I've just finished up a 5F2A Princeton using the Hoffman schematic. I am using 6N6C and ECC83 tubes. PT is 220-0-220 diode rectified to get 315vdc.
When powered up, if I move the volume control more than about 20% off 0, the amp oscillates. The volume and tone pot varies the frequency of the oscillation.
The oscillation sounds like a trumpet / farty kind of sound.
I tried swapping the OT Primary leads and it oscillated from switch on and in a much more brutal way.
I poked around with chopsticks and rerouted the shielded cables (they are terminated one end only to the pot ground.
Firstly, do my voltages look ok?
B+ = 315
Bottom of 6V6 470R screen resistor (B) = 275
Top of 100K 12AX7 V1-B Anode Resister (C) = 270
6V6:
anode/plate = 300
Screen = 277
Grid = 3.5
Cathode = 20
Cathode R = 470
12AX7/ECC83
V1-A
Anode/plate = 260
Grid = 0
Cathode = 3.5
Cathode R = 1.5k
V1-B
Anode/plate = 250
Grid = 0
Cathode = 3.5
Cathode R = 1.5k
When powered up, if I move the volume control more than about 20% off 0, the amp oscillates. The volume and tone pot varies the frequency of the oscillation.
The oscillation sounds like a trumpet / farty kind of sound.
I tried swapping the OT Primary leads and it oscillated from switch on and in a much more brutal way.
I poked around with chopsticks and rerouted the shielded cables (they are terminated one end only to the pot ground.
Firstly, do my voltages look ok?
B+ = 315
Bottom of 6V6 470R screen resistor (B) = 275
Top of 100K 12AX7 V1-B Anode Resister (C) = 270
6V6:
anode/plate = 300
Screen = 277
Grid = 3.5
Cathode = 20
Cathode R = 470
12AX7/ECC83
V1-A
Anode/plate = 260
Grid = 0
Cathode = 3.5
Cathode R = 1.5k
V1-B
Anode/plate = 250
Grid = 0
Cathode = 3.5
Cathode R = 1.5k
Attachments
Just on the subject of the OT, I used one with a 5k impedance. Google seemed to say this was ok for the 6V6.
With 3.5 V the 12AX7's cathode voltage are on the much too high side, I'd say. But this is in contradiction to the small 260 - 250 = 10 V difference across the plate resistors, corresponding to just 0.1 mA plate current. This leaves two options:
- measurement errors
- wrong plate and/or cathode resistors.
The tube itself doesn't appear to be faulty at this point, as for a (negatively biased) triode cathode and plate currents are expected to be identical.
Best regards!
- measurement errors
- wrong plate and/or cathode resistors.
The tube itself doesn't appear to be faulty at this point, as for a (negatively biased) triode cathode and plate currents are expected to be identical.
Best regards!
Layout!
The huge signal from the OT plate lead is sneaking-back to the volume/tone area.
Hoffman has good drawings of known-good layouts. Do yours as close to his as you can. Even if it means major re-work.
I have never had much success with the 12ax7. The mu is 100. I built a guitar amp and it oscillated badly despite being a no feedback amp. The second 12ax7 stage seemed to be feeding back into the first one somehow. The rails were decoupled separately so unless It was internal to the 12ax7 I don't know. I guess the 100 * 100 gain of the two 12ax7 stages which is 10000 was just far too much gain. A quick and simple bodge was to just replace 12ax7 with a 12au7 and that seemed to settle things down. The 12au7 is much smaller mu about 18.
I found layout was important keeping things as short as possible especially the input signal which should be screened where possible. Keep heater wires away from audio signal. Keep valves away from transformers. Route HVAC away from audio.
I have had troubles with hum but for some reason changing to DC never fixed it. SO AC heaters are ok if wires are routed well.
I found layout was important keeping things as short as possible especially the input signal which should be screened where possible. Keep heater wires away from audio signal. Keep valves away from transformers. Route HVAC away from audio.
I have had troubles with hum but for some reason changing to DC never fixed it. SO AC heaters are ok if wires are routed well.
Last edited:
Unlikely. You'd need a plate load of unfinite impedance, e.g. a sophisticated CCS, to get a gain of 100. With a common resistor it always is lower than that.
Most probably your issue was the layout, too.
Best regards!
I agree with Kay Pirinha and Printer2. A 12AX7 is almost completely cut off with the cathode 3.5V volts more positive than the grid, and this is reflected in the very low voltage drop across the 100k anode load (270 V - 260 V). That means only 0.1 mA is flowing through each of the triodes, instead of the expected 1 mA.
Have you checked - with power off - the resistance from the 12AX7's pin 3 to ground, and also pin 8 to ground?
I bet you will measure something much bigger than the expected 1.5k. Either those 1.5k resistors aren't actually 1.5k (maybe they're 15k?), or there is a missing ground wire, or a bad solder joint.
This may not be the only problem you are facing, but clearly, DC operating conditions for the 12AX7 are very wrong, so that needs to be fixed before further trouble-shooting is done.
-Gnobuddy
Yes, that; but he says it oscillates. HIGH grid drive tries to take grid positive, can't, and the input R-C charges-down to a negative voltage.
Ha, but he shows zero at grid? Well, maybe it quits oscillating when he puts the probe on the grid, capacitive loading. With two meters he could watch the cathode voltage recover while he pokes the grid.
The measurements were taken with the volume set to just under the oscillating point IIRC. I'll try take another set with volume definitely at 0 when I get back (I'm outta town at the mo).
Thanks, I'll make those measurements. I did check that the resistors were indeed 1k5 but I shall check the run from 3 and 8 to ground to be sure there isn't a missing solder blob...
In the meantime, I will also tidy up the wiring to be certain that the Outputs are well clear of the Inputs.
Am I right that the heater leads should be routed through the centre of the valve socket rather than each lead coming around either side?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.