Transistor exploded can partly read it any guesses what it might be ?

[ESXi]

yo,

Got a vacuum here a German bosch gas 50 professional with pic16c621a-041/p airflow-vacuum sensor and electromagnetic dedusting function.
The brains are running (the pic). Still need to test the other sensors but a chip is exploded I did not see this at first because it was covered in epoxy resin..
Seems gate of this transistor that is blown probably NPN that enables electromagnetic dedusting function gate is connected to PIC on pin 7(RB1) and 8(RB2) bridged for probably slightly higher pull-up signal? Datasheet says "Input/output pin. Internal software programmable weak pull-up".

Anyways I carefully took apart whats left of the chip. I can't find anything on digikey or google. Probably my mistake or maybe it is impossible to find out again because there is no BOM available and the board came with epoxy on a few parts from the factory so no way of finding the chip numbers unless you get the resin remove first...

Here is a picture of it.
As far as i can make up of whats left of the laser engraving.

Any ideas to what this can be?

I would figure I first try to get my hands on this transistor if not possible, then I need to look for a replacement board or Frankenstein is together... (last option)

[DavidAlfa]

Post quality pictures of the board.
Is it directly connected to the Pic?
Try to track the connections.

[ESXi]

Post quality pictures of the board.
Is it directly connected to the Pic?
Try to track the connections.

Sure here is the board to note that I didn't screw up these tracks, some smd cap was blown out and some other guy tried to repair it.
What I found in the meanwhile was that the DPST switch one of of the internal switches was burnt up fully locked. I have detached them again and did the sandpaper thingy.. I tried to look for another switch also no success.. So let just say never use the switch again keep it in the same mode.

I did test some more, the indicator led goes on if I restrict the airflow with my hand for 90+ % after 10 seconds~ roughly the led comes on at this point it should power the "speaker" which makes a knocking sound with 24vdc (it gets 24vdc) and it should do the electromagnetic cleaning procedure.
Speaker doesn't do anything don't see 24vdc on the output neither.

But the red LED confirms that the PIC is running, checked pin 7(RB1) and 8(RB2) on PIC gets 5.5 vdc.

board.
https://imgur.com/a/Yp8kALk

[Kleinstein]

I may still have a similar / same broken PCB (from GAS25, the smaller brother). I can't see on the foto were the blown transistor was suppose to sit.
With a bit of luck it is not blown off my board (with mechanical damage - something heavy hit the switch).

[ESXi]

I may still have a similar / same broken PCB (from GAS25, the smaller brother).

Does it also has this electromagnetic dedusting feature? Maybe it's using the same transistor who knows. But yeah you have to dug out the transistor from it's epoxy resin blob preventing it from "flapping around in the breeze" lol.

I can't see on the foto were the blown transistor was suppose to sit.

This is the location bottom picture left corner see red circle. It's under the potential meter.

With a bit of luck it is not blown off my board (with mechanical damage - something heavy hit the switch).

Ah that sucks, well this vacuum cleaner it's my dads his cleaner from work. It's a old one and has seen A LOT of work so I was not surprised to see the switch being bad and wipers being scuffed up.
With all this dust and sand it sees these parts wear faster. I was disappointed to not see a DPST switch that was fully glued/sealed together (just press fit with clips). Small dust particles that went inside is I think not the root cause but also not helping to prolong the lifespan of it.

[Neomys Sapiens]

If asking for help in such matters, I tink it would be good practice to do as much work as possible yourself and to provide the most of information that you have.
Measure the pitch of the pins for example. Post those pictures in full resolution and not 3-in-one.

[fzabkar]

It's either a lo-sat PNP transistor or a P-channel MOSFET with BCE or GDS pinout.

[ESXi]

If asking for help in such matters, I think it would be good practice to do as much work as possible yourself and to provide the most of information that you have.
Measure the pitch of the pins for example. Post those pictures in full resolution and not 3-in-one.

fullres board images
https://mega.nz/file/8BYD0CCJ#ChkmD5N06SeYRlYU97Pqp_zsG_PnKjiFwmQAgBXjzd0
or
https://imgur.com/a/nGmGjE5

I did some practice, already repaired 720p DLP projector old one and a Ilyama touch screen monitor (first time QFN without proper flux only resin core lead free = hell, pb yes it is & no microscope).
I know that I need to dig deeper. I want to pull out my scope but last time I did some scope shit sth wasn't right I need to practice more with Arduino or other prototype first. So I can scan all these pic io ports with the scope for activity on PORTB when the indicator led goes on if airpreausre sensor resistance is changed for prolonged period. I think when the led goes on pic RIP instruction is in cleaning routine and is pulling up PORTB I/O signal from here I can slowly trace down the rabbit hole, any other ideas maybe.

[rsjsouza]

The packaging reminds me of the ancient BC147/148/149 series, although the pinout does not match (neither the probable age of the equipment).
https://duckduckgo.com/?q=bc149&t=fpas&iax=images&ia=images

[Kleinstein]

This is a relatively new product. So the old BC147 are out. The case is similar, but this are definitely more modern parts. I would also expect higher power, as for low power they would have used SMD.  It is odd to see a so many THT parts, but this could help with the heat.  So as others have already noted more like a power part.  To switch the cleaning part, this would need quite some power - it is just a simple electromagnet, more build to a cost and not for efficiency.  Are there really 24 V ? , the power may be directly mains.

[fzabkar]

Unless I'm cross-eyed, the emitter of the transistor is connected to the Vdd supply of the PIC, so it is switching 5V, not 24V.


Edit: Now I'm confused. The TO-220 packaged device is an FT1214MH, which is a TRIAC.

https://alltransistors.com/scr/view.php?pdf=FT1214MH

So is the mystery part also a TRIAC???

[DavidAlfa]

What L and N are? 220V? Or 24Vac?
L seems to go to GND through the switch??
Also L goes to the pin near the misterious component, through what is seems a current transformer, so it must be AC.
Maybe a BTA600 triac was there.

[ESXi]

Unless I'm cross-eyed, the emitter of the transistor is connected to the Vdd supply of the PIC, so it is switching 5V, not 24V.


Edit: Now I'm confused. The TO-220 packaged device is an FT1214MH, which is a TRIAC.

https://alltransistors.com/scr/view.php?pdf=FT1214MH

So is the mystery part also a TRIAC???

I think it might be. But not so sure, I will need to draw a full schematic for myself and you guys because trust me in this thing it's a spaghetti of wires and this electromagnetic cleaning it seems almost as if it switches to motor off from L and throws all this movement and the energy in the coils above the filters thus cleaning the dust? Idk. Bosch isn't going to tell us so we'll have to figure it out ourselves.
And yes, the board gets 220vac and no buck converts or whatever I think some small regulators for 5.5vdc (PIC cpu) and 24vdc for hammer, not sure if there are any more power rails.

The packaging reminds me of the ancient BC147/148/149 series, although the pinout does not match (neither the probable age of the equipment).
https://duckduckgo.com/?q=bc149&t=fpas&iax=images&ia=images

Close match bt I think it's notthis part. I will go further making a full reverse schematic at least trying to.
This would make it much easier to repair and it will probably help people since Bosch was just too dicky to even release it it's just one single layer...

Unless I'm cross-eyed, the emitter of the transistor is connected to the Vdd supply of the PIC, so it is switching 5V, not 24V.

Edit: Now I'm confused. The TO-220 packaged device is an FT1214MH, which is a TRIAC.

https://alltransistors.com/scr/view.php?pdf=FT1214MH

So is the mystery part also a TRIAC???

I think it is because for the size you would say uhm that it is a TO package /w heatsink. But it's not
Little delayed, my car broke down have to fix it and some other shit. Anyways I have made the "boardview" 8-bit greyscale easy on the eyes.

What L and N are? 220V? Or 24Vac?
L seems to go to GND through the switch??
Also L goes to the pin near the misterious component, through what is seems a current transformer, so it must be AC.
Maybe a BTA600 triac was there.

L and N are 220, yes and the 24VDC is for the hammering speaker it's not a piezoelectric speaker sticker on it says 24VDC, hooked it up to my bench power supply and it did nothing at all. The board itself it looks like it has no "Hot and Cold" part. It's all on one single layer board the 220vac in and the other voltage lines for the PIC (5.5v, 24vdc).

This speaker / hammer idk wtf it actually is. No numbers can't find anything about it anyways here is what it should do it's just a device to let the user know that it's in the cleaning routine.
https://youtu.be/L6wYK1rVd3Q?t=4

[fzabkar]

https://mastergrad.com/forums/t136782-stroitelnyy-pylesos-interskol-pu-32-1200-nebolshoy-obzor-glazami-polzovatelya/?page=2&order=desc

Z0410MF, STMicroelectronics, TRIAC, 600V, Gate Trigger 1.3V, 4A, TO202-3:
https://docs.rs-online.com/576f/0900766b80ed29f8.pdf

[DavidAlfa]

Hats off, great job

[ESXi]

https://mastergrad.com/forums/t136782-stroitelnyy-pylesos-interskol-pu-32-1200-nebolshoy-obzor-glazami-polzovatelya/?page=2&order=desc

Z0410MF, STMicroelectronics, TRIAC, 600V, Gate Trigger 1.3V, 4A, TO202-3:
https://docs.rs-online.com/576f/0900766b80ed29f8.pdf

https://media.mastergrad.com/thumbnails/forum_post_attachment/46/da/7c/46da7c753e60897277a99a0abb129412.jpeg

https://s3-eu-west-1.amazonaws.com/enrgtech-images/full/70f1a4326f86fa289925729610415e80bdafb680.jpg

Well, damn got to admit that was amazing!
Only thing missing is bypass smd capasitor on the backside of the pcb, and V7 Z0410MF, great! So the failure point was the Z0410MF shorted and it pop'd also the smd bypass capacitor.



I was playing with cv2, wanted to make a CAD file of the data points extracted from cv2 results. I can try it, would be fun if a get a image trace to cad poc running.
test /w py

[fzabkar]

https://www.interskol.ru/storage/docs/products/PU_20.1000_32.1200345.1400_s_blokom_upr._.pdf

V7 = Triac Z 0410 MF 4A 600V 25mA
C6 = CHIP-C 22nF/10% X7R 0805 50V

[ESXi]

https://www.interskol.ru/storage/docs/products/PU_20.1000_32.1200345.1400_s_blokom_upr._.pdf

V7 = Triac Z 0410 MF 4A 600V 25mA
C6 = CHIP-C 22nF/10% X7R 0805 50V

100% 
I'll post back here when I have ordered and soldered the new parts on the board. 

[ESXi]

Can confirm, tried to replace

V7 and C6 (22nF, yes) both correct even from diffrent version PCB (smaller vacuum) parts ARE in fact the same. Still didn't work somehow. Ordered a brand new board from Bosch and tossed the doomed one, will install and test. The only thing that was different was C6 wasn't 0805 but think it was 1206 or 1806 don't know about tolerances (caliper measurement 3mm~ in lenght). Yes I just took the hot air gun at the new board and took it off to verify. No warranty anyways I am the warranty.. Will update on what was the issue, I hope the coils are still fine.

Edit:
It's so wierd I can't find the problem I measured all parts.
Rotary wiper switch works, mains switch works all components none is shorted. But I do see changes in the design. They removed the 24vdc speaker control and sense, and removed the potentiometer just jumped and changed some resistor value. Will try this on the old board, want to know what the issue is. The new board works, and behaves different, still yet unable to find it. Maybe dead zener diode? I can't test it.

I found out it switches on the manual cleaning function (that the Triac is used for) when PIC detects R plane shorted by pull down resistors (560k Ohms 5%) to PIC pic16c621a-041/p pin 9 and 10. Dead PIC maybe??

Edit2:
I fixed the old board but had to modify it, remove components and replace values, never would have guessed this if Bosch never sent me the updated PCB.
Here are the changes I made to repair the old board.