Hi,
I finished first channel. When working on second channel, I messed up polarity, I accidentally connected VCC+ from PSU to VCC- AMP PCB and VCC- from PSU to VCC+ AMP PCB. C2, C4, C5 and C6 are gone. From C5 there was also some smoke going outside. It took about 10 seconds until I disconnected PSU. Is there any chances that I could destroy all transistors? J112, 557,547 and also TOSHIBA drivers were installed
Hi Amadeus00
I am sorry for you
I suppose you mean C3 and not C2.
it is difficult to say if you damage transistors. I suggest that you measure them for possible short between their pins first. Bjt in diode mode of your multimeter should measure about 0.6 to 0.7VDC between b-c and b-e with polarity based on npn and pnp types. Jfet and mosfet should measure some resistance between pins.
Fab
Any sot23 to dip adapter will do the job.
Google it or look at ebay. 50 pieces will cost about 2euro.
Hi tmas
That will not be enough. I recommend that you perform also section 8.11.2 where transconductance at various current (at least 20ma and 50ma) is checked.
Fab
I easily could have prevented this trouble... first real issue in 6 years...
However I tested a speaker and swapped the speaker wires on a live amp.
Normally no issue but the speaker + wire slipped from my hand and touched the shell of my (grounded) soldering station. My right channel went silent and I had a earth leakage trip.
Disassembled the right channel and measured
- power supply ok
- TP11-12 and 9-10 are 0mV
- TP7 psu + is 100mV (instead of 2,5V)
- TP8 psu - is 28mV
- Bias is zero
- I saw a white stripe from speaker + to the R27
Replace small transistors Q1-4 and J112’s first? Hope that’s all...
Could appreciate some help on repairing
Regards, Albert
However I tested a speaker and swapped the speaker wires on a live amp.
Normally no issue but the speaker + wire slipped from my hand and touched the shell of my (grounded) soldering station. My right channel went silent and I had a earth leakage trip.
Disassembled the right channel and measured
- power supply ok
- TP11-12 and 9-10 are 0mV
- TP7 psu + is 100mV (instead of 2,5V)
- TP8 psu - is 28mV
- Bias is zero
- I saw a white stripe from speaker + to the R27
Replace small transistors Q1-4 and J112’s first? Hope that’s all...
Could appreciate some help on repairing
Regards, Albert
Hi fab,
I have been studying section and have a few questions:
1) Is the 15V V5 on the print a -15V supply? It is not labeled -15 on the schematic.
2) On Table 8.11.2.1. Are the various VGSN and VGSP measurements always taken from Points A,B to ground in the test circuit? Or the voltage at the actual MOSFET gate and source pins on the device itself?
3) What does the (10 Ohms) in the measurement table mean? I do not see a 10 Ohm resistor anywhere in the test circuit. Not sure what is being referred to here?
4) When "Matching" the devices. What measurement specifically is being targeted for matching? What tolerance is considered acceptable for the devices to be "Matched"
5) When "Matching" am I trying to match the +/- device in a pair or the pairs to each other in the R/L channels?
Hi Asanden
Sorry to hear that. Hot swapping is always very risky and Murphy’s law always happen...
If you had high voltage injected into the circuit then it is probably the worst thing that can happen.....then all small transistors are possibly gone and if not stressed too much anyway...so change them all.
While you have theses transistors removed you can measure the resistors to ensure they are not damaged too. Some of them are in parallel so you will measure the equivalent resistor value....
For the mosfet drivers and output Mosfet you can check if there are any short between pins. Diodes D1 to D4 may also be failed short and if so could have saved your output mosfet..
And R27 is gone for sure. It should be the first thing to check because it can be only that too
C3 and c4 are rated only 16VDC so they may be damaged too.
Good luck
Fab
Last edited:
Hi tmas
1) V5 Polarity is properly shown so it is a negative polarity of course in relation to ground.
2) Points A,B to ground, voltage across gate resistor is negligible.
3) 10 ohms could read 22 ohms (mistake of the manual) but 10 ohms can be used too.
4) it is not intended for matching but to see if the device has a very good chance to be genuine instead of fake. The parameter checked is transconductance ( section 8.11.2.2) or Yfs in datasheet. You can use it for a “reasonable” matching too at a certain extend
Fab
Last edited:
I highly recommend the use of "safety" connectors that have no exposed contacts. Example: https://spot.staubli.com/fileadmin/spot/22.2656_de.pdf
Nikosokey,
Are you saying that the devices in the second photo are not genuine Toshiba parts, but function as well as the original Toshibas?
2oz copper on the boards, relevant traces are crosstalk-optimized (mainly between LT4320 and mosfets)
When the LT4320 daughterboards are working as they should I could upload the gerber files so that diy'ers who want to can send them to a PCB-house of their choice.
The "mainboards" could be used with TO-220 resistors on a heatsink or via an extra connector to small capacitancemultipliers (one positive, one negative). The capmx-boards could be mounted on the main-heatsink for example. Each amp-board will be connected via a molex-type connector.
My USSA5-boards are waiting for their powersupply. More progress on saturday.
When the LT4320 daughterboards are working as they should I could upload the gerber files so that diy'ers who want to can send them to a PCB-house of their choice.
The "mainboards" could be used with TO-220 resistors on a heatsink or via an extra connector to small capacitancemultipliers (one positive, one negative). The capmx-boards could be mounted on the main-heatsink for example. Each amp-board will be connected via a molex-type connector.
My USSA5-boards are waiting for their powersupply. More progress on saturday.