1954 Gibson GA-77 issue/design flaw?

[jukelemon]

Hi all.


I have a continuing issue with my GA77 that has me stumped.


Some background...I have owned this amp for many years, serves as my main gigging amp,  and have played thru it > 1000 hours easily.  I know it well :)

I have fixed several things thru the years per the usual wear BUT there is one component that fails almost every year it seems.


This component is the cathode bypass cap on the 6l6's.


This amp has the unusual design whereas the standby is not breaking the B+ but instead breaking the cathode resistor/cap to ground connection.

When I got the amp, the cap had already been replaced with a 50uf/50v sprague.  It lasted maybe 25 hours and took the amp down during a show.

50v is usually plenty for a bypass cap so I assumed a bad cap from the factory and got another one.  It lasted about the same.

So given there is no voltage spec on the schematic, the next one was a 50uf/150v.  Well...that lasted about 50 hours and failed at a show..again.

So, over the years I have replaced these caps so many times I have lost count.  I typically just do it as if its an oil change so I am not left dead in the water at a show.

Despite this...it happened again this past Friday at a show and I am a bit over it lol.

Sooo...what can be going on?  Is the design such that no cap will work like this and I should just convert the wiring over to the typical standby/Fender convention?

When I say the caps fails...it blows the end off lol.  I mean...it really fails.

Thanks all.

[shooter]

when it "fails" does it spill it's guts?


have you tried NOT using SB? 


my quick guess, with the SB OPEN, the cap charges to the B+ rail, or the cathode it setting at B+, then you hit the switch n the cap "sees" B+ long enough to blow it's guts.

[SEL49]

Make this simple change to the 25K connection. Replace the cap with a 50µF/150V. Never have to replace it again.

[Willabe]

Make this simple change to the 25K connection. Replace the cap with a 50µF/150V. Never have to replace it again.

What does this do that the bypass cap wont blow now?

What channel do you plug into most of the time, 1 or 2?

[jukelemon]

when it "fails" does it spill it's guts?


have you tried NOT using SB? 


my quick guess, with the SB OPEN, the cap charges to the B+ rail, or the cathode it setting at B+, then you hit the switch n the cap "sees" B+ long enough to blow it's guts.

Yeah, it spills/literally the top bursts open.

Never have tried to not use SB.

The cap is charged always no?  The standby is just taking the cathodes to ground no?  How does the addition of the cathodes place stress on the cap AND why is there such a voltage demand that is blows BUT only after hours and hours of use?  It is almost like the cap cannot handle the in rush or duty cycle?

[jukelemon]

Make this simple change to the 25K connection. Replace the cap with a 50µF/150V. Never have to replace it again.

Thanks.

Can you explain why?  Is it because the current design is actually picking up B rail voltages??  I never even thought about that.  Why is there even B rail voltages connected to the cathode/bypass cap? Why is the 25k even needed then?

[jukelemon]

Make this simple change to the 25K connection. Replace the cap with a 50µF/150V. Never have to replace it again.

What does this do that the bypass cap wont blow now?

What channel do you plug into most of the time, 1 or 2?

I play in Channel 1 mostly.

[jukelemon]

So...in reading some other forums it seems that this design, albeit very odd, is in effort to reduce the current spike when going off Standby.

And the 25k/150ohm connection forms a divider so even if the screens are getting ~300v the actual voltage to the cap/cathode resistor is very small?

Ugh...which further confuses me because at this stage 150v would be more than sufficient as a spec given the divider is what....1.8 volts?


And..I have looked at some other 77s and they have a 12k resistor going from the ON terminals of the switch to the cathodes.  It is not even on the schematic but I have seen pics of several that have it.

Mine does not have this 12k resistor.  Could that be the issue? 

Attaching pic as an example.

[jukelemon]

I think I may have figured the issue out.

My amp, and like many others it seems on the web, have used a more common wiring method for the bypass cap as opposed to following the actual schematic of the 54 whereas both the Cathode resistor and the 50uf cap are switched in and out via the power/SB switch.

Now..what I was thinking is that I go back to how this design can function being that the cathode Resistor/Cap is connected to ~300v from the screens BUT also from a 25k resistor thus creating a voltage divider and thus allowing the 50uf cap to really only see a small voltage.  And with that, the cap really did not need to be 150v but that's an aside.

BUT...this amp that I have and many others it seems have moved the bypass cap off the switch and permanently connected to the cathodes and it's own ground connection but still connected to the screen supply.

I just confirmed this on my amp i.e. when in SB the cathodes and this this 150v cap is actually seeing 316vdc.  Way over the spec and so every time I would leave the amp on standby this cap was getting pummeled.  When measured at the cathode side of the cathode resistor (where the cap really should live), voltage is 0 and per design until the SB is moved to ON.

This is the only conclusion I can come up with that makes sense and also explain why every 54 GA-77 likely runs thru these all the time i.e. the cap is never mounted right.

Thoughts?

[HotBluePlates]

Thoughts?

I would personally leave the voltage divider of Screen --> 25kΩ --> 150Ω --> Ground, and not use the standby switch.  Just go from "Off" to "On" instead.

The 25kΩ resistor is there to help cool off the idle bias, as the current through the 25kΩ adds to the cathode-current pulled through the cathode resistor to yield a slightly larger bias voltage.

    - No-cathode-current through the 150Ω should cause the voltage at the + of the cap to fall when the standby is open.

I would probably just turn down the amp Volume if taking a break between sets.

[jukelemon]

Thoughts?

I would personally leave the voltage divider of Screen --> 25kΩ --> 150Ω --> Ground, and not use the standby switch.  Just go from "Off" to "On" instead.

The 25kΩ resistor is there to help cool off the idle bias, as the current through the 25kΩ adds to the cathode-current pulled through the cathode resistor to yield a slightly larger bias voltage.

    - No-cathode-current through the 150Ω should cause the voltage at the + of the cap to fall when the standby is open.

I would probably just turn down the amp Volume if taking a break between sets.

Thanks

Do you disagree with my last findings?  Seems like the smoking gun to me.

[jukelemon]

Thoughts?

I would personally leave the voltage divider of Screen --> 25kΩ --> 150Ω --> Ground, and not use the standby switch.  Just go from "Off" to "On" instead.

The 25kΩ resistor is there to help cool off the idle bias, as the current through the 25kΩ adds to the cathode-current pulled through the cathode resistor to yield a slightly larger bias voltage.

    - No-cathode-current through the 150Ω should cause the voltage at the + of the cap to fall when the standby is open.

I would probably just turn down the amp Volume if taking a break between sets.

Hey there.

So..wired the amp back to original design including the 12k resistor that ran from the 150/50uf connection to the cathode off the switch.  This 12k is not on the schem BUT is in all the 1954 77s.

Amp is back/healthy BUT the damn switch pops now lol.  I have read that the 12k was designed to eliminate/reduce the pop but it sure doesn't lol.

So...one could (like you said) just never run on SB or I guess I could go back to the way it was and just get a higher rated cap to be able to deal with the 300+ vdc screen supply that is hitting the bypass cap when in SB.

[jukelemon]

Hi all!  Hope everyone is well.

Well...the cathode cap blew again...at a show...out of town.

I cannot get used to not using the Standy By.  It is just too much of a habit.

Sooooo I am thinking of just rewiring this to a more standard HT type Stand By design like a Fender.

With that said, can you all see any harm in:

1- breaking the connection between the B+ rail and the cathode RC connection and
2- removing the 25k resistor entirely and
3- rewiring the SB to a HT Off/On and
4- just wiring the Cathode resistor and bypass cap like a typical early 50s design

The bypass cap would go back to a 50v in this above revision.

I am just so over the blowing of this cap lol.


Thank you.

[passaloutre]

I cannot get used to not using the Standy By.  It is just too much of a habit.

Sounds like you also have a habit of replacing cathode bypass caps. Removing the standby switch would break both habits.

If you insist on using the standby switch (there is no reason for it), leave the power tubes conducting but move the standby switch to ground the grid of the phase inverter. Leave the first section of the switch as is on the mains circuit. On the second section, connect the common pole to the grid of 6J5 (pin 5), connect the "off" and "sb" positions to ground, and leave the "on" position not connected to anything.

Or, y'know, just don't use the standby.

I've been gigging 20ish times a year for a decade now and never thought to reach for a standby switch. If I need a mute, I step on my tuner pedal or turn the guitar down. The power tubes don't mind idling while you take your set break.

[DummyLoad]

The 1953 & Vanguard models did it right - Shorting the 5881/6L6GA grids together post PI coupling caps. Mod yours to mimic that. That was stupid place to put a standby switch on a cathode bias amp. The three variants; 53/54/Vanguard plans all use the same switch pattern (2P3T), use that 53/Vanguard plan and forgo all the heartache and pops.


--Pete

[jukelemon]

The 1953 & Vanguard models did it right - Shorting the 5881/6L6GA grids together post PI coupling caps. Mod yours to mimic that. That was stupid place to put a standby switch on a cathode bias amp. The three variants; 53/54/Vanguard plans all use the same switch pattern (2P3T), use that 53/Vanguard plan and forgo all the heartache and pops.


--Pete

Hey Pete.

Please attach a schem to what you mean.  The Vanguard is the GA-77 and I though the 54 was the first year?

Thanks

[jukelemon]

I cannot get used to not using the Standy By.  It is just too much of a habit.

Sounds like you also have a habit of replacing cathode bypass caps. Removing the standby switch would break both habits.

If you insist on using the standby switch (there is no reason for it), leave the power tubes conducting but move the standby switch to ground the grid of the phase inverter. Leave the first section of the switch as is on the mains circuit. On the second section, connect the common pole to the grid of 6J5 (pin 5), connect the "off" and "sb" positions to ground, and leave the "on" position not connected to anything.

Or, y'know, just don't use the standby.

I've been gigging 20ish times a year for a decade now and never thought to reach for a standby switch. If I need a mute, I step on my tuner pedal or turn the guitar down. The power tubes don't mind idling while you take your set break.

Gotcha and yeah, same here.  Except I have always used amps with SBs and it's just a thing in the memory lol.

[jukelemon]

The 1953 & Vanguard models did it right - Shorting the 5881/6L6GA grids together post PI coupling caps. Mod yours to mimic that. That was stupid place to put a standby switch on a cathode bias amp. The three variants; 53/54/Vanguard plans all use the same switch pattern (2P3T), use that 53/Vanguard plan and forgo all the heartache and pops.


--Pete

Hey Pete.

Please attach a schem to what you mean.  The Vanguard is the GA-77 and I though the 54 was the first year?

Thanks

Nevermind...I found it in my Gibson Master book.  Thanks and good idea.