Blackstar HT60 PCB leaking voltage

[Mick Bailey]

Just got in a Blackstar HT60. No signs of damage or burning to the PCB, but it had both the phase-inverter MOSFETS shorted. I thought this was going to be a straightforward fix. No other components damaged, so replaced both MOSFETS and TR2 failed instantly. I removed it and powered up the amp on the variac and monitored the voltage on the pads, then noticed TR2's 47K drain resistor was dropping 330v. Removed the only other connected component (coupling cap to the EL34) and still getting the voltage drop.

So that's 330v dropped across a resistor that's not connected at one end. Probing around the board shows high voltages between unconnected areas. Between the MOSFET pads reads 700 ohms, so what this becomes under stress is another matter.

Anyhow, the entire PI area is conductive, plus there are abnormal switching conditions coming out of the PIC.

I wonder if this is a one-off, or is this going to be one of many? Just like the Marshall DSL/TSL boards?

[Rhodesplyr]

I can't comment specifically on your problem, but I have fixed a couple of these. Though I will admit that they sound good when working, my general impression was that these amps were going to be trouble down the line, based on the design, component choices, and build.

[alexradium]

There are many of these, I think it's one of the first production, it was also in their forum.
Try at least to cut the board with dremel between the 3 legs of each mosfet.

[Mick Bailey]

That did come to mind. I've since removed more components and there's voltage leaking all over, not just between the MOSFET pads. There's even 7v on the grids of the preamp tubes. I'd be interested to see any more info on what people have done with these - couldn't see anything on their forum, though.

Sadly, this is the third Blackstar this week and another one from the same owner is coming in on Monday.

EDIT: I drilled a couple of 2mm diameter holes with 5mm spacing part way through the board and put my insulation tester between them to test the substrate. At 1kv it reads higher than my tester goes up to (100M Ohm) Further tests between pads shows the conductivity exists in straight lines between points of highest potential difference. My conclusion is that the board is tracked internally between the through-plated holes. I drilled between the MOSFET pads and then slotted them, and there's a little improvement. I think I'll work out a strategy for slotting and drilling to see if I can fix it.

I asked Blackstar for the price of a board and they wont supply one as I am not authorized. My customer has already taken it to their recommended repair facility and they wanted £450 to replace the board. Insanity for something that's clock-radio quality.

[Rhodesplyr]

Insanity for something that's clock-radio quality.

I warned the owner of the one I worked on that it would probably work for a few years and then become a maintenance/repair nightmare, like the electronics in my mother's Volvo.

[Mick Bailey]

I'd almost forgotten to update the thread to where I got;

I drilled a 1.5mm hole between the MOSFET pads (that's TR1 and TR2) and then another hole either side of these just inside the silk screen outline for the transistors. I then used a piercing saw to cut between them, resulting in slots between the pads at right-angles to the line of the pins. You could chain drill them and use a scalpel to open them up instead. I then cut a slot centred between the two transistors by drilling two 3.5mm holes and filing between them. The resulting slot is about 6mm long, again at right angles to the pins. I lifted the ends of R140 and R141 on the MOSFET side and drilled 3.5mm holes between each pair of resistor pads, close to the lifted ends. With C122 and C123 removed, holes were drilled between the mounting pads - I forget the size, but just enough to clear the pads. Then I took an imaginary line between C123 +ve lead and the ceramic resistor pad nearby and drilled a 3.5mm hole midway between the two (flipping the board to make sure there wasn't a track in the way).

Before refitting new transistors I temporarily reinstalled the board and checked the MOSFET pads for unwanted voltage, as well as checked for any drop across R140 and R141. Everything was perfect so I installed new IRF830s and everything was the good. All voltages back to normal and the PIC switching is now fine.

[alexradium]

all this work because those BS mofos thought a tube PI is too expensive!

[SoulFetish]

I'd almost forgotten to update the thread to where I got;

I drilled a 1.5mm hole between the MOSFET pads (that's TR1 and TR2) and then another hole either side of these just inside the silk screen outline for the transistors. I then used a piercing saw to cut between them, resulting in slots between the pads at right-angles to the line of the pins. You could chain drill them and use a scalpel to open them up instead. I then cut a slot centred between the two transistors by drilling two 3.5mm holes and filing between them. The resulting slot is about 6mm long, again at right angles to the pins. I lifted the ends of R140 and R141 on the MOSFET side and drilled 3.5mm holes between each pair of resistor pads, close to the lifted ends. With C122 and C123 removed, holes were drilled between the mounting pads - I forget the size, but just enough to clear the pads. Then I took an imaginary line between C123 +ve lead and the ceramic resistor pad nearby and drilled a 3.5mm hole midway between the two (flipping the board to make sure there wasn't a track in the way).

Before refitting new transistors I temporarily reinstalled the board and checked the MOSFET pads for unwanted voltage, as well as checked for any drop across R140 and R141. Everything was perfect so I installed new IRF830s and everything was the good. All voltages back to normal and the PIC switching is now fine.

Good Thinking! It's like the answer was right there the whole time!
But..., just in case someone reading this has only has only had the opportunity to work on one blackstar, and were following right along until their imagination began to fail them terribly right around

imaginary line between C123 +ve lead...

I don't suppose you have any pictures of what you did?

[Mick Bailey]

I should have taken pics (even for my own records) but I was in such a hurry to get it back together I neglected to take any. No doubt I'll be seeing another and will take the opportunity then