Low bias voltage throughout- is it the quad cap? [Update: It was the quad cap.]

Hi Bob/anyone,

I had an issue last week after running the amp for a long period, where I heard a quiet "pop" and then the amp slowly became noisy over the next 5-10 seconds. I let it run for about 3 minutes as it was still playing, then shut the amp off and took out the multimeter, turned it on and monitored for stability. Bias voltages on all 4 points were 100-150 mV, far from the 400mV they should be. However the amp still semi-worked even at that, just was noisy.

My first assumption is the quad cap, and I found this post: https://dynacotubeaudio.forumotion.com/t1053-blowing-rectifier-tubes and tried to measure capacitance on the quad sections based on that with the multimeter (which does have a capacitance feature) but results were inconclusive or low, in the range of 10-50 nF. This either means that the capacitance is actually very low, or I did it wrong. If someone could explain the procedure (which points to test, I just went from ground to the leads) for this then that might help.

The other possibility is the rectifier tube, but I figured that would make more of a show if it went and I didn't see anything, nor do I see anything when it's turned on again, seems normal. However the rectifier is the original Mullard 5AR4 still, so I wouldn't be surprised.

Thanks for any help diagnosing this!

Forgot to mention—the amp is a VTA mod of an original ST-70, and worked great with regular use for about 4 months prior to this.

I would remove the 4 output tubes and the 5AR4 tube and check the voltage on pin 5 of the output tubes. With the bias pots turned all the way to the right you should have approx negative 70 volts and when you turn the bias pots to the left, the negative voltage should drop to approx negative 20 volts more or less. Let us know what you measure.

Sal

With Bias pots turned all the way to the right: -30.0 V

All the way to the left: -57.2 V

(or those could be reversed depending on what you mean by right and left; I took right to mean fully counter-clockwise, and left turned fully clockwise)

I'm guessing we're measuring the PT directly and those are a bit low?

-57.2 volts should be OK, hopefully Roy or Bob will give their opinions on how much negative voltage is coming out of the VTA board, but from what you measured, that should be enough to bias the tubes.

Next, I would measure the B+ on the quad capacitor, should be around 475-500 volts on the first section. If you B+ is low, your bias will be low. Does your quad cap get hot pretty quickly?

Sal

And that would have to be with rectifier plugged in, but probably still not the output tubes, correct?

Measure the B+ without the output tubes, then with the output tubes. Make sure you turn off your amp with you install your output tubes.

432V with tubes out, 425V with tubes in. Seems a tad low.

I'm not sure about the heat on the quad cap, never really paid attention to that. Also I know it is one of the old-style quad section can caps, not a newer design with better caps housed in a can.

Do you have another 5AR4 on hand to try out and measure the voltages again with the tubes in? Voltages without the tubes seems low but pretty close with the tubes in. Per the dynaco schematic, on pin 3 of every output tube you should measure 410vdc give or take..

I do not, nope. Just the one. Might be a good idea to pick up another anyway as it's the only "old" tube of the bunch. I'll measure on Pin 3 when I have some more time and get back to you, and see if I can find another 5AR4 to test with. Thanks for all your help so far.

calinet6
all your voltages are in the correct ranges, normal ST70 will have around 425v B+, bias range -60v to -30v.
the actual bias setting on the VTA board will vary depending on the output tubes, but you can typically measure around -35vdc on pin 6 of the output tubes, and that should give about 400mv across the 10 ohm bias resistors (40ma of current).
Voltage to the VTA board should be around 380-400vdc after going thru a 2.2K 2w resistor on the last section of the quad cap.
Current draw to the VTA board should be around 10ma, which is 22v across that 2.2K resistor.
Maybe you popped a cap on the VTA board if the voltage was too high, or the output tubes were removed and the amp powered up (which will cause B+ to up to 460-480vdc due to no load on the power supply).
I'd recommend a new 5AR4 tube, and a new quad cap.

Thanks for your help Sal & tubes4hifi, I got a new quad cap from Dynakitparts - the upgraded modern one with the 4 discreet caps just mounted in the can. http://dynakitparts.com/store/product.aspx?id=273 Should have got that one to begin with

Also got a new Ruby 5AR4 for good measure and everything is working and sounding wonderfully. It might be my imagination but I think it sounds a little better than before.

The biggest clue that the quad cap had gone bad was a low 120Hz hum along with the low bias voltages. The "noise" I talked about in the original post was indeed 120Hz hum once I measured it.

I say... "Just say no to quad caps" I've been running these hand hacked 800V+ cap boards for about a year with great results. I hope you can see this photo, Photobucket stopped working for me and this is my first try with Flickr.

" alt=""/>

Finally, say bye bye to that quad cap. At best their marginal. No one knows how to plug the hole LOL so this is why people still use them. Probably the same logic behind using two triode switches.

People, you can always polish the cap can and leave it in place. You can gut the cap and install caps inside it or you can remove it and install a plug in its place.

Decent power supply wakes up the Dynaco and the stock values in the can just aren't good enough.