Champ OT 5K or 8K?

[texwest]

hey guys, one of my Champ style builds that I sold to a guy blew the OT.  He was running it pretty hard a lot.  The OT was the cheap magnetic components one from CE dist.
Really nothing seemed out of whack except the OT - the tube, the caps, the voltages seem ok.  It didn't even blow the fuse but ran at very very low volume.  Kind of a mystery why it went bad.

I've replaced it with the Hammond 1750C and replaced the filter caps and cathode cap just in case.  It's now working and sounds great.
I have it wired with the 8K primary winding.   I'm wondering which winding put less stress on the OT?  The 5k or 8k?  Since he runs it hard A LOT maybe I should use the one that will make it last longer this time.

Thanks

[shooter]

you need more info than what we have.
PA tube used

plate VDC of PA
Cathode VDC of PA


but if you want to skip all the fun math "A typical SE CHAMP with 1 6V6 around 300vdc plate OT is ~8K"

[texwest]

6v6gta running at   ~350 vdc   cathode 20 vdc

Another bit of info is that we had it wired with a V1B .68uf cathode cap to make it more gainy.  It crossed my mind that this extra juice might have something to do with the failure.

[shooter]

I forgot Rk, Cathode resistor
do the ohms law, check against the datasheet
play REALLY hard while monitoring the cathode VDC, the speaker VAC, do more math for power at the tube and audio power at the speaker.


all that data adds up to minimum smoke, or at least, "Dude, If I leave it the way you like it's, schedule another fail in 3 months" 

[tubeswell]

Tetrode and pentode plate current (and grid curves, transconductance, plate resistance and gain) is affected by screen voltage. If screen voltage is too high, higher loads will result in excessive screen current under big signal conditions if you've set up the output tube to idle at Pmax. Keeping the screen about 50V below the plate idle voltage is one way of keeping a brake on this. Also Rg2 is important if you're going to drive the tube hard.

[texwest]

Tetrode and pentode plate current (and grid curves, transconductance, plate resistance and gain) is affected by screen voltage. If screen voltage is too high, higher loads will result in excessive screen current under big signal conditions if you've set up the output tube to idle at Pmax. Keeping the screen about 50V below the plate idle voltage is one way of keeping a brake on this. Also Rg2 is important if you're going to drive the tube hard.

Thanks a lot for that post. The screen voltage is about 30 volts less than the plate.  And I was not using a screen grid resistor on it.  That seems like a really good idea. I'm going to add one.

But theoretically could excess screen voltage cause the output transformer to fail?

[tubeswell]

But theoretically could excess screen voltage cause the output transformer to fail?

Not screen voltage, but screen current - and indirectly (through excess tube current and red-plating/thermal runaway - . But you'd probably see the signs (like deformed screen grid support rods). Depends how the OT failed. Maybe they didn't have a speaker connected (causing flyback transient voltage in the OT puncturing the winding insulation)? Hard to say from this distance

[PRR]

Running the OT without a load, LOUD, is perhaps the most likely cause of Champ OT failure. Ask Randy Bachman.

[texwest]

The owner says he didn't run it without a load or use a different cabinet or use an attenuator.  Still wondering what caused the failure. I read somewhere that sometimes arcing inside a tube can cause an OT failure???

[shooter]

Still wondering

quit 
when you fix broke stuff for a living there are just too many "wonder what......"
the better solution is to know/measure ALL the parameters, under MOST operating conditions.  That way you know, you did your part insuring it was "spec'd correctly"

[texwest]

Well I built the amp for the guy so I don't want a repeat of this.  I'm having to cover the cost of fixing it right now.  It's pretty much a spec champ except for the v-1b cathode cap that's added.   That makes it a hot rod.  I think I'm going to add a grid stopper resistor and a screen grid resistor. 

[sluckey]

Suck it up and pay for it. Maybe build it stock next time.

[texwest]

Suck it up and pay for it. Maybe build it stock next time.

Of course that's what I'm already doing. The amp is 8 months old so I offered to fix it.  Now I'm trying to learn from what happened, but all I know is that the OT failed and I can't find any reason why at this point.  I've built many amps and this is the first OT failure I've had.  Maybe it's just a fluke and it was a bad part to begin with.  I've heard of a lot of people putting a cap on v1b and that's the only real difference here.  If no one else has anything to add, I appreciate everyone's help. Thanks

[tubeswell]

A cathode bypass cap on the output stage won’t cause the OT to fail. Something else caused that, more likely a no-load situation or a buggered output tube

[sluckey]

It's unlikely that the cap on V1B cathode caused the OT to fail. Most likely playing the amp hard is the cause. The OT is always at risk when running an amp wide open, especially so if the speaker becomes disconnected even for a couple seconds while playing a power chord. Adding a 3KV diode connected from plate to ground will protect the OT from high voltage "flyback" pulses due to a disconnected speaker. You could use three series connected 1N4007s. Many bigger amps have this diode protection.

New OTs are rarely defective from the manufacturer, but it could happen. I've never seen one. Is your OT from a good quality manufacturer? Is this amp just a head version being used with an external speaker cab?

I hope my "suck it up" comment did not offend you. I meant no offense, but many times I don't choose my words carefully. Just ask anyone here.   

Maybe caution the owner that the most likely cause of failure was playing the amp hard. If he continues to do so it's likely to happen again.

[shooter]

OT failed and I can't find any reason

once it's up and working,
take all the normal DC measurements and post, maybe someone can spot something "off"
get a scope, post all the signal amplitudes with the amp pushed hard, post, maybe someone can spot something "off"
otherwise you can never really know much of what's going on electronically

[texwest]

It's unlikely that the cap on V1B cathode caused the OT to fail. Most likely playing the amp hard is the cause. The OT is always at risk when running an amp wide open, especially so if the speaker becomes disconnected even for a couple seconds while playing a power chord. Adding a 3KV diode connected from plate to ground will protect the OT from high voltage "flyback" pulses due to a disconnected speaker. You could use three series connected 1N4007s. Many bigger amps have this diode protection.

New OTs are rarely defective from the manufacturer, but it could happen. I've never seen one. Is your OT from a good quality manufacturer? Is this amp just a head version being used with an external speaker cab?

I hope my "suck it up" comment did not offend you. I meant no offense, but many times I don't choose my words carefully. Just ask anyone here.   

Maybe caution the owner that the most likely cause of failure was playing the amp hard. If he continues to do so it's likely to happen again.

Thanks.  I really wanted to know if that v1b cathode cap could cause problems.   

[shooter]

Lets say V1 magically put out a 5Vrms AC signal, some small milliseconds later the guitarist would know. (and should shutdown)
the rest of the circuit right of V1 would "clip" that 5V signal, convert it to a squarewave and hit the PA tube.
my understanding, tubes don't like square waves for long periods of time, so there is a slim chance the PA tube goes south, taking the OT with it.
my understanding, the PA tube is fine, so a V1 fail without a PA tube fail is REAL slim to none.


a no-load condition, mentioned before, will almost always take out an OT being hit with a changing signal.

[PRR]

> Adding a 3KV diode connected from plate to ground will protect the OT from high voltage "flyback"

In push-pull, yes, sorta. But isn't this amp single-ended? It can only conduct when it breaks-down. If not rated for repeated breakdown the diode is likely to die.

A heavy dummy load seems like the best answer. 50r 10W across the 8r tap.

[sluckey]

> Adding a 3KV diode connected from plate to ground will protect the OT from high voltage "flyback"

In push-pull, yes, sorta. But isn't this amp single-ended? It can only conduct when it breaks-down. If not rated for repeated breakdown the diode is likely to die.

A heavy dummy load seems like the best answer. 50r 10W across the 8r tap.

Sorry. I didn't realize there was a difference between SE and PP.

[texwest]

Okay so it sounds like putting the three diodes on the plate isn't a good idea?  It sounds like what he's describing is to put a 50 ohm resistor on the 8 ohm tap? But this transformer only has a 4 ohm tap so it doesn't seem like that would be possible?

[sluckey]

The 50Ω load will work on the 4Ω tap also. Not trying to match any loads. Just trying to prevent an open circuit.

[PRR]

> diodes on the plate isn't a good idea?

The normal plate swing may be 350V, down to 50V, up to 650V, and back.

The dangerous swing is 350V, down to 50V, up to 3500V, zapp. (Below shows only 300V B+ and more damping, "only" 2,056V kick.)

If we had a solid source of 650V-700V, we could rig a diode to there so it never goes past "double", which is only a bit past "normal".

In push-pull we have a trick. Tying the -other-side of the CT winding to zero clamps the peak to twice the supply. You see this on a LOT of commercial amps 1970s-onward (when hi-hi-hi-voltage diodes/stacks got affordable).