what is this unknown component

[moose]

Hi
This component is off my Fisher & Paykel dishwasher controller PCB.
Based on how the circuit it is in operates:
when pin 1 is hi, pin 4 will connect with pin 2&3 which connect to GND. Pin 4 is the low end from a small fan.
similarly with pin 8 & 5...pin 5 connects to 6&7 when 8 is hi.

how would one even describe it so as to look up parts suppliers websites?

thanks in advance
moshe

[bob91343]

It looks like a dual transistor.  Or maybe a logic element that behaves as one.

[moose]

Hi Bob....yes, i've just discovered that a dual channel n-mosfet has the right pinouts...now to find a replacement!
thanks for your input.

[MathWizard]

SMD codes, what can I say, I guess we're lucky if there's any labeling

[amyk]

Could it be 8205 series dual NMOS? What voltages and currents are expected?

[Unixon]

Looks like a dual MOSFET soldered upside-down, usually pins 1,8 are drains and pins 4,5 are gates.
Google suggests Hitachi HAT2050T

[moose]

yes..on the board it is upsidedown to the photo so the hitachi & renesas fit the profile.
But can you tell me how you managed to identify it?

[Unixon]

yes..on the board it is upsidedown to the photo so the hitachi & renesas fit the profile.

No, unfortunately it doesn't. Look at pin 1 mark, pinout is different, Drain<->Gate pins are swapped.

[Unixon]

Can you show a larger piece of the board?
Now it seems to me that the pins were misidentified, resistors next to pins 4,5 are probably pull-ups, then it fits standard pinout for this package.
SOT-23 to the left is probably BAV99 and SOD-323 to the upper-left is some kind of diode or zener.

[moose]

pins 4 & 5 go to MCU pins, pins 1 & 8 are the low side of attached peripherals so Hitachi & Renesas are correct.
To satisfy your curiosity here's the bigger picture.

I would still like to know how on earth you managed to ID it.

[Unixon]

To satisfy your curiosity here's the bigger picture.

Oh, it's not for curiosity, but rather for context. Surrounding parts may show if a hypothesis is likely true or not.
Here it all makes sense, polarity, protection diodes, pull-ups, 40V at rail and N-MOSFET rated for 100Vds.

I would still like to know how on earth you managed to ID it.

So, either 2050 or 4625 in top marking may indicate model number and the initial guess is that it is a dual MOSFET and we know the package is TSSOP8, which is not a hugely popular package for these. So we're making a 2nd guess that there's not a lot of parts where these three come together, so we're trying to check if "model+package+device type" search gets us where we want.

[richnormand]

@ moose

on your photo below the damaged chip is another one that looks similar. Is it the same number? If so does it measure the same? Can you remove it and compare the tw out of circuit?
Again assuming it is the same number chip.

[moose]

yes..there are 3 chips the same...1 below the other. in the photo..in fact what i did was replace the faulty chip with (an unshorted) one from another board.

The biggest problem with working on this controller is that i cannot test it when its powered up as it is in an inaccessible enclosure. And to rig up a jib to test it OFF the machine means i have to set up a skeleton machine on a worktable else the MCU simply reports errors because nothing is connected that it tests on startup.
(Plus it also has 240V coming in on 2 pairs of connections)...its a very nice dishwasher until you have to start working on it! 

[amyk]

Unfortunately I can't find any seller's pictures to confirm but I think Unixon got this one.