine is unstable at best and putting on the cover or temp changes really effects the offset. The output cap will block the DC but to get the most voltage swing you need low offset.
That being said, you would have to have a pretty dramatic imbalance to lose that much output. So, if you are say under a volt of offset when warm this is probably not the issue.
What's your DC offset (before the output cap) when it's warm and the output goes down?
It may be that the offset is zero when cold, zero offset let's it swing more volts and the offset is drifting as it warms up..which reduces the output swing.
It wire some probes up so you can watch the bias and offset as it warms up. Mine is unstable at best and putting on the cover or temp changes really effects the offset. The output cap will block the DC but to get the most voltage swing you need low offset.
That being said, you would have to have a pretty dramatic imbalance to lose that much output. So, if you are say under a volt of offset when warm this is probably not the issue.
That's a good one. When I test the bias and DC offset it's stable. I left the probe on R13 while putting a signal on the input and DC offset swings with the attenuator from 0V at step 1 to 17V at step 24.
That is ~32 times gain. If it can swing rail to rail and using +/-24V PSU that's a 48V pk-pk limit. 48V/2.83=17VRMS. VoRMSmax/Gain(X)=max input RMS signal to preamp before the output clips, i.e. 17/32=0.53VRMS. Can your warmed up preamp make good looking output sinewave with about 0.5VRMS (1.4Vpk-pk) input signal (after the pot)? If yes, your build is in spec. If not, its to debug. (I know nothing about BA-3 practice but those relations are the general basics). 🙂
I bypassed the attenuator altogether. When I put in 1Vp-p in I get 10Vp-p out (gain of 20dB). If I put much more than 1V in the wave starts to clip. So starting cold I get a clean wave and 0V offset. As time goes by I can watch offset slowly rise(0-16V over ten minutes), the wave clip, and voltage go down.
I have seen several builders (generg for one) reporting actual gain in stock form round 8-9x. So more like 18db.
I believe the explanation is, that the jfet input stage is actually not operating with unity gain, but more like 0.5x.