• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

KT88 Cathode Current Runaway/Oscillations?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
M

mbeards

Member
Joined 2012
Hi everyone,

I just completed a mono pair of KT88 Ultralinear amps. I have them on the bench for first power up and testing and I am having an odd issue with both of the amps just built.

The amps are fixed biased adjustable via a pot and when ever I increase the current through the tubes to get to the operating points. Both tubes run away. I have cathode ammeters on the KT88's and can see them suddenly jump off the scale and skip about after increasing the idle current by a few milliamps. I tun the bias pot back down and the tubes are still in runaway/oscillating.

This is the second pair of these I have built after a few minor layout changes with the first working flawlessly.

Interestingly, this pair I added 120ohm screen resistors on the UL taps to push them on the safer side of stability.

Plate voltage and screen voltage is 550V. This will go down to more like 515V or so if the KT88's are properly biased (like the first pair built), but I cant get there due to the "runaway"

Any suggestions?

-Matt
 
sajti

sajti

Member
Joined 2003
mbeards said:
I am using Electro Harmonix. I think they are one of the better sounding offerings although I have not tried them all.

What would paralleling the 100K down to 50K achieve?
Click to expand...

Yes. I found thermal runaway with some 6550s as well. But lower grid resistor solved it. My experience that the Svetlana 6550 was the best with high value grid resistors.

I think that 100k is the maximum recommended with fixed bias. Higher voltage means more dissipation, which is the reason of the thermal runaway.

Sajti
 
Last edited:
M

mbeards

Member
Joined 2012
So I just tried paralleling the 100k down to 50K and the problem was a little bit better. I could now stop the runaway by turing the bias back with minimal current through the tube where as before they would still runaway and I would have to shut the amp off.
 
sajti

sajti

Member
Joined 2003
mbeards said:
So I just tried paralleling the 100k down to 50K and the problem was a little bit better. I could now stop the runaway by turing the bias back with minimal current through the tube where as before they would still runaway and I would have to shut the amp off.
Click to expand...

In this case You have check for oscillation. 50k must be low enough for any possible voltages. Did you connect any load on the output?

Sajti
 
sajti

sajti

Member
Joined 2003
mbeards said:
yes 8 ohm 200 watt dummy resistor. I don't want to run it too long like that and burn anything up
Click to expand...

Keep the dummy, and put parallel some cheap 22-33ohms 0.1-0.25W resistor. In case of a hevay oscillation this resistor will burn. If it happens, You can be sure, that Your amplifier is oscillating.

What kind of gate stopper are You using for the G2?
Is the stoppers are close to the pins?

Sajti
 
M

mbeards

Member
Joined 2012
For the screen resistors on G2 im using some of the PRP audio series resistors. They are soldered right on the tube pins.

Maybe try bypassing them all together like my first set I made?

I do have a global feedback network of about 15dB form the OPT 8 ohm tap to the cathode of the V1 6SN7
 
sajti

sajti

Member
Joined 2003
mbeards said:
For the screen resistors on G2 im using some of the PRP audio series resistors. They are soldered right on the tube pins.

Maybe try bypassing them all together like my first set I made?

I do have a global feedback network of about 15dB form the OPT 8 ohm tap to the cathode of the V1 6SN7
Click to expand...

What value is the G2 stopper? I found (not with KT88/6550, but with PL509) that insufficient grid stopper results ultra heavy oscillation. The tube push to the current limit the 3A power supply :)

Try to remove the feedback, and try the setting once again.

Sajti
 
Last edited:
sajti

sajti

Member
Joined 2003
mbeards said:
Feedback removed the amp is now stable. Same feedback setup as the first pair and same OPT so Im a little confused. Should I try reducing the amount of feedback?
Click to expand...

Are You sure that You don't set it opposite way (positive feedback)? If the output tubes are turned off (no bias), not enough gain for the oscillation. As You set minimum current, the oscillation is start immediately.
Check if the output windings are connected properly.

Sajti
 
M

mbeards

Member
Joined 2012
Feedback on this design is taken from the 8ohm speaker tap and sent to the cathode of the first 6SN7 tube. The feedback circuit is just a 15K ohm resistor in parallel with a 68pf capacitor. The 6SN7 is cathode biased via a 1K ohm resistor to ground. It is essentially a modified a modified Williamson Design and follows much of what is on the literature of the more modern Williamson style amplifiers.

Positive feedback had occurred to me too (aka incorrectly wired feedback loop) Ill need to check the phasing of input vs. output but again it is configured exactly the same as the other so It cannot be positive feedback (at least in the desired passband).

Ill try adding a large feedback capacitance to see If I can kill the oscillation with a dominate pole. Just hope it doesn't kill the performance too bad either!!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom