Hi there folks. After some lurking I figured I would introduce myself and ask a valve question.
I've been designing electronics for a living since the late 1960s - starting with germanium transistors and the early op-amps for audio and signal processing. Then I moved to digital design (old school TTL, 8051 style assembler and more recently VHDL for FPGAs) in my later years. I also do PCB layout as part of my job. Hey - you've got to love engineering as a career - I still enjoy it after all these years.
Anyway, now I'm half retired and I have gone back in time to the valve era, modifying and designing guitar amps. I bought a 6V6 Fender Champ 600 and cured the hum, added a tone stack and it's now on loan in a recording studio.
Then I fixed a Fender Pro Junior that was red plating its EL84s, removed the hum again, and gave that some decent tone controls and a light overdrive effect.
Recently I designed a twin 6V6 single ended 8W amp combining the Princeton and Angela designs with some tweaks of my own. That runs a 12 inch Greenback and sounds just great to my ears. I play what I call "late night blues" at home so it needs to sound good at low levels.
Finally I repeated that class A circuit as a 3W dual EL90 using a pentode OP Transformer from a 1957 radio
That also sounds great (to me at least).
So, that's where I am right now. But here's the question:
Next project: Imagine a single valve guitar amp. Lets start with an ECL86 perhaps. I already have a 1950s pentode OPT and a 200v/6.3v power transformer. This won't have enough gain to work well.
So - big question: How about applying a tiny amount of positive feedback to raise the gain - without causing oscillation? Anybody done anything as mad as this before..
Cheers
Pete
I've been designing electronics for a living since the late 1960s - starting with germanium transistors and the early op-amps for audio and signal processing. Then I moved to digital design (old school TTL, 8051 style assembler and more recently VHDL for FPGAs) in my later years. I also do PCB layout as part of my job. Hey - you've got to love engineering as a career - I still enjoy it after all these years.
Anyway, now I'm half retired and I have gone back in time to the valve era, modifying and designing guitar amps. I bought a 6V6 Fender Champ 600 and cured the hum, added a tone stack and it's now on loan in a recording studio.
Then I fixed a Fender Pro Junior that was red plating its EL84s, removed the hum again, and gave that some decent tone controls and a light overdrive effect.
Recently I designed a twin 6V6 single ended 8W amp combining the Princeton and Angela designs with some tweaks of my own. That runs a 12 inch Greenback and sounds just great to my ears. I play what I call "late night blues" at home so it needs to sound good at low levels.
Finally I repeated that class A circuit as a 3W dual EL90 using a pentode OP Transformer from a 1957 radio
That also sounds great (to me at least).So, that's where I am right now. But here's the question:
Next project: Imagine a single valve guitar amp. Lets start with an ECL86 perhaps. I already have a 1950s pentode OPT and a 200v/6.3v power transformer. This won't have enough gain to work well.
So - big question: How about applying a tiny amount of positive feedback to raise the gain - without causing oscillation? Anybody done anything as mad as this before..
Cheers
Pete
Welcome Pete!
You have a lot more experience than I, but I'll make a contribution in respect of your latest project nonetheless. It'll give your thread a nudge at least!
I constructed a single ECL82 valve amp and find it has there's more than enough gain in to suit my guitar playing. Then again, I'm a bit of a three chord wonder!
So don't know about the need for positive feedback - I'll leave that for the experts!. My amp has selectable negative feedback and provides ample loudness with or without it in circuit.
You have a lot more experience than I, but I'll make a contribution in respect of your latest project nonetheless. It'll give your thread a nudge at least!
I constructed a single ECL82 valve amp and find it has there's more than enough gain in to suit my guitar playing. Then again, I'm a bit of a three chord wonder!
So don't know about the need for positive feedback - I'll leave that for the experts!. My amp has selectable negative feedback and provides ample loudness with or without it in circuit.
There are two kinds of feedback: negative or degenerative, and positive or regenerative. Both can be current or voltage loops.
Usually a positive voltage or current positive feedback is inside a negative global negative feedback, and has been used extensively in the era when obtaining high gain from a single tube was mandatory, and opamps chips were only a dream. I can help, as I am making a tube amplifier with positive current feedback as in the 50's with good results.
Usually a positive voltage or current positive feedback is inside a negative global negative feedback, and has been used extensively in the era when obtaining high gain from a single tube was mandatory, and opamps chips were only a dream. I can help, as I am making a tube amplifier with positive current feedback as in the 50's with good results.
Wish I could have said that! I'll return to practising my B flat chord!
Hi Osvaldo, may I ask you how you managed it? I know from old (1950ies) German magazines that positive current feedback serves for minimizing an amplifier's output impedance . One may even make it negative when exaggerating the PFB. The main issue appears to be the varying loudspeaker impedance, which makes finding the optimal FB rate almost impossible. After all, I suspect that this kind of FB is not what the OP is after
. Best regards!
You can manage the loop response to maintain PCF only at low frequencies around the bass resonant frequency of the loudspeaker(s). In my current project, I used a current transformer with the primary in series with load, and the secondary tuned with the parasitic winding capacitance and (possibly) any other added at hoc and that it becomes inoperative or NCF at medium and high end. This idea is robbed from industrial environments where current transformer are used extensively.
This is partially my design currently working pretty fine. In the schematic there is omitted a 22Ω in parallel of the transformer's secondary. No cap still, but I don't know in the final project may be one of them.
Also the idea is robbed from here.I found scanned the original article.
This is partially my design currently working pretty fine. In the schematic there is omitted a 22Ω in parallel of the transformer's secondary. No cap still, but I don't know in the final project may be one of them.
Also the idea is robbed from here.I found scanned the original article.
Attachments
Last edited:
Wow, there's some brave design going on out in the big world
When I got my first 6V6 design going and it came time to close the NFB loop - I had a nasty feeling the feedback could be positive not negative, so I applied the feedback through a large resistor, and - yes the speaker signal got larger not smaller. So it was positive feedback not negative.
But the sky didn't fall in. A gentle application of PFB did indeed raise the amp gain. So I put that idea in the back of my head for future use..
It's interesting that a triode/pentode single valve circuit might have reasonable gain for guitar - at least for a low power amp. That is encouraging.
So, my follow up question is: Do I need an expensive ECL86 in that case or could I get by with a cheaper ECL82 I wonder...
Cheers
Pete
When I got my first 6V6 design going and it came time to close the NFB loop - I had a nasty feeling the feedback could be positive not negative, so I applied the feedback through a large resistor, and - yes the speaker signal got larger not smaller. So it was positive feedback not negative.
But the sky didn't fall in. A gentle application of PFB did indeed raise the amp gain. So I put that idea in the back of my head for future use..
It's interesting that a triode/pentode single valve circuit might have reasonable gain for guitar - at least for a low power amp. That is encouraging.
So, my follow up question is: Do I need an expensive ECL86 in that case or could I get by with a cheaper ECL82 I wonder...
Cheers
Pete
Hi, OK that project is interesting.
There is an article from the 1950s (from Bendix USA) where they used light PFB inside a larger NFB loop. So this has been done before.
I'm just thinking of PFB with no outer stabilising NFB loop... so that might have more risk.
Pete
Lucky you! My family is currently huddled together around my valve amplifier for warmth!
Hi Pete,
the main differences are the pentode's lower plate dissipation rating (7 W vs. 9 W) and the triode's lower µ (70 vs. 100). Perhaps you might not even notice it in practice
.The ECL82's pro is the much better availability and it's price. But you also might opt forthe series heated PCL86, which - still - is a lot cheaper than the ECL86.
Best regards!
Hi there
Unfortunately my power transformer means it has to be 6.3v heater. However the ECL82 looks like a contender from what you say. I will go look at eBay and see what's available.
At present this is just a fun project to use up left over components. I kinda like the idea of a one valve amp
Unfortunately my power transformer means it has to be 6.3v heater. However the ECL82 looks like a contender from what you say. I will go look at eBay and see what's available.
At present this is just a fun project to use up left over components. I kinda like the idea of a one valve amp
Watford Valves Watford Valves product search :: ecl86 currently have a Mazda ECL86 @ £30 + VAT and an RCA ECL82 @ £12 + VAT.
It might be worth considering the ECL86 for its greater potential power output.
I tried it in my amp, but it seem to be unusable. All set becomes unstable with signal.
Kay, I sent a PM. Didn't you receive it?
I wonder what your evidence for that is Alastair? My stereo amp incorporates four original Mullard ECL86 valves that are still fully operational after 50 years and still sound good to me!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.