anatech said:
This concept interests me at present, I've considered what Tubelab said. Others seem to agree.
When I get time, I'll clip some feedback onto the output stage cathode and play. I would like to try putting the OPT secondary into the cathode circuit.
Hi Indm,
I must admit that I'm working with a SS output stage and a variety of voltage amplifiers. P-P tube output may respond well but I'm not so sure of myself when talking about a single ended output (not my favorite).
-Chris
I must admit that I'm working with a SS output stage and a variety of voltage amplifiers. P-P tube output may respond well but I'm not so sure of myself when talking about a single ended output (not my favorite).
-Chris
My OPT secondary signal was maybe 8dB lower than the cathode, I could see no point sending it there. (I put it in the cathode circuit anyway with mixed results) As I was trying to figure out how to do it, I started reading this. http://www.tubecad.com/march2001/ on partial feedback, sending the output stage's plate to its grid.
One variation had the feedback resistor act as the driver's plate load. I pulled the drivers Ra from the rail and attached it to the output's plate. Gain went down 6dB. Ok.
This sounds a lot like local feedback. Ie, perhaps more control, and it sounds happy about the fact. Possibly a little more musical.
The jury is still out, but I think the damping factor improvement has tightened the bass a little. I am not certain whether I prefer the treble, but it seems clearer and better controlled. More low level detail seems uncovered.
I think I'll live with this for a while and tweak it. There could be potential.
One variation had the feedback resistor act as the driver's plate load. I pulled the drivers Ra from the rail and attached it to the output's plate. Gain went down 6dB. Ok.
This sounds a lot like local feedback. Ie, perhaps more control, and it sounds happy about the fact. Possibly a little more musical.
The jury is still out, but I think the damping factor improvement has tightened the bass a little. I am not certain whether I prefer the treble, but it seems clearer and better controlled. More low level detail seems uncovered.
I think I'll live with this for a while and tweak it. There could be potential.
The few times I tried Plate-to-grid the results were marginal and it was eventually removed. In the regions it's trustworthy Spice sims showed for me that for the first few dB of return the 2nd came down nicely with little effect on the higher harmonics. Soon after the 3rd rose with each additional application. Measurements seem to indicate higher ones come up as well, though that's working from memory.
Another interesting option if the OPT has a 16 ohm tap is to ground the 8 ohm and use the remaining two to provide both cathode and global feedback. Presumably the cathode feedback by linearizing the oiutput reduces the global requirement. I've tried it and it works but the jury's still out on how well.
Another interesting option if the OPT has a 16 ohm tap is to ground the 8 ohm and use the remaining two to provide both cathode and global feedback. Presumably the cathode feedback by linearizing the oiutput reduces the global requirement. I've tried it and it works but the jury's still out on how well.
With a couple of days to listen, I am thinking of changing. I dislike the new overload characteristic (I often listen close to the mark), even though there now seems to be more headroom. There is something from the feedback that I do want to keep, not sure what.
I want to include the secondary, I may have to take it to my driver's cathode (though I am biasing it with LED's at present). Since the OPT and the driver/output grid relationship seem the two big issues in a SET amp, well, I just don't know yet.
I want to include the secondary, I may have to take it to my driver's cathode (though I am biasing it with LED's at present). Since the OPT and the driver/output grid relationship seem the two big issues in a SET amp, well, I just don't know yet.
Well,
The entire point of feedback is that you never want to get into clipping. Feedback will overdrive the circuit under these conditions and make things much worse.
lndm, you need much more power if you are running that close. You are almost always losing your program peaks. A little feedback helps as long as you don't clip.
-Chris
The entire point of feedback is that you never want to get into clipping. Feedback will overdrive the circuit under these conditions and make things much worse.
lndm, you need much more power if you are running that close. You are almost always losing your program peaks. A little feedback helps as long as you don't clip.
-Chris
Agreed, and I do wish I was running 845's (plural 
). My peaks do get kicked around a bit but probably less than 99% of amps out there driven to that point.
My idea of a fix will be horn related not power related. My GEC kt66's are babied in my amp as they are too hard to replace.
Anyway, too many things on at the moment for the horns. I will still play with the feedback and listen at various levels.
). My peaks do get kicked around a bit but probably less than 99% of amps out there driven to that point.My idea of a fix will be horn related not power related. My GEC kt66's are babied in my amp as they are too hard to replace.
Anyway, too many things on at the moment for the horns. I will still play with the feedback and listen at various levels.
Hi lndm,
In that case, make sure you aren't overdriving the amp, even a little, in order to evaluate your feedback levels.
Horns would be loud with your KT66's. BTW, you can use 7581A's also. There are some new 6L6GC's that would stand up well (JJ & EH).
-Chris
In that case, make sure you aren't overdriving the amp, even a little, in order to evaluate your feedback levels.
Horns would be loud with your KT66's. BTW, you can use 7581A's also. There are some new 6L6GC's that would stand up well (JJ & EH).
-Chris
- Status
- Not open for further replies.