Okay.. i need help here. Basically the amp is working. Music is playing through the speaker nicely BUT there is something i can only describe as low frequency oscillation. The music mutes every 1 second and plays normally again.. so like 1-2Hz oscillation. Imagine the second hand on a clock: the music mutes everytime the hand moves and plays again everytime it stops.
I am going to try to remove the feedback resistors (the ones stretching from the output stage plates to the input LTP and see if it plays normally. Any other suggestion what to try? Schematic is on post #58. The only difference is that i am not using the 10pF caps parallel to the 270k feedback resistors. No global feedback yet.
Is this what is called motorboating?
I am going to try to remove the feedback resistors (the ones stretching from the output stage plates to the input LTP and see if it plays normally. Any other suggestion what to try? Schematic is on post #58. The only difference is that i am not using the 10pF caps parallel to the 270k feedback resistors. No global feedback yet.
Is this what is called motorboating?
Last edited:
I have tried these but the oscillation still occurs:
1. Remove both 270k feedback resistor
2. Put 100R decoupling resistor between the supply for the final stage and driver stage. There is 10uF filter cap on the driver stage.
Something interesting happened when i unplug the power to the HT transformer and the -12V supply for the CCS tail (they are connected to the same power plug, the 12V heater supply is separate plug), the music plays smoothly with no oscillation.. that is until the HT power supply reservoir caps fully discharged. I am using simple SS bridge rectifier and 2400uF reservoir cap. No RC filter, it all goes straight to the tubes. Help?
1. Remove both 270k feedback resistor
2. Put 100R decoupling resistor between the supply for the final stage and driver stage. There is 10uF filter cap on the driver stage.
Something interesting happened when i unplug the power to the HT transformer and the -12V supply for the CCS tail (they are connected to the same power plug, the 12V heater supply is separate plug), the music plays smoothly with no oscillation.. that is until the HT power supply reservoir caps fully discharged. I am using simple SS bridge rectifier and 2400uF reservoir cap. No RC filter, it all goes straight to the tubes. Help?
Possibly the 100R is not enough to separate the drivers from the finals. I had a similar low freq oscillation in an amp where the drivers and finals were separated by a 150 ohm choke. A change in current through the finals leads to a small change in HT which is not fully filtered out when it arrives to the driver.
In my case I had the same uF before and after the choke, possibly resulting in a reverse current from the driver cap back to the finals cap. A diode in between fixed it. In your case more R?
In my case I had the same uF before and after the choke, possibly resulting in a reverse current from the driver cap back to the finals cap. A diode in between fixed it. In your case more R?
My best sounding diy amp so far! Has full authority on bass just like SS amp and i haven't put any global feedback yet. Thanks Parafeed813, your suggestion is correct. I need to fully decouple the driver from final stage. No more motorboating, even with the plate to plate feedback resistors connected.
My B+ for the final isn't as high as my plan (300VDC) but i blame my local power company for that. I am not getting the 220VAC mains i am supposed to. Try 190VAC.. sheeshh! But even with that problem, i am getting the volume i need.
My B+ for the final isn't as high as my plan (300VDC) but i blame my local power company for that. I am not getting the 220VAC mains i am supposed to. Try 190VAC.. sheeshh! But even with that problem, i am getting the volume i need.
If you're referring to distortion measurement, can't do that. I don't have the tool for it.. 
The most sophisticated tool i have is a pocket oscilloscope
I am interested in seeing square wave output test though. Maybe will do that this coming weekend. That's something i can do with a smartphone app and my pocket oscillator.
Meanwhile, here's a youtube video:
https://youtu.be/O7-RUVDvBjM
Of course, cellphone mic+youtube can't do no judge on how it actually sounds.
I'm using a separate boost converter module to power the driver+input stage. This cures the motorboating. It's amazing how there is no hummm although i'm not using any LC nor RC filtering. There was none also when it still suffered from motorboating. I have always been intrigued by the excellent PSRR from symmetrical topology like this or Baby Huey.. now i can hear it by myself.
The most sophisticated tool i have is a pocket oscilloscope I am interested in seeing square wave output test though. Maybe will do that this coming weekend. That's something i can do with a smartphone app and my pocket oscillator.
Meanwhile, here's a youtube video:
https://youtu.be/O7-RUVDvBjM
Of course, cellphone mic+youtube can't do no judge on how it actually sounds.
I'm using a separate boost converter module to power the driver+input stage. This cures the motorboating. It's amazing how there is no hummm although i'm not using any LC nor RC filtering. There was none also when it still suffered from motorboating. I have always been intrigued by the excellent PSRR from symmetrical topology like this or Baby Huey.. now i can hear it by myself.
Finished the other channel and of course it works and sounds nice also. Next step is to find an enclosure. Meanwhile, i played with LTSpice a bit and realized i should have made the input stage cascodes. Oh well, perhaps next project.. Maybe a universal push pull input+driver board.
Attachments
Maybe it is.. I haven't really test that loud. Any reason why you suspect it would? Perhaps i can avoid the same mistake if i am to design another amp. Pretty sure with my listening room size, i won't be clipping the amp, though. With my cellphone as source at full volume, it's already too loud (and still sounds nice).. and phones don't output 2Vpp like this amp is designed to handle.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.