Gibson Thor rebuild

Gibson, Valco, Gretsch, and more

Re: Gibson Thor rebuild

Postby foose on Thu Nov 03, 2016 3:10 am

ken,

the abnormally high voltages were present before i replaced all the caps today....as for the resistor....i should have known something was up! but you see, im actually a machinist by trade, so i sort of run by-the-print, and i am only beginning to ascend the learning curve on amplifiers. and that bit of jargon you just unleashed on me will have me re-reading it for several hours while i google every other word, i am sure. but thats what i need, and i thank you for it! ill change back the other resistor tomorrow, and honestly im going to retrace the front panel. i rewired the whole thing, and the odds are good i missed something. its funny, you try to be as thorough as possible, but like i always have said, "if it doesnt work, its probably because of something you did." so ill figure out that one tomorrow maybe. thanks again for the reply, i appreciate it!

Brian
foose
Peasant
 
Posts: 16
Joined: Sun Oct 09, 2016 2:13 pm

Re: Gibson Thor rebuild

Postby Vortexion on Fri Nov 04, 2016 3:51 pm

Brian, re. your measurement of +15V at the grids of the Phase Inverter (V2) - I've just had one of those "How did I forget THAT?" moments. Here goes...

The type of Phase Inverter you've got there is a "Long-Tail Pair" - a design that's often used in musical instrument amplifiers. What I forgot is that, with a Long-Tail Pair, you CAN'T take a voltage reading directly from the grid(s) using an ordinary test meter (i.e. a standard Moving needle or Digital Multimeter), because the input impedance of the meter will load down the very high input impedance of the grid, draw a tiny bit of current and thus give rise to a false, higher-than-expected reading. So my guess is that this is most probably why you're seeing this bizarre +15V whenever you stick your meter probe on those pins.

To get its quoted voltage measurements for pins 5 and 8, Gibson's techs would have used a device called a "Vacuum Tube Volt Meter" (VTVM), which has an enormously high input impedance and thus would hardly load down the grid at all. The later, solid-state equivalent device is called a FET-VOM. Old VTVMs and FET-VOMs turn up fairly affordably on eBay, but to be honest, they're not the sort of kit you're going to be using an awful lot! If you did ever want to buy one though, I'd go for the VTVM, because FET-VOMs are rather delicate little beasts and easy to kill. A VTVM, by comparison, will withstand a lot of electrical abuse. Of course it will - it's got a VALVE in it! :mrgreen:
Vortexion
Newt
 
Posts: 1251
Joined: Thu Jan 10, 2008 4:28 pm
Location: Norwich, UK

Re: Gibson Thor rebuild

Postby foose on Fri Nov 04, 2016 8:00 pm

ok.....so perhaps those pins are ok where theyre at. perhaps not. i havent had time to retrace my front panel or put that 560K resistor back in there....my @$*&% amplifier on that old juke box cut out yesterday and all the normal tricks (smacking it) have stopped working. i have to get back in there and check my connections because something is obviously loose....i have a sneaking suspicion i know what though...my tubes are getting hot but i think the 6.3 for the heaters is conked out, which is a blade-and-socket connection coming from another unit in the juke, and i think the socket part isnt contacting my blades properly.

anyhow, i guess until i can start fixing stuff instead of breaking stuff, i better cool it for a while. be more methodical....take more time. check things more thoroughly...(maybe even twice!) stop hitting things to make them work. you know...that sort of stuff.

i hope this curve starts to slope a little more soon. but truly, the more i learn the less i know.

thanks again for the help, ill be back as soon as i have something else to report!

Brian Foose
foose
Peasant
 
Posts: 16
Joined: Sun Oct 09, 2016 2:13 pm

Re: Gibson Thor rebuild

Postby Vortexion on Fri Nov 04, 2016 8:29 pm

foose wrote:be more methodical....take more time. check things more thoroughly...(maybe even twice!)

I like the sound of that.

foose wrote:stop hitting things to make them work.

Steady on there - that's my principal diagnostic method!

foose wrote:truly, the more i learn the less i know.

Ah well, that never changes. And if you keep doing it for long enough, eventually you forget half of it.

Finally, remember the Two Golden Rules when working with High Voltage equipment:

Always work naked.
Always work drunk.

:wink:
Vortexion
Newt
 
Posts: 1251
Joined: Thu Jan 10, 2008 4:28 pm
Location: Norwich, UK

Re: Gibson Thor rebuild

Postby foose on Sun Nov 13, 2016 1:33 pm

well its just been a wildly productive weekend!

i managed to get my jukebox amp running again, this time better than ever (broken wire and a blown tube) so i tackled the Gibson once again. i found that i had put a cap in the wrong spot on the treble control, and sure enough she fired right up!

i had put a 1Meg audio taper pot in for the bass control as previously suggested, since you cant find 2 meg audio taper pots anymore. everything sounded pretty good, but i thought id put the 2 meg in there after all, just to see how it sounded. the difference is actually negligible, so the 2 meg will probably stay in there.

the only other thing is that the amp isnt very loud. you can crank all of the controls, and its barely loud enough to feed back. i mean if you were playing against an acoustic guitar it would be great, but its certainly not going to keep up with a drum kit. with the volume cranked all the way up the sound quality is slightly distorted, as you would expect, but it isnt bad....it makes me wonder if perhaps thats just all the volume this amp ever put out. again, the sound quality is good, but it just doesnt seem to punch you in the chest like i feel a bass amp should. do you guys have any tips or tricks for increasing/testing output of an amp? or am i expecting too much from a little amplifier from the late sixties?

thanks again for everyones help and patience! im sure ill be back on with another project very soon, ive always got my eyes peeled for the next thing.

Brian Foose
foose
Peasant
 
Posts: 16
Joined: Sun Oct 09, 2016 2:13 pm

Re: Gibson Thor rebuild

Postby Vortexion on Sun Nov 13, 2016 1:43 pm

Good news that it's working now, but a 50W amp should be good and loud - I used to play bass through a VOX AC50 and the whole street hated me! :mrgreen: So there must be some reason why the thing isn't working at full-bore yet.

Did you say that you'd tested the output tubes? If they were worn out, that would explain a lot. And what about all those voltages that were way above what they should be - have they settled down, or are they still OTT?

ADDED LATER:

...and did you remember to put that 560k resistor back into the NFB loop?
Vortexion
Newt
 
Posts: 1251
Joined: Thu Jan 10, 2008 4:28 pm
Location: Norwich, UK

Re: Gibson Thor rebuild

Postby foose on Thu Nov 17, 2016 8:26 pm

i did remember to replace the 560k resistor...but thanks for the reminder!

i was initially so pumped that the amp worked again, i just buttoned everything up and didnt check a dang thing, so today i pulled the chassis back out and re-ran my tests...everything was excellent EXCEPT for pins 3 and 4 on V3 and V4....i should expect 420 volts on pin 3 and 415 on pin 4, i measured 467 and 454 respectively (on both tubes)...i looked at the schematic and reasoned that my 450 volt tap from T1 must be running high, and sure enough i measured 472 volts after choke L1, as it goes into the 1K/2W resistor...my guess is that there must be something going on with that choke? im not sure about how transformers typically fail, but in my mind i would expect a reduced output if the transformer had dropped a few windings....i cant imagine a scenario where it would be able to produce extra voltage, but again, im nearly a complete novice and i dont know about the typical failure modes of components. what i mean to say is that, assuming T1 is operating properly and outputting the correct voltage, the only other thing that could have an effect on voltage would be the choke, does that sound reasonable? i have not desoldered the choke and done a resistance measurement across it yet...perhaps not the best indicator of a faulty choke, but its all i know on how to test it....once my caps bleed down a little bit maybe ill do that.

thanks again for your thoughts!
Brian Foose
foose
Peasant
 
Posts: 16
Joined: Sun Oct 09, 2016 2:13 pm

Re: Gibson Thor rebuild

Postby Vortexion on Thu Nov 17, 2016 10:15 pm

Hi Brian,

I've got some good(ish) news for you. I've just taken another look at the Thor schematic, and (don't ask me why I didn't see this before! :Doh: ) in the Voltage Chart, the 420V quoted for the plates (Pin 3) of the output tubes (V3 and V4) is WRONG. If 450V was measurable at the centre-tap of the output transformer primary (as is clearly shown on the diagram itself), the voltage could not possibly drop to 420V by the time it reached the output tube plates, because there is no significant resistance in that path to create such a voltage drop (you can ignore the DC resistance of the transformer primary winding: it'll be very low). You can see this on the schematics of many tube amplifiers, such as this Fender Bassman circuit:

Image

You'll see that the voltage marked just to the right of the Standby switch is +425V. If you now look at the voltage marked at Pin 3 of the 6L6 output tubes, it too is +425V.

Consequently, the Gibson Thor Voltage Chart should say "450V" (or very near) for pin 3 of the output tubes. So the good news is that your 467V isn't as massively over the top as you thought. In fact, it's only about 3.8% more than it should be. Hooray!

Before we go doubting the choke, we need to establish whether the power supply transformer is putting out the right AC voltages. Unfortunately, Gibson hasn't provided us with an AC volts figure for the HT secondary winding - and besides, even if they had, would it be trustworthy? HA! :roll:

However, we can check what's coming out of the heater winding. With all of the tubes in their sockets and warmed up, slap your test meter across that heater winding and check what AC it's putting out. One thing to keep in mind is that, in many parts of the USA, mains voltage has increased since the 1960s.

Let's talk output tubes. Sorry to be asking you again, but are you SURE that the amplifier's output tubes are in full working order? If they were worn out, they would hardly be drawing any current, and thus the voltage at the plates would be higher than expected. One way to check for this is to do the following:

Remove both output tubes, switch on the amp and check the DC voltages again. If they are noticably higher than when the tubes are in circuit, then we know that those tubes are drawing current when they're in the amp. If the voltage turns out about the same regardless of whether they're in the circuit or not, then either the tubes are duds or they've been biased so cold as to be in cut-off (and hence are being prevented from drawing current).
Vortexion
Newt
 
Posts: 1251
Joined: Thu Jan 10, 2008 4:28 pm
Location: Norwich, UK

Re: Gibson Thor rebuild

Postby foose on Fri Nov 18, 2016 9:18 pm

i have tested my output tubes on my relatively ancient precision instruments "tubemaster 10-12" and both tubes came out good....they were also bought brand new and matched. i also tested the original 6ca7's and they also tested good....i notice no difference in the volume when switching out either set of output tubes.....

my heater voltage is 6.7 VAC on both output tubes.

with tubes removed im getting 506 VDC, so the tubes are definitely drawing current when theyre in there.

so.....crap. everything is coming up just fine again. and yet the amp is pleasant when its cranked all the way up. i feel like it should easily keep up with a drum kit but i have my doubts.

this just in....on a whim i reversed my speaker leads (they had been cut and not marked for polarity)....the volume got slightly better, but still not what id expect. we are still poking around at things though....
foose
Peasant
 
Posts: 16
Joined: Sun Oct 09, 2016 2:13 pm

Re: Gibson Thor rebuild

Postby Vortexion on Sat Nov 19, 2016 11:09 am

Thanks for checking out those things, Brian. Good to know that the output tubes are healthy.

It's fair to assume that the power supply transformer is getting a rather higher mains voltage than it was originally designed for, but not disastrously so: 6.7VAC at the heaters is just grazing the upper limit of Mullard's recommended working range (which is +/- 7% either side of 6.3VAC).

Curious to hear that reversing the speaker connections should apparently result in a slight volume boost. If the speakers are in a sealed cabinet, I can see that being a possibility. If the cab is open or vented at all, I wouldn't expect it to make any difference.

BTW... you know the easy way to check the polarity of a speaker, don't you? Disconnect it from the amp and connect a small 1.5 Volt battery across the speaker terminals (just for a second or so - don't keep it there). If the cone moves forward, the positive terminal is touching the positive end of the battery. If it moves backwards... well, you get the gist.

Anyway, that doesn't solve your low volume issue. Given that we know your output tubes are healthy, the fact that you can crank the thing up full and it still remains clean and civilised suggests that either:

a) the output tubes are biased way too cold
or
b) there just isn't enough signal getting through to the output stage.

The best way to test for (a) is to measure how much current is passing through each output tube when the amp is on but the volume control is down, so no signal is being amplified. This is known as the "quiescent current" or "idle current". To do this, you can either buy a current measuring device, such as a "Bias-Rite" - while easy to use, these are rather expensive - or you can do the following, which is much cheaper and cleverer:

1. Switch off the amp, unplug it from the mains and ensure the caps are discharged.

2. Check that the Pin 6 tag on each output tube socket is not connected to anything (I mention this because, with this being a point-to-point amp, it's possible that the tag for Pin 6 might be being used as an anchor point for something). If it is being used, do not proceed any further.

3. If (and ONLY IF) Pin 6 is unused, disconnect the ground wire from Pin 8 of each socket.

4. Obtain two fine-tolerance 1 Ohm resistors (standard 1% metal film types will suffice). Solder one of these resistors between Pins 6 and 8 of each tube socket (being very careful to keep it clear of Pin 7, which is one of the heater terminals :shock: ).

5. Re-connect the ground wires, soldering them to Pin 6 of each socket.

If you've done this, you have now put a "current sensing" resistor between cathode and ground on each output tube. Now to proceed with the test itself:

6. Set your meter to a low DC Volts range (we'll be measuring millivolts here). Clip the negative probe to chassis ground and clip the positive probe to Pin 8 of one of the output tube sockets. BIG SHOUTY WARNING: DO NOT TRY TO HOLD THE PROBE ONTO THE Pin 8 TAG BY HAND - IT'S TOO RISKY. ALSO, BE VERY CAREFUL TO ENSURE THAT THE PROBE CLIP IS ABSOLUTELY SECURE, NOT LIABLE TO SLIP OFF AND NOT AT RISK OF SHORTING TO ANY OTHER PIN. YOU HAVE BEEN WARNED! :old:

7. Make sure the amp's volume knob is turned down and that nothing is plugged into the instrument socket.

8. Plug in the amplifer, switch it on and wait for it to warm up and stabilise for about 10-15 minutes.

9. Look at the reading on your test meter and make a note of it. Whatever it says in millivolts is equal to how many milliAMPS are being drawn through that tube.

10. Switch off the amp, discharge the caps etc... and do it all over again with the positive probe on Pin 8 of the other output tube socket.

Assuming that the Plate voltage is still around +467V as you quoted earlier, and the tubes are being properly biased, you should be seeing something between 27 to 37 milliamps per tube.

27mA would be a "cool" but usable bias (equivalent to 50% plate dissipation)
32.5mA would be a middling sort of bias (equivalent to 60% plate dissipation)
37mA would be a "hot" bias (equivalent to 70% plate dissipation)

If you're getting a figure much below 27mA per tube, then the output tubes are biased too cold, which would result in low output and lots of headroom (so making the amp very hard to push into distortion).


If the tubes appear to be well biased, then we have to consider situation (b) - i.e. there's not enough signal getting through to the output tube grids..

The problem is that testing this idea really requires a signal generator and either an oscilloscope or a meter capable of showing peak-to-peak AC voltage, to see the voltage amplitude of the signal at various stages of the circuit.

Beyond this, without being able to actually get hands-on with the amp itself, I'm afraid I'm running out of ideas! :dunno:
Vortexion
Newt
 
Posts: 1251
Joined: Thu Jan 10, 2008 4:28 pm
Location: Norwich, UK

Re: Gibson Thor rebuild

Postby foose on Sat Nov 19, 2016 1:54 pm

holy smokes....i can tell i have a long, long way to go in my education...

while poking around last night we decided to test resistance across the speakers and found about 12.5 ohms...im probably going to display my incredible ignorance here, but arent they supposed to be 8 or possibly 16 if theyre in series? 12.5 just seemed like an odd measurement...not that i think my volume issue is completely tied to the speakers because as youve said, with the volume cranked i cant seem to push the amp into distorting. i just thought id get your input on the speakers themselves possibly adding to the problem.

that said, we went ahead and got a case of beer last night and proceeded to rock out the basement for a couple hours, and as it turns out the Thor kept up with everyone pretty well. i believe theres still more work to do on it, so i will try the amperage sensing resistor you described and see what i get.

i do have an oscilloscope, but its also old as heck...hasnt been calibrated in 25 years or so...i got it from the old tech ed teacher from my highschool. it does still fire up and basically works, but i dont know that id trust it for any real measurements. who knows, maybe ill hook it up at some point and see what i can see.

thanks again for all the help youve been giving me...your knowledge is seriously impressive and i cant thank you enough for taking the time to help a novice like me fuddle his way through the flat part of the curve.
foose
Peasant
 
Posts: 16
Joined: Sun Oct 09, 2016 2:13 pm

Re: Gibson Thor rebuild

Postby Vortexion on Sat Nov 19, 2016 2:31 pm

No need to worry about the DC resistance of the speakers being lower than their nominal impedance - that's quite normal. DC resistance and AC impedance are different animals, and that 12.5 Ohms sounds typical to me.

When you do that mod to measure the idle current, if you do find that it's out of whack, get back to me and we'll start investigating the bias supply circuit. Also, if you fancy, we can make a very simple mod to it, to enable you to have adjustable bias - always a useful thing with modern output tubes, which tend to be more fussy than the sturdy beasts we had in the good ol' days.
As for my knowledge... Well, I knows what I knows, and there's still a lot I don't! :mrgreen: - but I'm very grateful for the compliment. :cheers:
Vortexion
Newt
 
Posts: 1251
Joined: Thu Jan 10, 2008 4:28 pm
Location: Norwich, UK

Re: Gibson Thor rebuild

Postby foose on Fri Nov 25, 2016 8:25 pm

hey! just an update....basically to say i havent done anything yet...holidays and all....hopefully early next week or possibly even this weekend! ill let you know what i figure out!

hope you had a good thanksgiving and whatnot, i managed to get in there, eat, and back out in 1 hr 12 minutes. personal best.

im not a fan of family gatherings.

Brian
foose
Peasant
 
Posts: 16
Joined: Sun Oct 09, 2016 2:13 pm

Previous

Return to Miscellaneous American Amplifiers

Who is online

Users browsing this forum: No registered users and 1 guest