Hoffman 5F6A Single Channel Rectifier Tube Arcing

[jonyoungyi]

Hey all,

Just wired up a Hoffman 5F6a and removed the first input (220uf/820R, etc), keeping second input, bright channel.

Unfortunately, when I fired her up, I immediately noticed a rapidly growing, loud buzz/hum, so loud that I had to switch back to standby. While the sound increases, I also noticed that my rectifier was arcing terribly! (Purple flashes).

The rectifier tube is good, as I've pulled it from my working AB763 Deluxe Reverb. Even after I saw the arcing, I plugged it back into the AB763 and it worked fine.

I've wired the filter caps the "normal way" with the 1st stage being 2x 22uf 500v in parallel. To save space I've put the caps on spacers sitting above the PT. The other filter caps are nested under the pots, attached to my ground bus.

Other culprit could be a 9v AC to DC transformer that I wired up to the power switch. (going to put digital reverb inside).

PT = Mojo762
OT = Classictone 40-18092 (50/60 watt 4.2k 4 ohm/8 ohm/ 16 ohm)
Choke = Classictone 40-18003 (4 Hy, 90mA DC)

I've built 3 amps now and have never had this happen to me, so I'm not really sure where to start.

If someone could point me in the right direction that would be great, thank you!

[sluckey]

Filter caps installed backwards?

Pull output tubes. Recto still arcing?

With amp turned off measure resistance from rectifier socket pin 8 to ground. Flip the standby switch both ways. What have you?

[jonyoungyi]

Thanks for the reply sluckey! (and for helping me put a MV in my AB763. I saw your schematic/post haha)

Filter cap orientation looked good to me. It would have exploded if it was backwards?

I'll have to measure rectifier pin 8 to ground after I get home from my day job.

Will post my findings, thank you!

[eleventeen]

"filter caps the "normal way" with the 1st stage being 2x 22uf 500v in parallel."

This has never happened to me, but tube rectifiers have a "maximum capacitance at first node" rating. What do have there, a 5Y3 or GZ34 or 5U4? You've wired 2 qty 22 ufd in parallel. So, if both of those caps are "rich", they could be nearly 30 ufd each and the combo could be nearly 60 ufd; you are expecting about 40 (which is about the limit, if I recall correctly) 

One o' them Russian GZ34's? There have been spurious issues with those, depending upon brand and date and mfr. I cannot state which ones are the hinky ones.

You are right, the e-caps would blow if hooked up backwards. My suggestion is simply to snip the + lead of ONE of those 22 ufds and do a super-fast stdby-switch flip test (just like you did) to see if the rectifier tube still arcs.

[jonyoungyi]

Hey eleventeen,

The schematic calls for a 5AR4/GZ34. I put a JJ GZ34s in there and noticed the arching.

I just checked the gz34s spec sheet and looks like 60uf is my limit. I ordered a Sovtek 5AR4 recently, so maybe that'll do? Unless, it's a bad tube right out the box. I believe it's brand new, so hopefully the mfr date isn't too far off.

Hmm... I suppose I can also measure the capacitance I'm getting from the parallel 22uf. So as long as it's below 60uf, I should be fine, right? If it's even close to 60uf, I suppose that's the culprit!

Overall, it seems like my filter caps/power is the issue here, huh?

[eleventeen]

In a case like this, I think you just go "snip" on the extra cap and check it out.


There was a thread in these parts some time ago about hinky GZ34's.


When I say this has not happened to me, I tend not to use GZ34s...only because I have tons of 5Y3's and 5U4's lying around. If I have a tranny where I need to get every volt I can out of it, then I use a 5V4. Oh yeah, also because I am cheap. A 5V4 has maybe only 5 volts more drop than a GZ34, 5 volts = nothing in tubeland. You can buy NOS US made 5V4 RCA/GE/Westinghouse/Sylvania for $10 on ebay. If I was in the habit of building amps that wanted GZ34, I would buy 3-4 of them. I've gotten mine in misc used gear or at flea markets.


Even if I am working on an amp that uses GZ34, I will put a 5Y3 in there during service just in case something blows up the rectifier. I'd rather blow up a cheap 5Y3 than a pricey GZ34, and, as a bonus, I am giving the thing somewhat lower B+ while servicing.


What's the difference? The indirectly heated cathode. It produces slower warmup (people like that on their expensive power tubes) and much less voltage drop than 5Y3/5U4, (20 volts vs 50 or even 60 volts) but it also places the cathode and heater much closer to the plate(s). Closer = more tendency to arc. More caps = more inrush current when the thing first ramps up = more tendency to breakover = arc. Just kind of common sense. I can't guarantee it's your problem.

[tubeswell]

This is why you should install SS protection diodes between the PT's high tension winding ends and the rectifier tube plate pins.

[jonyoungyi]

Eleventeen,

Oh, that's good to know! 5v doesn't seem all that much. My first amp was a Champ, so 5Y3. I forget what I paid for it, but I'm sure it was a lot less than what I apid for the 5AR4 I just ordered =X

Tubeswell,

For sure, that's the first thing I'm going to do when I get a chance at the workbench. Luckily, it looks like my rectifier tube is "okay"... unless I just arced the life out of the thing and it's hanging on for dear life haha

I was going to use this layout for the diodes. Comes up first in the google search and the guy clearly labeled it as a 5AR4/GZ34 rectifier, but the college ruled paper is throwing me off, so hopefully this checks out hahaha Looks good to me though!

[tubeswell]

Yes, that's correct. It will work for any 5Y3, 5U4, 5R4, 5V4, 5AR4 type etc where the plate pins are on 4 and 6. Use 1000V 1A diodes minimum. 1.5kV 1A or 2kV 1A is even better

[jonyoungyi]

tubeswell,

Awesome, thanks for the confirmation! I've a bunch of 1N4007 diodes, so looks like I'm at the minimum!

[2deaf]

Surmise all you want, but by all means do what Sluckey says in Reply #1 before you do anything else. 

[tubeswell]

Yes, the protection diodes won't stop whatever is causing the arcing rectifier tube, but they could prevent a shorted tube taking out the reservoir filter cap and the PT.


If the rectifier tube has arced, it will probably go on arcing, because the arcing leaves carbon deposits that make a current pathway. You should probably ditch that rectifier tube, even tho it looks like it may be working in another amp.

[jonyoungyi]

Alright,

So I pulled the output tubes and... NO ARCING! I put the output tubes back in and.... ARCING!

So, that lead me to inspect my 8 pin wiring. Low and behold... the pin 4 wire going to the 3w 470r resistor broke loose from the socket lug! It's solid core, so it looks like it's connected.

After I re-soldered that wire back in, it wasn't arcing WITH the output tubes in!

Thanks, Sluckey!

But now, I'm not getting any sound from my speaker, no hum, no nothing. Fortunately, I've been here before! I also wired an impedance selector switch, so I'll have to double check my wiring, there, also.

2deaf, thanks for the extra confirmation!
tubeswell, that's good to know! I had no idea that the carbon deposits would make a pathway! I should get some of those 5Y3 to test with... =X