Screen grid resistors and tubes keep blowing

[Metatron_X]

This amp has been blowing screen grid resistors on the same side, three times since I've owned it - brand new from the factory, for the past 6 years. It's a Brunetti 059, 120W, EL34, blackface.

When I first purchased the amp it blew a fuse and tube withing the first two weeks of owning it. I took it in, because it was under warranty and I wanted to make sure it wasn't a more complicated issue. I'm just now realizing that there might be a bigger issue there since it keeps happening on the same side every time and the tubes are blowing prematurely. This thing blows through tubes faster than any amp I've owned. Is there a common cause for regular screen grid resistor blowout? It's a 1K, 4W, 1% screen grid resistor. I've changed the resistor myself twice and I've done plenty of work on prior amps (mods etc) I'm just not good at tracing specific issues. I understand the safety precautions.

Honestly this amp has been a cancer as long as I've owned it and help from Italy has been nearly nonexistent unless I want to ship it there and back on my dime. When I got it back from the first tech that I took it to years ago the foot switches started breaking, one after another, three in all. I sent it back to Italy to have it fixed. Brunetti claimed that someone took it apart and reassembled it incorrectly... If not the amp tech then who? That fiasco cost me $200 and the amp has never really been "right" from day one. About six months ago it developed a new issue where the lead channel sounds like it has an octave note at all times and the breakup is squishy and farty.

If anyone has any ideas where to start looking I'd appreciate knowing.

 

[Malcolm Irving]

Suspect the output transformer might have a problem on one 'side' of the primary winding only. Could you measure the DC resistance of the OT primary from the centre-tap to either plate (everything powered off and filter caps discharged)?

 

[Metatron_X]

Suspect the output transformer might have a problem on one 'side' of the primary winding only. Could you measure the DC resistance of the OT primary from the centre-tap to either plate (everything powered off and filter caps discharged)?

Thanks for responding. Yes, I'll take that measurement as soon as I get a chance and report back. I forgot to mention what prompted this: Yesterday the two EL34 output tubes on the right side, corresponding to the usual screen grid resistor blowout, were red-plating... both tubes simultaneously. Also, this amp eats through fuses too, as one might expect from the other symptoms.

 

[forum_crawler]

Your power outlet may be letting the gremlins in as well. I would also look into a power conditioner of some sort.

 

[VaughnC]

Sounds like a bias issue.

Did you confirm you have the same bias voltage reading at the control grids of all tubes?

Did you check the idle bias current on each tube? Without idle cathode current readings, you'd just be guessing.

 

[Metatron_X]

Your power outlet may be letting the gremlins in as well. I would also look into a power conditioner of some sort.

I did consider that but wouldn't my power outlets also effect other tube amps and other equipment if this were the case? Also, I've had these same problems with the same amp in several (at least 3) different homes I've lived in over the past 5 yrs (my wife was in grad school so we've moved around alot).

 

[drewl]

You may also have a bad coupling cap from the phase inverter to the output tubes.

That could mess with the bias, and is easy and cheap to replace.

 

[Blue Strat]

Sounds like a bias issue.

Did you confirm you have the same bias voltage reading at the control grids of all tubes?

Did you check the idle bias current on each tube? Without idle cathode current readings, you'd just be guessing.

+1000. Red plating indicates a bias issue. If you have a bias problem for the one pair of tubes, this will continue happening forever...until it's fixed.

Could be a leaky cap from the PI or other issues. Start by checking the pin 5 voltage on all power tubes. It should be approximately -35 volts for all of them.

 

[Metatron_X]

My readings for the OT are as follows. In case you aren't familiar with this amp (Brunetti 059 blackface) it has two power sections 120W EL34 and 20W EL84.

Red --> White (Pc) EL34: 22.2
Red --> Black (Pd) EL34: 22.9

Red --> Yellow (Pa) EL84: 85.1
Red --> Beige (Pb) EL84: 76.4

I'll check the cap values of the coupling caps and see if one or more is off and pin 5 voltage for the EL34's too

 

[Metatron_X]

Correction:
Grid readings for output valves 1-4 facing the amp left --> right
LV1 Grid: -40
LV2 Grid: -40
RV3 Grid: -37 > Red-plate
RV4 Grid: -37 > Red-plate

I have not found any bad caps yet but I have to take the board out to get to some of the test points. This is a double sided PCB and I can't tell where all of the traces are going on the other side, not to mention I'm working without a schematic which REALLY sucks since I'm not super proficient at this, lol.

 

[Metatron_X]

Just played it for 30 minutes before it started to hum and showed any signs of red-plating...

As soon as the plates started burning I put it in standby, flipped it over and took new readings, the plate current is exactly the same as above.
I should also add that the lead channel anomaly where it has the octave effect and mushy distortion doesn't happen until after about 30min of playing too... seem like they could be related issues but this is the first time I've seen any red-plating in the tubes.

-------------------

Turned it off, let it cool for about 15min then turned it on and played for another 45... nothing happened this time, no red-plate. I'm going to clean the tube sockets with Deoxit and then play it for a while to see what happens.

Intermittent problems, bluh.

 

[VaughnC]

Idle plate current for output valves 1-4 facing the amp left --> right
LV1: -40
LV2: -40
RV3: -37 > Red-plate
RV4: -37 > Red-plate

I have not found any bad caps yet but I have to take the board out to get to some of the test points. This is a double sided PCB and I can't tell where all of the traces are going on the other side, not to mention I'm working without a schematic which REALLY sucks since I'm not super proficient at this, lol.

How are you measuring the plate current? Those look like control grid voltage readings...and, if so, the bias voltage isn't high enough. Looks like a problem in the bias supply, or the bias pot needs adjusted, and/or the bias range resistor needs changed.

You also need to measure the cathode current of each tube to tell what's going on...the grid voltage(s) won't tell you that.

I should have asked previously...but I'm assuming the amp is fixed bias and not cathode biased?

 

[Metatron_X]

How are you measuring the plate current? Those look like control grid voltage readings...and, if so, the bias voltage isn't high enough. Looks like a problem in the bias supply, or the bias pot needs adjusted, and/or the bias range resistor needs changed.

You also need to measure the cathode current of each tube to tell what's going on...the grid voltage(s) won't tell you that.

I should have asked previously...but I'm assuming the amp is fixed bias and not cathode biased?

I'm sorry, those ARE the grid current (oops, voltage) readings from pin 5 to ground on all of the power tubes. I'll have plate voltage and bias readings for all of those tubes tomorrow.
There is a single bias adjustment pot for all four EL34's. The two EL84's do not have a bias pot.

I'm not really proficient with regard to diagnosing and terminology so I'm kind of fumbling through this as much as I do everything. I've done quite a few mods on amps I've owned, I can trace circuits, re-caped and overhauled an amp by replacing stated values, switched plate and cathode resistor values and coupling caps for different gain structures, modify tone stacks, bias tubes, I'm a good solderer, etc but I'm not an expert, otherwise I wouldn't be asking for help. It has been a few years since I've had to be under the hood so I'm having to refresh as I go along too, so please bear with me.

I feel like I've been so utterly screwed on this amplifier since day one and the last tech I took it to fleeced me royally so I want to, at least, get the problem narrowed down to a point where I have a general idea what the problem is so that I don't get screwed again if I take it to another tech - This amp is a money pit. I've heard so many horror stories about every tech in Portland (OR) except two and one of them isn't doing amp work any more so that leaves one tech in all of Portland with no reputation that I know of - good or bad. I don't know what it is about this amp but apparently buying it meant that A) I have really deep pockets B) I love to throw money away C) I'm stupid and deserve to be punished... none of which is true but that's how I have been treated since I dared to save up for and purchase what I thought was a good investment in a quality product, boy was I wrong! I have never had so many problems with something I paid so much for and with no recourse whatsoever. This amp has been a bain, my Crate Blue Voodoo is more reliable than this amp. Next time I will be buying a US made amp that has an explicit warranty and that I can get serviced in the US... Soldono? Time to start saving again.

Sorry for the rant... The whole reason I started working on my own amps is because of what what I've gone through after buying this amp.

 

[Blue Strat]

The reading from pin 5 is voltage, not current.

The fact that the voltage on one pair is different from the other pair is concerning. Could be an electrically leaky coupling cap from the PI. In a situation like this where the problem only occurs after heat up time, you have to take readings WHEN the problem is happening, not before or after.

I would just replace that cap for starters. It's a $2 part and I'm sure your time is worth more than $2 for the time invested so far.

 

[VaughnC]

I agree with Mike, the bias voltage should be the same on all 4 EL34's. Being a PCB amp, I'm not certain how the bias voltage is getting to the EL34's in that particular amp...so maybe there might be a bad ribbon cable or something that's accounting for the 3v difference between the left & right pairs. Still, it seems like all 4 tubes might be biased too hot...but measuring the cathode currents would be the only way to tell.

 

[Metatron_X]

I agree with Mike, the bias voltage should be the same on all 4 EL34's. Being a PCB amp, I'm not certain how the bias voltage is getting to the EL34's in that particular amp...so maybe there might be a bad ribbon cable or something that's accounting for the 3v difference between the left & right pairs. Still, it seems like all 4 tubes might be biased too hot...but measuring the cathode currents would be the only way to tell.

I don't have new tubes right now so I'm taking readings with the old tubes. I'm using an older model of weber biasrite with a single socket. I spent last night moving the tubes around, the readings follow the tubes but the weird thing is that the RV3 tube (below) kept climbing to 60 in the LV1 position but stops at 20 in every other position... none of the other tubes do this in any position - probably an anomaly. For the final readings I made sure the same tubes used for the grid readings were back in the same positions.

TungSol EL34B - with roughly 60-80 hrs of use

Plate voltage:
LV1 PlateV: 458
LV2 PlateV: 459
RV3 PlateV: 460 > Red-plate
RV4 PlateV: 453 > Red-plate

Cathode Current:
LV1 C: 29
LV2 C: 32
RV3 C: 17 > Red-plate (This one tube climbs to 20 from 14 or 17 over a couple of minutes and usually stops but I have seen it slowly climb as high as 60)
RV4 C: 18.7 > Red-plate

From these readings and the fact that the readings follow the tube, it seems apparent that these tubes are bad, I've heard there was a run of bad Tung-Sol EL34B's so I'm not surprised. When I got them I got two with red bases and two with brown bases as a matched set. The brown base tubes are both the burning tubes - the red-plating doesn't always happen with consistency. This doesn't explain, however, why this amp usually burns through tubes quickly and usually in the V3 & V4 position, which is where the screen grid resistors are blowing out too.

 

[EADGBE]

You might want to try some tubes other than Sovteks. Maybe try some Groove Tubes tested JJ or some Shuguang EL34s. Also maybe take the amp to a good tech to find out why it eats tubes.

 

[Metatron_X]

You might want to try some tubes other than Sovteks. Maybe try some Groove Tubes tested JJ or some Shuguang EL34s. Also maybe take the amp to a good tech to find out why it eats tubes.

Not sure where Sovtek came into the thread, I've never used sovtek and have not mentioned them once, thanks for the advice though. I usually buy tubes from a reputable resaler who tests, burns them in and matches them - same as gruv-toobs without re-branding. I prefer JJ, Svetlana and Tung-Sol brands.

I'm going to change out the PI coupling caps and see if that helps, if not then I'll take it in and have someone look at it if I can find someone reliable and who knows what they are doing.

 

[Metatron_X]

I agree with Mike, the bias voltage should be the same on all 4 EL34's. Being a PCB amp, I'm not certain how the bias voltage is getting to the EL34's in that particular amp...so maybe there might be a bad ribbon cable or something that's accounting for the 3v difference between the left & right pairs. Still, it seems like all 4 tubes might be biased too hot...but measuring the cathode currents would be the only way to tell.

Thanks again for your input,
The only ribbon cables are: one to connect the channel tone stacks board to the preamp board and one to the effects loop/speaker impedance/rear controls to the preamp board. All connections from the preamp board to the power amp board are soldered wires.