Gibson GA5-T crestline B+ low and bias questions

[DrGonz78]

So I am bringing this Gibson GA5-T back to life and have some questions. First off the schematic calls for 250v on the plates and my voltage is at about 205v. So obviously that is low and is part of the problem I believe. First off I did reform the caps by running the amp at 10-20VAC which put 20-40vDC or so on the caps, I did this for while. I am pretty sure reforming was necessary and I plan on recapping once I get the amp up to speed a bit more. Anyhow the R18 is the cathode bias resistor and is supposed to be 270ohms. I measured 259ohms on R18, so I am pretty sure the bias might be running a bit hot bringing down the plate voltage. Also, I measure 220k on R17 and 215k on R16.

Now I am not sure how to check the bias on the output tubes in this instance and would appreciate help on that matter. Just to note I have replace C1 and C9(cathode bias capacitor) as both were just croaked out. The amp does sound a bit better after the few things I have done but I am not sure how this amp is supposed to sound. It does not sound very good. Before replacing C1 and C9 the low frequencies were sounding like cross over distortion or something of that nature. I have attached a similar schematic to my amp and it is 100% the same. There are small differences but it is very very much the same. Any help is always appreciated and eye opening.

[DrGonz78]

Just wanted to update my numbers there... Sorry. R17 measures 215k and R16 measures 220k. Also, plate voltage is at 209v and the cathodes for the 6AQ5's is 13.28v and schematic is 15v.

[Leo_Gnardo]

I'd be wary of those maybe-not-reformed-enough filter caps. For parts that must be approaching 60 years old I'd just replace them. Filter caps leaking current heat up all by themselves and drop the supply voltage. The cathode resistor at 259 is close enough to 270 ohms but something to consider is modern line voltage tends to run high, figure about 10V more than in the mid 50's. That means your cathodes are running hotter & output tubes drawing even more current.

One quick way of guesstimating output tube current is measure the voltage across the cathode resistor, calculate I=V/R then divide the answer by 2 because 2 tubes are feeding current into that resistor. That's your plate plus screen grid current.

My check for plate current is the shunt method. Set your meter for milliamps, 100 or 200 mA scale whichever you have. Clip one probe to the OT center tap and the other to an output tube plate. We're assuming the ammeter's resistance is a small fraction of the resistance of the OT winding, otherwise this won't be very accurate, but it often is. The meter will give you the plate current for the tube it's attached to. Then do the same for the other tube.

Yet another: measure resistance of the OT windings from center tap to each plate lead. Then after warming up the amp, measure voltage across these same points. Use I=V/R again and there's your plate current.

Old interstage caps like to leak. If the voltage on output tube grids isn't very close to zero, you know you need to replace the caps bringing signal to those grids. And maybe the rest of them too. If the grid voltage on any of the pre tubes is dragged upwards by leaky caps, they're not going to behave right either. Have a cap replacement fest and I'll bet that GA5-T will work a charm.

[Chuck H]

I'd say that all the electrolytic caps get replaced. C12 (death cap) should be checked for leakage. Your measured resistances don't seem so far off as to cause that much voltage drop. It's possible you have a bad power tube. You can plug in just one tube at a time and measure the current of each. You can check the current by dividing the voltage across the cathode resistor by it's value. If both tubes seem similar take the same measurement with both tubes installed. Then measure the voltage from a plate pin to the top of the cathode resistor and multiply that by your current reading and that will give you the bias in watts. You can pull the rectifier tube and check the actual PT secondary voltage at pins 1 and 6 on the empty socket.

[DrGonz78]

Great info guys! Yeah I already removed the death cap and have a 3 prong grounded cable installed. I always do that first really. To know if a death cap is leaky you measure for any AC voltage across the cap correct?

I will definitely recap first before going further on this one. The amp has a wax & paper 20uf/20uf 350v and another single 20uf 350 caps to replace. With the multi-section one I can just get two 20uf 350v(or 450v) caps and join the negative sides together with each positive end going to where it was originally was connected correct? Have not had to do a cap job replacing multi-sectioned ones with singles.

[Leo_Gnardo]

I just wouldn't trust any "death cap" that's more or less old as I am. You did right by going grounded and kicking that old thing to the waste bin. I dont' see any need to have those caps in an amp even if it isn't grounded. All they do is give you an annoying shock if they're working right. Or a potentially deadly shock if they're leaky or shorted. Who needs either one.

Sometimes replacing a multisection with individual caps gives you the opportunity to land the negative end of those caps somewhere that makes better sense than the manufacturer's original. For instance the cap serving the preamp, usually the last one on the PS string, you can park its negative end closer to the input section. The caps handling first and second stage, at or very near the spot where PT center taps are.

I just worked on an early 60's GA-6 the other day. The guy who worked on it before me replaced the multicap with singles, twisted all 3 cap negatives into a bundle and landed that in the preamp zone. Then he wondered why it buzzed like a hive of bees on espresso. I removed the bundle of caps that were dangling from a wire tie, chose new caps double the uF of the original, and brought their negative poles to ground as I described. Hum dropped by a huge amount. Not totally gone but very acceptable. And the new caps are mounted much more sturdy - not likely to break loose & cause hi voltage mayhem.

[DrGonz78]

Okay so should I go from 20uf to 40uf and will that help reduce some hum? Also the cap serving the preamp section has it's negative connection closer to that area, so I get that idea better now thanks.

[Enzo]

FIND OUT where any hum comes from. You may need to recap the amp, but before you do, have you measured B+ ripple?

I'll say it again, this is just a guitar amp, not precision lab gear. 259ohms instead of 270 ohms is within 5% and not a problem. We all the time put a 1 ohm resistor in the cathode circuit and measure voltage across it to calculate current. Well you have 259 ohms in the circuit, and Ohm's Law still applies, use a calculator. Dissipation is found by current through the tube times voltage across it. measure the voltage on the cathode and subtract it from the plate voltage. That gives the voltage ACROSS the tube. Calculate tube current by voltage drop across the 259 ohms and Ohm's Law. Multiply together for dissipation in watts. That is how you calculate it on any tube amp, cathode bias doesn't change that. If it had fixed bias instead, the cathode would be grounded, so the voltage on the plate is the same as the voltage across the tube, but we use the same method either way.

How are your screen voltages and downstream B+?

Your B+ is low, is it being dragged down? That tube current figure should tell you. You put new cathode bypass caps in, OK, but did you measure the voltage at the power tube grids? They ought to have about zero volts on them, but if C7 or C8 is leaky, then a positive voltage will be there, and that will cause excess tube current and drag down your B+. B+ cann aalso measure low if it is not well filtered. You report hum? Is the B+ rippled in huge amount?

[Chuck H]

did you measure the voltage at the power tube grids? They ought to have about zero volts on them, but if C7 or C8 is leaky, then a positive voltage will be there, and that will cause excess tube current and drag down your B+. B+ cann aalso measure low if it is not well filtered. You report hum? Is the B+ rippled in huge amount?

Drat! I missed that one. I've only had two Gibsons in front of me but I've read that leaky caps are really common for them.

[Leo_Gnardo]

Okay so should I go from 20uf to 40uf and will that help reduce some hum? Also the cap serving the preamp section has it's negative connection closer to that area, so I get that idea better now thanks.

There are some who insist the "sound" of the amp will change, or you'll wreck your rectifier tube. Those who insist the way is was made is the only way. Think about this: manufacturers were cost-cutting back then too. Larger caps meant less profit or forcing the price up a couple bucks, now your amps' not competing well on price. Plus the smaller amps were considered to be "student amps", their makers and marketers didn't respect their own products the way we do now, Champs and all their kin.

Rectifier tubes are surprisingly resilient but they do have their limits. I wouldn't load one up with hundreds of microfarads of filtering. To keep hum down, hardly a problem going from 20 to 40 or 47 or 50.

Some experimenting with clip leads, you can pick a minimum-hum landing for the preamp filter.

Purists who insist on "the way the factory did it" better accept whatever hum & buzz arise from that without complaints. Most of my customers are recording or playing live thru PA systems; the less racket from the amp (not including the "talent" ) the better. Not one of them has moaned about "it's not exactly the same." Usually the review is "thanks, my amp sounds so much better."

[DrGonz78]

The good, bad and ugly. Well to answer some of Enzo's questions, I have a ton of B+ ripple. I measured 66.9vac (C11A) and 57vac (C11B) on each section of filtering for the B+ voltage. Before ever powering up the amp the first time I tested the capacitance on each filter the C13 test about 27uf and was much healthier than C11A or C11B. C11A/B both had no reading of any good capacitance on the meter and that is why I was so reserved to reform the caps for as long as I did. So C11 multi-section is toast. I have not tested the capacitance readings on the caps since I first ran the amp because the caps simply do not drain below 7vdc. Since there is voltage that can't be drained I would dare check it with the meter now, but I bet it's not looking very good.

I am getting 198vDC on the screens of the power tubes. After C13 feeding the B+ to the preamp section I barely measure any significant ripple so that is good, voltage at C13 is 163v. Plate on V2 is 110v, plate on V1A is 123v, and plate on V1B is 112v. Also good is the voltage at the grids of the power tubes is 0 vac or 0 vdc. Going to recap the beast and will return with the outcome on the amp later.

Thanks for helping me along with this repair and for answering all my questions so in depth. I am really starting to understand these things way better every time I get help from the MEF members.

[Leo_Gnardo]

Going to recap the beast and will return with the outcome on the amp later.

I think you'll be delighted with the results. In the future, you'll skip over the attempts to reform ancient filter caps, save time & frustration, whack a new collection of caps both filter & interstage in there without hesitation.

[telenut62]

The GA-5 I had was so quiet, I couldn't hear if it was on.....it had the major caps replaced.

[DrGonz78]

So yes I got this one all fixed up now and I stuck with 22uf filter caps. I had read some other threads out there where the tech had commented on how quiet this amp is after recapping. After fixing the amp I was amazed how quiet the amp is and it made me double check that it was actually on too. I did have one problem and that was the blown speaker that came with the amp. Oh well after replacing the speaker the amp sounds great!! Thanks again!

[Leo_Gnardo]

Glad it worked out. Yesterday must have been national blown speaker day. I took one out of a '65 Princeton Reverb, not a reissue. As a consolation prize I found Sam Hutton's initials on the floor of the box. Sam was Fender's prize tweed & tolex man. He continued working for individuals in his retirement, re-tweeding & tolexing amps into the 1990's. If you see what looks like a $ sign, with 2 vertical strokes, usually yellow wax-pencil, that's Sam.