Love this. Simple enough for me to play around with. I’d love to know more about how you physically attached the tube socket to your breadboard. Thanks for sharing!
The bias changes when you shut off the signal generator because you omitted the input cap shown in your last video. In vacuum tubes operated at such low voltages/currents, the grid tends to collect electrons from the cathode, and that current across the 1M ground reference resistor creates a negative votlage at the grid - just like grid leak biasing. If you watch the measurements on your scope, you'll see when the signal generator is turned off, the input goes from ~-28mV to ~-162mV.
@@bbugl That explains the change in average voltage while signal is applied, but does not account for the change in plate voltage when the signal generator is turned on/off with zero amplitude.
At low voltages, tubes have a lot of forward grid current. When you shut off the generator it no longer holds the grid at 0V, allowing grid current to flow in the 1M grid leak, generating a negative voltage at the grid which biases the tube a little colder
I realize there is some interest in low voltage vacuum tube stuff on the Interwebs, but it's going to work much better if you use tubes that were designed and characterized for that. There's a series of so-called space charge tubes that were used in car radios, for example, that work properly on very low B+ voltages.
This was a quick & dirty experiment for fun, basically so I could have students in the lab try something and have it be safe. If you really want to make a practical low-voltage tube amp, look up "space charge" tubes -- they're designed to work happily at low voltages. In particular there's a bunch designed to work at 12 V because that's what a car battery gives you.
Is channel 1 input impedance set to 50 ohms? The signal generator is expecting to drive 50ohms for the amplitude value to be correct. Also, you're not using a probe on channel 2, so what's the scope's input impedance? Is this loading your plate down a bit?
There's two 6 V filaments wired in series with a center tap at pin 9 and the ends at pins 4 and 5. So you can tie 4 and 5 together to 6 V and to pin 9 to ground (or vice versa), in which case you have them wired in parallel, or you can ignore pin 9 and put 12 V between pins 4 and 5, which splits the 12 V with half the voltage for one filament and half for he other. So in retrospect I could have run the heater off my 6 volt bench supply and used the full 20 V from my bench supply for the B+. I didn't think about that until I read your comment. ;)
Valves are really sensitive to heater variation and the heater is fragile. It is not unreasonable to use a constant current supply. Burnt out filaments are a valve killer. Too little heat will mean the cathode cannot emit electrons as it should.
The way I look at this amplifier is that it is a classic cathode bias circuit. The voltage drop across the cathode resistor is too low, therefore the grid bias is too high resulting in high side clipping. Try raising the resistance of the cathode resistor. A rheostat might help to work it out. The grid leak resistor is very small relative to the grid impedance; It does have an effect on the grid bias , but it can be considered as a short to ground. We are securely in never never land with 12 volts on the plate... Or maybe I am a totally uneducated (true!) moron.
![Just 2 Transistors! DIY Overdrive Goes From Crunch To Fuzz! [Breadboard DIY Guitar Pedal Build]](http://i.ytimg.com/vi/Uf-NqNtaeTM/mqdefault.jpg)
I'm certainly going to build one of these! Thank you!
Enjoy!
Love this. Simple enough for me to play around with. I’d love to know more about how you physically attached the tube socket to your breadboard. Thanks for sharing!
The bias changes when you shut off the signal generator because you omitted the input cap shown in your last video.
In vacuum tubes operated at such low voltages/currents, the grid tends to collect electrons from the cathode, and that current across the 1M ground reference resistor creates a negative votlage at the grid - just like grid leak biasing. If you watch the measurements on your scope, you'll see when the signal generator is turned off, the input goes from ~-28mV to ~-162mV.
Good point
It coud also be from the distortion, as the upper half wave is lower than it would be if it was linear. Therefore the average would be lower.
@@bbugl That explains the change in average voltage while signal is applied, but does not account for the change in plate voltage when the signal generator is turned on/off with zero amplitude.
@@gregkrobinson true.
@@misterhat5823 Might want to re-read the thread - you're agreeing with me.
At low voltages, tubes have a lot of forward grid current. When you shut off the generator it no longer holds the grid at 0V, allowing grid current to flow in the 1M grid leak, generating a negative voltage at the grid which biases the tube a little colder
I would love to see the input and output spectra compared.
I realize there is some interest in low voltage vacuum tube stuff on the Interwebs, but it's going to work much better if you use tubes that were designed and characterized for that. There's a series of so-called space charge tubes that were used in car radios, for example, that work properly on very low B+ voltages.
Could this be used for a micro guitar amp?
This was a quick & dirty experiment for fun, basically so I could have students in the lab try something and have it be safe. If you really want to make a practical low-voltage tube amp, look up "space charge" tubes -- they're designed to work happily at low voltages. In particular there's a bunch designed to work at 12 V because that's what a car battery gives you.
Is channel 1 input impedance set to 50 ohms? The signal generator is expecting to drive 50ohms for the amplitude value to be correct. Also, you're not using a probe on channel 2, so what's the scope's input impedance? Is this loading your plate down a bit?
Some of your measurements will be easier to see with AC coupling on the scope.
BTW: Have you ever had a look at the Korg nu-tube?
Not really, but it's on the list of things to think about... have you had a look and if so what do you think?
@@Lantertronics I had only a look at the website. I think, it's worth to have a look at.
I thought the heater took 4.5-6.0v (from 'cool' to 'hot'). Can they take 12v ? is there an advantage running the heater more than 5V ?
There's two 6 V filaments wired in series with a center tap at pin 9 and the ends at pins 4 and 5. So you can tie 4 and 5 together to 6 V and to pin 9 to ground (or vice versa), in which case you have them wired in parallel, or you can ignore pin 9 and put 12 V between pins 4 and 5, which splits the 12 V with half the voltage for one filament and half for he other.
So in retrospect I could have run the heater off my 6 volt bench supply and used the full 20 V from my bench supply for the B+. I didn't think about that until I read your comment. ;)
Valves are really sensitive to heater variation and the heater is fragile. It is not unreasonable to use a constant current supply. Burnt out filaments are a valve killer. Too little heat will mean the cathode cannot emit electrons as it should.
@@Lantertronics That's why it's called a *12*AU7.
The way I look at this amplifier is that it is a classic cathode bias circuit. The voltage drop across the cathode resistor is too low, therefore the grid bias is too high resulting in high side clipping. Try raising the resistance of the cathode resistor. A rheostat might help to work it out. The grid leak resistor is very small relative to the grid impedance; It does have an effect on the grid bias , but it can be considered as a short to ground. We are securely in never never land with 12 volts on the plate... Or maybe I am a totally uneducated (true!) moron.