Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. The sunn still shines online!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Quick Positive Feedback Notes

3/6/2004 1:42 AM
Gil Ayan
Quick Positive Feedback Notes
Hi all,  
Just thought I'd follow up on the positive feedback conversation that was started in the other thread, the *umble one.  
I did do some quick testing yesterday, taking the ouput of the second OD stage and feeding it to the input of the 1st one. I didn't spend too much time on it, but I have to say that I was somewhat disaappointed. I tried to keep things under control, so I used only as much feedback as I could get away before getting the infamous "whistling" effect. Depending upon the size of the cap used, the extra singal obtained could be annoying (mosquito buzz using a small, 20 pF, cap), or just like turning the presence control up a tad (47 pF). I did pad the thing down with a resistor...  
My conclusion is that it's really not that great an effect after all. The extra sizzle achieved by applying the feedback can be also implemented by tweaking the network located at the input to the overdrive - which varies depending upon the schematic one uses, but it is always a fixed resistor followed by a trimpot at the input of V2A). I couldn't get things to sound quite like an amp that's inherently unstable due to lead dress, and with additional stages (like adding the Fender style reverb send and mixer stages).  
One thing that can be done is to let more highs into the overdrive circuit, and then dump some highs at the output. When low gain is used, the sound is more compressed and singing this way, and when higher gain settings are used, the thing starts to resemble a Mesa Boogie. Either way, it kind of negates the principle of the *umble design, which is more of an open and dynamic sound.  
I have to conclude that the positive feedback identified in the Mark III, via the 3.3Meg//20pF network, has to be negligible... especially, as another friend pointed out, since at the input of the overdrive in the Boogie, there is a 47pF cap from grid to ground. I believe that Boogies were that way for the sake of simplicity only - rather than swtching out the above mentioned network when in lead mode.  
If anyone happens to try something like this and obtain other results, I would be interested in hearing about it.  
Out for now,  
3/11/2004 11:40 PM

Don't give up too easily. I suspect the best +FB circuit would roll off the extreme (mosquito zone)trebles and a little sub bass. WRT the Boogie circuit, I bet that the 47pF is there to reduce the loop gain at high frequencies, since that 10pF would tend to increase the loop gain there and could (and probably did, necessitating the 47pF cap) cause enough gain to oscillate ultrasonically....  
Here's an idea I had about building it into the actual circuit (the added bits are in red):  
C ya,  
3/12/2004 3:15 AM
Gil Ayan

That looks interesting. I have not tried anything like that, so I am just speculating here. I believe with the positive feedback you suggest, you will pretty much be able to increase the overall gain and with that, it would increase the amount of "distortion" as well -- at least that's what I was getting with my previous experiments which I wrote about the other day.  
However, what I experienced with the "inherently unstable" amp was like when an amp is cranked: you let go of the strings and they start ringing all by themsleves. But, the amp still sounded "clean," although it was being played at livingroom volumes. One thing I DIDN'T like about that amp was that the feedback was no louder than the regular notes, maybe even a tad lower in volume, which felt a little awkward. Usually, when a guitar starts feeding back, it gets somewhat louder, like it's going to runaway.  
Keep messing with your concept though, maybe you have something there!  
3/12/2004 4:17 AM
Andrew M

I know of a Norman Crowhurst book that has a chapter on this topic. I seem to recall that using positive feedback within a negative feedback loop somehow allowed more negative feedback and less distortion, but I never dug through the math he provided. His goals were more hi-fi oriented but this might help you some...  
The book was called "High-fidelity circuit design" [by] Norman H. Crowhurst and George Fletcher Cooper. (1956). I'll check it out from the library and have another look. I could scan the relevant material if anyone is interested.  
Andrew M
3/12/2004 3:14 PM

I haven't tried anything yet lacking time, but I have the intention to implement something similar to Dutch's circuit.  
One modification will be additional lowpass filtering in the loop. Basically I just want the "fundamentals only" to be "positively enhanced".  
3/12/2004 6:53 PM
Gil Ayan

On 3/12/2004 3:14 PM, Teo said:  
[QUOTE]Basically I just want the "fundamentals only" to be "positively enhanced".  
Really? I think upper even order harmonics is what's one likes to hear in a guitar feeding back. If yout want more fundamentals only, it would be like turning the gain up in the amp.  
3/13/2004 6:59 AM
Steve A.

    I remember you had an article on your site which discussed parasitic feedback effects, both positive and negative. You described an experiment using live leads within a chassis to increase and decrease the level of hum. If that article is still there maybe you can post a link to it.  
    IMO the positive feedback tricks are better suited for a simpler preamp that uses only 3 triodes (like a Wreck). I think that the ODS has plenty of gain already so there is no need to throw a joker into the deck, so to speak. But the experimental circuit in Dutch's post does sound interesting... ;)  
Steve Ahola

  Page 1 of 3 Next> Last Page>>