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

[Chuck H]

Why 200W?
With 8 (!) KT88s and 500V plate voltage, 400V screen voltage this amp should deliver in excess of 400W (provided that PT and OT support that power).

Right. Thank you. The guy still won't notice

[Chuck H]

The largest matched assembly of 10watters I have on hand is 2x350ohm, for a combined 175 ohm 20watt. I also have a 2.7k/270r pair of 10 Watts I could use, for a total of 245r, 20w. Would his be safe considering the imbalanced resistance of the pair? That would change the handling of each individual resistor, correct?

Well since Helmy was good enough to illuminate the wattage issue (not sure why I got 200W stuck in my head) you're going to need a higher rating on the resistance. 220 ohms will draw about 16W and 470 about 8W. So if you land in between a 20W resistor is probably ok. I'd go 25W or a pair of 20W in parallel.

So what do you have for 20W? 25W? You can stack rating by putting resistors in series too. That is, two 150ohm 10W resistors in series is 300ohms at 20W. I prefer not to use series arrangements for this circuit if only because it's just another possible weak link in a safety chain. That said, I would do it if it's what I happened to have on hand. So using lower values in series could be an option for your inventory too.

[Helmholtz]

The largest matched assembly of 10watters I have on hand is 2x350ohm, for a combined 175 ohm 20watt. I also have a 2.7k/270r pair of 10 Watts I could use, for a total of 245r, 20w. Would his be safe considering the imbalanced resistance of the pair? That would change the handling of each individual resistor, correct?

A 400W amp produces around 57 Vrms across an 8 Ohm load before clipping (clipping increases RMS output voltage). Square the voltage and divide by load resistance to find resistor dissipation.

[Mr_bibbles]

Unfortunately the hugest I have is 10w. 270, 330, and 350 at the only low ohmage ones. This leaves me with very little options, so perhaps I should just mouser the correct parts.

Also, you can stack wattage handling with a series assembly?!? Would the amperage cross each resistor with its full power in sequence? I trust your expertise but am just surprised.

[Mr_bibbles]

Maximum load resistance for this amp is 8, so (57x57)/8 = 406, increase it a sprinkle to account more a maximum clipped load and somewhere around 425-470 is appropriate?

[Helmholtz]

Maximum load resistance for this amp is 8, so (57x57)/8 = 406, increase it a sprinkle to account more a maximum clipped load and somewhere around 425-470 is appropriate?

Clipped output power and voltage are hard to predict. It strongly depends on power supply sag. Marshall states 170W clipped power for their 100W amps.

My advice above was meant for calculating the dissipation of the base load resistors.

[Chuck H]

Also, you can stack wattage handling with a series assembly?!? Would the amperage cross each resistor with its full power in sequence? I trust your expertise but am just surprised.

I'm unclear by the way you worded this I think. I'll say it like this. In a series arrangement the resistors share the current relative to the voltage across their resistance. So it's like I said above about the 150ohm 10W resistors. The two in series would be good for 300 ohms at 20W. If the resistances are different then the load would be shared unequally with the lower value resistor carrying more of the current. In direct proportion actually. So if a 90ohm resistor is in series with a 10ohm resistor the 90ohm resistor will carry 10% of the current and the 10ohm resistor the other 90%.

[Mr_bibbles]

A bit of a necro-post, but the same amp came back from the customer and we've finally re-opened here in Chicago after being corona-closed for months.

Blew another Kt88, white topped, in a DIFFERENT position than last time. Main fuse blown, small soot mark on TUBE-SIDE of the socket... doesn't look like an arc, more like something just got really hot there.

I replaced the tube with a matched replacement 6 total) and the fuse, booted it up no problem. Sent a 1k sin wave through it at reasonable volume for 6 hours, stable, no issue..... I had asked the customer to test it first with no pedals, but he didn't listen and played an active bass through 2 drive pedals, possibly doing the damage by overloading it??? His band seems like doomy / stoner metal stuff where volume is key, so this is in the realm of possibility.

If this is the case, is there any protections / modifications I can do to let the customer run wild with input gain? It has input resistors for hi/lo jacks already, but beyond that I wouldn't know where to start. Coupling caps and PI feeds sound like a place to start, but perhaps there's an old standard fix?? Thanks guys.

[The Dude]

I doubt a hot input would damage power tubes. It would just cause early clipping if anything. I don't have time to re-read the thread, but was bias checked? Is B+ reasonable. If bias and B+ are reasonable, I would simply expect a bad tube from the get go. Modern tube QC isn't what it used to be. I recently went through 3 batches of output tubes on an amp repair before I got a good set.

[g1]

So when you say the tube 'white topped' you mean loss of vacuum like when a tube cracks?
If so, I guess it's possible that the loss of vacuum caused the overheating at the socket as well.
If there was another similar occurrence (white topping) earlier in the thread, can you point us there? It's been awhile.

edit: never mind, I see the very first post was about a tube 'white-topping'. Weird.

[Chuck H]

Is it possible the owner is physically abusing the amp as well as sonically? Like tossing the amp hot right after a gig onto the steel floor of a cold truck or van and then driving down bumpy roads, etc.

[Helmholtz]

I'm unclear by the way you worded this I think. I'll say it like this. In a series arrangement the resistors share the current relative to the voltage across their resistance. So it's like I said above about the 150ohm 10W resistors. The two in series would be good for 300 ohms at 20W. If the resistances are different then the load would be shared unequally with the lower value resistor carrying more of the current. In direct proportion actually. So if a 90ohm resistor is in series with a 10ohm resistor the 90ohm resistor will carry 10% of the current and the 10ohm resistor the other 90%.

Just noticed there seems to be some misconception here. In a series wiring there can be only one common current (I) and power is distributed as P = I² * R, so higher resistance takes more power.
In a parallel arrangement all resistors see the same voltage (V) and resistor power/dissipation is given as P = V²/R, so lower resistance gets/takes more power.

[Mr_bibbles]

I doubt a hot input would damage power tubes. It would just cause early clipping if anything. I don't have time to re-read the thread, but was bias checked? Is B+ reasonable. If bias and B+ are reasonable, I would simply expect a bad tube from the get go. Modern tube QC isn't what it used to be. I recently went through 3 batches of output tubes on an amp repair before I got a good set.

Thanks for chiming in Dude! Always appreciate your advice. I have checked the bias for each tube individually any time I changed something in the amp. Everything is well within spec, biased around 60%. B+ all comes up normal (500vish), but as it is a boutique amp I have no schematic for reference. There is a sketch of the supply on page 3 for further info! Also, 2 tubes from this Genalex set have white-topped, in separate instances and before it made it here the tubes blew and were replaced by someone else. This leads me to believe it's an amplifier issue.

[Mr_bibbles]

Is it possible the owner is physically abusing the amp as well as sonically? Like tossing the amp hot right after a gig onto the steel floor of a cold truck or van and then driving down bumpy roads, etc.

Oh it's possible, but I let him know about careful transport just in case... but we also know that the amp works for 20-30 minutes, begins to crackle, and then pops a tube. Because I've run it with a loud 1k sin wave for hours, it doesn't seem like something that fails under "normal" use.

After crawling over it time after time, suspecting my own work, suspecting the amp-makers work, suspecting faulty parts... what is the possibility that an overloaded input could pop a tube? Or what else could be "Serial white-topping" these poor innocent KT88's?

[Mr_bibbles]

Just noticed there seems to be some misconception here. In a series wiring there can be only one common current (I) and power is distributed as P = I² * R, so higher resistance takes more power.
In a parallel arrangement all resistors see the same voltage (V) and resistor power/dissipation is given as P = V²/R, so lower resistance gets/takes more power.

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