You could, I guess, you don't have any Mosfets on hand, even through hole version?
You could even remove the mosfet and rig a test circuit up.
Max.

 

Unfortunately no, not N-Channel. I only have a P-Channel and that is only rated for 60v.

 

So after some testing of the transistors on there, it seems that 3 are bad and 1 is working still. However, for the sake of it, i will replace all 4. The way i have been testing them is with the resistance mode on my multimeter and then probing from base to collector, collector to emitter. Some are conducting both ways, and some are just not conducting on the base to collector. there are 3 NPN transistors and 1 PNP there. I have found the datasheets of all 3 different transistors used and have found replacements that match the specs exactly. So i will get those ordered and a replacement MOSFET today.

Obviously there may well be other components that are broken here but i think that replacing the transistors and that broken resistor is the best thing to do right now, would you agree? Would you keep probing to find anything else that is broken?

Also, that resistor i link previously, would you use that as a replacement on this board to that one that has a broken lead?

The parts just in case you're interested:

NPN Transistors:
http://uk.farnell.com/diodes-inc/bc846a/transistor-npn-65v-0-1a-sot23/dp/1902483
http://uk.farnell.com/on-semiconductor/bcw65clt1g/transistor-bipol-npn-32v/dp/2535613

PNP Transistor:
http://uk.farnell.com/nxp/bc856a/transistor-pnp-65v-0-1a-sot23/dp/1907653

 

Check and replace as much as you can.
Then use the schematic to trouble shoot if still problem, not sure of the resistor question?
Max.

 

The only other components are small smt resistors and caps which i can't tell if an unmarked resistor is bad apart from if it shows a very high resistance through the meter which i've gone around checking them all. So i think i can only assume that the transistors are bad.

The resistor that i speak of is the one that is broken off that i shown a photo of earlier, measured on the meter i can confirm its 18k, so i'm planning on replacing that with a 5W resistor of the same value. May be over rating on the wattage but better to be safe than sorry.

This is the resistor i'm replacing it with - http://uk.farnell.com/panasonic-ele...j183/resistor-metal-oxide-18kohm-5/dp/2324614

 

I have replaced all of the transistors, mosfet and that resistor. No joy.

I have taken a reading of 1VDC on the motor output connector when it should be on.

 

What is the 8 pin IC?
Do you have a semi or Complete drawing?
Max.

 

It is a microcontroller, i've had it off and inspected all of the pins and the pads, continuity tested its routes, everything seems fine to my eye. However, i don't know exactly how it is layed out in terms of what functions it is doing. As far as i can workout is, its taking a ADC pin and monitoring the motors speed. Then it is controlling the speed, if it gets too high (maybe) through that mofset, i assume anyway. I need to investigate further on how this MCU is doing this.

The datasheet for the part is: http://www.nxp.com/files/microcontrollers/doc/data_sheet/MC9S08QD4.pdf

 

I, for some reason, didn't expect this to happen and it scared the living day lights out of me.

This is precisely why you DO NOT do it this way. Capacitors have no external resistance curtailing the flow of current at a given rate, and will dump their maximum rated voltage at whatever current they can deliver.

Instead, use a resistor and a bulb/led. this way you can not only discharge it safely, but have an indicator as to when it's discharged.

 

This is precisely why you DO NOT do it this way. Capacitors have no external resistance curtailing the flow of current at a given rate, and will dump their maximum rated voltage at whatever current they can deliver.

Instead, use a resistor and a bulb/led. this way you can not only discharge it safely, but have an indicator as to when it's discharged.

I have since found this out through reading articles regarding this. Thank you, really appreciate your input on that topic!

 

i assume anyway. I need to investigate further on how this MCU is doing this.

It makes it tricky when a μp is used as the firmware is unknown and probably a security fuse blown.
If I had to guess, it would be something in the current sense circuit.
Max.

 

It makes it tricky when a μp is used as the firmware is unknown and probably a security fuse blown.
If I had to guess, it would be something in the current sense circuit.
Max.

I'm not sure how much futher i can go without schematics that show the values on each cap and small resistor.

 

There should also be a low voltage supply derived from the HV DC in some form or another. Have you explored this part of the circuit, I would not expect any more than 0v and 5v on VDD/VSS.
Max.

 

Indeed, that MCU runs off of a maximum of 5.5v, the datasheet species this. So therefor the circuit is being brought down to at least 5.5v. However, i'm failing to see any sort of regulated drop down, i really hope they haven't done this through means of a voltage divider, but it genuinely wouldn't surprise me for one second.Mr Dyson strikes again.

I need to investigate and dig deeper to workout what is bringing the voltage down. To do this i will be probing the board for DC power to the MCU while it is turned on, if it receiving power like expected i'm then expecting to be checking that it isn't (for some reason) in a constant reset process. If this is fine then the only two options i see being wrong would be the driving of the mosfet just is not working OR the monitoring of the motor speed is broken for some reason, either pulled high or low through a short maybe....

I'm out pretty much all day today, so it may be later on that i get any results back to you.

 

Don't hold the resistor with your hand or fingers, you can try gluing it at the end of a Popsicle stick or some other insulated material. but only if you are using 20 volts or less capacitor.

 

Well in this case its a 400v cap.

 

Measure the actual pins and ensure the 5v is present. In some cases where the current demand is very low, a resistor and zener is used to drop the voltage. A voltage regulator IC would not be used to bring 300vdc down.
There does not appear to be any kind of SMPS for it.
Max.

 

Right okay. Well from my mental recollection, those three 18k resistors adding up to 54KOhms, then goes through a SOT23 IC, which is a double diode package. I'm not sure if its a zenor diode IC or just a normal diode IC. I will investigate when i get home tonight.

 

Just done some measuring here, the voltage on the MCU is 750mV. Obviously not enough.

The power is coming from those 3 18k resistors, its being dropped down to 9v, it then is connected through a SMT resistor, then smoothed with a cap down to GND before it enters the MCU VDD pin. So, i'm thinking that the SMT resistor is broken. No obvious signs of this though.

Do you think it is wise to drain the cap and measure the resistance of this resistor? What would you expect it to be? I'm thinking, maybe around the figure of 390Ohms. But this is with my assuming that the datasheet shows a current draw of 12mA on maximum stress on all pins BUT i'm not factoring in the zener diode, i'm not 100% sure how much this would be expected to drop on the line either.

So if i measure any resistance over the 390Ohms calculated, i think this resistor is bad.

 

It would be nice to know the precise method of providing the 5vdc, I would expect a some kind of regulating device of some kind in there at the end of the series string.
Obviously this seems at all or at least partially the cause of the malfunction.
Obviously the bulk of the voltage is dropped across the series resistors, what do you measure across them.?
Max.