And you only pay for them once (not to say a tube will not last a long time) and you can shove the circuit in any little space not worrying about heat.
Me too. What a fantastic job! KMG did what so many have tried and failed to do. He seems to have overlooked nothing of importance.
The only flaw I could find is a relatively minor one: there is nothing to emulate the rather low (internal) anode resistance of a triode, so the output impedance of KMG's MOSFET triode is higher than a real half-12AX7 with the same 100k external anode load. With a 100k anode load, the real valve will have Zo about 40k, while the MOSFET pseudo-triode will have something close to 100k.
That means that if you load the output with a frequency-dependent impedance (say a tone control circuit), then KMG's MOSFET triode will end up with a different frequency response compared to a real half-12AX7. The output will drop more at those frequencies where the load is heavier.
I think this could be fixed with a second LND150 direct-coupled to the drain of the first, acting as a source-follower. Take the output from its source, put an actual 39k resistor in series, and there you go, even better triode emulation. It will cost a few pennies more for the second LND150 and two more resistors.
But KMGs sound-clips suggest his "triode" works pretty darn well, so maybe this would just be gilding the lily?
Indeed!The soft bottom clip in the simulation looks quite similar to the FET clipping I observed with my scope. Much more civilised than op-amp clipping
Pete
The soft clipping at the bottom of the output waveform is from the diode that simulates grid current flow, clamping the gate on positive input signal swings.
Meantime, the soft clipping at the top of the output waveform is from the five diodes in series at the source of the LND150. The fifth of those (1N4148) is to protect the low-voltage Schottky diodes from reverse voltage.
-Gnobuddy
Ciao Pete
Honestly I didn't read all the thread (because at the moment I've only few time)
if you didn't see this before I think it will be of interest
Link to the project --> The Squirrel Monkey One Tube Guitar Amp
Franco
Last edited:
Squirrel Monkey is a great start toward a useful practice amp. I ended up sticking a typical Fender input tube stage on it, as the basic SM does not have a lot of gain. I also played a lot with the feedback loop, and ended up with just a pot there. It's easy to work on such a simple circuit, so you can have a lot of fun getting it just where you want it...
Did you have any trouble getting the pint-sized monkey stable? Many (maybe most) of the Squirrel Monkey build descriptions I found online ended in hair-pulling and frustration, because of relentless oscillation problems.
There is a Super Squirrel Monkey out there, too. Someone stuck what he described as a 12AX7 cathode follower into the middle of a Squirrel Monkey to create it. I think he actually meant a common-cathode gain stage direct-coupled to a cathode follower stage, a la Fender Bassman and it's derivatives.
Given the stability problems of the original Squirrel Monkey, I'm amazed someone managed to add even more gain into the middle of the signal chain, and still get the thing stable.
Many years ago I was sufficiently intrigued by the Squirrel Monkey to buy a couple of 6AF11s off Ebay, along with a pair of Compactron sockets. After reading about all the issues with the design, I set them aside somewhere. Maybe one day.
Bigger picture, the "one tube guitar amp" idea seems to have surfaced many times in many people's minds, but I've never heard of an implementation that turned into a particularly good or useful amp. The gain is always too low, and the sound usually isn't particularly "tubey" when clean, or rich when overdriven. It seems that several gain stages in a chain, each one adding a little distortion, creates a richer tone than you can get from one or two heavily driven stages.
I think this is also the reason why so many distortion and overdrive pedals sound lifeless and static - most of them use a single op-amp stage to generate all the distortion, and that tends to create a predictable, monotonous, unvarying sound. The few pedals that produce rich overdrive invariably turn out to have multiple distorting stages and an AC-coupled design that causes signal-level-dependent changes in the duty cycle of the output waveform, like the old Wampler Plexi-Drive and the Runoff Groove "Three Legged Dog".
-Gnobuddy
Even with the added gain of an AX7 on the front end I've had no oscillation problems. A 470K grid stopper on the last triode stage and on the output triode was essential to make it stable, though.
My other useful tweak was to simplify the feedback stage to just a 50K pot, with the cathode resistor on that last triode at 39K. This results in really nice overload, and a really fun ability to tune the growliness with the feedback control. Output transformer is an inexpensive Champ replacement part.
I also put a Master Volume pot between the last triode and the pentode so that I can overload the chain of triodes but keep the overall loudness down in the interests of familial harmony. Crank up the Master Volume, though and it sings!
My other useful tweak was to simplify the feedback stage to just a 50K pot, with the cathode resistor on that last triode at 39K. This results in really nice overload, and a really fun ability to tune the growliness with the feedback control. Output transformer is an inexpensive Champ replacement part.
I also put a Master Volume pot between the last triode and the pentode so that I can overload the chain of triodes but keep the overall loudness down in the interests of familial harmony. Crank up the Master Volume, though and it sings!
Yes, I gave up trying to make a single ECL82 into a decent guitar amp, but as detailed here ...
The one valve guitar amp is dead, so ....
... I reworked the idea with a pair of ECL82s (pentodes in parallel) and a FET front end, I got a very sweet sounding 4 watt "twin".
The one valve guitar amp is dead, so ....
... I reworked the idea with a pair of ECL82s (pentodes in parallel) and a FET front end, I got a very sweet sounding 4 watt "twin".
Please DELETE my post #170 it was sent wrongly (THANKS)
---
May be ECL82 is a better amp
however the 6AF11 tube has some follower
Link: ページが見つかりません - TENTEC 電子工作メモ帳
---
Link: Percolator 2W Tube Guitar Amp (valve amp) - Zeppelin Design Labs
Link: YouTube
Link: https://www.zeppelindesignlabs.com/wp-content/uploads/docs/Zeppelin_Perc_ASSEMBLY%20INSTRUCTIONS.pdf
Link: https://www.zeppelindesignlabs.com/wp-content/uploads/docs/Zeppelin_Perc_OWNERS%20MANUAL.pdf
Link: https://www.zeppelindesignlabs.com/wp-content/uploads/docs/PERCOLATOR%20Electronics%20Only%20addendum.pdf
K
---
May be ECL82 is a better amp
however the 6AF11 tube has some follower
Link: ページが見つかりません - TENTEC 電子工作メモ帳
---
Link: Percolator 2W Tube Guitar Amp (valve amp) - Zeppelin Design Labs
Link: YouTube
Link: https://www.zeppelindesignlabs.com/wp-content/uploads/docs/Zeppelin_Perc_ASSEMBLY%20INSTRUCTIONS.pdf
Link: https://www.zeppelindesignlabs.com/wp-content/uploads/docs/Zeppelin_Perc_OWNERS%20MANUAL.pdf
Link: https://www.zeppelindesignlabs.com/wp-content/uploads/docs/PERCOLATOR%20Electronics%20Only%20addendum.pdf
K
Last edited:
Ciao Pete
at the moment this seems the less expensive tube
6AF11 TUBE . MIXED BRAND. NOS / NIB. CB12 | eBay
not sure, but this seems the right socket (I tried the socket on the tube when I got it at a pair of HamFest)
lot of 4x compactron tube socket NEW ! 12 pin base for 6KD6 6C10 6LF6 6JS6 etc. | eBay
Franco
at the moment this seems the less expensive tube
6AF11 TUBE . MIXED BRAND. NOS / NIB. CB12 | eBay
not sure, but this seems the right socket (I tried the socket on the tube when I got it at a pair of HamFest)
lot of 4x compactron tube socket NEW ! 12 pin base for 6KD6 6C10 6LF6 6JS6 etc. | eBay
Franco
Last edited:
There are several tragic tales of 6AF11 misfortunes in this thread: 6AF11 Push Pull Amp
What's particularly noteworthy is that many of the tales of unstoppable 6AF11 oscillations are from people with plenty of experience designing and building tube guitar amps. Darryl Hoy, for instance, who was a commercial Australian manufacturer of valve guitar amps back in the 1970s, and who still churns out home-built valve guitar amps at a staggering rate.
-Gnobuddy
What's particularly noteworthy is that many of the tales of unstoppable 6AF11 oscillations are from people with plenty of experience designing and building tube guitar amps. Darryl Hoy, for instance, who was a commercial Australian manufacturer of valve guitar amps back in the 1970s, and who still churns out home-built valve guitar amps at a staggering rate.
-Gnobuddy
The PP story may be tragic, but the SE story isn't much better. I've read a few Internet tales of failed SE builds too, Squirrel Monkeys that could never be tamed long enough to use as an amplifier.
As an example, look at the schematic Adam Reed posted, and it tells a tale of the things he had to do to get his SE version stable. Enormous grid-stopper resistors are used to roll off high-frequency response. The triodes have huge unbypassed cathode resistors to lower voltage gain to the point where there is barely any voltage gain. The whole schematic screams "This design is entirely about fighting oscillation first and foremost, at the cost of gain, output power, and everything else."
And Mr. Reed was one of the successful ones - he got his amp to actually work in the end, though he had to throttle the gain of the 6AFll so low that he ended up adding another tube at the input to bring the gain back up to usable levels.
Every indication I've seen is that designing a stable three-stage guitar amp around a 6AF11 always results in a compromised design.
There's a nice little power pentode (I think it's actually a beam tetrode) in the 6AF11. But there's nothing special about it, and other similar-power NOS pentodes and beam tetrodes are also available.
The question then becomes, why use the 6AF11 at all, with its oscillation problems, exotic socket, and enormous heater power consumption? Why not use a common 12AX7 for two gain stages, and pick a nice NOS low-powered pentode for the output stage, achieving the same 3-stage SE guitar amp without the design compromises?
I think the lure of the 6AF11 "one bottle guitar amp" is a bit like the lure of trying to build a perpetual motion machine. There is a part of our brain that just likes the idea very much, even though logic and science and engineering and bitter historical experience all say "Sorry, it just won't work."
-Gnobuddy
I forgot to say
If I were to add a tube to the 6AF11 I'd like to add a 6112
for apparence
Link: JJS.at : Electronic Projects : Class AB Subminiature Tube Guitar Amp
K
If I were to add a tube to the 6AF11 I'd like to add a 6112
for apparence
Link: JJS.at : Electronic Projects : Class AB Subminiature Tube Guitar Amp
K
You're probably right. But there is a tiny silly corner of my brain that wants to. I think it's the same silly corner of my brain that made me jump off the roof of my parents house when I was a boy!Ciao Gnobuddy
OK, received, you'll never go to try to build one
K
-Gnobuddy
Pretty good analysis of my R&D process, Gnobuddy!
However...the only part of it that was relevant to killing the oscillation was the bit about the huge grid stoppers.
From there I fiddled with the cathode resistor values in the interest of getting the thing to break up the way I wanted it to. I did all of this listening thru headphones, and it took a lot of experimentation to get sweet sounding bluesy breakup with my old Telecaster.
I have no arguments with anyone who'd prefer to do it with more conventional tubes; even a chain of 12AX7s (including the output in Class A) has been done a bunch, and can sound really great.
I just saw that Squirrel Monkey write-up, and once I started I couldn't stop til I had it working the way I wanted it...this was actually my first success at really voicing an amp to my tastes after cook-booking several that I didn't care for, or had to be run at ear-bleed levels (like my 5F6A Bassman clone) to sound decent.
cheers
Adam in Ashland, Oregon
However...the only part of it that was relevant to killing the oscillation was the bit about the huge grid stoppers.
From there I fiddled with the cathode resistor values in the interest of getting the thing to break up the way I wanted it to. I did all of this listening thru headphones, and it took a lot of experimentation to get sweet sounding bluesy breakup with my old Telecaster.
I have no arguments with anyone who'd prefer to do it with more conventional tubes; even a chain of 12AX7s (including the output in Class A) has been done a bunch, and can sound really great.
I just saw that Squirrel Monkey write-up, and once I started I couldn't stop til I had it working the way I wanted it...this was actually my first success at really voicing an amp to my tastes after cook-booking several that I didn't care for, or had to be run at ear-bleed levels (like my 5F6A Bassman clone) to sound decent.
cheers
Adam in Ashland, Oregon
Congratulations!
I've been using a graphic EQ pedal for voicing clean tones for a year or two now. Dial in the "tone" with the EQ, then measure the frequency response of the EQ pedal, then attempt to duplicate it with RC networks. LTSpice is a great help here.
I found this approach was much faster than the usual "change this cap, change that resistor, add this cap, rinse, repeat". It also led me to EQ curves I would never have found by the usual method - like the small-budget solid state guitar amp I built as a gift for a friend, which sounded harsh until the graphic EQ pedals showed me that a shallow notch of a couple of dB at around 800 Hz took most of the harshness away, and a little bass boost down low helped warm up the tone.
I was then able to duplicate that EQ curve with the notch and bass boost using a mistuned Twin-Tee RC filter tweaked with LTSpice until it produced the right frequency response.
This process seems to work fairly well for clean tones, but for me, the headaches start when trying to voice the overdriven tones. There seem to be so many variables there that I still haven't found a good way to do it.
Sorry for any false conclusions I jumped to regarding your specific flavour of Squirrel Monkey!
(Though it certainly is illuminating to take the original Squirrel Monkey schematic, and draw the original-design triode load lines on the 6AF11 data sheet curves. Woo, we're operating down in the weeds!)
-Gnobuddy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.