Damaged mosfets on Duex?

dc42

@educatingsavvas you should use the PWM1 output on the DueX for the laser control, not the E2 heater output. The PWM signal levels are 5V so make sure that your adapter can tolerate that.

EducatingSavvas

@dc42 Yep I tried that. The input on the laser adapter did specify it could accept 0-24v logical control.

I think the ground for the laser head power and PWM are shared on the adapter - and was later told by the company to not connect ground to mosfet ground - instead going directly to PSU ground.

EducatingSavvas

@dc42 So I've found this: https://www.ebay.co.uk/itm/143987360827?hash=item218650c43b:g:lQQAAOSw3FZgW3Y1 and this kit for de-soldering: https://www.ebay.co.uk/itm/234019166593?hash=item367ca11981:g:qCoAAOSwqpJgw5i8
I think I'll give this a try.

dc42

@educatingsavvas you will need solder wick too, to clean the low melting point solder from the pads after you remove the chip.

EducatingSavvas

@dc42 Hiya - just wanted to follow up that I've replaced 74HCT02 driver chip U10 - it seems to go well and I was very careful. The duex-E4 seem to be behaving correctly but I'm finding that E2 and E3 are still having issues. E2 stays on and doesn't switch and E3 can be switched but the LED fades very slowly. Have I damaged that same chip while installing - or do I need to also change their mosfets as well?

Many thanks, Savvas

dc42

@educatingsavvas for E2 it sounds as though the mosfet may be damaged. Check this by measuring the voltage between the gate and source pins. It should be 5V when the mosfet should be on, and 0V when it should be off. The signal comes from the 74HCT02 through a series resistor.

For E2 my guess is that there is a bad solder joint, either on the 74HCT02 output pin that drives the E3 mosfet, or on the series resistor between that pin and the mosfet gate, or on the gate of the mosfet.

EducatingSavvas

@dc42 I tried using a usb microscope to see the solder process a little more easily, and there's a bridge that I can't seem to get rid off. I guess I can still use the board but without the E2-E4 pins. Needless to say I bought a new Duex.

dc42

@educatingsavvas have you checked the schematic? Maybe those pins are supposed to be connected together and there is a trace under that 'bridge'.

EducatingSavvas

@dc42 I can't seem to find that - is this something I'll have to look at on KiCad, or is there a Jpeg somewhere?

dc42

@educatingsavvas just look at the schematic. Pins 2 and 3 are connected together on both 74HCT02 chips on the DueX.

EducatingSavvas

@dc42

I checked the mosfets with the voltmeter and it seems E2 and E3 are damaged. There is no voltage between gate and source on either - and checking between gate and drain turns the LED for E2 turns on. E3 takes a while to turn on sometimes. I've found replacement AOD4184A on chanzon on aliexpress that ship from the UK - there doesn't seem to any on eBay. It would be good if the spare parts documentation included part numbers from RS Components or CCP Farnell - the product codes don't seem to bring anything up on their websites. Thanks again for your help.

dc42

@educatingsavvas you can also use this one https://uk.rs-online.com/web/p/mosfets/8259364/?sra=pmpn. That's the one we use for the bed heater mosfet, which needs a higher specification than the other heater mosfets. However:

and checking between gate and drain turns the LED for E2 turns on

That suggests to me that the mosfet is OK but either the output pin of the 74HCT02 is not connected, or the gate series resistor is not connected.