Hoffman Amplifiers Tube Amplifier Forum
Other Stuff => Solid State => Topic started by: G._Hoffman on May 22, 2013, 05:03:04 pm
-
So, we have a Fishman Loudbox here in the shop we use for testing guitars and the like, and over the years the output has gradually fallen. A lot. Upon visual inspection, the only thing I found wrong was the three caps in the picture. C201 and C202 are stacked in parallel to the secondary of the power transformer, and C203 is bypassing the main reservoir cap on the +33V positive rail. So, those are obviously getting replaced, but the voltage on the rails is about right (+36V, with nothing going through the system and the mute switch on). That reservoir cap looks fine, but the esr appears to be drastically different from the negative rail (I don't have an ESR meter, but when measuring the resistance they wildly bounce around differently, so there is something different between them, which is enough to tell me something is wrong with one of them, and I'm going to pick the one that caused the bypass cap to scorch the PCB. Though the problem could be the bypass cap). At any rate, I'm replacing the electrolytic caps and the ceramic caps with over rated caps (100V instead of 50V), and hopefully that is the problem.
If the caps AREN'T the problem, any suggestions on where else to look? It is a active tri-amped system, with the highs and mids being class A opamps, and the bass using cascading BJT's in class AB (three BJT's per side). I'm a Fishman dealer, so I really can't share the schematic they sent me (though there is one out there on the Internet that someone else posted), but I'm assuming the problem has to be in the power amp or the power supply. And since it appears to be all three amps (high, mid, and low), I'm kind of assuming the power supply. What do you think?
Gabriel
-
I'm a Fishman dealer, so I really can't share the schematic they sent me
Nice photo.
-
SOMEthing sure happened to those little caps, but I can't see how that would hurt power.
As a "dealer", isn't there some factory support?
-
SOMEthing sure happened to those little caps, but I can't see how that would hurt power.
As a "dealer", isn't there some factory support?
It's about ten years old, so well out of warranty.
Gabriel
-
I'm a Fishman dealer, so I really can't share the schematic they sent me
Nice photo.
Yeah, I was pretty surprised for an iPhone shot.
Gabriel
-
Seriously, that's good detail for a cell phone. Point and shoot cameras are probably on the way out.
-
Seriously, that's good detail for a cell phone. Point and shoot cameras are probably on the way out.
Nah. The optics on the camera still don't look great. They are fine for relaying information, but that's about it. If you want really good pictures, you need to have decent optics, not just a pinhole camera.
The caps didn't fix it. So now, I have to take it home over labor day weekend and run through the whole thing with a meter and a scope.
Bother.
Well, maybe it's a regulator for the preamp power supply.
Gabriel
-
Checking the power supply is always a good place to start. Good luck.
-
OK, so I've finally been able to take some time to do more on this thing. The voltages all look fine. There are two bi-polar supply voltages; ±33V and ±14V. I've checked them both through-out the amp, and the 33V is right around 36V, both plus and minus; and the 14V is at 14.44 both positive and negative. My next step is to put a signal at the input, and run through with an O-scope probe and see which semiconductor isn't doing it's job, right? I was really hoping I'd find the supply voltages were off somewhere; it'd have made life so much easier, but the power supply appears to be in good shape.
Gabriel
-
sluckey: Yeah, I was pretty surprised for an iPhone shot.
iphone camera (type) elements are used as an array component in the construction of high resolution drone imaging systems. An array of several hundred create a virtual imaging component.
Saw that on an OPB segment.
Silverfox.
-
OK, I'm working on this thing again (it hasn't been a high priority, obviously!) For a week or so, I've got another one in my lab which works properly, though, so right now I've can compare the two. So far, all I've done is to compare all the DC voltages at all the semi-conductors (I'm going to fire up the O-Scope this weekend, and put signal through them both).
There are only two areas with significantly different voltages. One is part of the mute switch circuit (it mutes the amps for the mids and highs speakers), but as it's regulating the voltage to the mid and high chip amps, and the voltage going to the chips is the same in both amps, and it is using bi-polar transistors, I think that is within the normal tolerance range.
The one that I thing IS a concern is the jFET at the input. On the amp which is working, the gate sits at 0V, but on the amp which doesn't work it is at -0.7V. I can't show the whole schematic, since I was given it by a company I have a relationship with, BUT I've drawn the part of the amp I'm concerned about. I'm thinking that whatever blew up the caps in the first post must have damaged the jFET as well. Makes sense? (The voltage on top is from the good amp, and the one in parenthesis is the damaged amp.)
-
OK, a bit of poking around with a 1K sine wave and a O-Scope, and I'm pretty sure it's not the jFET. The jFET on both amps is a pretty straight forward unity gain buffer. It then goes to a quad op amp for the TMB and volume controls, and some where in THERE it all goes to shit. The first stage is fine, but by the second stage (a, I believe it is called, "balanced T" treble/bass network - pretty sure I saw it in Merlin's preamp book!) is all wrong. I've measured all the resistors, and they are either in spec, or both amps measure the same and I can see the other resistors that are in parallel to drop them below spec - i.e., the resistors are all right. Do I bother to check all the caps for capacitance, or do I just assume the op amp got blasted when the spike blasted the power supply caps? It does seem odd that the first stage of the op amp (a 0.1-100x gain stage, with the volume control in the feedback loop) is fine, but one or more of the other 3 in the chip are fubar (though, they aren't a dead short, since they aren't drawing down the power supply rails!)
Gabriel