Blown screen resistors and more

[titser_marco]

Actually, the JJs burned the last screen resistor.


In any case, would you know if the TAD 6L6GCs are STR spec?

[Stevem]

Yup, any tube can fail and a lot of times that's the result, but many times it's not the fault of the tube but poor connections in the socket.

I am not sure about the TADs, but I think the Tung Sols are.

[Phil_S]

Was doing some research and found this discussion somewhere else:
http://www.fenderforum.com/forum.html/searchthreadgateway.php?topic_number=740085
Based on this, it seems to be that a way to alleviate this problem without having to lower B+ is to use a higher screen resistor (1k instead of 470ohms) but still keep it at 1w so that the resistor gives instead of the tubes.

Thoughts?

Yes, I think this is not particularly relevant to solving the problem. This discussion is about fine tuning something that many people can't hear.

This resistor is there to kill oscillation more than anything else. It is good to have, but it is not meant as tube protection. I'm doubtful a change from 470 to 1K produces any meaningful or audible result. Increasing the wattage rating of that resistor will make it less likely to go open, however, I am having quite a bit of difficulty imagining that a half-watt resistor isn't good enough. The upper end of the relevant range for a 6L6 screen is around 20mA. A voltage drop of 25V across that screen stopper would be required to be at the half-watt limit. I seriously doubt you'll be seeing that sort of voltage drop. That said, many people use 1W or 2W rated resistors because they dissipate the heat better and/or because the supply voltage is over the rating spec for the resistor. (Most half-watters are rated for 250V or 350V service.)

Even so, I think you are looking at the wrong thing and this will not solve your problem. Those screen stoppers went open because either a) bad tubes or b) something else. It is doubtful that all tubes (or 3/4) went bad at exactly the same time. I wish I had a thought about b) but I don't have enough information to formulate a guess.

I suggest it isn't necessary to concern yourself with whether the tubes are STR or whatnot. That won't solve the problem either.

You've got a fundamental problem that is causing excessive current draw. That's what needs to be found.

[Firestorm]

I'm confused (nothing new there) . You said you were modding this Bassman. In what ways? You said it was working with the GE tubes. Working right; sounding good? You said that after the swap to the Sovteks, the screen resistors "lit up and burned." That makes it sound like they caught fire; did they? You said the remaining tube lit more brightly when you put the switch on standby. Do you really mean that or was it when you switched from standby to "play"?

Then you put the old GE tubes back in and it worked? With three open(?) screen resistors?

And the B+: you measured 559VDC. Tubes in or out? (Even with them in, if three screen resistors opened up, they would be trying to run as triodes and probably couldn't draw much current).

The PT was replaced by a previous owner. With what? Can you read any numbers off it?

Finally, you're focusing too much on voltage at the moment. Things don't burn up from voltage, per se, they burn up from current. So as Phil said, there's a current draw problem. With all the burning up of screen resistors, I'd be wary of the condition of the sockets.

[Phil_S]

You've got a fundamental problem that is causing excessive current draw. That's what needs to be found.

I know it's a bit odd to quote myself...IOW, there is probably a short somewhere.

[Firestorm]

Right. I don't see how a tube swap takes out screen resistors unless the screen shorts to the cathode or the suppressor plates. You'd see that. And I wouldn't expect the tubes to work fine in another amp afterward. Something screwy as yet undiscovered.

[pdf64]

Maybe the tube sockets would benefit from re tensioning?

[Phil_S]

I guess then, I'm left with two options

A. Lower the B+

The schematic says fully engaged the amp supplies 500VDC to the output transformer. Whether it's a good design, I can't say. What's there should be able to withstand the spec'd voltage. Plate voltage will be under 500V.

B. Get tubes that can take the high B+

6L6GC in UL operation should be OK with plates in the 490-ish range. The UL taps (let's assume 70%) should keep the screens at 350-ish which creates an appropriate operating environment for the tubes. While you don't want Russian 6p3s tubes, most any "real" 6L6GC should be fine.

For Option A, I can think of the following:

A1. Use a dropping resistor to drop the B+ to about 440-460v.
- What value should I use? What wattage? Can I just put it just before the standby switch?

A2. According to some sites, I can also use a Zener diode.
Again, what value should I use? I saw some using 5W Zeners from CT to ground, but I was wondering what voltage rating the Zener should be. And could someone confirm where exactly to put this? And are there any disadvantages to this method? My local tech said that this generally doesn't work for high-voltage environments and that I risk having high voltage swings.

This doesn't seem to be relevant to the problem. Fix the amp before you start fiddling with mods.

A3. Get a new PT.
Getting a new PT is definitely my long-term solution, but I would want to be able to use the amp while waiting for the new PT.

Meh...see above. This is the design. See if you can live with it. A new PT is going to be pricey and I'm doubtful it will fix the problem related to the burnt screen resistors. Those will fry just and nicely at 450 B+.

B1. Get KT66 tubes.
Heard that they were essentially more robust 6L6GCs but I wanted to verify this.

B2. Get KT88 / 6550 tubes.
Given the high B+, I looked around for datasheets on these tubes and it seems that they may be able to take the high B+. Do I need to do any mods before putting these in?

A tube change, if that's what you want to do, might be worth doing after you get the amp running right in stock format. Changing things around will hamper your troubleshooting efforts.

Assuming their pin out is OK, it is really just a matter of getting the bias right. By OK, I mean pin 1 is sometimes used as a tie point for components. If any of those other tubes use pin 1, you've got to accommodate that.

In general, I suggest sticking with the amp as is and concentrating on finding and fixing the problem, not doing the stuff on this list. IMO, there is Option C: fix the amp without modifying it. In the end, you'll be glad you did.

Do a few things to help us help you. Some pictures of the chassis with close ups of the problem areas are in order. A clear description of the problem(s) you've observed (without suggesting any cause) might also be in order. Was there any smoke? Have you disconnected the output transformer to test for any open or fused windings?

[titser_marco]

Thanks for all the replies. This reaffirms that I went to the right place to learn about tube amp electronics for guitar.

In any case, maybe this needs to be clarified as I am not sure if I had said it above. I got this amp about two weeks ago and it was working, but I decided to take a look just to see the innards. Lo and behold, the seller essentially sold me an amp that has been overhauled. By this, I mean the following had been done:

- Chassis was now an aluminum one instead of the original heavy steel Fender chassis. Tube sockets were not original as well
- The original Fender board has been replaced by terminal strips
- PT was now something that was made by "Philippine Transformers Inc" which led me to think that it is not original

Clearly, I should have been more careful with this purchase as this is a bum deal. I then figured, oh well, I might as well use this amp to learn tube electronics and stuff.

So I started working on the preamp section trying to modify it to D-type amp specs and was able to do it successfully using the schematicsheaven website. At that time, it didn't occur to me to check the voltages that were being thrown at the output stage.

Now the problem really started when I swapped the Sovteks in place of the original GEs. I didn't change anything at this point, just to be clear. When I plugged all four in, the two leftmost tubes' screen resistors burned up.


I then tried the JJs and then the third screen resistor blew up. I turned it off and then the last one went off. This was the only time I bothered checking B+ and I discovered that it was hovering around 580. Checked it again the day after and it was at 600v.

So I guess you're all right: I need to get the power and output sections working first and I've already called on my amp repair guy to rewind the PT as well as the choke since it seems that it was taken down as well by all this shorting. I disconnected the OT and checked the primaries and they were still all OK.

While waiting for that, do you think it's a good idea to take out all the power tube sockets, clean them up, test, and then rewire? I will try to send pictures the soonest I can.

[Firestorm]

Unfortunately, this not an overhauled Fender, it's a poor copy with pirated trademarks. Prevalent in parts of Asia as I recall.

[titser_marco]

Unfortunately, this not an overhauled Fender, it's a poor copy with pirated trademarks. Prevalent in parts of Asia as I recall.

Could have been that, this much is true. In any case, I think I'll just try making lemonade out of this rather annoying lemon and learn as much as I can.

Addendum: Googled right after I read it and boom, found the link below. She sockets are just like the ones on mine:

http://www.tdpri.com/forum/amp-central- ... r-amp.html

[Phil_S]

So, it's a bad copy! Finally we get an idea of what we are dealing with.

If the sockets are tight and no evidence of arcing, really there is nothing to be done. Don't look for things to do that don't need doing. You are likely to create a problem. If there is a carbon track from arcing, the socket needs to be replaced.

[titser_marco]

So, it's a bad copy! Finally we get an idea of what we are dealing with.

If the sockets are tight and no evidence of arcing, really there is nothing to be done. Don't look for things to do that don't need doing. You are likely to create a problem. If there is a carbon track from arcing, the socket needs to be replaced.

A carbon track along the wiring side of the socket or the tube side? Is there any way I can still use the sockets after cleaning them?

[pdf64]

6L6GC in UL operation should be OK with plates in the 490-ish range. The UL taps (let's assume 70%) should keep the screens at 350-ish which creates an appropriate operating environment for the tubes

Phil, I don't see how you get that?
Under static conditions, UL screen grids are up at VB+.
Under drive, their instantaneous voltage will be between their plate voltage and VB+, but closer to VB+.
I think UL taps are usually at 40% of the impedance between CT and plates.

[Phil_S]

This is going to be a minor hijack of the thread. Apologies for doing it.

I think I misunderstand UL winding. I am operating with the understanding that 40% means the tap is at the point where 40% of the turns are on one side of the tap and 60% on the other side. (This is with respect to each half from the CT.) Since there is a direct relationship between turns and voltage, if the end of the winding puts out 300V, the 40% tap should put out 120V (or is it 180V). Where did I go wrong?

I just attempted some Google research, but found information that goes over my head. It all seems to come back to this article: http://www.r-type.org/articles/art-140.htm In that article, the winding information is at the bottom. It looks to me like the UL tap is at 20% of the turns.

I'm having difficulty understanding how impedance is a factor here. Even after allowing for some interaction between the tube, the plate, the screen, and the transformer, it seems to me the turns between the plate supply (end of the winding) and the UL screen tap are going to keep the screen somewhat below the plate voltage.

Please help me to understand this. I see that I should not have posted what I did about UL.

Thanks.