I have just built up a 80 watt valve amplifier.
On the front end I have two 12ax7's.
One of the 12ax7's has a cathode voltage of 4 volts and the other 2.5 volts.
Both cathode resistors are 1k and the anode resistors are the same at 47k.
Is this a normal spread ?
They are both new valves.
On the front end I have two 12ax7's.
One of the 12ax7's has a cathode voltage of 4 volts and the other 2.5 volts.
Both cathode resistors are 1k and the anode resistors are the same at 47k.
Is this a normal spread ?
They are both new valves.
Last edited:
😕
in my amp ecc803/802 are as simple srpp input.
U Rk between
: same tube +-0,01V
: each tube +-0,1V
when i used triode 6l6 cathode follower, they were Rk +-0,3V (not paired)
normal cheap 0,6w resistors
not a advertisement on JJ, 803 are bit harsh treble 😀
Hi!
This deviation is not normal if the tubes are ok. Actually both values seem a bit strange for a regular autobiased stage with 1k and 47k. That would mean they draw 2.5 and 4mA respectively. With a 47k plate resistor this would require a very high B+ voltage.
Did you check that the circuit is wired ok? All resistors have the value which they should have? What is the B+ voltage and the voltages at the plates?
Best regards
Thomas
This deviation is not normal if the tubes are ok. Actually both values seem a bit strange for a regular autobiased stage with 1k and 47k. That would mean they draw 2.5 and 4mA respectively. With a 47k plate resistor this would require a very high B+ voltage.
Did you check that the circuit is wired ok? All resistors have the value which they should have? What is the B+ voltage and the voltages at the plates?
Best regards
Thomas
The circuit is definitely ok.
The B+ is 200 volts.
What happened was I was getting HF oscillation at certain positions of the gain and volume pots.
One of the 12ax7's had a anode voltage of 30 volts which was driving a phase splitter, but once I raised it to 80 volts the problems went away.
Cracked it.
I had a look through valve boxes and one is marked 12au7 and the other is 12ax7.
Somehow I must have got a 12au7 in my 12ax7 pile.
I had a look through valve boxes and one is marked 12au7 and the other is 12ax7.
Somehow I must have got a 12au7 in my 12ax7 pile.
With Ra = 47 k and Rk = 1k you need some 500 V as +Ub to get 2.5 V at the cathode. With +Ub = 200 V Ik is some 1 mA.😕
Also, with 47k Ra the linearity is far from optimum.
I usually have 220k as Ra.
I was trying to keep the anode resistor low as possible as it is driving a tone control stage next.
I have ordered some more 12ax7's and will look at the circuit again when they arrive.
The amplifier is playing as I type and sounds good.
The datasheet says 1.2mA current and a grid voltage of 2 volts for the 12ax7.
This gives 1k5 for cathode resistor and 82k for the anode resistor.
I am using 1k for cathode resistor and 47k for the anode resistor so I am not too far out.
I have ordered some more 12ax7's and will look at the circuit again when they arrive.
The amplifier is playing as I type and sounds good.
The datasheet says 1.2mA current and a grid voltage of 2 volts for the 12ax7.
This gives 1k5 for cathode resistor and 82k for the anode resistor.
I am using 1k for cathode resistor and 47k for the anode resistor so I am not too far out.
Last edited:
If lower output impedance is what you're after, why not retain the 12AU7, which has a lower anode resistance then the 12AX7 anyway. You should optimise the anode and cathode resistor in that case though.
The ECC83/12AX7 is not very good at driving anything, unless it has a very high impedance. A low anode resistance just ensures high(ish) distortion.nigelwright7557 said:
Usual rule of thumb for good linearity would be a plate load resistance of about 3x rp. I would also aim for a plate voltage of 100V or better - large signal swings modulate transconductance, and at low plate voltages you may run out of headroom. Who needs unnecessary distortion?
That's a 37.5% deviation between sections, and that is definitely wrong. I've seen some deviation from design nominal values, but nothing so drastic as that.
The design itself is hideous (unless this is a guitar amp) as 47K for 12AX7s is a very steep loadline, given a nominal Rp= 90K. For best linearity when using triodes, you need that passive plate load to be at least 2XRp, and more is better. The best performance form 12AX7s usually involves either higher than normal DC rail voltages or active plate loads. (One reason to prefer 6SL7s: lower gain, but the Rp= 40K is more manageable.)
Most complaints about inferior sonic performance from 12AX7s involves plate loading that's way too heavy for the type.
A 47K passive plate load would be more appropriate for types like the 6SN7 or 6FQ7.
- Status
- Not open for further replies.