I'm building a pre-amp for bass guitar, utilizing two 12ax7's. I'm fairly new to (tube) amp design, but did a lot of reading learned a lot on this forum. I am however struggeling with the second stage of the preamp. It's a direct (DC) coupled recovery stage/ low impedance driver. The recovery stage was easy, but I fail at the cathode follower.
The gain stage, biased at -1.5V, gives about 144V quiescent voltage at the anode and therefor the grid of the CF. I calculated the CF's cathode resistor should be about 120kOhm to get a voltage drop of about 145V for biasing. Looks good on paper, but after building it, I'm measuring around 180V at the cathode and there's no way I can figure to lower it. I've tried different values of Rk, but it's just lowering it by a couple of volts. It's getting nowhere near 145. I've double checked the wiring, stared infinitly at the schematics, but I'm obviously missing something.
Can anyone help me to understand why my design is failing? Pictures included ofcourse:
The gain stage, biased at -1.5V, gives about 144V quiescent voltage at the anode and therefor the grid of the CF. I calculated the CF's cathode resistor should be about 120kOhm to get a voltage drop of about 145V for biasing. Looks good on paper, but after building it, I'm measuring around 180V at the cathode and there's no way I can figure to lower it. I've tried different values of Rk, but it's just lowering it by a couple of volts. It's getting nowhere near 145. I've double checked the wiring, stared infinitly at the schematics, but I'm obviously missing something.
Can anyone help me to understand why my design is failing? Pictures included ofcourse:
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
The CF is doing what it's supposed to do. The issue is the plate voltage of the gain stage. Welcome to the world of real tubes that have tolerances.
Try swapping a couple of tubes in to see how that varies.
One note: your circuit is set for early tube death. Be kind and wire a reverse-biased diode between grid and cathode of the CF. I hope you've respected maximum cathode-to-heater voltages...
Try swapping a couple of tubes in to see how that varies.One note: your circuit is set for early tube death. Be kind and wire a reverse-biased diode between grid and cathode of the CF. I hope you've respected maximum cathode-to-heater voltages...
Thanks for the quick reply!
I left the diode out for clarity. Also forgot to mention I raised the heater supply
But I fail understand what mechanism causes the voltages I'm measuring. Even lowering Rk to 82kOhm doesn't do much. So in my mind, the CF is biased at 30V, which offcourse is rediculous for a 12ax7
I left the diode out for clarity. Also forgot to mention I raised the heater supply
But I fail understand what mechanism causes the voltages I'm measuring. Even lowering Rk to 82kOhm doesn't do much. So in my mind, the CF is biased at 30V, which offcourse is rediculous for a 12ax7
If the CF is cathode biased at 145V, then the voltage measured from the cathode to the plate would be 195V. Remember, the tube references everything to its own cathode. The grid would be biased at -1V relative to the cathode, so looking at the plate curves, there would be about 2mA flowing through the tube and Ra3. So Ra3 = 145 / 0.002 = 72.5K.
In short, the job of the cathode follower is to force the cathode to follow its grid, you can't change the cathode voltage by changing the cathode resistor, this just changes the current thru the tube . . . until you abuse it !
To set another cathode voltage all that you can do is to set another grid voltage by altering either the current or the plate resistor of the first tube.
Yves.
To set another cathode voltage all that you can do is to set another grid voltage by altering either the current or the plate resistor of the first tube.
Yves.
If I understood right you said that the voltage at the anode (1st) is 144 V and at the cathode (2nd) is 180 V. This is impossible situation and I think that if the voltage readings are really these, then the cathode follower is possibly oscillating.
You could check this with oscilloscope. However, you can avoid the oscillation by inserting a series resistor of some 1k at the grid of the CTF.
Can you confirm the actual voltage readings from the anode of the 1st. 12AX7 and at the cathode of 2nd half ?
If you want to adjust the voltage level of the first anode and the bias of the cathode follower, this is done by fine tuning the cathode resistor of the 1st 12AX7.
You could check this with oscilloscope. However, you can avoid the oscillation by inserting a series resistor of some 1k at the grid of the CTF.
Can you confirm the actual voltage readings from the anode of the 1st. 12AX7 and at the cathode of 2nd half ?
If you want to adjust the voltage level of the first anode and the bias of the cathode follower, this is done by fine tuning the cathode resistor of the 1st 12AX7.
This is what you should get:
In addition:
Gain = 37.7 dB, THD > 1 % @ 13 Vrms out.
Frequency response ( -1 dB ) 12 Hz...>100 kHz.
An externally hosted image should be here but it was not working when we last tested it.
In addition:
Gain = 37.7 dB, THD > 1 % @ 13 Vrms out.
Frequency response ( -1 dB ) 12 Hz...>100 kHz.
Thanks a million guys. I really appreciate it. Here's how I've been progressing
First, I followed artosalo's schematic (thanks for that) and I added the 1k between anode V1 and grid V2 as suggested. Next, I measured between V2's grid and cathode. 1 volt! That's more like it. But I couldn't dial in (pot) a lot more potential between the two. So I swapped the tube with a new Tung-Sol and I was able to set the potential to 1.5 volts with ease. Seems tube nr.1 has issues.
And then I checked what payitforwardeddie and SY wrote about my meter's impedance. It's a simple cheap multimeter, so input impedance won't be a lot more then 1M, and thus effectively lowering Ra1 from 470k to about 320k!
It's my first stage with such a high Ra, so that's a big rookie mistake I guess.
So, unexpected tube behaviour combined with a cheap multimeter was what's going on. Problem solved. Sounds great! 37.8 dB of measured gain. I couldn't have done it without your help.
First, I followed artosalo's schematic (thanks for that) and I added the 1k between anode V1 and grid V2 as suggested. Next, I measured between V2's grid and cathode. 1 volt! That's more like it. But I couldn't dial in (pot) a lot more potential between the two. So I swapped the tube with a new Tung-Sol and I was able to set the potential to 1.5 volts with ease. Seems tube nr.1 has issues
.And then I checked what payitforwardeddie and SY wrote about my meter's impedance. It's a simple cheap multimeter, so input impedance won't be a lot more then 1M, and thus effectively lowering Ra1 from 470k to about 320k!
It's my first stage with such a high Ra, so that's a big rookie mistake I guess
.So, unexpected tube behaviour combined with a cheap multimeter was what's going on. Problem solved. Sounds great! 37.8 dB of measured gain. I couldn't have done it without your help.
Last edited:
And it isn't just cheap meters, every meter has an impedance. You just must remain aware of it. When you measure B+ voltage or put your meter across a 1.5k resistor, it essentially remains invisible to the circuit. But when your meter is in series with a 470k resistor, all of a sudden its impedance is a significant amount to the circuit. MY trusted Fluke meter will give confusing readings in such a situation too.
This is what we call "learning from experience".
Remember - there is only one stupid question - the one you DON'T ask.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.