Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: Joneg44 on December 12, 2024, 04:19:05 pm
-
Hi to all - thanks in advance for your patience and wisdom. This is my first attempt at an amp repair. KNow this is a lot of words and questions. <smile>
The back story - my guitarist son bought this amp pre-covid from my brother's friend for $300. On the first day of use, smoke poured out and that was that. He had other priorities and other amps to use. He noted wax dripping, talked to a friend who told him a transformer melted. He was really happy with the sound - briefly.
It sat until now in my home. I suddenly have some time and decided to dig in. Have limited electronics and hardware skills but am working on that. Have reviewed a ton of videos and amp basics things. Am cautious on safety elements and understand how to be safe. Have gear including multimeters, variac, current limiter, tube tester, scope, chop stick(s), cap discharge thingy.
QC date is March 28, 1973 which lines up with transformer date codes - qc person is J Castillo - no serial number that I can see. Utah speakers, 4 ohm 100w RMS output marked on panel. (Yeah I have seen and understood previous threads about increased line voltage increase magically reclassed circuit from 85w to 100w)
Questions:
First and foremost - schematic and layout would help me greatly and it is confusing to me to figure out. Can you point me in the right direction?
1. The transformer that melted was the vibrato tranformer - my guitar teacher says only use orginal fender transformers but it looks like only new after market are easily available today - seems like that shouldn't be an issue - whadd'ya think? My guitar teach also says some other problem melted the ttx and figure that out.
2. From my limited understanding, am guessing a shorted tube on the vibrato circuit could be a potential culprit. Thoughts?
3. Another thought might be a failed coupling cap (feel free to laugh heartily at any of my noob guesses and redirect - am learning)
4. The power cord has duct tape by plug - have not taken off yet but might that be playing into it. (Looked like 10amp fuse was in 3amp fuse holder) Could this be part of the mystery? Assume I should just replace cord. (looks like aux plug is wired in reverse polarity) Thoughts?
5. My plan is figure out and correct root cause for the meting transformer, test and replace tubes as needed, recap filter caps, test other caps and resistors and replace as needed and leave well enough alone - no mods planned.
6. The fiber board is a "wavy" not sure if that is an issue?
7. Looks like caps can be a philosphical discussion - for the filter caps am planning on FT and am not sure what to use if I need to replace the coupling caps - thoughts on either welcome.
Pictures are here on imageshack: https://imageshack.com/a/ef517/1 - let me know if you have any difficulty accessing or want additional pix.
Thanks again and have a superduper great day!
-
Welcome!
I don't think the transformers 'melted'. Old Fender iron does that with what looks like wax dripping off the outside. But are still fine.
It may be that a power tube went out and took the OT (output trans) or the PT (power trans) with it.
There is no vibrato trans. PT, OT, choke, and reverb trans.
Keep all those preamp tubes, they may still be good. There looks to be some RCA and Bugle boy's. Just set them aside.
The power tubes look like they may have gone bad, especially the 1 on the left in the pic, closest to the preamp tubes. Tubes should have what they call 'getter' sprayed onto the inside of the tube. It's silver in color. It absorbs gasses inside the tube. If the tubes seal has broken, the getter turns white. Tubes dead.
Big mistake with a 10A fuse!
Put your nose right up very close to the transformers 1 at a time and smell for a burnt smell. Look for burnt resistors, they will smoke if they get overloaded from current draw.
If no burnt smell from the iron set, no burnt R's, pull all the tubes, plug into the vari and light bulb limiter. Slowly bring up to 120acv.
If light bulb goes back dim, turn off amp, set limiter aside, you don't need it now for a while. Now without limiter bring up on vari to 120acv, measure B+ leads that go to the rectifier diodes for acv, measure with gator clip meter leads, 1 to B+ red, the other to ground.
Then measure heaters, green wires, same way. And measure for the bias acv.
Post all these please.
-
You will not like my anwer.
You're dealing with something big as your first repair.
A Twin is not at all an amp for the novice DIYer.
You're getting ready to replace parts unnecessarily.
You obviously lack training at the outset, reading your questions that mix up all the circuits and their functions.
Your guitar teacher is definitely not giving you good advice. A reverb transformer doesn't have to be a Fender original. But this part doesn't seem to need replacing.
You should study more seriously how an amp works before putting your hands on it.
Buy this book first, you will learn what you need to know. Easy to read.
The Tube Amp Book by Aspen Pittman. . The best invest to who want to learn and fix tubes amp.
-
Thanks Latole and Willabe.
Willabe - Like your answer because it gives me a starting point. Tubes are pulled. Transformers sniffed - no burnt smell. Power tube getters are all mainly silver but some darker streaking or patches. Resistors look fine. The reason we thought the transformer melted is because a waxy substance dripped into the cabinet from the reverb trans (thanks for correcting my terminology on trans.) I have to wait until my Limiter bulb and base come in tomorrow to go any further. Duct tape is off of power cord - the sheath was pulled out from the plug but the wires look fine will test and use shrink tube. Thanks for your quick response. Will report back with with the readings you requested when I have them.
Latole - Your answer is fine. Am more a hands on kinda guy that learns by doing, while reading. Am reading quite a bit to re-understand fundamental electronics and am starting to understand amps. Appreciate the resource you suggested. Was not really planning on replacing parts unnecessarily but thanks for the warning and advice. You accurately identify that I am not well trained and am not fluent in terminology. Aspen Pittman looks like he was very well respected in the field and very accomplished.
-
I admire people like you who are willing to learn and are able to take feedback in a positive way.
I wish you good luck with this repair job, because I believe you'll succeed.
-
Yes that is a big, somewhat complex circuit (though not by modern standards) to work on for your first repair. What you do have going for you though is that it looks like it's never been serviced, which means it should match the schematic and you don't need to worry about someone else's bad modifications or attempts at repairs. It's also one of the most popular amps ever so whatever the issue is, is probably nothing new. Willabe's advice is a great start, definitely start there.
I believe the schematic you're looking for is this one - https://el34world.com/charts/Schematics/files/Fender/Fender_twin_reverb_sf_100_schem.pdf (https://el34world.com/charts/Schematics/files/Fender/Fender_twin_reverb_sf_100_schem.pdf)
For a layout, I found this one on TDPRI although I can't speak to its accuracy.
(https://i.ibb.co/x7TNRHL/100-W-Mast-Vol-SF-Twin-Rev-LAYOUT-original.jpg)
Honestly also the standard AB763 Twin Reverb schematic and layout https://el34world.com/charts/Schematics/files/Fender/Fender_twin_reverb_ab763_schem.pdf (https://el34world.com/charts/Schematics/files/Fender/Fender_twin_reverb_ab763_schem.pdf) will get you 90+% of the way there, it would probably just be missing a few Silverface tweaks like master volume and some parts value differences.
-
I have this yellow one :
-
Latole and TitaniumValhalla appreciate the guidance on schematics and layout. Gives me something to wrap my head around. Have a great weekend!
-
Tubes are pulled. Transformers sniffed - no burnt smell. Power tube getters are all mainly silver but some darker streaking or patches. Resistors look fine.
This is all very good, but something caused the smoke. Don't throw those power tubes out, set them aside for now.
Your going to need a whole new set of tubes for that amp as you move forward. Get a matched quad of 6L6GC's.
The reason we thought the transformer melted is because a waxy substance dripped into the cabinet from the reverb trans.
Yes, old Fender iron can/does that.
I have to wait until my Limiter bulb and base come in tomorrow to go any further.
Has to be an old style incandescent bulb, can't be LED or compact florescent. For that amps power/current draw, you want a 100w bulb.
-
Willabe - tubes ordered and old ones set aside. The bulb is 250w incandescent photo lamp. Will post previously requested meter readings once in hand. Thanks.
-
Bulb is too big.
The lower the wattage the more protection it gives.
For a 4 x 6L6GC 100w bulb is good.
-
Bulb is too big.
The lower the wattage the more protection it gives.
For a 4 x 6L6GC 100w bulb is a good.
I use 100 watts incandescent bulb for nay amp .
-
Bulb is too big.
The lower the wattage the more protection it gives.
For a 4 x 6L6GC 100w bulb is good.
I use 100 watts incandescent bulb for nay amp.
You can, it will work, but on smaller amps, 2 x 6V6's, I use a 60w, and on a 1 x 6V6, I use a 40w I got a hold of. :laugh:
-
Am reading quite a bit to re-understand fundamental electronics and am starting to understand amps.
Here's a few links that will help you, if you don't know about these yet;
Merlin;
https://www.valvewizard.co.uk/ (https://www.valvewizard.co.uk/)
Rob Robinette;
https://robrobinette.com/Amp_Stuff.htm (https://robrobinette.com/Amp_Stuff.htm)
Randell Aiken;
https://www.aikenamps.com/index.php/white-papers (https://www.aikenamps.com/index.php/white-papers)
And Sluckey's web page, has many proven amp builds, but read his scrapbook, not long but great information, very clear drawings, some high lited.
https://www.sluckeyamps.com/misc/Amp_Scrapbook.pdf (https://www.sluckeyamps.com/misc/Amp_Scrapbook.pdf)
https://www.sluckeyamps.com/index.htm (https://www.sluckeyamps.com/index.htm)
And Uncle Doug can be helpful on youtube. Helps to see it many times.
-
Willabe said
measure B+ leads that go to the rectifier diodes for acv, measure with gator clip meter leads, 1 to B+ red, the other to ground.
Then measure heaters, green wires, same way. And measure for the bias acv.
Post all these please.
There was no drama with power up on limiter so moved to variac only. All measurements at 120acv from Variac.
B+: acv = 341 (same on both)
Heaters: acv = 3.63 and 3.49
Bias: acv = 62
Willabe and Latole - thanks for the advice about to too high a wattage on the bulb for limiter.
Willabe - thanks for the reference/reading materials. Started with some of Rob Robinette's stuff a while back and will continue. Am working through a reference suggested somewhere here: "Electronics for Physicists." Am looking forward to Sluckey's material. Will dig into Merlin and Randell too.
-
Here is my plan forward - please chime in if I am being foolish somewhere:
1. Read about failure modes that can cause smoke in tube amp and continue electronics 101 refresher
2. Continue on the route that Willabe implies may be a failure mode candidate - bad power tube - will have a tube tester and new 6L6GCs mid-week
3. Will check vdc and vac from rectifier and bias to relevant tube sockets using schematics and layout y'all kindly provided
4. Will check back here when I have results of above - have a great week
Only question - looks like bias acv from TR1 is above 20% "tolerance." Assume that transformer windings are stable over time and it likely has been like that but do you gurus see it differently? (Will look for something near -52vac going to balance thingy.)
-
1st, snip out the .047uF/600v 'death cap' on the ground switch before you do anything else. With the 3 wire cord you don't really need the ground switch any more. Just snip it, you don't need to heat up the iron.
Read about failure modes that can cause smoke in tube amp and continue electronics 101 refresher.
When something draws too much current it over heats, and will burn up, causing smoke. That's why we use fuses, to limit the current in case of failure. That's why putting in a 10A fuse where it's supposed to only be 3A's is a really bad idea. :BangHead: :cussing:
That amp looks to be in original condition mostly. Power tubes look stock, electrolytic caps (e-cap) look stock and must be changed.
A tube can have it's screen grid short out internally and take out the screen grid R that's mounted on the tube socket. Pins 4 to 6. Measure the resistance of all 4 and post the #'s you get. Should be 470 ohms. Those R's are carbon comp and will start to smoke when heated up and can/will catch fire.
And you need to change out every e-cap. The electrolytic paste inside dries out over the years. They can short out to ground and draw massive current. After 20/30 years, their done, taking a chance leaving them in. F&T's are very good caps at good price, many guys use those, I do. (See at the very end.)
And the B+ dcv power supply dog house board has a few dcv dropping R's on it, 1 of those could have smoked. When you change the B+ filter caps change those R's, their CC (carbon comp) to flame proof metal film R's. Same for the 6L6GC screen grid R's. And the 1K5 grid stopper R on the tube socket, pins 1 to 5. A lot of heat there and if the tube shorts could cause a fire with CC.
will have a tube tester
Buying a tube tester can be/is a discussion on it's own. I don't know much about them, never owned 1. Many of the used 1's need to be calibrated.
Will check vdc and vac from rectifier and bias to relevant tube sockets using schematics and layout y'all kindly provided.
I'd rebuild that rectifier board too. Those solid state (SS) diodes are pretty old. Use new 1N4007's. Ultra fast are nice, UF1N4007, they switch on for a shorter time and so are a little quieter because of that. For pretty much same price now a days, I use them, not a must though. Don't forget to change the e-cap in the bias balance circuit.
An amp needs a full load, all tubes in, to give the correct acv/dcv measurements. Unloaded all readings will be high, sometimes very high.
(Will look for something near -52vac going to balance thingy.)
That amp has a -bias balance pot, not a -bias adjust pot. It it to try and get both sides of the 6L6's that are working in push/pull to draw the same current on each side. There's 2 6L6GC tubes on each side. Class AB PP, when 1 side is on, push, the other side is off resting.
I would add a 25KL (linear) pot after the 15K R that is hooked up to the -bias balance pot's wiper. Then you can set the -bias balance and set the overall -bias. Read up on negative bias.
And when you check/measure for -bias, you always measure for it at the tube socket pin. Always! DO NOT put power tubes in until you know for certain you have enough of a -dcv at the tube socket.
Only question - looks like bias acv from TR1 is above 20% "tolerance." Assume that transformer windings are stable over time and it likely has been like that but do you gurus see it differently?
We don't use the term 'gurus' here. :laugh:
And TR1 we call PT, power transformer.
Tube amp transformers are very stable the years. Your getting a high reading because the PT is not loaded, it's not doing any work, not under any strain. Think of a car going up a steep hill, engine is under strain, it will slow the engine down.
That 20% tolerance note, in most/pretty much all cases, is way larger then what you really see.
Our host Doug, has the F&T caps in his on line store along with many/most of the parts you'll need. And his shipping is very fast.
https://hoffmanamps.com/MyStore/catalog/parts4.htm (https://hoffmanamps.com/MyStore/catalog/parts4.htm)
-
And your going to want to put in 1 ohm R's between the 6L6GC's cathode (K) and ground.
I just disconnect the ground braid from pin 8 on the power tube socket and solder the 1 ohm R to pin 8 and the other end to the disconnected end of the ground braid.
Then you can safely measure the power tubes current draw without the high dcv B+. You really want this.
Doug sells them.
-
Willabe -
Thanks very much - am clear on all your suggestions and will do. The tube tester failed 1 tube on emission test and showed no shorts but is an old 1964 Knight 600 (good / bad / short tester) have not tested the components or calibrated. Kinda moot with new tubes coming.
Resistance between power tube sockets pin 4-6:
1 = 499 ohm
2 - 515 ohm
3 - 506 ohm
4 - 507 ohm
The 10A fuse was previous owners *idea* - my son played the amp three hours before smoke. Also upon further review - missed it in my excitement to discharge caps, see some carbon and corrosion on inside of doghouse that seems an artifact of "smoke" and it is right by the reverb trans so it explains why my son thought it was coming from trans - same general vicinity. (image attached)
Will put a part list together and get from our host Doug. In addition to your suggestions, am also going to try to generally follow Doug's post on servicing a Fender Amp... and make a dummy load for testing.
-
Resistance between power tube sockets pin 4-6:
1 = 499 ohm
2 - 515 ohm
3 - 506 ohm
4 - 507 ohm
Those are fine, just wanted to see if 1 was open. They only drifted up a little, wouldn't have hurt anything.
The 10A fuse was previous owners *idea* - my son played the amp three hours before smoke.
Yes, I understood that.
... - missed it in my excitement to discharge caps, see some carbon and corrosion on inside of doghouse that seems an artifact of "smoke" and it is right by the reverb trans so it explains why my son thought it was coming from trans
By that loose screw up on the side, looks like smoke. That must have been from 1 of the B+ dropping R's, measure them. What are they?
Will put a part list together and get from our host Doug. In addition to your suggestions, am also going to try to generally follow Doug's post on servicing a Fender Amp... and make a dummy load for testing.
Ok.
But you don't absolutely need a dummy load, I don't use one. That's just so you can plug in a guitar and bang away as you look at a scope. So no load loud volume.
Edit; And for reading acv's and looking for at what stage distortion starts, but you can use a signal tracer amp for that.
Some times you need to hear if there's strange noises, pots are scratchy, squeals/oscillation, etc.
-
Personally, I think a dummy load is a good idea, if you want to inject a signal and monitor it with a scope, or even trace it with a listening amp.
You don't have to listen to the drone of the signal, and can test signal/voltages along the chain.
-
... if you want to inject a signal and monitor it with a scope,
Of course you do. Rarely do you really need to inject a signal, only when you've hit a wall as a last resort. Whole lot of ways to trouble shoot a problem without a scope.
He's got a lot of things to do with this amp before he needs a dummy load.
Cross 1 bridge at a time.
-
AllNewman and Willabe - AllNewman - thanks for chiming in on dummy load. Wilabe not only are you right about one step at a time and the list of to dos growing - but I can be like a squirrel in a cage full of nuts so appreciate your step-wise focus. The conversation on when and for what a dummy load is useful as I learn but am setting aside dummy load for now. Thanks again.
Willabe -
You wrote:
That must have been from 1 of the B+ dropping R's, measure them. What are they?
R closest to Transformers 2.4 k ohms
R furthest from Transformers 10.4 k ohms
Consolidating parts list now and will order kit suggested so far. Death cap clipped.
-
R closest to Transformers 2.4 k ohms.
R furthest from Transformers 10.4 k ohms.
Those are both correct. But do change them out for flame proof metal film (MF) R's.
Death cap clipped.
Good.
-
OK - got some parts in and before I melt lead want to check on some substitutions I made based on what the internets told me:
ecaps 20 uF 500vdc > 22 uf 500vdc
ecaps 50 uF 70 vdc > 47 uF 100vdc
rectifier diodes UF1N4007 > UF4007
If these substitutions are not the right idea, let me know and I will reorder as needed. Thanks for your help.
-
OK - got some parts in and before I melt lead want to check on some substitutions I made based on what the internets told me:
ecaps 20 uF 500vdc > 22 uf 500vdc
ecaps 50 uF 70 vdc > 47 uF 100vdc
rectifier diodes UF1N4007 > UF4007
If these substitutions are not the right idea, let me know and I will reorder as needed. Thanks for your help.
All good. Green light. Let's get this beast roaring again.
-
Hi all - after holidays. some house projects, and some parts ordering snafus am caught up with myself.
Added resistors for bias measurement and replaced several that were out of spec.
Rebuilt rectifier circuit
(https://imagizer.imageshack.com/img922/865/OAoH0I.jpg)
Dog house - recapped and resistors replaced
(https://imagizer.imageshack.com/img923/5438/MVGTYi.jpg)
Added bias pot to the balance circuit
(https://imagizer.imageshack.com/img923/9355/jHmufo.jpg)
Tested preamp tubes and 3 of the 6 needed to be replaced and are on order.
My guess at next steps are (thoughts welcome):
1. smoke test on current limiter/variac without 6L6s
2. after successful smoke test, put in 6L6s and bias them
3. after 6L6s are biased, put in preamp tubes and bring up slowly on variac and current limiter - smoke test
4. if all good, provide a signal and see if it emerges at output
Questions
1. Robinette describes procedure for biasing using the resistors newly installed - assume following that with due care is how I will bias. Is that right?
2. The bias pot I got had four legs - I assumed I only needed two on one side - one to resistor and one soldered to the balance pot. I clipped the other two off - OK?
3. Are there steps I am missing etc.?
Some notes to other n00bs:
1. Part of my part snafu was ordering resistor values based on layout or schematic. The best reference is the color coding of the existing resistor since the values can vary.
2. The Fender schematics are confusing for rectifier diodes. The schematic shows the rectifier anode and cathode correctly oriented by the symbol but for some reason put + plus and - minus signs in opposite direction.
3. I burned the insulation on a wire and needed to replace. There seems to be reasonable consensus that you can use 20 awg wire with teflon insulation to replace.
-
Check bias before installing the 6L6 tubes, adjust it for max negative then carry on with installing the tubes. But where did you get that little bias trim pot? How is it hooked up in circuit? Looks like you'll have to make sure your bias circuiit is working and adjusting properly on all four tube sockets prior to doing anything further.
-
Thanks mrresistor. Here is the bias pot I used.
https://darrenriley.com/homepage/fender-bias-pot-25k-linear-pcb-0027873000/
Let me know your thoughts. Basically soldered one “leg” to balance pot and connected resistor on same side. Snipped off other leg and connector on the other side.
-
1. smoke test on current limiter/variac without 6L6s.
2. after successful smoke test, put in 6L6s and bias them.
3. after 6L6s are biased, put in preamp tubes and bring up slowly on variac and current limiter - smoke test.
1. No variac, just light bulb limiter, LBL. 100w bulb for 4 x 6L6GB/GC. Make sure speaker is plugged in for all these tests.
Read this using the LBL for this build start up, read reply #16 1st, then the 1st post, then reply's #8, #10.
https://el34world.com/Forum/index.php?topic=32424.0 (https://el34world.com/Forum/index.php?topic=32424.0)
2. No, test/check -bias without power cubes. Check for -dcv at each power tube socket. turn the -bias pot FCW and FCCW to check for dcv swing, post those -dcv's here. You want the -dcv you end up with setting the -bias to hopefully be close the the middle of the -bias pot, so you have some wiggle room to put in a hotter/colder set of tubes some day.
If -dcv verified, set the -dcv for the most -dcv, then put tubes in. Power up amp, no LBL, set -bias.
3. No variac, only LBL.
4. if all good, provide a signal and see if it emerges at output.
Just plug in a guitar.
1. Robinette describes procedure for biasing using the resistors newly installed - assume following that with due care is how I will bias.
You put in 1 ohm current sensing R's on the power tube cathodes (K)? If you have matched tubes, then clip your meter leads across a 1 ohm R, set meter for dcv. 1dcv = 1mA of tube current. Find an online bias calculator, set -bias of tubes for 60% to maybe 70%. I like 60%.
2. The bias pot I got had four legs - I assumed I only needed two on one side - one to resistor and one soldered to the balance pot. I clipped the other two off - OK?
I'm thinking the 4th leg of that pot was for mounting? No connection to the pots resistor element.
The pot gets wired up to 1 outside leg and the pots wiper, so it's an adjustable R.
1. Part of my part snafu was ordering resistor values based on layout or schematic. The best reference is the color coding of the existing resistor since the values can vary.
No. That's backwards.
That can be the least accurate way to check a R's value. Colors can fade, and new R value colors can be pretty useless. The old R's, when new or haven't faded, those color were good.
As carbon composition (CC) R's age they often drift, usually they go up in value. So go by the schematic. Plus, it's a 20% tolerance in Fender amps on R values. So if they pulled a higher value R out of the bin when wiring up the amp and then it drifted up, it can be way high.
You don't know if someones been in there and changed a R. You look at the schematic, you can double check by looking at the layout, if you have 1, then try to match it up by the colors. You can then try to measure the R with your meter set to resistance, but the CC R might have drifted way up. and that 20% tolerance in Fender amps on R values.
All in all, go by the schematic.
3. I burned the insulation on a wire and needed to replace. There seems to be reasonable consensus that you can use 20 awg wire with teflon insulation to replace.
Depends on what the wire is for, depends on the current load. Most of the wire in a tube amp only carries the ac signal voltage, and it's low current. Fender used 22 awg for that.
-
https://darrenriley.com/homepage/fender-bias-pot-25k-linear-pcb-0027873000/ (https://darrenriley.com/homepage/fender-bias-pot-25k-linear-pcb-0027873000/)
Let me know your thoughts. Basically soldered one “leg” to balance pot and connected resistor on same side. Snipped off other leg and connector on the other side.
Don't cut off anything until you figure out that pot.
Use your meter, set to resistance to figure our what's what. Find the 2 ends of the pot, and find the pots wiper.
Looks like 2 of the legs are the same thing. They put the extra leg to solder it to the pcb for strength and stability.
-
Thanks Willabe. Understand your guidance and next steps.
(Though I already cut off the 'extra' leg on pot - but will verify with meter and replace if I screwed it up.)
-
Bias readings at pin 5 without tubes
Tube FCCW FCW
1&2 -70.6. -53.3
3&4. -67.4 -45.3
Will plan on installing power tubes and biasing later unless these readings concern people.
-
Checking my work against the thread - missed a couple of things suggested by Willabe and am looping back to complete (once parts come in) before going any further.
1. screen grid and grid stopper Rs
2. small ecaps - 25 uf / 25 v
in the words of the Bard - "doh!"
-
Not a "hard" amp to work on, but it's a pretty huge circuit and repair costs can get quite high taking it to a tech who sounds like he doesn't know what he's talking about to begin with and potentially ripping you off (no such thing as a "vibrato" transformer).
Sounds like you got a pretty good handle and safety on what you're doing so far. Basically replace all the electrolytics, crumbling carbon comp resistors in the high voltage sections and you're good to go for the next 40 years.
My .02c - while you're doing this overhaul, might not be a bad idea to give the normal channel its own unique voicing so it's something you'd actually use versus the Vibrato Channel.
You can follow the Robinette mod page , but summarized is this:
220k anode resistor.
.022 tone and coupling caps.
470p to 560p treble peaker cap.
All of these raise the distortion characteristics dramatically and keep it controlled and tight, somewhere between a Dumble and a Marshall.
Using no master volume however, you're going to find it's still going to be quite loud.
It's possible to install a post phase inverter Lar Mar style Master volume, but that's quite a bit of labor still.
In either case, do the "Fritz Mod" which connects the Normal Channel to the Vibrato Channel - you do this mainly to gain vibrato and reverb function (nice), but at the price of a volume cut.
That's not necessarily a bad thing! We typically want our clean and dirty channels around the same volume and in practice this is very useful! You can crank the volume/gain pot as a result and really get some ripping tones.
It's actually one of the easiest, most dramatic mods you can do in a mid-60s Fender circuit and is totally reversible.
-
Thanks for chiming in GlideOn -
As an update, completed recapping and replacing screen grid and grid stop resistors. Bias pot in largest negative resistance position. Added tubes back.
Powered up on LBL - all good, removed LBL
Powered up - first issue was I was getting no dcv reading across the sensing resistor, thought I had a measurement issue.
Measured plate to cathode voltage (measuring plate pin to cathode pin) was about 457vdc on each socket
Measured range of bias current was still -70 to -53 and returned to highest -70 setting.
Plugged in guitar - and it played through speaker but unbiased sound.
Powered down and up multiple times trying to figure out why there was 0 vdc on k ground.
Poking around with my chop stick either broke or dislodged two of the sensing Rs - removed 1 (not enough lead to solder back to pin) and resoldered ground to pin. Reattached other one.
Tried to measure bias current directly with Nobsound 8-Pin Dual Bias Current Probes Tester Meter for EL34 KT88 6L6 6V6 6550 Vacuum Tube Amp Amplifier (2*Meter + 2*CT1-C, Cathode Current) - which I had bought though I had not planned to use - they also read 0 mA.
Took out bias probes and restarted amp again (screwed up and did not have speakers plugged in - noticed after turning off standby and reading high plate voltage 550v) shut down - probably on in this mode 30 seconds, unplugged, restarted with speakers plugged in.
Was starting to remeasure plate to cathode voltage pin to pin and caught something that caused a momentary arc, led indicator went dark and it was two step forward and three steps back kinda President's day.
Other items for your kind consideration:
Replacing ecaps - one of the ecaps was 5mfd 50vdc - I used a 25mfd 25vdc cap I had while I waited for a NOS 5mfd 50vdc.
Guess the question is what is my next step...
-
Upon further review, tripped a breaker. My next steps are to work from power transformer through the circuit. (Have tubes out again)
LED lights. There is B+ voltage. LBL is not lighting up. Will post again if I have questions.
-
Too much negative pin5 voltage could mean no current is flowing thru tube.
Either on or off limiter fully turn on amp and turn bias knob to allow more current to flow and check across 1ohm resistor with meter.
Some current should flow at -53, but if not may need to go to -40 to -53.
-
Thanks Calboy - looks like maybe I need to figure out a different resistor after bias pot since at last measurement my bias pot voltage range was:
Tube FCCW FCW
1&2 -70.6. -53.3
3&4. -67.4 -45.3
Also looks like I burned up heater ground resistors at LED indicator. Am also going to redo my 1 ohm resistors so the measurement is further from the socket.
-
Did the heater 100ohm resistors blow when something arc'd on power tube sockets when taking a measurement?
Couple things.
Take vdc readings to ground.
Those 1ohm resistors on pin8 might not be sufficient, although they may be fine. My 1ohm 1% resistors are physically larger. I had an issue with 1ohm resistors that smaller size.
Make sure screen voltage is making it to pin4.
-
Yes Calboy - heater resistors appear to have burned when the arc happened at the tube socket. Thanks for the thoughts though on the sensing Rs.
-
JoneG44 I would advise to get a pair of hook leads for your meter so you can connect to pins of tube sockets while powered down and caps drained so that the chance of shorting out a pair of adjacent pins of a tube is minimized. They look like this also notice that the metal end of the hook is encapsulated in non-conductive plastic. If you use regular meter leads to probe around you should have insulation sleeves on them so that only the very tip is exposed.. And you have to have a very steady hand so it's best not to use those on a live circuit fired up power tube.
-
Also many of us have more than one meter and more than one set of hooked leads so we can hook up and observe the voltages of pin 3 and pin 8 of a 6V6/6L6 at the same time.
-
Mresistor - thanks. I caught the arc as I was unhooking hooked leads - should have powered down first. Getting second set and secondary multimeter. Thanks for the suggestion.
Latole - got my copy of The Tube Amp Book by Aspen Pittman a couple weeks ago. Have found it a useful resource. Was a bit expensive forthe hard cover and I hope his loved ones get a reasonable slice of that from the publisher. At times the book reads a bit like an infomercial for GrooveTubes but that is a very minor distraction and fair given Mr. Pittman's interests and the other good content. The collection of layouts and schematics is impressive as are the practical tips and troubleshooting advice.
All - do others have suggestions for books that are useful on the topics of guitar amps and ac/dc electronics? (Am continuing up the learning curve with an eye toward doing a new build as my next project - but I'd like to understand better what and how I am building versus just sticking things on a PCB and in a chassis.)
-
With PT heater leads disconnected from LED, Rs, and heater wires. And with no tubes and no load.
One PT heater lead has 17.5 vac and the other is 23.5 vac. Should I be concerned they are different?
-
You're probably measuring the filament voltage incorrectly. The correct way is to measure between the two green wires, ie, one probe on a green wire and the other probe on the other green wire.
-
SEL49 - you are correct. And thank you. measured each one separately one probe on it and the other probe on ground. Thanks. Measured correctly got 6.9vac which makes sense.
Guess it makes sense to me now looking a schematic after your guidance - two legs from transformer - led, hum resistors, and heating filaments all in parallel at each socket.
-
Just as an update. Concluded that the range for the bias circuit was too cold (thanks to Calboy.) Better understand how the bias circuit works from Doug's content. (thanks Doug)
Concluded that the likely "fixed bias" spec was either -42vdc or -52vdc (have conflicting layouts/schematics.)
The voltage coming of the rectifier diode was -80vdc versus schematic of -63vdc.
Replaced a couple of resistors (trial and error) so that now my bias range at pin 5 is roughly -69vdc to -28vdc which brackets the spec.
Next will bias the amp.
-
OK - biased the amp using Robinette calc for Tube Dissipation Using Cathode Resistor Voltage Drop. (see attached pdf for final setting)
Plugged in guitar. Definite improvement from last time - order of magnitude better - but not there yet.
High notes sounded good. Mid-range there was some subtle percussive pop/whoosh as strings were fretted. Strong distortion on low E string. Voice recorder recording here - the percussive popping is not clear but the low note distortion is clear. (https://whyp.it/tracks/262069/fsfr010325)
Done for the day but have a couple of thoughts (redirect me as you feel fit)
1. Maybe low end distortion is blown speaker. (Will test this with another known good cabinet tomorrow)
2. Maybe I didn't bias correctly. (Maybe someone will review my bias calc to see if it looks right.)
3. Something else.
Any thoughts are welcome.
-
I'm sure you've done this already, but make sure every mechanical connection is sound. Go through and tighten ever nut and screw in the amp.
I've had tubes that weren't necessarily microphonic, but I could pull them out and give them a shake or tap them with my fingernail and hear a sizzle.
-
Hi,
I'm chiming in late, and my very casual observation suggests to me to mention you might wish to replace a couple of things (if you haven't already)
from your Pic it looks like you haven't replaced the Bias Cap (silver can mallory 50µF 70V) or the 3.3KΩ resistor from the pot to the bias supply board.
There's also a 68K from the bias pot to the coupling cap - I'd suggest that for stability and longevity, that you replace these with some vishay pro1/2 series resistors (https://www.vishay.com/docs/28729/pr010203.pdf) Not that there aren't other brands out there, but I have found these to be the most reliable metal films with a good tight tolerance.
Everyone has a preference - I am not saying my preference is better, only a suggestion to use a quality resistor in this application. (i.e. through all the areas in the amp where component drift is more than 5%( despite fender's 10% tolerance from back in the day, we have more advanced manufacturing processes these days, so take advantage of this).
Edit: For sh* and giggles, I've got a 73 quad reverb on my bench right now, and apart from the filter caps and valves, it's alarmingly unmolested.
Don't feel like you're trying to summit Everest - these things are easily tamed. Like eating an Elephant.... you can do it - One bite at a time.
I'll attach a couple pics of it for you to reaffirm your progress on your twin... Best of luck - You've been given excellent advice already by others.
-
With PT heater leads disconnected from LED, Rs, and heater wires. And with no tubes and no load.
One PT heater lead has 17.5 vac and the other is 23.5 vac. Should I be concerned they are different?
Like someone said, this is not the way to measure heater voltage. But this is an excellent way to confirm that you have elevated heaters. Note the 6V difference? I'd say there is (roughly) a nice 6VAC floating on 20.5 volts of DC. Very many amps have elevated heater voltages due to tube specs, arc avoiding etc.
- Mike, another noob
Edit: Schemas show only an artificial ground. So, are these really elevated, because "right" AC readings to ground should be -3.15 and +3.15? :w2:
-
74-335 - thanks for chiming in and the affirming the progress.
from your Pic it looks like you haven't replaced the Bias Cap (silver can mallory 50µF 70V) or the 3.3KΩ resistor from the pot to the bias supply board.
Good eyes - that pic is a bit old at this point. Have replaced both.
There's also a 68K from the bias pot to the coupling cap - I'd suggest that for stability and longevity, that you replace these with some vishay pro1/2 series resistors (https://www.vishay.com/docs/28729/pr010203.pdf) Not that there aren't other brands out there, but I have found these to be the most reliable metal films with a good tight tolerance.
Have not yet replaced this - seems worth doing while it is on the bench.
Thanks again.
-
stw307 - thanks for your post. Yeah I got sorted on the heater measurement and ended up at 6.9vdc at the sockets. Us n00bs gotta stick together so thanks again!
-
Update:
Following AlNewman's suggestion tightened up everything. A couple PT bolts needed snugging and a couple of pre-amp tube sockets were looser that the other sockets.
I changed out 7025 to 12AY7 in the first socket and cleaned that socket.) Swapped 12AT7 to another 12AT7 on sixth socket and cleaned that socket.
Fired it up again and here are the differences day to day:
Clean channel:
1. The tube changes and tightening things up definitely cleaned up a good amount of distortion on low end.
2. When compared to reverb channel - the clean channel has half or less of the volume of the reverb channel.
Reverb channel
1. Has the expected twin reverb loudness.
2. Reverb and tremolo work as advertised
3. Seems like lows are less distorted
Feels like I am getting closer. But always welcome input.
My next steps are:
1. chopstick the thang to see if there are any loose connections
2. clean remaining sockets and pots (did have some noise on one or two when turned.)
3. only have one preamp tube at this point that was not replaced - will likely see if swapping it with one of the new tubes tells me anything.
4. See if replacing any of the preamp metal tube covers affects sound.
-
Have you checked the board for conductivity yet? The super waxy mid 70s Silverface boards seem to be the most susceptible to becoming conductive. Check for DC on the fiber board leaking around the eyelets where there is B+.
-
Testing the Normal Channel with a 12AY7 may not give accurate results because the "Y" version makes much less gain than the "X" version or a 7025.
You really should put in a 12AX7 to verify that the distortion on the Normal Channel has been fixed.
-
ac427v - will put back the 12ax7
TitaniumValhalla - will test the board for conductivity.
Thanks you both.
-
Plodding along - preamp tube sockets and pots cleaned, tested board and there was no leakage on board around B+. Replaced the 12AY7 with the 7025. verified blue blob caps are not leaking dcv.
Did some more listening afterward.
- Reverb channel remains louder than Normal channel
- Reverb channel highs are nice and sparkely
- Normal channel highs are muddy
- Normal and Reverb channels have distortion on E and A string which goes away-ish when bass at 1. The distortion is more pronounced on the Normal channel.
Will continue tomorrow. Will drag another cabinet over to rule out speakers. Will study the schematic and chop stick the connections and think of some logical approach to problem isolation. :dontknow:
-
Carbon comp resistors have a tendency to drift. Old caps have a tendency to drift. Almost always they drift to higher values, meaning an out of spec resistor or cap will often tend towards a muddier tone.
Resistors can usually be tested in circuit, caps not so much.
For the rattle, have you tried new power tubes, or just pre amp tubes?
-
Thanks again AlNewman
For the rattle, have you tried new power tubes, or just pre amp tubes?
Yeah new power tubes in, Willabe suggested that and put in new matched set. But since I did have an arc at the tube socket while they were in, maybe I'll pull a pair at a time to test.
Almost always they drift to higher values, meaning an out of spec resistor or cap will often tend towards a muddier tone.
Resistors can usually be tested in circuit, caps not so much.
Have previously tested and replaced out of spec Rs - will double check.
On capacitors - understood. How far out of spec warrants replacement or is it just replace if not at spec?
-
Going through layout and schematic. I do not understand what this section is and what it is doing. Symbols are R and solder bridge.
Also it appears to be "flashing" under its covering - normal or not? (Not sure either a resistor or a solder bridge should flash)
-
That is your tremolo “roach” - a light bulb on one side and a light dependent resistor on the other, shrink wrapped together. The light should strobe with the tremolo and the resistance on the other side should rise and fall with it.
-
Also it appears to be "flashing" under its covering - normal or not?
Normal.
That is your tremolo “roach” - a light bulb on one side and a light dependent resistor on the other, shrink wrapped together. The light should strobe with the tremolo and the resistance on the other side should rise and fall with it.
-
Finally back at it pulling and checking caps. Quick question.
Have a .047 cap measuring .0514 which is close to 10% tolerance.
Do I only change the cap when clearly outside of tolerance or for ones that are close like this should I swap it them out?
-
.047 cap measuring .0514 which is close to 10% tolerance.
Same difference in a guitar amp.
Taken any set of caps from a factory run and test them and they'll be all over the place 'within tolerance'. So situation normal
Also, these size caps typically function in electrical circuits as filters of one sort or another that work in conjunction with a range of resistances (i.e. in R/C or C/R filters) and so the end aim - which is usually some frequency roll-off point or other, is usually achievable with no audible difference detectable to the human ear. Guitar amp technology is a 'ballpark' game. You'd have to change the caps values by a factor of 10 or more to notice changes