So here’s a fun one (AD-100 clone woes)

[Helmholtz]

Thanks for further explanation! Would it be worth it to remove these entirely to see if they are in any way causing an issue?

Resistors wired across the speaker output can protect the OT and plates from HV spikes when no speaker is connected. But they can't protect the screens from melting when the speaker connection fails. They generally have no negative influence on tube operation.
You might use two 470R/10W cement resistors wired in parallel.

BTW, did you ever exchange tube sockets?

[Mr_bibbles]

Resistors wired across the speaker output can protect the OT and plates from HV spikes when no speaker is connected. But they can't protect the screens from melting when the speaker connection fails. They generally have no negative influence on tube operation.
You might use two 470R/10W cement resistors wired in parallel.

BTW, did you ever exchange tube sockets?

Cool, thanks! Currently I have 2 pairs of 2 1k's at 5watts, I'll change it to your recommendation.

And for your second question, I did replace a couple tube sockets that were damaged, but because the tubes were failing in DIFFERENT positions, this couldn't be the cause right? The one I left in didn't have an "arc scorch" or carbon residue,, but rather a small black discoloration from heat on the tube-side.

[Helmholtz]

Currently I have 2 pairs of 2 1k's at 5watts, I'll change it to your recommendation.

If that means 4 x 1k/5W in parallel, that would result in a total of 250R/20W and is just fine.

[g1]

Sorry if I missed it, but any chance this unit uses the spring & cap type retainers for the power tubes?

[Mr_bibbles]

Sorry if I missed it, but any chance this unit uses the spring & cap type retainers for the power tubes?

They do not, but I think it was Chuck who brought up earlier that the Genalex KT88s have metal bases if you're thinking about a possible arc on the tube socket side...

There is no spring'n'cap though, and I don't think that these tubes are being damaged in a mechanical way if that's what you're thinking.

To make things more clear, the amp works fine for the customer for about 20 minutes, it begins to crackle, the customer keeps playing on it and then a fuse pops and a tube loses vacuum. So, obviously, thermal tests are in order, but I ran a sin wave through it for 6 hours without seeing any damage or any crackling / qualitative issues at all. This is why after 3 trips on my bench I'm starting to suspect that it's being overloaded with the customers setup: active bass cranked though two different distortion circuits... this isn't a "tested and true" design, maybe it's vulnerable to parasitic oscillation above hearing range after that much harmonic content gets added? To be honest I'm grasping at straws.... it works great here for hours and hours and then pops as soon as the customer plays it.

[g1]

Saw this pic in Orange 200 thread so I thought I'd pass it on... (white-topped power tube)

https://music-electronics-forum.com/...413#post908413

[Enzo]

I can't think of anything electrical that would crack a tube. Even when bias goes away completely I have seen the glass start to melt right over a hole in the plate. But that doesn't generally shatter the tube.

Don't play a sine wave through it for burn in, play music. The amp isn't designed for sine waves, and the customer certainly never plays a sine wave through it for hours. Plug a CD player into it.

Can you visit the guy's gig? he may not be whacking the tubes with his spare guitar cord storage, but who says he doesn't have a sweaty drummer flicking sweat off that lands on a hot tube.

[Mr_bibbles]

So apparently the Building got back in touch after a year or so of building it, discovering that he has incorrectly rated the filament current for 8 KT88's. I have the spec sheet from Heyboer for the PT-500 but no information about AC supply is there besides the normal 6.3v rating.

Considering a full tube complement is like 15 amps, this sounds probable. Does anyone have experience with under-rated filament supplies physically destroying tubes? Would a filament transformer be a decent solution if the PT is found to be under-rated?

Danke

[The Dude]

Measure the filament voltage.

[Mr_bibbles]

6vAC, measured across the indicator bulb.

[The Dude]

That's only slightly low. I doubt it would damage tubes. Does the transformer get hot? Can you upload the transformer datasheet and tube compliment for your amp?

[Helmholtz]

5% down is more likely to increase filament life.

[Chuck H]

6vAC, measured across the indicator bulb.

The indicator bulb is often the first stop for the supply. There may be greater losses at the tubes themselves with the last in the circuit being the lowest voltage. If the filament leads to the tubes are of an only adequate gauge you might be able to improve the situation by rewiring the filaments with stupid big wire and a little less twisting. But I'd be more concerned with winding overload. Dudes suggestion to monitor transformer heat is probably fine.

[Mr_bibbles]

First, I'd say the transformer does get a little hotter than normal, but not 'burn your fingertips off' hot. It's pretty close to 2 fat KT88's giving off heat too so I never considered this an issue before...


So I let the amp heat up for a good 15 minutes while I was on the phone, re-measured the Filiment at the Bulb and got 6.3VAC, .3 Volts higher than last time.

At the furthest of 6 (out of 8 possible) KT88's currently installed, we have 6.13VAC.

Help me understand though, if our hypothesis is that our Transformer cannot supply enough current, are we measuring the voltage so that we can see it's relation to the amperage? And we're measuring it at different points to see what the relation is to draw? That's my guess but a lot of this is very new to me.

I've supplied a Spec sheet but there is no data about Amperage.

From your replies I see that I'm missing something big about heater draw.. something that is probably right in front of my face...

Thanks as always!

[Helmholtz]

So apparently the Building got back in touch after a year or so of building it, discovering that he has incorrectly rated the filament current for 8 KT88's. I have the spec sheet from Heyboer for the PT-500 but no information about AC supply is there besides the normal 6.3v rating.

Considering a full tube complement is like 15 amps, this sounds probable. Does anyone have experience with under-rated filament supplies physically destroying tubes? Would a filament transformer be a decent solution if the PT is found to be under-rated?

Danke

I don't understand the first part: Who discovered that the filament supply is underrated? Your voltage results don't seem to confirm this.

Anyway your measured heater voltages are well within tolerance and are certainly not the reason for the cracked tubes.
BTW, heater voltage varies proportionally to mains voltage. And it's normal for mains voltage to vary by 5% or so at different times of the day