I have a Technics SU-G91 integrated amp that im doing a pre-out modification on. I removed the amp, then when i turned it on to test it, i noticed a smell and some smoke coming out of it. Long story short i discovered that 2 resistors look blown (R771 and R772). I had originally removed the amp and heatsink, not realizing there is 3 transistors connected to the heatsink. Would this have caused the resistors to blow? I put the heatsink and transistors back in, now my question is what will happen if i turn it on with blown resistors? Since these are in the amp section which is removed, will the rest of the unit function like it should? Will it cause a fire? What am i dealin with here? I have attached a photo. I have the schematic but the file is too big to upload. Thanks guys!
Attachments
Here's the schematic. https://www.hifiengine.com/manual_library/technics/su-g91.shtml
Those 10R 1/2W resistors are in series with each other, and R774 (also 10R), and the
regulator transistor Q701's collector, so check that and also the D705 Zener diode and R774.
My guess is that the resistors were just discolored from years of running hot, probably also R774.
If Q701's emitter voltage is 14.7VDC when put back together, this circuit is probably working ok.
If so, replace the three 10Rs with new ones if you want.
Last edited:
R771, 772 and 774 are in series and form a resistive dropper for the +15V regulator, and sure enough, R774 also looks toasty.
You've probably got an excessive current draw problem on +15V. Make sure Q701 is not shorted (bad news if so), and check for any direct shorts to ground on that line. Unfortunately this voltage is quite widely used throughout the amp, and the problem could be anywhere. I am spotting 16 V rated 10µ caps on the supplies of the electronic volume IC530 (TC9212P), that seems a bit tight to me. If C534 had decided not to support 14.7 V any more in old age, that could explain the problem. It would also act as overvoltage protection in case Q701 ever shorted out, possibly protecting the electronics.
Some poking around with the multimeter is in order.
You've probably got an excessive current draw problem on +15V. Make sure Q701 is not shorted (bad news if so), and check for any direct shorts to ground on that line. Unfortunately this voltage is quite widely used throughout the amp, and the problem could be anywhere. I am spotting 16 V rated 10µ caps on the supplies of the electronic volume IC530 (TC9212P), that seems a bit tight to me. If C534 had decided not to support 14.7 V any more in old age, that could explain the problem. It would also act as overvoltage protection in case Q701 ever shorted out, possibly protecting the electronics.
Some poking around with the multimeter is in order.
I dont really have a way to check them, I just want to know what will/could happen if I power the unit on?
sgrossklass - if the transistor Q701 is bad, cant I just replace it? What would be a good replacement for C534, and any other caps that might not be up to par? How would I go about testing the resistors and transistors with the multimeter? What do I set the meter at?
well I turned it on and nothing happened. Just didnt power on at all. No spark, no display, nothing. Should I replace those bad resistors?
These probably didn't burn up for no reason though, so finding the fault first would be a good idea. BTW, the voltage readings on the schematic look suspect to me - why would they use three half watt, 10 ohm resistors in series to drop a mere 2 V, which translates to 133 mW? I would expect something closer to 500-750 mW, which would be a 3.9-4.7 V drop or 130-150 mA.
You probably have no +15V at all right now, and as it happens the +6.2 V that the digital section runs on is derived from this voltage (via Q705 / D706), hence no lights. So your problem could be there as well - is R708 still looking good?
So you have a multimeter but literally no clue how to test a resistor even? Go watch a multimeter tutorial or two first.
(Spoiler: You may find the resistance range is involved.)
Crude in-circuit testing guidelines:
1. A resistor can never measure higher than its actual value. (So if your measurement is well above what it's supposed to be, it's probably toast.)
2. Transistors can be inspected for shorts using resistance or continuity testing.
3. A transistor contains two pn junctions, which can be checked with the diode test. Japanese transistors tend to be ECB (/BCE) pinout, labeling assumes a 2S prefix, 2SA and 2SB are pnp, 2SC and 2SD are npn - if in doubt consult the datasheet. So you should be seeing two typical silicon diodes either facing towards the base (pnp) or away from it (npn).
Q701 is a 2SD1761, Rohm, Vceo = 60 V, Ic = 3 A, fT = 8 MHz, Pc = 30 W, TO-220F (fully insulated plastic package, not a regular TO-220 where the tab commects to the collector), Cob = 90 pF. Crosses to 2SC3690, 2SC3746, 2SC3851 (Mouser has these), 2SD1586. A 60 V type that has 62 V going into it, maybe up to 64 V on powerup (when +15V is still at zero)... seems a bit tight to me.
If you can get nothing else, a TIP31A (but preferably B or C grade) would also kinda get the job done, but you would need an insulating (e.g. mica) washer or silpad for TO-220.
You should also have some plain white silicone-based heatsink compound (thermal grease) at hand, as it needs to be reapplied whenever the heatsink is remounted. (Hint: Alcohol is well suited for removing old thermal grease.)
You probably have no +15V at all right now, and as it happens the +6.2 V that the digital section runs on is derived from this voltage (via Q705 / D706), hence no lights. So your problem could be there as well - is R708 still looking good?
So you have a multimeter but literally no clue how to test a resistor even? Go watch a multimeter tutorial or two first.
(Spoiler: You may find the resistance range is involved.)
Crude in-circuit testing guidelines:
1. A resistor can never measure higher than its actual value. (So if your measurement is well above what it's supposed to be, it's probably toast.)
2. Transistors can be inspected for shorts using resistance or continuity testing.
3. A transistor contains two pn junctions, which can be checked with the diode test. Japanese transistors tend to be ECB (/BCE) pinout, labeling assumes a 2S prefix, 2SA and 2SB are pnp, 2SC and 2SD are npn - if in doubt consult the datasheet. So you should be seeing two typical silicon diodes either facing towards the base (pnp) or away from it (npn).
Q701 is a 2SD1761, Rohm, Vceo = 60 V, Ic = 3 A, fT = 8 MHz, Pc = 30 W, TO-220F (fully insulated plastic package, not a regular TO-220 where the tab commects to the collector), Cob = 90 pF. Crosses to 2SC3690, 2SC3746, 2SC3851 (Mouser has these), 2SD1586. A 60 V type that has 62 V going into it, maybe up to 64 V on powerup (when +15V is still at zero)... seems a bit tight to me.
If you can get nothing else, a TIP31A (but preferably B or C grade) would also kinda get the job done, but you would need an insulating (e.g. mica) washer or silpad for TO-220.
You should also have some plain white silicone-based heatsink compound (thermal grease) at hand, as it needs to be reapplied whenever the heatsink is remounted. (Hint: Alcohol is well suited for removing old thermal grease.)
I think the reason the resistors blew is after I removed the amplifier IC, I removed the whole heatsink with it, transistors and all (not realizing they needed to be there) and powered it on...thats when the resistors fried. Would that cause that?
Plug-in transistors, interesting design. If Q701, 708, 710 are just left out, absolutely nothing should happen. R773 will get a bit warmer than usual, but that's about it. That's assuming the 10n cap, C703, has a high enough voltage rating and does not suddenly decide to break down when subjected to 4 V more than usual.
It seems worth checking that neither C703 nor zener D705 are shorted. A replacement for C703 better be rated at about 100 V, which does not seem to be asking too much for a 10n X7R ceramic.
Other than that, I fail to see how the resistors would have let the magic smoke escape unless you had somehow created a short. Really baffling.
It seems worth checking that neither C703 nor zener D705 are shorted. A replacement for C703 better be rated at about 100 V, which does not seem to be asking too much for a 10n X7R ceramic.
Other than that, I fail to see how the resistors would have let the magic smoke escape unless you had somehow created a short. Really baffling.
Get six new 10 ohm, 1/2 watt resistors, put 3 in (not too close to the board if you can help it), and have an eye on them when powering up. (Got clip leads? Having a look at voltage drop over all three could be handy. As stated, expect 4-5 V.) Sounds like you might get lucky and there'll be lights and no smoke.
Otherwise you'll be no worse off than before.
Otherwise you'll be no worse off than before.
- Status
- Not open for further replies.