Hi,
I wanted to build a tube tremolo for a small amp, but the only double triode that I had available was the 12AU7, mainly because of the low gain.
Well, the 12AU7 has a much lower gain than the 12AX7, which requires some extra stages until the LFO starts. I used LTspice to make some simulations and tested it with the small amp I'm building and it sounds good.
According to Merlin a three stage CR network requires a gain of 29, so I just added another one, but it was quite unstable and the oscillation was only starting at the settings right in the middle of the potentiometer. So, I decided to add another stage and make things better. I also had to increase the resistors and caps, mainly because of the parts I'm using.
Because I couldn't find an affordable switched potentiometer I decided to use the end of the pot to turn the effect off. This way at one end the tremolo stops, and the other end gives the lowest oscillation.
In case you're interested here is my final schematic:
I wanted to build a tube tremolo for a small amp, but the only double triode that I had available was the 12AU7, mainly because of the low gain.
Well, the 12AU7 has a much lower gain than the 12AX7, which requires some extra stages until the LFO starts. I used LTspice to make some simulations and tested it with the small amp I'm building and it sounds good.
According to Merlin a three stage CR network requires a gain of 29, so I just added another one, but it was quite unstable and the oscillation was only starting at the settings right in the middle of the potentiometer. So, I decided to add another stage and make things better. I also had to increase the resistors and caps, mainly because of the parts I'm using.
Because I couldn't find an affordable switched potentiometer I decided to use the end of the pot to turn the effect off. This way at one end the tremolo stops, and the other end gives the lowest oscillation.
In case you're interested here is my final schematic:
Clever.
The drawback is less variation of frequency when you vary just one resistor. The 3-stage network, one R variable, gives near 3rd-root relation between R and F. 5 stages would be near 5th-root.
Assuming a 10:1 range of R even works, 3-stage will vary about 2.1:1, 5-stage about 1.55:1. So I guess it isn't a huge difference.
Back before they had LTspice and YouTube:
https://dalmura.com.au/static/thetapered%20phase%20shift%20oscillator.pdf
https://dalmura.com.au/static/phase%20shift%20oscillators%201941.pdf
https://dalmura.com.au/static/thetapered%20phase%20shift%20oscillator.pdf
https://dalmura.com.au/static/phase%20shift%20oscillators%201941.pdf
Another option to get more gain on the oscillator would be to add a solid-state, directly-coupled buffer stage off the triode's plate, and add a bootstrap to increase the stage's gain, like Merlin's example at the end of this page:
The Valve Wizard
You can use a MOSFET just like the triode in his example.
I might try that to put some 6J6 tubes I have lying around to some use.
The Valve Wizard
You can use a MOSFET just like the triode in his example.
I might try that to put some 6J6 tubes I have lying around to some use.
Trobbins, thank you for the papers! never crossed my mind look for publications other than in amp foruns.
blackcorvo, this time I'm gona go with the two CR pairs, much easier to implement. The drawback is not that bad. I really thought it changed a lot in my video. In the end it was really fast.
blackcorvo, this time I'm gona go with the two CR pairs, much easier to implement. The drawback is not that bad. I really thought it changed a lot in my video. In the end it was really fast.
That would be too easy! As I said, I wanted to use the 12AU7.
While searching about this subject I found several threads,where the person just tried to use the 12AU7 in a 12AX7 tremolo circuit and it didn't work. Here, I tried to show that with some minor changes it will work.
There is no need for more gain, and it's not a pedal, so I don't want all the possible speed and depth settings, just a simple tremolo to play one song or two.
I'm just always thinking of ways to implement both glass and silicon in designs together, but that's just me! It's not as much about shortcomings, but more about trying to keep tubes only where they're absolutely needed, or simply desired.
For example, inspired by Thomas' amp here, I ended up drawing a schematic for an amp with a tremolo circuit, but instead of using tubes for the LFO, it uses DIYStompboxes' Vibratone/Vibracaster solid-state LFO, with 2 transistors. This was mostly because it would be easier for me to implement, rather than adding another tube (I want to keep it simple, with 2x PCL82's, which is also why I'm considering a transformer phase inverter instead of adding more glass).
In the end, Thomas' solution was both elegant and effective.
Going to do the same to my Lamington SOLO, plus a Fender Champ 12 reverb-from-the-output
The DIY stomp thread is here
That's the most likely cause
Two years later I'm stealing the LFO from the '68 playmaster circuit
but I'll be playing with modulating the cathode of one half of the 12AX7, a bit like the PM-125 transistor amp (the tremolo was probably the only good thing about that amp
)
from Playmaster
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.