AC filaments too hot!

[JBeast]

I blew a fuse on the AC filament line and realized today that I had my power tubes -KT77- biased too hot (facepalm) - I'm not sure what I did but it was set around 10ma hotter than it should have been for the plate voltage and I know not to do that if I want to keep my tubes for a while - I am not always as observant as I should be. The plate voltage is reading around 480V and the KT77's were biased around 45-50ma. I readjusted them down to about 35ma to keep them in the mid to high range but not over. They seem ok with full range of sound still.

Amp = Crate BV120H (Blew Doodoo), KT77

I have a burnt area of the board under the AC Filament fuse that popped and the power tube filament wires look burnt where they meet the board. Also, there are a couple burn marks on the solder joints where the filament traces connect via 13 pin header to the power tube board. I guess what I'm wondering is: Is there anything I should be concerned about that I'm overlooking? It seems like the only problem was that I had somehow over-biased my power tubes but the fact that the filament wires and traces look a little toasted too has me concerned. Is it normal for the KT77 to burn the filament wires and pop a fuse because the bias was set too hot? I've read on a few forums that KT77 can draw more voltage through the filaments than the PT can safely supply with some amps.

The clean channel wasn't exhibiting the crackling or the volume drop symptoms but the clean channel draws it's filament current from a separate DC line while the 12AX7 before the CH2 tone stack and the PI both draw from the same AC filament line the power tubes draw from. My guess is that the crackling was the sound of the fuse burning before it blew but it could have been the power tubes choking as well - although they don't seem to have problems after my readjustment. I have some used EL34's I'll try too but I don't have any new tubes to test out.

I replaced the fuse, readjusted the bias and everything sounds fine but I'm concerned about how hot the filament wires got and the blown fuse. I'm still quite the newb with regards to this tube amp stuff so I'm not sure if I caused the problem with the over-bias or if the KT77's are frying my amp as this is the first and only amp I have ever worked on.

I've noticed this amp runs a little hotter in general than it's original specs. The schem says B+ = 460V but mine runs around 480V. I went over all of the relevant test points on the schem relating to the preamp and power tubes and they all seem consistent with respect to spec and each other and they are all consistently on the high side compared to the schematic specs. I'm not burning through preamp tubes or anything and the PI is fine too...

[Enzo]

Correlation is not causality. Bias has nothing to do with heater wiring. You noted that your tubes were biased a little bit hotter than you wanted, then you also noted the heater wires were burning up. Separate issues.

The amp was designed to use four 6L6 tubes. That tube draws 0.9 amps of heater current each - 3.6A total. Your KT77 tubes draw 1.4A per tube on the heaters. 5.6A total. You are now drawing 2A extra through the heater wiring and all its connectors. Plus when the tubes are cold, as when furst turned on, their resistance is lower, so the current draw is even higher until they get warm. That is most likely why your heater fuse opened. Bias has nothing to do with it.

The weakest link in this is going to be the connectors, including the fuse clips. They were probably already marginal, but drawing over 50% more current through them is going to stress them further, and they burn up. Ribbon connectors are especially at risk. And the transformer. Most transformers are wound with some extra capacity, but this 2A increase is asking a lot. With all tubes installed and the amp running, measure the 6vAC at the tube sockets to see if it has dropped from the increased draw. If you find 5.9v or something, the transformer is struggling.


These are guitar amps, not precision circuits. NASA can send highly sophisticated circuits into space on their satellites, we have crude basic circuits to distort guitar sounds. There are not critical adjustments in there. Bias is not critical. People may obsess over a half a milliamp trying to get EXACTLY 70% dissipation - or whatever - but that is not necessary for the amp, that is only necessary for the owner's notions. Having your bias currents a few milliamps one way or the other is not going to damage the amp. Sure, if the bias is way too hot, your tubes will not last as long, but it won;t be night and day. But basically, if the bias is such that the tubes are not red plating or so cold as to pinch off the sound, then the bias is acceptable. AMps like the 5150 come from the factory running at something like 14ma - and there is no adjustment.

[Enzo]

Oh, and your voltage. 480v instead of 460? 20 volts high is an increase of only 4% over 460v. Old Fender schematics used to say right on them, voltages are for reference only and may vary as much as 20%.

If you have 120v mains voltage and your B+ is 480v, that is a 4 to 1 ratio, so for every volt your mains changes, your B+ changes 4 volts the same direction, so a 5v rise in maains voltage will mean a 20v rise in the B+. None of the voltages are regulated, and your mains certainly is not. Here, my mains are sometimes close to 125vAC, while on a hot summer day when every air conditioner in town is running, that voltage can sag to under 110v. Think of the range of B+ I can have with that.

These are not precision circuits.

[JBeast]

I measured the filament voltages and this is what I got:

PreV1 (DC): 6.20vdc
PreV2 (DC): 6.19vdc
PreV3 (AC): 5.95vac
PreV4 (AC): 5.80vac - Phase Inverter

PwrV1&2 (AC): 5.70vac
PwrV3&4 (AC): 5.92vac

[JBeast]

You are definitely correct in your comments with regards to the component quality. I've had a few fail on me in the past year or so and I've been upgrading as I go along. Enzo, you helped me out a few months ago with a similar issue I was having with the power tube filaments not getting power. I had the same KT77's in there at that time. It looks like the tiny traces, tiny solder joints and tiny header pins have a hard time handling the current these KT77's draw. So, if the heater current isn't too low and it seems as though the PT is handling the current ok is it safe to assume that with a bit of reinforcement (component replacement - better fuse clip - and hard filament wiring) that I can get away with using the KT77's? I've pretty much overhauled the preamp, voicing and distortion characteristics to suit my taste using the KT77.

[JBeast]

Oh, and your voltage. 480v instead of 460? 20 volts high is an increase of only 4% over 460v. Old Fender schematics used to say right on them, voltages are for reference only and may vary as much as 20%.

If you have 120v mains voltage and your B+ is 480v, that is a 4 to 1 ratio, so for every volt your mains changes, your B+ changes 4 volts the same direction, so a 5v rise in maains voltage will mean a 20v rise in the B+. None of the voltages are regulated, and your mains certainly is not. Here, my mains are sometimes close to 125vAC, while on a hot summer day when every air conditioner in town is running, that voltage can sag to under 110v. Think of the range of B+ I can have with that.

These are not precision circuits.

This explains why my tube amps seem so temperamental despite new/old tubes. One day they sound loud, proud and growl with conviction and the next day sound a little squishy and not so rigidly defined - they have a human like element to them, lol.

[Enzo]

There are so many factors. That is one. Others are relative humidity in the air, temperature makes a difference. Loudness has a large effect on your ears, so if you happen to turn it up louder one day compared to another it will sound different. Look up "Fletcher-Munson Curves" for insight. If there are drapes in the room and they are open one day and drawn the next, that has an effect on room acoustics. Even your mood has an effect, among other things it can affect you you address the guitar, how aggressively you pick, etc.

Lots of factors.

[JBeast]

With regards to the voltages I posted below, does it appear that the PT is capable of handling the filament consumption despite poor quality components? I've been looking up tutorials and other posts about how to figure this out myself but I'm not getting any straight answers from my search engine queries. As in: Which specifications do I need to look at in order to determine that my PT is capable of handling a specific amount of current draw? I see tube spec sheets showing how many mA are required to support a specific tube but I'm not sure how to translate the tube specs to determine the answer. So, I know based on my readings that the KT77's are drawing 5.7vac but the spec sheet for JJ-KT77 says the requirement is 6.3v... How do I know what the PT can handle?

[Enzo]

What poor quality components?

The transformer puts out 6.3v at its rated current load. The tube heaters do not draw a voltage from the transformer, they draw current. Due to resistance in the winding and elsewhere, the more current you draw, the more the voltage will drop. The very fat that your tubes have dragged the 6.3v down to 5.7v is telling you the tubes are more load than the transformer can really handle. Like if your car is pulling a trailer and you can't get it over 40mph, then you are towing too much trailer. ACtually, a better example is the power in your wall sockets. ON a hot humid summer day, when every air condityioner in the state is running on high, the current draw is so high, the voltage on the power lines drops. If that gets bad enough, you have a "brown out."


Next to that 6.3v heater on the spec sheet, there ought to be a current rating. That is what I was talking about in post #2, second paragraph.

[Tom Phillips]

… does it appear that the PT is capable of handling the filament consumption despite poor quality components? I've been looking up tutorials and other posts about how to figure this out myself but I'm not getting any straight answers from my search engine queries. As in: Which specifications do I need to look at in order to determine that my PT is capable of handling a specific amount of current draw? I see tube spec sheets showing how many mA are required to support a specific tube but I'm not sure how to translate the tube specs to determine the answer…

I believe that you are specifically talking about heater current winding spec of your PT. If you were designing from scratch you would add up all the current drawn by the tubes that you plan to use and then pick a PT with sufficient capability. However, since you don’t have the PT specs, your best answer lies in Enzo’s post #2 of this thread. In your case you would add up the total heater current required to run all the tubes that the amp was originally designed to use. You would then assume that Crate would not pay for a PT any larger than necessary to run the stock BV120H amp. It then follows, again based on the current draw that Enzo posted, that your transformer is overloaded and running hotter than intended. This is further supported by the low heater voltages that you measured. This would not usually cause a burnt looking PCB or wires. That may be caused by bad solder joints, loose fuse clips or something else that raises the resistance of the current path. Can you post photos of the overheated area?
Tom

[JBeast]

Thanks again for your advice Enzo, I appreciate it.
I was going partially on your original comments above "The weakest link in this is going to be the connectors, including the fuse clips. They were probably already marginal," Also, in my experience with working on this amp everything has pretty much been marginal at best". I've found a couple 1/4 watt carbon comp resistors that drifted significantly in value and I've had to replace the the filter caps because the bass became really mushy and one of the filter caps was microphonic. I've been using this thing as a learning experience.

I somehow missed the part in your original post where you said that if the voltages were around 5.9 or less that it's straining, sorry if I re-asked a question you already answered. I think I'm beginning to understand. Looks like I'll stick to 6L6 tubes in this amp because I can't justify plugging in another $275 for the Mercury Magnetics replacement. This thing is really hard to work on due to the thin, close, traces lifting and the thin, brittle, single sided PC board. I always have the sense that this amp is ready for the scrapyard but I can't bring myself to get rid of it as long as I am actively using it.

[JBeast]

Oops, somehow I missed the part in Enzo's original post where he said "If you find 5.9v or something, the transformer is struggling." Not sure how I missed that or if I just didn't remember it was stated... sorry for asking a question that was already answered.

Ok, so, in reality it is difficult to know the true specs of the PT without it being explicit?
I'm beginning to understand now. I guess I was a bit confused because of my own bias in thinking with regards to the fact that EL34's work in this amp. I ran Eurotubes integrated quad in this thing for a few years and it seemed to handle it without the burning or fuse issue. Because of this I assumed that it was meant to handle EL34 types as well - Fallacy in logic on my part.

Here are a couple pics of the burnt areas all relating to the filaments.

[Tom Phillips]

The "burnt" areas in the photos you showed don't look too serious at this time. The look is typical of PC Board areas that get too hot from adjacent parts that are dissipating lots of heat. My opinion is that the fuse clip and wire connection points are marginal. It will help is you switch back to 6L6s. Also do a close inspection and make sure that the solder joints in the area shown are good and that the fuse clips are tight.I think that the amp should sound good with 6L6s if the bias is set at the correct value. Cheers, Tom

[JBeast]

Tom,
Thanks for the advice, I appreciate it. I'll rework those solder joints, get some 6L6's and throw them in there instead of messing around with trying to modify it for more power. Mercury Mag. has a replacement PT for $275+ but it's more than I want to throw into this thing as it's pretty much just a kicker and project amp and it's not high enough build quality to make it worth the money - I've been using it as a tool to learn troubleshooting, repairs and mods mostly.

[JBeast]

I just thought of this: Couldn't I add an auxiliary filament transformer?
One of these? http://www.tubesandmore.com/products/P-T166Q6
or these? http://www.tubesandmore.com/products/P-T167S6
or these? http://www.alliedelec.com/search/pro...x?SKU=70218457

Would I have to isolate the two power supplies from each other?

I could bolt it onto the chassis, run the power lines to the same place as the PT and the 6.3V lines to the same joints where the PT filament lines are soldered in, no?
Are there any disadvantages or caveats to this idea? The solution seems workable, simple, inexpensive and would allow me to run EL34 and KT77 tube types.