Unfortunately you have a bigger problem - the PCB is badly scorched, and burnt PCB tends to become conductive.
The blown part looks like a resistor to me. If it was a zener thered be nothing left except maybe a small hollow glass tube.
The blown part looks like a resistor to me. If it was a zener thered be nothing left except maybe a small hollow glass tube.
jaycee said:
Not only that,but it looks like the burnt part might have scorched through some of the traces! 🙄
In that third pic,it looks like it burnt through some traces,down into the fiberglass of the board.
a thorough cleaning,and some jumper wires might fix it,if the board hasn't become conductive..You could try to drill/cut out the charred sections,if it's not a multilayer board.
Well I seemed to have got a bit sidetracked today, but pulling the PCB was pretty easy as you can leave the FET's on the heatsinks... I was going to try the lightbulb test but I'd like to leave the caps discharged (for now) to be on the safe side. 🙂
What I I got lots of photo's from the top and bottom, I could then stitch them together and at least see the board layout from both sides then... I should be able to do this tomorrow, as I'm about to get some sleep now.
Here's some shots I got of the back, hopefully it might make things a little clearer... I should be able to get some better shots tomorrow with a SLR and create some board layout shots... It's still annoying having no schematic tho, and I have had no reply from Gemini so far...
Left (burnt part near tank cap):
Middle:
Right:
I got a few close ups of the burnt back traces if that's of any use? The tracks have lifted a bit, but I don't mind using jumper wires. They are thin tracks that look like they are intended for a small current... Key word being "intended" I think...
I would have thought so, but there only seems to be this part of the circuit once on the circuit... Perhaps it's something to do with power or bridging.... This amp has a bridge switch on the back, plus a separate speakon connector for the bridged output.... Kind of a long shot tho.bigwill said:
Well, looking at the back it makes things a little easier as you can see the traces.ACD said:
What I I got lots of photo's from the top and bottom, I could then stitch them together and at least see the board layout from both sides then... I should be able to do this tomorrow, as I'm about to get some sleep now.
I hope the burnt parts aren't conductive, would giving them a good clean help things? The tank cap is in the way, but I can remove it in order to get rid of all the black stuff...jaycee said:
It's 2 sided. Luckily all the traces on the back look ok, apart from the traces of the parts that got toasted...DigitalJunkie said:
Not only that,but it looks like the burnt part might have scorched through some of the traces! 🙄
In that third pic,it looks like it burnt through some traces,down into the fiberglass of the board.
You could try to drill/cut out the charred sections,if it's not a multilayer board.
Here's some shots I got of the back, hopefully it might make things a little clearer... I should be able to get some better shots tomorrow with a SLR and create some board layout shots... It's still annoying having no schematic tho, and I have had no reply from Gemini so far...
Left (burnt part near tank cap):
An externally hosted image should be here but it was not working when we last tested it.
Middle:
An externally hosted image should be here but it was not working when we last tested it.
Right:
An externally hosted image should be here but it was not working when we last tested it.
I got a few close ups of the burnt back traces if that's of any use? The tracks have lifted a bit, but I don't mind using jumper wires. They are thin tracks that look like they are intended for a small current... Key word being "intended" I think...
Good news!
Gemini pulled thru and sent a good quality schematic!
Here it is:
http://hh.exofire.net/diya/XPM_600_900_1200_sman.pdf
It looks like one of the parts was a zener and the other was a resistor...
Now, I managed to get a good quality shot of the board:
Now, I've also annotated it and added known components around the burnt parts... I'm pretty sure I've guessed the burnt parts correctly, but if someone could confirm this then this would be good.
I've uploaded tham at full resolution, hence thumbnails.
Now, it looks like the zener actually steps down from high voltage (90v?) to 15v for the op amps. Is it possible the zener got overloaded with too much current? Surely something like a 7815 would be better for this task?
as for the resistor burning out, I haven't a clue how that happened... maybe when the zener blew it passed on more than 15v to the rest of the circuit?
Anyway, shall I just replace these 2 parts and try the lightbulb test?
Also, am I right in assuming the rails of this amp are 90v as the tank caps are rated for 100v?
Gemini pulled thru and sent a good quality schematic!
Here it is:
http://hh.exofire.net/diya/XPM_600_900_1200_sman.pdf
It looks like one of the parts was a zener and the other was a resistor...
Now, I managed to get a good quality shot of the board:
An externally hosted image should be here but it was not working when we last tested it.
Now, I've also annotated it and added known components around the burnt parts... I'm pretty sure I've guessed the burnt parts correctly, but if someone could confirm this then this would be good.
An externally hosted image should be here but it was not working when we last tested it.
I've uploaded tham at full resolution, hence thumbnails.
Now, it looks like the zener actually steps down from high voltage (90v?) to 15v for the op amps. Is it possible the zener got overloaded with too much current? Surely something like a 7815 would be better for this task?
as for the resistor burning out, I haven't a clue how that happened... maybe when the zener blew it passed on more than 15v to the rest of the circuit?
Anyway, shall I just replace these 2 parts and try the lightbulb test?
Also, am I right in assuming the rails of this amp are 90v as the tank caps are rated for 100v?
Hi Mike, I am looking, are you sure it's R69 220ohm thats burnt, it's opposite number on other channel is R35. Just need to know, give me a bit of time and I will try and piece the evidence together.
I can actually see "R" silkscreen before it's burned away, so it's definitely a resistor. I'm 90% sure... I checked the components around it and it only makes sense if it is R69.Mooly said:
I've looked at the other side of the PCB, and you are right, R35 is in the same place as R69.
O.K. Mike, First off, I can't see yet where the current needed to zap that zener has come from. The 15V rails go off somewhere, the only reference I can see is a plug ref X12 at the top left above Opamp. Is this unused ? . I think the only way to takle this is slowly/logically and in stages. ZD9 and ZD 10 replace. Use 15Volt 1.3Watt Zeners (Maplin or C.P.C.) TLO72 replace. TLO82 is O.K. to use even a 4558 / 1458 etc will work to get you up and running. R69 is in a different part of circuit (This is what puzzles me), the only way this could go up is with a major failure in poweramp. Check initially with meter on low ohms, (use other channel as comparison) Q17 and Q18 the MJE340/350 pair, then Q 15, finally Q13/14. Check also D10 and D15 and R 62 & 55 both 47ohms and R77/78 100ohms. Untill you have better idea whats happened use the bulb if you need to power it up.
Yes. They've left room for pins on the board, but there aren't any. In fact the part of the board I've marked "+15v 0v -15v" is marked X12 on the other side... I'm not really sure where the voltage gets stepped down to +-15v really...Mooly said:
I already have a TLO72 handy, but no zeners, so I'll pick up some at maplins tomorrow.
How do i go about checking these with a multimeter?: Q17 and Q18 the MJE340/350 pair, then Q 15, finally Q13/14
Do I simply measure the resistance from the middle leg to the outside ones?
Mike, the zeners do the stepping down and are fed from the 4.7K 2watt resistors. To zap a diode like that needs a lot of energy, far more than the resistors can give. This is the problem, were you working on the amp at the time or was anything connected anywhere 😉 .Do a google search for "Testing transistors". I would suggest you replace all burnt bits inc R69 then measure the transistors on "diode test" on your meter. Compare them with the other channel. If one is short or very leaky it will show but this is not foolproof. Transistor failure can be subtle and takes experience and a feel for interpreting the results. Get a couple of cheap transistors from Maplin an N.P.N and P.N.P. and measure them. All silicon transistors (like this anyway) will test the same whether power or small signal. Using meter on diode test it will ( should ) show the volt drop across the junction e.g.For an NPN Red lead on Base,Black lead on emiter, meter will show about 630, this is actually showing 630mv dropped not ohms. Check also D13,14 and ZD12,13,15,16 and Q19/20. This is a potential low impedance( high current path) to R69.
Good luck Karl
Good luck Karl
Gotcha, the Q semi's are like 2 diodes in one package... I guess it would be best to measure them out of circuit, but I guess as I have a working side to compare to this should help... Right, off to maplins!
I recommend Rapid if you're going to buy parts. Maplin are now pretty useless with anything, more so in the stores.
http://www.rapidonline.com/
http://www.rapidonline.com/
I haven't used this amp in a while, it was mostly a backup. I recently lent out my main amp to someone, so I was quite surprised to find one of the channels wasn't working but I haven't used it in a while. It would sometimes crackle, so I haven't plugged up any speakers since. It wasn't until I looked under the cover and saw all the black stuff that I thought it would be best not to use it. Basically, it could be anything... I know in the manual it says not to set it to bridged when it's turn on, and I've never done this, but it's possible someone else has (I sometimes lend this amp out to a friend of mine...)Mooly said:
I know they're not the best, I normally use farnell, but they are handy when you only need a few small things and don't want to have to wait for the postie. 🙂jaycee said:
Right, I was looking to see which way round to put the zener, and I actually thing I may have incorrectly assumed it was a zener after all!
I was hunting around and found both ZD09 and ZD10 intact. This makes sense as I doubt the amp would work on either channel without the +-15v rails for the opamps.
Anyway, I've updated my picture with more parts, and I think instead of ZD10, R55 makes more sense:
An externally hosted image should be here but it was not working when we last tested it.
I've looked around the circuit, and cannot see R55 anywhere else, but it could be hiding... 😀 Also, it seems to be twinned with R46 is you see what I mean...
The good news, the circuits appear to be mirrored in the schematics, so I'll start testing the diodes in a minute...
Also, I have both of these resistors, my 220R one is 2W, and my 47R is only a really tiny 1/8W one... Is it worth just trying to solder them in place, and see what happens? I have a feeling the 1/8W won't be meaty enough....
I've also got a load of 7W 10R resistors, so i could series wire them for 50R just for testing...
Makes more sense now, just a power amp fault. Do you mean you can't find R55 on the PCB. No problem, measure from components you can find, C39 to FET.s for e.g. and see if it reads 47ohms . Value not critical for testing, you can even link it out. Won't do any harm, the 220 ohm would still go up if fault still present. It is used to isolate input stage supply to help improve power supply rejection. Check all transistors and diodes for shorts, am guessing the FET,s are O.K. as I can't see any fuses and if these were short the mains fuse would pop. Just give each one the once over with meter checking for any abnormally low reading.
Have fun
Have fun
I will. 😀
When I said I couldn't find R55 it is because I think it the burned out component by the TL072. If I managed to find R55 somewhere else, then it means my burned component is not R55... Process of elimination basically. 🙂
Ok, I'll check all the soldering with a torch + magnifying glass and clean off any flux as the soldering is a little messy in places. I'll use long jumper wires to I can wire in the resistors from the top, and put them somewhere where it will be safe if they decide to go bang, away from the PCB.
My next post won't be 'till Monday now, as I'm about to leave for London for the weekend, but I just want to say thanks for the help... This is only the 2nd amp I've attempted to fix, I must say it's going better now I know what parts have blown... 😎
When I said I couldn't find R55 it is because I think it the burned out component by the TL072. If I managed to find R55 somewhere else, then it means my burned component is not R55... Process of elimination basically. 🙂
Ok, I'll check all the soldering with a torch + magnifying glass and clean off any flux as the soldering is a little messy in places. I'll use long jumper wires to I can wire in the resistors from the top, and put them somewhere where it will be safe if they decide to go bang, away from the PCB.
My next post won't be 'till Monday now, as I'm about to leave for London for the weekend, but I just want to say thanks for the help... This is only the 2nd amp I've attempted to fix, I must say it's going better now I know what parts have blown... 😎
- Status
- Not open for further replies.