'73 MKii Superlead bias issues

[flintrock]

Hey guys, I have a 1973 MKII S/L on the bench with some wacky bias going on. I retubed it with KT77's (which was in it previously). When the amp came in it was at 60ma bias, too hot. Checked the original screen resistors and of course they were drifted way off spec. Replaced with all new 1K 5W 1% and now when I try to take a bias reading with my bias probe it's off the meters scale and I have some crackling going on in the speaker with the volumes down that wasn't there before. All the resistors in the bias circuit measure good, but theres some shady topside resistor replacement done before I got this amp. I am going to replace all of those. The two bias caps look fine, there ITT's 10uf 160V, figure I might as well change those why the board is out. Trimmer looks ok but could be questionable as welll since it is original. Grid resitors check out fine 1.5K, as well as the pair of 220K's. Plate voltage is 520V with no tubes installed. Anything else I should check why I am in there? And will the components I am going to replace get this back in line?

[ampgeek]

Welcome Flintrock!

I am not very familiar with that circuit and didn't pull up a schematic so I will speak only in general terms here.

What range of bias voltages do you see when you sweep the bias pot from end to end? Guessing that it ought to be somewhere in the ~-30VDC to ~-50VDC range.

Good luck!
Dave O.

[flintrock]

I was at -60 to -90 before I changed the screens. Now I'm over -200, off the 200ma scale. I think I'm gonna change the 47k bias resistor to 56k to calm it down a bit.

[ampgeek]

Just to clarify, are you reporting voltage or current?

You need to check the bias voltage at the control grid, g1, of the power tube (which controls the current flow through the tube) to best determine the "health" of you bias supply system.

Let us know!
Dave O.

[flintrock]

With tubes installed...510 plate voltage and -43.9 neg bias on pin 5.

[flintrock]

Initial readings were ma via my bias probe. Last reading is in mv on pin 5.

[flintrock]

Range on bias pot is -36 to -46 mv.

[ampgeek]

Looks like a major problem in the bias supply system. Normally, you would be seeing a MUCH higher negative voltage on the control grids. You are reporting millivolts. I was expecting to see those values in volts. Those low bias voltage values are putting the tubes into uber over conduction as your probe is already telling you.

I would look closely at the bias supply system all the way up to the tube control grid.

Dave O.

[flintrock]

Range on bias pot is -36 to -46 mv.

[vibratoking]

-36 to -46 mv.

Since you posted it twice, it must be in mV. As stated, you have a major problem in your bias supply circuitry. Measurements at that level would seem to indicate that you are measuring noise and not any 'real' voltage. Check for an open circuit from the negative rail in the bias supply or a short to ground.

[flintrock]

Doublechecking it is volts. No idea why it posted twice or why I wrote mv. Not really trusting those bias caps.

[ampgeek]

That indicates that the bias voltage supply is probably good. Divide and conquer! Onward to something else.

Take a look at the power tube screen current next. Measure the voltage drop across the screen resistors for each tube and divide by the resistance (1000 ohm IIRC) to get current. Try to do this quickly (and carefully!) so you don't meltdown your power tubes.

Dave O.

[flintrock]

Meter was bad. With new JJ KT77's I am only able to get down to -59ma using the probe. Plate volts are at 438.

[ampgeek]

That sounds just about right! Remember, some of that 59 ma (note: it is not a negative value) you are measuring is coming from the screen grid contribution to the current that you are measuring at the cathode. Maybe 4 to 6 ma.

If you measure the voltage drop across the screen grid resistors, you will know that value and can, generally, use it in later calculations.

KT-77's look to be good up to 32W max for plate dissipation. A good target is 75% of that max value...or...24W (assuming Class AB-1 toplogy). If we subtract 4 ma for the screens, you are sitting at ~23W. So....just about perfect in my book!

You might be able to squeeze a few more ma of current through the plates if you knew the screen grid current flow.

Cheers,
Dave O.

[flintrock]

Changed the bias resistor out to 56k and the range is much better now. About halfway through the trimpot travel is perfect. Looking at the power cord coming in, it is far from correct. It has been changed to a 3 prong but the ground isn't even connected to the chassis! I'll get that fixed, but the polarity switch is still hooked up, minus the death cap. I thought these were supposed to be bypassed and taken out of the circuit. Here's some pics maybe someone knows the correct way to rewire this mess.

[IMG768]http://i1000.photobucket.com/albums/af122/flintguitar/001_zps66763630.jpg[/img]

[IMG768]http://i1000.photobucket.com/albums/af122/flintguitar/003_zps219b464d.jpg[/img]