Burning Screen Resistor

[bal704]

I've got a 150k resistor, followed by a diode leading into my bias circuitry. I've got voltage before and after the 150k resistor, but nothing after the diode. The diode stripe is towards the voltage input, as seen below. I tried to replicate the circuit below. I'll triple check it.

[sluckey]

What does "A" connect to ? Is there a bridge rectifier used for the B+?

[bal704]

Yea...bridge rectifier in my case. I Connected ‘A’ to one of the ac inputs to the rectifier.

[martin manning]

Steve is on the right track... There is no negative voltage available when a bridge rectifier is used. You will need to capacitively couple the bias supply to the HT ala JCM900.

[bal704]

Alright...let me digest that and make some changes to the circuit. Thanks.

[martin manning]

Simple to implement with your existing layout. I'm surprised you didn't burn up a lot more than a screen resistor.

[bal704]

I rewired my bias circuit to match this one (at the bottom right):

http://www.valvewizard.co.uk/bias.html

I now have negative bias voltage with no tubes installed. I adjusted my bias adjust pot to -45v (with no tubes) as a starting point. When I plugged my 6l6's back in, I had sound and no smoke so far....

These are supposedly matched tubes, but one had bias of 20mA and the other 40mA, so I'll try different tubes. Somebody mentioned previously that 470 ohm resistors weren't big enough for EL34's (which is my preferred tube in this amp), but I do want the ability to run 6L6's also. Should I change the screen resistors to a different value, or just leave the 470 ohm?

It's possible I did some unseen damage here....what other parts should I be looking at for possible damage?

[nworbetan]

The screen resistor value and being able to run 6L6 comfortably are both at least a little dependent on what your B+ is.

For example I have a '59 Bassman reissue (virtually the same as a JTM 45) with a B+ just shy of 500 with a solid state rectifier plugged into the socket. Setting the bias at -53v put the plates dissipating 18 watts, or 60% of max. With a lower B+ -45v may be enough for 6L6, or maybe not quite enough.

I don't know the math for the screen resistors well enough to give an example of that, but I'm sure it's similar to the bias situation: 470r might be good to go or it might be crossing into unreliable territory with modern EL34s, screen voltage depending.

[tubeswell]

Classic 50W and 100W Marshalls that run EL34s typically use 1k to 2k2 screen grid resistors rated at 7W (Welwyn ceramic green 'cigar shaped' resistors are 7Wers)

[Roe]

some early 50w marshalls used 470R individual screen resistors and a shared 110ohms choke. These amps are very hard on the tubes, and marshall (and most techs) recommend 1k screen resistors with el34s. You may use smaller values IFF the choke has high resistance (e.g. 690 ohms like the RS deluxe choke) or iff the voltages are low. PErhaps 470R may work at 400v or so. At 470+v you need 1k.

[bal704]

I swapped the 470R resistors for 1K resistors. I also put in my matched set of EL34's. Like my matched set of 6L6's, one tube had a bias of about 20mA, the other about 40mA. This is through bias jacks and 1R resistor. The same socket had the high/low bias on both 6L6 and EL34.

Hard to believe that 2 'matched' sets would have the same problem. What should I be looking for that may be causing this?

[tubeswell]

I swapped the 470R resistors for 1K resistors. I also put in my matched set of EL34's. Like my matched set of 6L6's, one tube had a bias of about 20mA, the other about 40mA. This is through bias jacks and 1R resistor. The same socket had the high/low bias on both 6L6 and EL34.

Hard to believe that 2 'matched' sets would have the same problem. What should I be looking for that may be causing this?

Did you try swapping each pair of tubes around?

[pdf64]

I'm building a JTM45 clone...

Yea...bridge rectifier in my case. I Connected ‘A’ to one of the ac inputs to the rectifier.

For future reference, just to note that it's confusing to describe something as being a clone when it's not

There are several different JTM45 schematics available, and some of the points of divergence apply to the screen grid supply arrangement, so even if it was a clone, the relevant schematic should be provided.
Due to its full wave bridge rectifier, as none of the regular JTM45 circuits can apply to your amp, it would be very beneficial if you were to supply an accurate schematic.
I acknowledge that will require a bit more effort on the part of the OP but this thread demonstrates that effort would have been well worth it.

How are you assessing the power tube currents?
If using current sensing resistors, bear in mind that the fault current that would be associated with the zero bias operation previously reported may have damaged them, such that their value is now outside spec.
1 ohm resistors are rather difficult to measure accurately, so if that's what's been used, then best to replace them.

[martin manning]

Even a half watt one ohm resistor will handle 700 mA...

[Stevem]

Yes, but it's right at .490 watts!