Blues Junior blown R47 & output tubes failed

[blackstone]

Spent a bit of time troubleshooting my amp, so I thought I'd share my findings.
Amp was cranked way up on a gig and it started to cut in and out then eventually fizzed out to nothing.
On inspection I noticed that R47 on power supply circuit feeding output tubes had burned up. Replaced resistor and tested to find that the EL84M's also had a problem - worked then fizzed out again. BTW these were almost new tubes & military grade, so quite a surprise!

OK could just replace tubes but I figure there must be a reason they failed, other than just cranking the amp which it should take anyway.

Checking the voltages I found that B+ was higher than fender spec. (329V) and would fluctuate up to 9V (331-339V on different tests) I'm guessing this is due to my mains supply fluctuation?
Is this normal?

With this in mind I predict that grid screen voltage Z must be reaching almost 320V on occasion, which can't be good since manufacturers tube spec. is 300V max.

OK so I put some old power tubes back in and crank the amp way up - all works fine but if you look at the power tubes while playing a chord you can see the grid screen glow orange - this, I think, is the reason for the tube failure, too much power going through the grid screen (pin9).

Btw. the amp was set to 240V and with the 27K bias mod.

Solution:
I have replace R47 with a 3.3K (3W) resistor (instead of 2.2K) to keep voltages lower, result is no more glow in grid screen when playing cranked to max.

Could have upped the grid stopper resistors instead, but I think this will work just as well? Comments?

I'm hoping that this will prevent any further trouble.

Wonder if anyone has had the same problem?
Hope this helps.

References:
http://www.ratvalveamps.com/fbj-lifesaver
http://www.aikenamps.com/JJ_EH_tube_tests.html

[PaisleyTube]

in any case: check the biasing!

[blackstone]

Yes, might be an idea to run a bit cooler, with the elevated B+ voltages. I did check the biasing when I changed the tubes, and it was running on the hot side, but I decided to leave it since I was using an M grade EL84.

Now that raises another question with B+ fluctuation I may not be getting an accurate measurement... maybe a bit of guess work? Best run a bit cooler anyway.

Thanks.

[xtian]

B+ will flux when you play the amp. Is that what you mean? Or is B+ changing while the amp is idle?

[blackstone]

No I mean at idle, B+ fluctuates.
On different tests I've seen up to 9V difference. Can only think it's my supply, or transformer temperature perhaps? Or could it be the filter caps? - no audible hum, quite a quiet amp in fact.

[xtian]

No I mean at idle, B+ fluctuates.
On different tests I've seen up to 9V difference. Can only think it's my supply, or transformer temperature perhaps? Or could it be the filter caps? - no audible hum, quite a quiet amp in fact.

9v is nothing. You'll see that much change while the amp warms up.

[blackstone]

Cool, glad to know it's normal then!
Guess I'd better allow for this though when calculating voltages, especially since the plate and screen voltages are already above manufacturers spec. of 300V max for a EL84.

[PaisleyTube]

EL84s can handle higher voltages very well, just bias them correctly......

[sixstringer]

It's probable the bad tube (even if it was new or near new) fried the grid resistor (R47 on the cream colored PC board version of the BJr) instead of the other way around.

Adding the Billm adjustable mod to a BJr is very worthwhile. I've done it to a number of BJrs, both green and cream board versions and it turns a Blues Junior in to a much more versatile, better sounding and louder amp. The stock BJr is biased (intentionally, to make it "bluesy") way over spec for EL84 and it makes the amp sound small and boxy. Many guys change the speaker trying to make it sound better but a more effective fix is to re-bias at a more reasonable level. There are many cool mods for the BJr but changing to an adjustable bias make the most impressive difference.
And, the tubes last longer too.