I have an old UK made Matrix M300 PA amp which has served me well for many years. It wasn't until I wired it to a speaker that had been bi-wirable but I forgot that I had changed it internally back to single wiring! It got very hot and was turned off quickly. Nothing was burnt or smoking, but it is now dead
I attach a pic of the Mosfets, some of which may have been replaced before I bought it at an Audiojumble. There are a pair of BUZ908/BUZ903 and a pair of BUZ901D/BUZ908D
I have removed one of 908's and crudely tested it using the diode setting of my DMM. I have assumed (possibly incorrectly) that pinout with the pins in the lower 1/3rd of the device are L:Gate R: Drain and Case is source.
I shorted the pins with the negative lead and held the Pos lead to the Drain, Neg lead to Source and touched the neg lead to the Gate momentarily and back to Source - No reading. Pos lead to Gate, Neg to Source - reading 302. With the Neg and Pos leads reversed and doing the same as above - No Reading i.e no forward voltage? I tried this test with the fet turned 90 degrees so that the pins were in relation to the first test LtoR and RtoL and got no readings at all.
I am thinking of replacing all the Mosfets with BUZ901/BUZ906 pairs. Would these be suitable?
I attach a pic of the Mosfets, some of which may have been replaced before I bought it at an Audiojumble. There are a pair of BUZ908/BUZ903 and a pair of BUZ901D/BUZ908D
I have removed one of 908's and crudely tested it using the diode setting of my DMM. I have assumed (possibly incorrectly) that pinout with the pins in the lower 1/3rd of the device are L:Gate R: Drain and Case is source.
I shorted the pins with the negative lead and held the Pos lead to the Drain, Neg lead to Source and touched the neg lead to the Gate momentarily and back to Source - No reading. Pos lead to Gate, Neg to Source - reading 302. With the Neg and Pos leads reversed and doing the same as above - No Reading i.e no forward voltage? I tried this test with the fet turned 90 degrees so that the pins were in relation to the first test LtoR and RtoL and got no readings at all.
I am thinking of replacing all the Mosfets with BUZ901/BUZ906 pairs. Would these be suitable?
An externally hosted image should be here but it was not working when we last tested it.
Do not "assume" , download the datasheet and be certain, otherwise the test is useless.
Why would you?
You have 3 elements: G , S , D ;measure from each of them to another, both ways, so 6 readings in total per transistor, post them here.
Label transistors with Sharpie so you identify them.
"No reading" is not a reading, state open / shorted / "a number" .
Maybe the transistors are fine (hope so 🙂 ), but "bad active parts" are not the only possible problems, you might have blown a fuse or wire ballast resistor or even scorched a track or connector or ....
Most 80s MosFet amps literally followed the suggested Hitachi circuit,
with minimal alterations if any at all, compare it to what you have.
If the same or very close (quite possible) , upload it here but relabelling parts to what you have.
As in: R1 R2 R3 might be R34 R52 R73 or whatever in yours and so on, draw what you actually have and repost.
Suggest you build a limiter lamp and plug amp there until deemed safe, turn it on without a speaker attached, all controls on 0, check for DC at the speaker output and whether both power rails reach the amplifier.
Agree with JMF (above) and Hitachi or Semelab Laterals I have used don't often fail rapidly due to overload or shorted outputs. They certainly get hot but due to their negative temp. coefficient, usually hang in, hot and stressed but still functioning such that fuses can be expected to work reliably.
The Hitachi (and similar) have gate failures when driving into hard clipping into low impedance loads.
The gate protection zeners in a Hafler DH500 will open up and then allow the gates to blow. This can usually be measured with an ohm meter from gate to source, it should be open on a good device, and will measure less than 10KΩ on damaged devices.
The gate protection zeners in a Hafler DH500 will open up and then allow the gates to blow. This can usually be measured with an ohm meter from gate to source, it should be open on a good device, and will measure less than 10KΩ on damaged devices.
Diode test is not correct for MOSFET diagnostics. You need to short gate and source and measure resistance between drain and source both ways round. A good device will be very high resistance.
I can't remember the pinout off hand, its worth you checking. The datasheet is easy to find. Exicon devices are worth looking at, could be cheaper.
I can't remember the pinout off hand, its worth you checking. The datasheet is easy to find. Exicon devices are worth looking at, could be cheaper.
Last edited:
I shorted the G&S, testing with - probe I get no reading on D or S. With + probe I get 302 or thereabouts on Drain.
- Status
- Not open for further replies.