Vox AC30-6/TB blosing T500mA HT fuses

[nevetslab]

I have a battle-scarred Vox AC30-6/TB blowing it's T500mA HT fuse. Initially, when it and it's partner in crime came in, both with labels stating NO Output, the one with the Greenback 12's worked fine, apart from it's Vibrato/Tremolo switches worn out, rotating past the stops and it's Tolex covering in shameful condition. This one didn't come up out of S/B. Both have their 5VAC heater wires directly coupled to the pwr xfmr secondary, and I'm switching the cathode circuit for Standby Mute, as I've done with all of our AC30-6/TB's. And, of course, moved the power supply wire to Pin 8 from Pin 2 where the factory had screwed up on the rectifier tube.

The Littelfuse 218.500HXP T500mA fuse I removed had the explosion appearance inside the glass tube. I replaced the unmarked 5AR4 rectifier tube after first replacing the fuse. It did power up, and I had output, but I also had a fair amount of hum that I didn't have on the other AC30. I replaced the rectifier tube with new J/J 5AR4. Stopped there to address the Tolex covering on the cabinet, as a fresh roll of black gaffer's tape arrived. Finished that task, cursing the roadies who seem to have it in for any Vox amp, and put the chassis back into the cabinet. I had checked my service notes from June 2018, to see I had then installed a fresh matched quad set of EL84's.

After powering it up, within 2 minutes, I heard the brief groan of the power transformer object, looked to see the S/B lamp had gone out, and the mains current had dropped down. HT fuse blew again. Sigh.........

I looked up a number of AC30-6/TB schematics, looking to see just where the HT fuse was located. Got several opinions from different revisions, stopped to ring out where it actually is. Sits between Pin 8 of the rectifier tube and the junction of the O/T Centertap, the input to the choke and the input to the 47uF/450V 1st stage supply filter (recent Nichicon VX series). Choke is a Drake Transformer 292-767 Choke, 4895 date code, DCR measures 437 ohms, which seems high to me, but don't know.

For the fuse to blow in that time, the fuse data sheet suggests around 200% current was flowing for that to open. It didn't leave black soot inside the replacement fuse, but still, I got somebody pulling too much current.

I haven't yet pulled the power tubes out or swapped them (I do have another matched set on hand). I thought about leaving the HT fuse out, and substituting one of my lab HV supplies to feed the plate circuit to see what's up. I can't say I'm a big fan of extracting the main amp board on these, having to pull all the preamp tubes, remove the control panel PCB, and unsnap all of the misaligned plastic standoffs that break.

ac3093pa.pdf

[Dave H]

I'd check the EL84 bias current. The hum could be caused by one side of the output stage pulling too much current. The DC current from the power supply is 250mA or more in normal operation but the RMS current in the fuse could be something like 400mA as the fuse is between the PT and first filter cap.

[nevetslab]

I'd check the EL84 bias current. The hum could be caused by one side of the output stage pulling too much current. The DC current from the power supply is 250mA or more in normal operation but the RMS current in the fuse could be something like 400mA as the fuse is between the PT and first filter cap.

I've gone thru the existing matched pair of EL84 tubes, first with an HP 712C supplying the HT supply voltage/current. The nominal idle current called out for an AC30-6/TB is 200mA, that being 10VDC across the 50 ohm cathode resistor. To achieve that, I dialed up the supply voltage to 340VDC @ 215mA from the 712C supply. I pulled the HP supply out of the set-up, and restored the 5AR4 to supplying the HT, with the T500mA fuse back into place. Using a current probe to look at both the current flowing into the O/T and that flowing thru the choke, there's about 12mA flowing thru the six preamp tubes, 20mA thru the screens and 175mA thru the power tube plates.

With the current probe on the output of the 5AR4 tube feeding the input of the HT Fuse, I'm getting 630mA peak current half-sine waveform. I'm not set up to measure the current into the 47uF 1st stage filter cap. Suffice it to say DC current resulting from the smoothing of that rectified current has very little AC ripple showing. The supply voltage is 337VDC, with 10.1VDC on the cathodes, and pulling 1.26A/138W @ 120VAC mains.

Under signal drive conditions, using burst pink noise, the signal content shows the current increase above the nominal DC plate current, hitting peak currents around 275-300mA.

The hum I'm hearing is 60Hz-based, with buzz, while I'm betting good balance in the two halves of the power tubes, so it's not imbalance current hum.

I swapped out the existing set of tubes with a new matched set, got the same results.

I still am a bit puzzled why finding twice the T500mA HT fuse has blown. I was expecting to find one or more of the power tubes suddenly go off the reservation to parts unknown, but haven't see that yet. So, I've got the near-6 month set back into place. Amp chassis is back on the bench.

The 5AR4's are rated for a max 250mA current load, which an AC 30 can draw, while I'm only seeing that on signal peaks, NOT steady state. With the amp in Standby...this case being the cathode load opened, the DC supply voltage climbs up to 410VDC. The fuse voltage rating is 250VAC, which is 353VDC. I've never known HT fuses in a similar circuit amongst all the amps that use them, they ARE being subject to higher voltages than they're rated for. I'm not yet willing to believe THAT is the case here and on other amps that sometimes come in with blackened HT fuses that blew, and not find the cause.

The power tubes sockets feel tight, not noticeably sloppy or loose-fitting, though the sockets do get pushed down into their cavities when the tubes are inserted. With the hold-down springs in place, they seem stable in this amp.

[pdf64]

Yes, a coupling cap or grid stopper to a power tube grid that goes bad can cause severe bias problems; with a vintage JMI, it’s the sort of thing that can overheat the PT and result in a wax deposit on the supporting chassis board underneath.
Alternatively if one tube stops conducting, eg bad tube or screen grid resistor, the other 3 will be way under biased, and counterintuitively, that can also overheat the PT.

[Helmholtz]

I've gone thru the existing matched pair of EL84 tubes, first with an HP 712C supplying the HT supply voltage/current. The nominal idle current called out for an AC30-6/TB is 200mA, that being 10VDC across the 50 ohm cathode resistor. To achieve that, I dialed up the supply voltage to 340VDC @ 215mA from the 712C supply. I pulled the HP supply out of the set-up, and restored the 5AR4 to supplying the HT, with the T500mA fuse back into place. Using a current probe to look at both the current flowing into the O/T and that flowing thru the choke, there's about 12mA flowing thru the six preamp tubes, 20mA thru the screens and 175mA thru the power tube plates.

With the current probe on the output of the 5AR4 tube feeding the input of the HT Fuse, I'm getting 630mA peak current half-sine waveform. I'm not set up to measure the current into the 47uF 1st stage filter cap. Suffice it to say DC current resulting from the smoothing of that rectified current has very little AC ripple showing. The supply voltage is 337VDC, with 10.1VDC on the cathodes, and pulling 1.26A/138W @ 120VAC mains.

Under signal drive conditions, using burst pink noise, the signal content shows the current increase above the nominal DC plate current, hitting peak currents around 275-300mA.

The hum I'm hearing is 60Hz-based, with buzz, while I'm betting good balance in the two halves of the power tubes, so it's not imbalance current hum.

I swapped out the existing set of tubes with a new matched set, got the same results.

I still am a bit puzzled why finding twice the T500mA HT fuse has blown. I was expecting to find one or more of the power tubes suddenly go off the reservation to parts unknown, but haven't see that yet. So, I've got the near-6 month set back into place. Amp chassis is back on the bench.

The 5AR4's are rated for a max 250mA current load, which an AC 30 can draw, while I'm only seeing that on signal peaks, NOT steady state. With the amp in Standby...this case being the cathode load opened, the DC supply voltage climbs up to 410VDC. The fuse voltage rating is 250VAC, which is 353VDC. I've never known HT fuses in a similar circuit amongst all the amps that use them, they ARE being subject to higher voltages than they're rated for. I'm not yet willing to believe THAT is the case here and on other amps that sometimes come in with blackened HT fuses that blew, and not find the cause.

The power tubes sockets feel tight, not noticeably sloppy or loose-fitting, though the sockets do get pushed down into their cavities when the tubes are inserted. With the hold-down springs in place, they seem stable in this amp.

If I understand correctly, there is a standby switch which opens the cathode circuit. If so, I wouldn't be surprized if there is some occasional arcing in EL84s, when suddenly exposed to 410V plate voltage. Especially as the cathode bypass cap acts as a momentary short when discharged. Also fuses may die by fatigue caused by repeated high current surges caused by switching on plate voltage at zero bias.

[nevetslab]

If I understand correctly, there is a standby switch which opens the cathode circuit. If so, I wouldn't be surprized if there is some occasional arcing in EL84s, when suddenly exposed to 410V plate voltage. Especially as the cathode bypass cap acts as a momentary short when discharged. Also fuses may die by fatigue caused by repeated high current surges caused by switching on plate voltage at zero bias.

While I wouldn't rule that out, I haven't noticed a steady stream of bad output tubes or blown HT fuses on all of our AC30-6/TB's. There are seven of them here, all still running (apart from this one that showed up with a blown HT fuse. Then did it again, but hasn't since. I may have installed a Fast Blow 500mA by mistake....already tossed it into the trash, so can't see that wasn't the case. Just haven't found WHAT caused the fault. yet.

Before changing the Standby switch from switching the Secondaries to the plates of the 5AR4 rectifiers, I went thru those tubes quite often, and likewise heard others with the same issue. I also had to correct the output wiring on them, moving the load wire from Pin 2 to the cathode / heater terminal @ Pin 8. There is more of a delay in ramping up the charge from the 5AR4 from original Standby circuit, where you're then charging up the first stage which also feeds the O/T primary and the plates of the EL84's, so yes, it's more of a shock. Silent going in and out of Standby with opening the cathode.

[Helmholtz]

While I wouldn't rule that out, I haven't noticed a steady stream of bad output tubes or blown HT fuses on all of our AC30-6/TB's. There are seven of them here, all still running (apart from this one that showed up with a blown HT fuse. Then did it again, but hasn't since. I may have installed a Fast Blow 500mA by mistake....already tossed it into the trash, so can't see that wasn't the case. Just haven't found WHAT caused the fault. yet.

Before changing the Standby switch from switching the Secondaries to the plates of the 5AR4 rectifiers, I went thru those tubes quite often, and likewise heard others with the same issue. I also had to correct the output wiring on them, moving the load wire from Pin 2 to the cathode / heater terminal @ Pin 8. There is more of a delay in ramping up the charge from the 5AR4 from original Standby circuit, where you're then charging up the first stage which also feeds the O/T primary and the plates of the EL84's, so yes, it's more of a shock. Silent going in and out of Standby with opening the cathode.

Short occasional arcing doesn't necessarily destroy a power tube but may blow the HT fuse. I just do not think that a cathode stand-by switch is a good idea with cathode biased amps, especially when the HT voltage at standby significantly exceeds the plate voltage limit.
Have you ever measured the EL84s' plate current turn-on surge?

[nevetslab]

Cathode switching voltage, plate current transient from S/B to Operate Mode

Short occasional arcing doesn't necessarily destroy a power tube but may blow the HT fuse. I just do not think that a cathode stand-by switch is a good idea with cathode biased amps, especially when the HT voltage at standby significantly exceeds the plate voltage limit.
Have you ever measured the EL84s' plate current turn-on surge?

In Standby mode, all of the tubes are energized...HT supply increases from 337VDC to 410VDC. The cathodes of the output stage tubes go from 32VDC in Standby Mode to 10VDC, while in that instant of switching, there is an instant of cathodes going to 0VDC, and the plate current goes up
accordingly. I just captured the switching transients both measuring the total plate current thru the O/T primary and that of the cathodes at the
Standby Switch.



After capturing the first image, I went back to change the current/Div from 50mA to 200mA/Div, turned off the cathode voltage trace and increased the phosphor brightness to find the peak current, running at 100mS/Div. Peak current on this was 1080mA total. Only the 220uF bypass cap across the 50 ohm cathode resistors are responsible for this high short duration peak current.

Note: in the first photo, the upper waveform is cathode Voltage, NOT current. My typo on the image. Went from 32VDC, briefly hit near 0V before rebounding to 10VDC

[drewl]

I removed the standby on my AC30, installed a master volume in it's place.
After correcting the rectifier wiring I haven't blown a rectifier for many years.

Also twisted the filament wires which reduced hum.

[TimmyP1955]

Don't forget to give the EL84s the thwack test to make sure there's not an intermittent short. I don't recall if I've had that issue with an EL84, but have had plenty of 6L6GC and some EL34 do it.

Many of the AC30/6TB have some hum owing to the placement of the heater wires. This can be cured by moving the wires around until you find the magic spot, and glue hot glue it down. Start with the controls at min and start on the PI end, then work your way to the front with the controls up. I often have to do it in three spots. (I'd post pics if I could.) Then of course make sure that the amp is still stable, and that the sound has not been compromised.

[nevetslab]

I removed the standby on my AC30, installed a master volume in it's place.
After correcting the rectifier wiring I haven't blown a rectifier for many years.

Also twisted the filament wires which reduced hum.

That sounds like a good idea with the master volume in place of the S/B switch! As long as 5AR4's remain as the rectifier, they slow-start before with the heaters before conducting and producing output voltage to be charged up. I hadn't before today checked to see just what the instantaneous switching current was when the cathodes are re-connected to the 50 ohm resistors/220 uF bypass caps. Thankfully that 1.1A current pulse is quite narrow. I suppose another solution would be to use that S/B switch to short the two AC-coupled driver signals for S/B, leaving the cathodes connected. It does burn current thru the power tubes, which was part of the intention of turning them off (while it does also burn current in the preamp tubes).

All the EL-84 data sheets I have list the max plate current as 65mA, and max anode dissipation at 12W, though we've all seen these tubes run way beyond the dissipation limit. Usually not more than 65mA. That's steady state though. This narrow switching pulse current coming out of Standby is in the uncharted territory region, being over 250mA for less than 10mS. The increase from 337VDC to 410VDC on the plates while NOT conducting current isn't exceeding the max plate voltage, and there's no current passing thru the tube.

The short duration current during closing the S/B switch to Operate is well withing the limits of the T500mA fuse graphs, where at 10mS, a T500mA fuse can withstand greater than 10A.

I did whack the sides of the tubes while it was running to see if there were any issues there.

Some other repair shop had taken this amp in without my knowledge, and cut the Radiohm Volume pots off the board, replacing them with lead-wire 24mm dia pots. That may be where I'm picking up hum/buzz, as nudging them with my chopsticks changed the noise some. More digging to do there.

[Dave H]

All the EL-84 data sheets I have list the max plate current as 65mA, and max anode dissipation at 12W, though we've all seen these tubes run way beyond the dissipation limit. Usually not more than 65mA. That's steady state though. This narrow switching pulse current coming out of Standby is in the uncharted territory region, being over 250mA...

Would this work? Leave the cathode cap connected permanently between the EL84 cathodes and ground and just switch out the cathode resistor for standby. That will keep the cathode cap charged so there's no zero bias situation when switching from 'standby' to 'on'. You'd have to make sure the cap voltage rating was high enough to withstand the increased voltage in standby or wire a 47k? resistor across the cap to keep the voltage down and allow a small current through the tubes. (Anyway, AC30s don't need no stinking standby switch )

[Helmholtz]

The increase from 337VDC to 410VDC on the plates while NOT conducting current isn't exceeding the max plate voltage, and there's no current passing thru the tube.

The tubes see the full 410V during the high current peak, so the 300V max rating is exceeded. The no-current plate voltage limit doesn't apply here.

The short duration current during closing the S/B switch to Operate is well withing the limits of the T500mA fuse graphs, where at 10mS, a T500mA fuse can withstand greater than 10A.

The fuse won't notice much of the peak current as this is mainly supplied by the filter cap.


What about lifting the screens for stand-by?

[R.G.]

Reading all that raised a couple of possibilities in my mind. The cathode cap is a critical element in the biasing, and they take a lot of abuse. It's possible that the cathode cap is having some issues. Could be simple age or abuse making it have an intermittent high leakage.

Does the standby switch leave the cathode cap across the cathode resistors when it's open, or does the switch just interrupt the cathode resistors, leaving the cathode cap from cathodes to ground? In some situations there could be a lot of voltage across it when the standby switch is open. Also, it would be interesting to look at what happens to the cathode cap when standby is switched on. Even if it's left connected across the cathode resistors, so it's voltage is zero when the standby switch closes, the cap has to handle all of the "saturation" current from the output tubes. That's tough on an electro, and maybe even a fuse.

Just speculating.

[drewl]

Since the rectifier is a soft start delay and your customers may want the standby function, wire it up like the old Silvertone amps where standby just shorts the signal together after the PI caps.

The Tone Cut is already wired there.