Still Wrestling with a Crrate GT500H/VC50 bias

[stoneattic]

I'm still playing with that Crate GT500H that I was asking about HT fuses on a while back:

https://music-electronics-forum.com/...ad.php?t=47617

As described in the previous post it had a blown HT fuse. I replaced it with a fast acting fuse, figuring that was the safe play, started it with the current limiting light. Everything looked good so I used it for a jam and it worked fine for a while. Then the next time I turned it on no sound, so I pulled it apart, found the fuse was fine and decided maybe it has a flaky solder joint and re-flowed all the joints. Put it back together and played it for a bit. It seemed fine, then blew another fuse. I put in a slo-blo and again it was fine for a bit then blew the fuse again. Nothing else appears or tests bad. At this point I'm thinking one of the Sovteks EL84s are bad/intermittent.

So I ordered a quad set of JJs and tried those, again with the current limiting bulb and all looks good. I plug it directly in and measure the bias and am getting 370V on all four and 41-43mA. Which seems crazy high. I did check the bias on the Sovteks at some point and they appeared to be 2 sets of 2:

V 386 378 380 377
I 19 29 30 20

Much lower. Of course in my mind at least one of those is bad.

So do I go with the 41-43mA JJs as is or try to figure a way to bias them colder? From what I see on the schematic I don't think I can get much more negative voltage on the bias supply:

crate vintage club 50.pdf

Any thoughts?

[The Dude]

I'm betting your right with your initial assessment of an intermittent tube. Just curious, how did you measure bias?

[stoneattic]

I used an Amp-Head bias probe. These guys don't seem to be around anymore so here's the archive.org capture:

https://web.archive.org/web/20060211...products_id=45

The options I am seeing right now are jumper past R77 and live with what ever bias I end up with (-15.7v and what ever that ends up being for current), since as far as I can tell that's the max bias available, or add an additional circuit for more bias. I've come across a couple of mods/circuits to provide more negative bias that others have posted, but I'm a little reluctant to go through all that trouble if it's not necessary. I'm also unsure as to how much more negative voltage those mods/circuits will add.

[Jazz P Bass]

Not too sure why you are overly concerned about the bias voltage.

There are thousands of these amps running this way.

[stoneattic]

Not too sure why you are overly concerned about the bias voltage.

There are thousands of these amps running this way.

I'm not concerned about the voltage I'm measuring (370V), but I am about the current (41-43mA). It seems like every EL84 guitar amp goes over the 300V max for EL84s, but all the things I've seen say I want something like 23mA @ the 370v this is running at (70% dissipation).

My understanding is I need a higher -voltage bias supply to lower the current. Am I mistaken?

[alexradium]

I'm not concerned about the voltage I'm measuring (370V), but I am about the current (41-43mA). It seems like every EL84 guitar amp goes over the 300V max for EL84s, but all the things I've seen say I want something like 23mA @ the 370v this is running at (70% dissipation).

My understanding is I need a higher -voltage bias supply to lower the current. Am I mistaken?

in theory yes,in reality what affects more the anode current is the screen grid voltage,probably that JJ set you have is not correct or was for some particular amp,but in general all the fixed bias el84 class AB amps are the same,350/370V on B+ and -15V on grids.
I'd get another brand,maybe EH,from a good seller.
If you are measuring cathode current,you need to subtract the screen grid number,for sure its more than 5 mA,just measure the voltage across the 100 ohm resistor on each tube,that way you know if the tubes behave normally.

[Dave H]

So do I go with the 41-43mA JJs as is or try to figure a way to bias them colder? From what I see on the schematic I don't think I can get much more negative voltage on the bias supply

You could try replacing Jumper J03 with a diode (or 2 in series). That will add 0.7V (1.4V) to the bias voltage.

[stoneattic]

You could try replacing Jumper J03 with a diode (or 2 in series). That will add 0.7V (1.4V) to the bias voltage.

I can't wrap my head around this. I know you "lose" .7V when passing through a diode, but it doesn't seem intuitive, at least to me. So if you start with, in this case, +15.7V and pass through a diode you end up with +15.0V, but somehow starting with -15.7V and was passing through a diode to end up with -16.4V doesn't' seem right. It also seems too easy. What am I missing?

[52 Bill]

I can't wrap my head around this. I know you "lose" .7V when passing through a diode, but it doesn't seem intuitive, at least to me. So if you start with, in this case, +15.7V and pass through a diode you end up with +15.0V, but somehow starting with -15.7V and was passing through a diode to end up with -16.4V doesn't' seem right. It also seems too easy. What am I missing?

Perhaps if you look at it this way, -16.4 less 0.7 equals -15.7.

[Dave H]

I can't wrap my head around this. I know you "lose" .7V when passing through a diode, but it doesn't seem intuitive, at least to me. So if you start with, in this case, +15.7V and pass through a diode you end up with +15.0V, but somehow starting with -15.7V and was passing through a diode to end up with -16.4V doesn't' seem right. It also seems too easy. What am I missing?

On the schematic the cathode is at 0V and the grid is at -15V (say) which is a grid to cathode voltage of -15V. Now insert a diode between cathode and ground, the grid is still at -15V but the cathode is at +0.7V so the grid to cathode voltage is now -15.7V. Try it with a 1N4000 series diode, it'll work fine.

[stoneattic]

I noticed that the boxes that the tubes were in were marked "37" so I emailed the folks I purchased them from and asked what, if anything, that marking meant. They responded that they test them at 400V and that's the current across the plate. They said I can send them back to exchange for a lower bias set. So I am going that route and included what I measured, was hoping for, etc., in a note with the return. I'm very happy with their service so far, but I'm reluctant to say who it is until I receive the new set.

The diode trick has me very interest as I have a set of EIs that I bought for a Mesa Studio Caliber DC-2 many years ago and ended up not using because of high current draw. I tested them in this amp and saw the high current, actually a little higher than the JJs. Now I'm thinking the diode trick might help me be able to use them. I hope to experiment with that this weekend.

Thanks everyone!

[stoneattic]

I sent the tubes back as I mentioned in the last post and the folks from Tube Depot sent me a set with their 24 rating. It was a big improvement! I was measuring 28-29mA @ ~370V vs the 41-43mA @ 370V I was getting with their 37 rated tubes. Still hotter than what I wanted so I did the diode trick with a pair of 1N4001s and got the bias down to ~19-20mA @ ~380V! It might be a bit on the cold side, but assuming it's a linear relationship, going down to one diode will get me over 70% dissipation so I think I'll leave it there for now.

Thanks for the help!