Engl valve amp disaster

OK, I repaired an Engl Sovereign 100 recently, it's a 100 Watt combo using 4 x 6L6.

One of the JJ 6l6 had obviously overheated, and it's screen resistor was toast. The other 3 measured quite worn, so I replaced all 4 with personally matched JJ Teslas I have in stock along with the screen resistor. I also replaced some HT caps which were leaking electrolyte.

All the pre-amp valves (JJ Tesla ECC83) tested good.

The amp was fine for a few gigs, the owner was delighted, saying it sounded great. Then it apparently broke. Blew the mains fuse.

Customer replaced it (more than once? With the correct type? Dunno, it was missing when the amp came back.)

Anyway I as usual first took out all the valves and tested them.

4 out of the 6 pre-amp valves had open circuit heaters.

Now, this is one of the most difficult and time consuming valve amps I have ever had the misfortune to have dealt with. You can't see any of the components on the main board as they are between the PCB and the chassis. There are 3 other PCB's hard wired to this mainboard (2 front panel potentiometer PCBs and the rear PCB.) No plugs and sockets, all 4 PCBs have to come out of the chassis as one, along with all the umbillicals connecting them together, which is a total nightmare as they only just make it. The OPT is actually ON the main PCB, making things physically really difficult to remove the boards. All the power transformer windings are hard wired onto the main PCB, so you have to desolder all these first.

One problem is I can't find the correct schematic for the board I have, and I have a burnt resistor on a preamp cathode I can't identify.

But more to the point is the mode of failure here.

The only schematic I could find is the same one Engl sent me, and is not the right one for this particular model, but here it is for reference.

formatting link

Sorry this is so long winded, but I feel it is necessary to provide all the pre data.

Anyway, look at the output stage. There are 2 BY509 diodes on the Anodes of the 6L6 valves, going to the standby switch and then to ground. Only the amp I have here isn't quite like that. THIS one has the 2 BY509 diodes connected to the HEATER circuit, pin 2, on the output valves. The diodes are connected to pins 2 and 3 of the output valves. The standby switch connects all cathodes to ground, but is not connected to either diode.

Now, I found one of these diodes is shorted. Bummer. Meaning you get HT fed to all the heaters in the amp. Double bummer is that the heater winding has a 250 Ohm hum balance pot across it, connected to ground, the heater winding is otherwise in the air. So now this HT has a path through all the heaters to ground.

This balance pot is now open circuit all ways, so current must have flowed from HT, through the heaters, burning 4 of them out, through the pot, burning it out, which is when it stopped.

I also have the burnt resistor on a preamp valve cathode to ground, maybe this was an internal short, putting full HT on the cathode?

There is other damage I won't get into just now, but I just would like to understand why they put these diodes on the heater circuit, and if my synopsis/series of events here is correct.

Nightmare.

Reply to
Gareth Magennis
Loading thread data ...

Can you find a flea-bay "spares or repair" one with a different fault to compare with?

Reply to
N_Cook

Can you find a flea-bay "spares or repair" one with a different fault to compare with?

*********************************************************

Not really. I think this particular one might be quite rare, since even Engl can't yet come up with the schematic.

Gareth.

Reply to
Gareth Magennis

I don't see the reason either, other than the heater circuit being close to ground (but not quite). I guess you should be grateful it didn't take out the filament winding of the power transformer.

I would consider wiring it with the diodes connected directly to the standby switch, and isolate the filament per the schematic you have. Can't help but wonder if it wasn't a factory screw up.

Another possibility is isolate the anodes to those diodes with a 125 ohm resistor to the standby switch instead of using the balance pot as the ground return.

Crazy. Good luck.

Reply to
John-Del

I wonder if the new output arced causing the diode to short.

Reply to
John-Del

I wonder if the new output arced causing the diode to short.

*******************************

These are 4kV diodes, hard to find. I had to order them from Europe, Farnell, RS etc don't have stock.

Gareth.

Reply to
Gareth Magennis

** WRONG !!!!

Their rating is 11kV at 4mA and should never have ben used in the amp.

.... Phil

hard to find.

Reply to
Phil Allison

the

n

to

r.

d

** Connection to the heater circuit was done out of convenience and assumes there is a CT that goes to chassis - which of course does not exist.

The BY509 diodes are *intended* to "catch" any large flyback spikes, clampi ng them so the plate voltage on the 6L6s never goes negative which in turn limits the peak voltage at the opposite end of the OT to no more than doubl e the DC supply - sufficient for all normal operation.

However the BY 509 is rated at only 4mA and so is entirely inadequate for t he job. Stings of 1N4007s are often used for this or other internally serie sed diodes that add up to 4kV or more rated for at LEAST 250mA !!

BTW: Don't you juts HATE schematics that have no dots to show if lines draw n crossing at right angles are connected or not !!!

What a crock of shit !!!!!!

..... Phil

Reply to
Phil Allison

** WRONG !!!!

Their rating is 11kV at 4mA and should never have ben used in the amp.

.... Phil

hard to find.

************************************************

Ah, sorry Phil, the schematic is not of the board I have, and doesn't show the correct diode, I just assumed it did.

The diodes on the board I have are GP 02-40.

formatting link

This is, I believe, a 4kV diode.

Gareth.

Reply to
Gareth Magennis

** Like you assume everything else ....

** More importantly, it is rated at 250mA continuous.

Nevertheless it failed and caused all the damage you have there.

I suggest you repair the damage and fit new diodes wired DIRECT to chassis /ground. The use of fast diodes rated at 1 amp and 1kV or more is a good idea.

.... Phil

Reply to
Phil Allison

** Connection to the heater circuit was done out of convenience and assumes there is a CT that goes to chassis - which of course does not exist.

The BY509 diodes are *intended* to "catch" any large flyback spikes, clamping them so the plate voltage on the 6L6s never goes negative which in turn limits the peak voltage at the opposite end of the OT to no more than double the DC supply - sufficient for all normal operation.

However the BY 509 is rated at only 4mA and so is entirely inadequate for the job. Stings of 1N4007s are often used for this or other internally seriesed diodes that add up to 4kV or more rated for at LEAST 250mA !!

BTW: Don't you juts HATE schematics that have no dots to show if lines drawn crossing at right angles are connected or not !!!

What a crock of shit !!!!!!

..... Phil

**********************************************

"** Connection to the heater circuit was done out of convenience and assumes there is a CT that goes to chassis - which of course does not exist."

This makes more sense now.

Gareth.

Reply to
Gareth Magennis

Are you sure the right tubes are there? The wiring suggests the tube pins are not matching the surrounding wiring and components.

Reply to
Sjouke Burry

** Like you assume everything else ....

** More importantly, it is rated at 250mA continuous.

Nevertheless it failed and caused all the damage you have there.

I suggest you repair the damage and fit new diodes wired DIRECT to chassis /ground. The use of fast diodes rated at 1 amp and 1kV or more is a good idea.

.... Phil

*******************************************

OK, so I am still of the opinion that nothing could have caused that diode to fail other than the diode itself (?)

Cheers,

Gareth.

Reply to
Gareth Magennis

** NB: Have a good look ( and a sniff ) of the output valves - one of them may have internal arcing damage inside the plastic base between pins 3 and 2.

.... Phil

Reply to
Phil Allison

** There are a couple of scenarios* that allow the 4kV rating to be exceeded & if the spiking is at a high audio frequency, the diodes may dissipate enough heat during switch off to cause failure.

Fact is, such diode DO sometimes fail and ought not cause collateral damage.

" Like plugging a jack lead from speaker output back to an input causing massive NF oscillation with no speaker load.

.... Phil

Reply to
Phil Allison

** NB: Have a good look ( and a sniff ) of the output valves - one of them may have internal arcing damage inside the plastic base between pins 3 and 2.

.... Phil

**************************************

And this could have shorted the parallel 4kV diode between 3 & 2?

Or did the opposite occur?

Gareth.

Reply to
Gareth Magennis

** I only mentioned the possibility of valve arcing for completeness - cos it is another scenario that would cause similar damage.

Fix you Engl, come back when you have.

BTW

When the Krauts stuff something up - they do a thorough job of it.

.... Phil

Reply to
Phil Allison

** I only mentioned the possibility of valve arcing for completeness - cos it is another scenario that would cause similar damage.

Fix you Engl, come back when you have.

BTW

When the Krauts stuff something up - they do a thorough job of it.

.... Phil

*******************************************

I have fixed the Engl.

Some of the 35v ELectro caps on the pre-amp cathodes were shorted, so all 10 got replaced. Kind of confirms a large global overvoltage happened I think, along with the burnt cathode resistor on V5.

The 6l6 heaters all survived, but I guess they are somewhat more resilient.

The Engl tech reckons V5 shorted internally. Not quite sure how this would short the OPT diode to put HT on all the heaters, but who knows the real sequence of events.

I don't know what the customer did, he has only said he replaced the mains fuse with the correct one, but is otherwise keeping schtum.

I still don't have the schematics for this amp, maybe putting the protection diodes on the heater circuit instead of ground was a big design mistake. I have moved them to ground.

Cheers,

Gareth.

Reply to
Gareth Magennis

** I only mentioned the possibility of valve arcing for completeness - cos it is another scenario that would cause similar damage.

Fix you Engl, come back when you have.

BTW

When the Krauts stuff something up - they do a thorough job of it.

.... Phil

*******************************************

I have fixed the Engl.

Some of the 35v ELectro caps on the pre-amp cathodes were shorted, so all 10 got replaced. Kind of confirms a large global overvoltage happened I think, along with the burnt cathode resistor on V5.

The 6l6 heaters all survived, but I guess they are somewhat more resilient.

The Engl tech reckons V5 shorted internally. Not quite sure how this would short the OPT diode to put HT on all the heaters, but who knows the real sequence of events.

I don't know what the customer did, he has only said he replaced the mains fuse with the correct one, but is otherwise keeping schtum.

I still don't have the schematics for this amp, maybe putting the protection diodes on the heater circuit instead of ground was a big design mistake. I have moved them to ground.

Cheers,

Gareth.

*************************************************

One other thing, V5 is driving the effects send and return loop, so this is available to the outside world.

Gareth.

Reply to
Gareth Magennis

** That the HV diodes connect to the heater line ( instead of chassis ground ) likely did not cause the failure, but it sure caused all the damage.

FYI:

A tech working for the maker / importer is NO use to you when it comes to bad design problems. I see a lot of them these days in tube amps and have to modify circuits to effect a reliable repair.

I don't give a hoot what the amp maker or even the customer thinks of this cos *** I *** am the one liable for a repeat failure both financially and in terms of reputation.

.... Phil

Reply to
Phil Allison

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.