|ilyaa||9/14/2017 11:34 AM|
|fender bassbreaker 45 schematic|
client has already sent this one in once for warranty repair - he'd rather not again.
called fender and they are mum on documentation - just want you to send it back.
im 95% certain its the fet-controlled variable voltage - the screens are at 0V and grids at -3V. the fet is cold when the amp is running and there is a 7912 near the bias circuit that is reading all kinds of crazy things but not -12V! without the schematic though i dont wanna poke around blindly -
anyone have any thoughts? or have worked on these? or have a schematic?!
|nickb||9/14/2017 2:03 PM|
|If it's anything like the bassbreaker 18/30 the FET does indeed control the screen grid voltage. It's associated with a BUJ103 and 2N4401 that switch the voltage lower as required. What are the voltages on FET?|
|ilyaa||9/15/2017 12:19 AM|
|yeah its a 7N60NZ fet -|
about 60VDC on the gate and 0.04VDC on both the drain and the source.
and yes, there are two BUJ103s nearby it - i havent opened things up completely so i can only measure their collectors, which are both at about 60Vs. right nearby the fet is big 390 ohm 10 watt resistor - its got what i imagine is the voltage that wants to go to the screens (about 400VDC) on one side and then 0.04VDC on the other - continuity test shows that that 0.04VDC end is going to the drain of the fet.
my hunch is that the fet is dead - in an ideal world, i think the client wants me to do what i can to maybe bypass the whole variable voltage circuit so he can just have this amp without the failure-prone bells and whistles of the variable output - thoughts on that?
|nickb||9/15/2017 12:35 AM|
|The resistor is open. It has 400V across it. |
Also 60V across the gate to source is too much as the limit for the FET is 30V so it is likely dead too. The big question is why did this happen? Replace the parts and then try adding one power tube as a time in case one of them is bad.
|ilyaa||9/18/2017 10:54 AM|
|right on - that 390 was open|
replaced it - if hooked as normal in the circuit, it's clear that the FET is shorted because that resistor gets really hot and drops about 400V.
like i mentioned, this part of this amp has broken before - would there be any risk in bypassing the variable voltage regulator altogether? as in, removing the bad FET and just wiring the 390 guy directly to the screens? and tapping the -40V from where it's still healthy in the bias network? as is, the control grids are at -3V - pin 1 of the bias trimpot has -40V, though - if i just short that pin to the grids am i risking anything?
|nickb||9/18/2017 11:24 AM|
|Is the FET mounted on a heatsink with insulating washers? If so there might be short from the drain to the heatsink and then ground. Use your ohmmeter and measure the resistance from to drain to ground. If the FET were shorted internally I don't think it could cause a dead short to ground like that. Did you remove all the tubes as instructed to remove the possibility of a bad tube ( screen grid to cathode short)? Measure the resistance from source to ground|
Until you figure out what is causing the short you should not power it up or you'll blow that resistor again.
|J M Fahey||9/18/2017 11:49 AM|
|With caution, it sounds reasonable.|
I hate unknown parallel paths messing with my expected results, so I suggest you :
1) remove said 390 ohms resistor
2) remove FET
3) lift PCB mounted presumably 220k (or whetever they used) power tube grid reference resistor ends leading into the current complex tracking bias system, leave them "in the air" for now.
Follow tracks to make certain they connect nowhere else, grids *should* be floating when lifting grid resistors "far end" but recheck just in case.
4) Follow power tube screen tracks leaving power tube screen pins.
They "should" go through screen resistors into variable screen supply, again lift "far ends" and leave them "in the air" for now.
What we are doing is isolating/floating away power tube pins from "evil" regulators
Hey, this sounds like standard Republican propaganda
5) join screen resistor free ends, if too separateduse a piece of stiff wire, and connect them to a new "screen supply"
We *might* mod the current RC filter/dropper chain but since I have never seen the schematic, safest is to build a new dedicated RC node.
No big deal: mount somewhere, even by silicone gluing to chassis or on a terminal strip a 22uF x 400/450V capacitor for screen filtering duty, feed it with 390 to 1000 ohms 5 or 10W resistor coming from +V and use cap as screen voltage supply.
We are back 60 years using a fixed safe RC screen grid supply.
6) join grid resistor free ends and connectbthem to a suitable bias supply.
They must generate some -C voltage to feed the tracking regulator, so tap some unvariable source and add , say, a fixed resistor, a trimmer and a cap (22 x 100V?) to be able to set fixed bias at will.
Now we have *fully* travelled back to 1963, when Jim Marshall built a Tweed Bassman clone with stuff he had around in England
|nickb||9/18/2017 12:07 PM|
|Juan - The FET is used to vary the screen voltage and so provide the variable power output (1-45W), I suspect. Bad things might happen if the FET is bypassed as the bias voltage may to lowered too leading to baked tubes. It's a risky proposition IHMO. Better to figure out the actual problem.|
|ilyaa||9/18/2017 1:03 PM|
|good call again nick!|
measured 17 ohms from drain to ground - took the FET out and saw that a tiny bit of the heatsink had chipped/been melted away right around the screw-mount hole - put in a new, intact mica insulating pad instead and measuring open from drain to ground! now with everything hooked up back to normal (and a new 390 resistor in place) the voltages look more reasonable - in line with a dead FET, though.
i'm getting ~400VDC at the end of the 390 ohm. FET is reading 61V at the gate, 400V at the drain, and 59V at the source.
in your guys' opinion: worth just fixing the amp, repairing it to stock circuit and leaving it at that? im just concerned that if this has already happened twice it's bound to happen again....maybe the heatsink was the problem and fender warranty repair didnt catch it first time around??
|nickb||9/18/2017 1:36 PM|
|It was repaired once before but what for? It sounds like the the root cause here was mechanical i.e. swarf of some kind that ultimately led to a short. If cleaned up properly I don't see why that should happen again. |
The gate and source voltages are good now. I'm guessing it's set to low power. Turn up the power control and the source voltage should rise.
|J M Fahey||9/18/2017 3:54 PM|
I am not helping "repair" the variable power circuit but *remove* it in full, and wire screens and grids like in a conventional way.
So no risk involved, quite the contrary
Lacking schematics and with owner fed up with varipower and anything related, decision was alredy made (not by me) to remove it completely, my instructions aim exactly to that and nothing else.
No reinventing the wheel, quite the contrary, going back to basics.
I am quite certain Ilyaa can do that on his own but it never hurts to say te same, maybe in different words.
Only extra stuff I added, just in case, was to remove power dissipating elements (initially resistor and Fet) because by definition they must be connected to main HV supply and able to pull current from it, we donīt want that.
Canīt go further because of sheer lack of schematics; IF they were available might instead suggest some way to repair the original circuit, but meanwhile, my suggstion makes that amp usable and giggable ... which is currently impossible.
|nickb||9/19/2017 12:46 AM|
|Enzo||9/19/2017 1:29 AM|
|First time in the original problem was never really found, in all likelihood. I agree with nick, once it is fixed it will likely work.|
maybe the heatsink was the problem and fender warranty repair didnt catch it first time around??
In a "normal" tube amp, about the worst a bad tube does is burn out a screen resistor. In this apparently more is at risk, so if this were mine, I would be trying the power tubes in some other conventional amp before even thinking about trying them in this thing.
|J M Fahey||9/19/2017 5:30 AM|
|Also agree that if the burr was the problem, the original circuit should be given a new chance, it might work properly and for a long time.|
If not, and owner needs to play *this* weekend , the Plan B is already on the table
FWIW lots of times customers have been given a "5 minute kludge" late Friday or Saturday (I live just around a very popular Blues Club where everybody, including the Rolling Stones came to play) with instructions: "bring it back on Monday and leave it here full week, for a proper repair" ..... and they never ever bring it back.
When I meet them later and ask "did you have it repaired somewhere else?" typical answer is: "donīt worry Juan, I LIKE the way it works now"
Once I replaced a dead TL072 with an LM358, a lowly Op Amp which is better for DC, instrumentation, interface duty, but not for Audio, "it has no highs", because I had just run out of 072 (and of course job was urgent, it was Saturday night, and a Blues Harmonica player ***needed*** it to play 1 hour later)
I met him Months later: "I love the warm (no highs) / organic sound I have now".
|ilyaa||9/19/2017 12:31 PM|
|with the heatsink repaired and the resistor replaced the amp seems to be working fine, 100%.|
think id rather leave the circuit stock - even though I like the idea of going 'back to basics' on this one, in the interest of saving me time and the client money I think I will let it run as it was -
making full power and sounding pretty good! maybe once these schematics are made available we can take another look....
|g1||9/19/2017 12:57 PM|
|I'm not clear on whether the Fet also ended up being bad or not and if it was replaced?|
|ilyaa||9/19/2017 2:52 PM|
|FET was still good! left it in there nice and cozy on a new heatsink bed|
|lowell||7/26/2018 9:34 AM|
|Would you please post screen voltage after repair woth power knob at 45W? Or do you remember? I have one in that has plates at 437v and screens at 304v at 45W. And it was brought in for flabby low end. I'm suspecting Vscreen should be a volt or 2 below plates at 45W per a typical design.|
Without a schematic this is a tough one. Any thoughts on soldering the Gate to the Drain to eliminate the Output control?