Good idea, thanks Nigel.
Not knowing what the amp is supposed to have for a primary impedance I am guessing in that it should be around 8k. This should give me a transformer with turns ratio of 31.6 : 1
I have an extra filament transformer that reads 7VAC unloaded, this means I should read 220mV on the secondary when applying the 7VAC to the primary.
Now that I think about it I have a 25w 8k PP output transformer laying around that I can use for testing. When I get the time I am going to do all this and check back..........I would like to find the problem but I really hope it isn't the output transformer for his sake.
I removed the output transformer and put the 7VAC to the primary winding. The secondary 8 ohm tap reads 225mV giving me a 31 : 1 turns ratio as predicted. I also checked the 4 ohm tap (160mV) and the 16ohm tap (320mV) and the math calculates out correctly for them as well. So maybe the output transformer is fine.
I might go ahead and try my 8k pp in there anyway knowing that it's the same load impedance, although I am not very hopeful now the transformer is the problem. This one is a doozy to diagnose (for me anyway).
I might go ahead and try my 8k pp in there anyway knowing that it's the same load impedance, although I am not very hopeful now the transformer is the problem. This one is a doozy to diagnose (for me anyway).
Have you tried scoping each output of your phase inverter? If the waveform is corrupted on one half of the signal like the screenshots you posted, then the problem is on the PI side. If it's clean, then it's either the output tubes or the output tranny. FWIW, this phase inverter is not an LTP, it is a variation of a differential amplifier.
I thought the LTP was a variation of a differential amplifier
Yes of course, the LTP is a variation of a diff amp. But this PI is not exactly an LTP, it's not even a "real" diff amp circuit. Since you posted scope images of the output of both the normal channel being driven and the term channel being driven, and the "bad" side following the different channels, then the problem is on the same side. Checking the output of the PI on both sides and if the drive signal is corrupt there, then the problem is either the PI or before that. If it's clean, then the problem is after the PI, which is either output section or the output tranny itself. Just trying to help you narrow it down before having to replace the output tranny then finding out that it's not.
Hello Poacher,
I really appreciate the time you have spent and your knowledge you share in this thread, and I had no intention of butting heads. On first glance looking at the schematic I thought it was a LTP phase splitter, thanks for clarifying.
I am fairly certain it is either the PI or the finals...........but I am going to go back to square one and check each section again and continue on down the chain until I find it. My scope loads down the PI and pretty much any high impedance output section so it's tough to see on the screen.
I really appreciate the time you have spent and your knowledge you share in this thread, and I had no intention of butting heads. On first glance looking at the schematic I thought it was a LTP phase splitter, thanks for clarifying.
I am fairly certain it is either the PI or the finals...........but I am going to go back to square one and check each section again and continue on down the chain until I find it. My scope loads down the PI and pretty much any high impedance output section so it's tough to see on the screen.
Is that a typo.? 10k seems way too low for the input impedance of a scope. Connecting that almost anywhere in valve circuit would not give any sensible results. Did you mean 10M?
It's not a typo
I use my computer as an oscilloscope, I have a unity gain buffer in between what I test and the computer just for added protection against damaging my soundcard, the buffer has an input impedance of 10k. Yes I know I need real test equipment
Ah, I see. I think there should be a way of designing a buffer, using the right op-amp etc., to get an input impedance of 1M (or maybe more). But I am sorry I am no expert on that. Maybe some of the folks on here, who are knowledgeable about op-amps, could point the way?
It would have to withstand high input voltages though - maybe a potential divider first?
It would have to withstand high input voltages though - maybe a potential divider first?
No worries, I was just trying to clarify something with the hope that it will help you troubleshoot this problem. Just a thought, since a capacitor is the only missing component to turn it into a real LTP, how about adding this cap to ground and see if the problem goes away. Only one channel would be used, normal channel only and you would have to disconnect the trem side for now just to see if having the differential input causes this problem.
I tried this......post #18 I think.
Anyway this is how I see this phase inverter. The cap you speak of to make it a a LTP is already there. For example if you are using the tremolo channel it would C1 in the schematic found in post #1. And vice versa, if you use the normal channel it would be C18. These capacitors shunt AC to ground.
Had some more time to look at the amp and finally cured the ailment
Ended up being the 6BQ5's shared cathode resistor's bypass capacitor. I lifted the + end and used a cap I had laying around for testing and the problem solved. I hooked up the bad cap again and the problem came back, so I installed a brand new 220uF 50V cap and all is well, no more bad distortion and lack of power.........sounds much better too!
Thanks to all who helped me with this, much appreciated
Ended up being the 6BQ5's shared cathode resistor's bypass capacitor. I lifted the + end and used a cap I had laying around for testing and the problem solved. I hooked up the bad cap again and the problem came back, so I installed a brand new 220uF 50V cap and all is well, no more bad distortion and lack of power.........sounds much better too!
Thanks to all who helped me with this, much appreciated
Yeah it was a strange problem indeed. I thought it was the PI at first but it was fine when disconnected from power stage. So I had it narrowed down to something after the PI. Simulation showed that without a bypass capacitor on the 6BQ5's I would get clipping at the same output voltage I was seeing it clip (but symmetrical which mine wasn't) so I figured what the heck makes sense and presto. I didn't mind disconnecting the output transformer and checking the turns ratio because my buddy had a guy install an after market output transformer and this was a good opportunity to make sure it was correctly implemented. I have had output transformers that the hookup/wire diagrams that came with it were wrong, so now I double check transformers before they get wired up
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.