It is not easy to follow on the diagrams on a screen.
I was wrong saying this, looking closer and there is an SMPS so this would not work.
I was wrong saying this, looking closer and there is an SMPS so this would not work.
I guess I just have to start looking anywhere beyond where it fried itself in that solder trace.
One useful trick to know is that if you link the vbe multiplier out to force zero bias and provided the output transistors are good then should not see the output transistors draw current no matter what any offset fault voltage may be.
In an otherwise working amp it will all work and play music normally in that state, just with a little crossover distortion that may be so low you can't even tell.
In an otherwise working amp it will all work and play music normally in that state, just with a little crossover distortion that may be so low you can't even tell.
What I don't understand is why there isn't a single short between components, all the readings are within tolerance (checked other good channels for this)...
Obviously there is a short, otherwise it wouldn't blow a fuse every time I join those weak solder points together.
Obviously there is a short, otherwise it wouldn't blow a fuse every time I join those weak solder points together.
I guess I can't edit my post after too much time has elapsed.
I think I received bad transistors. Every time I plug it in, it shorts fuses. Been checking for the last 36 hours for a short, out of bounds resistance, diodes, zener diodes, held my flashlight behind the board to follow every trace to pad.
Tomorrow, I'll take the resistors out of it tomorrow to get them out of circuit and test them.
Is there some sort of handy chart that Sanken (SK) has for values on their transistors? They aren't shorting on the board, maybe low resistance or something. What the ___..
I think I received bad transistors. Every time I plug it in, it shorts fuses. Been checking for the last 36 hours for a short, out of bounds resistance, diodes, zener diodes, held my flashlight behind the board to follow every trace to pad.
Tomorrow, I'll take the resistors out of it tomorrow to get them out of circuit and test them.
Is there some sort of handy chart that Sanken (SK) has for values on their transistors? They aren't shorting on the board, maybe low resistance or something. What the ___..
Last edited:
Have you tried linking out the vbe multiplier as I suggested earlier?
Try running it up with the output transistors and see if that stops it blowing fuses. That is the simplest check.
The Sankens can only have basic tests done on them. Failure is 98% with them failing short circuit collector to emitter. Reading from base to emitter will probably show an indeterminate value higher than a non Darlington type transistor. Although there are two base/emitter drops to rad across the internal resistor shunting B/E of the actual output part will cause the indeterminate reading. The value of the resistor and the meter test current will all alter the reading.
Try running it up with the output transistors and see if that stops it blowing fuses. That is the simplest check.
The Sankens can only have basic tests done on them. Failure is 98% with them failing short circuit collector to emitter. Reading from base to emitter will probably show an indeterminate value higher than a non Darlington type transistor. Although there are two base/emitter drops to rad across the internal resistor shunting B/E of the actual output part will cause the indeterminate reading. The value of the resistor and the meter test current will all alter the reading.
You mean, without the output transistors? Yes, I'll be doing that tomorrow.
Can I link that cap and have no output transistor on there?
I'd like to install this sometime this year.
Sanken's failure rate is 98%! How do they stay in business?
I presume there are alternatives?
Can I link that cap and have no output transistor on there?
I'd like to install this sometime this year.
Sanken's failure rate is 98%! How do they stay in business?
I presume there are alternatives?
If you meant what I thought you meant (running without Transistors). Should I be taking all transistors or, the 2650's? or the 1647's? or the Bias?
Linking the caps will only help with bias but not the shorting, correct?
They are both "output transistors" correct?
I swear, I'm learning a lot. This thread is definitely one of my favorites, lots of good information here.
Basically, I would like to use it. 😎
Last edited:
Only the output transistors need be removed at this point. It will prove if that is where the current is going.
Linking the cap across the vbe multiplier (the bias transistor) will 100% stop good output transistors from conducting together and drawing current. It would not stop either transistor forcing current into a short on the output line.
Linking the cap across the vbe multiplier (the bias transistor) will 100% stop good output transistors from conducting together and drawing current. It would not stop either transistor forcing current into a short on the output line.
That link is at capacitor c7125 as stated on the first page of this thread, correct? I might put the old ones back on while waiting for the RMA and use the bias link.
Also, I found a bad SMT thermistor by the 2560s... 560MOhm while it's supposed to be 470Ohm. Another reason for the heat.
Also, I found a bad SMT thermistor by the 2560s... 560MOhm while it's supposed to be 470Ohm. Another reason for the heat.
From what I remember, the 1647's are the output transistors. I believe they have the lead directly back to the connector. Not hard to test this. Emitter out to pin 5.
I've been working on computers a long time, disassembling and reassembly, I've never seen a plastic insert to go between the heat sync and transistor.
Are they necessary? Is that why it's shorting? I noticed that the emitter out has continuity with the back pad.
I've been working on computers a long time, disassembling and reassembly, I've never seen a plastic insert to go between the heat sync and transistor.
Are they necessary? Is that why it's shorting? I noticed that the emitter out has continuity with the back pad.
The 2SD2560 and 2SB1647 are the output transistors.The metal back is also the collector and this has to be insulated via a suitable heat transmitting washer and also (if needed, some transistors are plastic) an insulating bush to prevent the bolt from touching the case.
When the transistor is fixed correctly there should be no continuity between the centre lead (the metal tab) and the heatsink.
If the emitter reads short to the transistor metal tab then the transistor is shorted.
When the transistor is fixed correctly there should be no continuity between the centre lead (the metal tab) and the heatsink.
If the emitter reads short to the transistor metal tab then the transistor is shorted.
That was the most 'amateur' thing I've done on this project and will take the day to attone for my grave mistake (while I wait another day for Amazon).
I really can't believe I did that. Wow.
I really can't believe I did that. Wow.
I just double checked the thermistor (R7238).
581 ohms, cold.
Spec: 470ohm. The rest of the channels also read within this range.
This is the same channel that read 90°C+ .. should I definitely be changing this?
Also, if it's PTC, shouldn't the voltage reduce (temperature decrease/resistance go up) if the thermistor gets hotter? That would say to me that it's not going it's job, right?
581 ohms, cold.
Spec: 470ohm. The rest of the channels also read within this range.
This is the same channel that read 90°C+ .. should I definitely be changing this?
Also, if it's PTC, shouldn't the voltage reduce (temperature decrease/resistance go up) if the thermistor gets hotter? That would say to me that it's not going it's job, right?
Last edited:
PTC means the resistance rises as the temperature rises. 470 ohms is shown along with 115C so I'm assuming it menas it would read around that value at that temperature.
The thermistor won't causing the problem with the amp but if you suspect a problem then remove them all and measure them all when cold. You can't measure them accurately in circuit.
The thermistor won't causing the problem with the amp but if you suspect a problem then remove them all and measure them all when cold. You can't measure them accurately in circuit.
PNP Ohm Right surround, + to emitter, - to collector, result: OL.
Good readings (Not OL) from the rest of the rest of the PNP and NPN transistors (IN circuit)
I am almost there! Amp clicked to full on without the 5ch board connected.. I'm guessing there is some transistor damage from the shorting.
Good readings (Not OL) from the rest of the rest of the PNP and NPN transistors (IN circuit)
I am almost there! Amp clicked to full on without the 5ch board connected.. I'm guessing there is some transistor damage from the shorting.
It seems that all the transistors that were on that board that got shorted are damaged. I'm getting sparks left and right, each time, another fuse.
Where I saw sparks, those resistors are "dead" (much different from spec now)
I'm getting a little pissed about this board.
What is the chance that I did damage all those transistors on that short when I didn't have the mica inserts in to protect from shorting?
I'm going to take all the transistors off the board and see what happens if I plug it in. If there is another spark, I'm going to have to look at getting a new board.
Where I saw sparks, those resistors are "dead" (much different from spec now)
I'm getting a little pissed about this board.
What is the chance that I did damage all those transistors on that short when I didn't have the mica inserts in to protect from shorting?
I'm going to take all the transistors off the board and see what happens if I plug it in. If there is another spark, I'm going to have to look at getting a new board.
Last edited:
After removing all the transistors, there were resistors that sparked. Removed them, put the board in no more sparks or blown fuses.
Every one of the 2SD2560 and 2SB1647 transistors are bad. My DMM and newly bought transistor tester have proved an hFE around 30-500 and it's supposed to be 5000
The transistors KTC3964 that give me a reading are 1.5k, spec sheet is hFE=500, the other ones either say bad or register it as a diode
Any other parts I should add to my order?
Every one of the 2SD2560 and 2SB1647 transistors are bad. My DMM and newly bought transistor tester have proved an hFE around 30-500 and it's supposed to be 5000
The transistors KTC3964 that give me a reading are 1.5k, spec sheet is hFE=500, the other ones either say bad or register it as a diode
Any other parts I should add to my order?