Brushless motor transistor board mystery parts and topology

[sonnytiger]

I took apart an electric power steering pump off of a newer car and was surprised to find a brush-less DC motor in it. I plan on using the motor part as generator, but I also took a look at the somewhat ruined drive electronics. There are two boards, one with control stuff on it and another with a metal backing that has the MOSFETs mounted on it. I was looking up these devices, trying to find something on them, but came up blank. As you will see in the picture, there are four different devices, one has no text on it at all, one has p2n04l03 and ak038 in smaller print on it, another has n0403 and ac 42 in smaller print on it and the final one has n0403 and ad042 in smaller print on it. This seems strange as you would think there would be all the same devices on it, unless it is something more than just the switching transistors. Other than these devices, there are some diodes and capacitors. My goal is to basically try and use it to spin the motor if I want to use it for that instead of as a generator, I am interested in finding out what topology it is using.

Thanks!

[free_electron]

protected mosfets. these sense overcurrent and sht themselves off. custom part made by infineon. not avaialble on the market

[Short Circuit]

Appears like a standard 3-phase bridge.
Lowside are probably logic level FETs (hence the 'L' in the partno) to allow for direct drive from 3.3 or 5V logic.
Not needed for the highside FETs as those need bootstrapped levelshifters anyway.

[c4757p]

I like that hefty SMD current shunt... anyone know if they are sold anywhere in low quantity?

[sonnytiger]

I like that hefty SMD current shunt... anyone know if they are sold anywhere in low quantity?

I thought maybe that's what it was, dunno where to get one though sorry.

[sonnytiger]

protected mosfets. these sense overcurrent and sht themselves off. custom part made by infineon. not avaialble on the market

Think its possible to drive it? I can't use the original circuit as it is probably non function (was soaked in water for some time.)

[Zero999]

The top right hand MOSFET also appears to have a crack in the case.

[sonnytiger]

The top right hand MOSFET also appears to have a crack in the case.

it must have been a hair or something cause i cant see it

[poorchava]

protected mosfets. these sense overcurrent and sht themselves off. custom part made by infineon. not avaialble on the market

Not exactly true:

Note, that those devices have tab connected to one trace, first lead is a single signal, and all the remaining 5 leads are connected to one fat trace. High power protected switches usually have more signal leads for diagnostic purposes (like external current-mirrored current output). Those are N-MOSFETs in7-lead packages. They do that because single leads in TO263 cannot handle such large currents. They don't do that for P-MOSFETs.

This must be from a 12V passenger car, because in 24V vehicles (like most trucks) power devices are typically rated for 55V and higher.

P2N04L03 = IPB160N04S2L-03

**n0402 according to Infineon's naming convention will be a 40V 2mOhm N-Mosfet. TO-263-style package means it will be IPB series. I would guess one of those:
IPB180N04S3-02
IPB160N04S4-02D

Infineon's naming format for automotive mosfets is following:
IPabbbcddee-ff
IP- infineon power mosfets
a - package symbol (B = D2PAK)
bbb - current rating
c - N for N-channel, P for P-channel
dd - rated voltage divided by 10
ee - semiconductor technology. S2 = optimos, S3 = optimos-t S4 = optimos-t2
ff-Rds in miliohms.

As for the high-current shunts I would bet on custom stuff from Isabellenhütte.

[amyk]

I was going to post the IPB160N04S2L-03 too... and I think the other one is the IPB160N04S2-03 (without the L - has higher Vgs(th) - extra margin to avoid shoot-through?). It looks like there's 3 of each.

GL{1-3} and GH{1-3} are obviously gate inputs for low and high.

[poorchava]

I was going to post the IPB160N04S2L-03 too... and I think the other one is the IPB160N04S2-03 (without the L - has higher Vgs(th) - extra margin to avoid shoot-through?). It looks like there's 3 of each.

GL{1-3} and GH{1-3} are obviously gate inputs for low and high.

You are right. For some reason I was looking for N0402 rather than N0403.

@OP: what sort of bridge drive do they use? A dedicated chip (Infineon does those too), a gate driver + CPU or some sort of discrete gate drive (most automotive ECUs I've seen have some sort of general usage charge pump to generate voltage above rails in order to be able to use N-mosfets for all possible tasks instead of p-channel ones.

[sonnytiger]

I was going to post the IPB160N04S2L-03 too... and I think the other one is the IPB160N04S2-03 (without the L - has higher Vgs(th) - extra margin to avoid shoot-through?). It looks like there's 3 of each.

GL{1-3} and GH{1-3} are obviously gate inputs for low and high.

You are right. For some reason I was looking for N0402 rather than N0403.

@OP: what sort of bridge drive do they use? A dedicated chip (Infineon does those too), a gate driver + CPU or some sort of discrete gate drive (most automotive ECUs I've seen have some sort of general usage charge pump to generate voltage above rails in order to be able to use N-mosfets for all possible tasks instead of p-channel ones.

yes it is from a passenger car, only a couple years old. The pump assembly had another board with drive electronics, including a large chip and a couple 16 pin packages. I wouldn't mind drawing out the schematic of the transistor board, but im not sure which pins are which. The tap on the devices would be the drain right? the single pin the gate and the other 5 the source?