PIE675T01E/09 induction hob fault c3

[vjekobalas]

I haven't been involved in electronics for donkey's years but I don't have the money
for a new hob and was hoping I could repair it.

Maybe someone can give a clue how to try to find the cause of the fault c3 .
I called Bosch and asked what c3 means as I couldn't find the fault code anywhere
 and they told me it indicates overheating. I tried switching power off/on but it made no difference
(fault occurred again).

The PIE675T01E/09 hob has 4 plates: front and back left pair and front and back right pair and 3 pcb's , one
under each of the two pairs of plates and one in the front for the buttons.
The two plates on the left work without faults.
The two plates on the right work such that if you set them to a low heat value (max is 9, low is somewhere around 3), they almost immediately
start flashing the fault code. If you turn them on to a higher value, they will work for a few seconds and start flashing the fault code.
If you turn the hob off, you can again try setting the temperature and the fault occurs again.

I have opened the hob, disconnected the plates, thermistors and mains power, removed the right pcb
(I presume the right pcb is associated to the two right plates which cause the
fault code to appear) . It is a Siemens 9000 156 089 ELIN-DCHA ADV pcb. On initial inspection, I don't see any visible
damage anywhere except the mica on top of one of the two plates causing the fault has a slight haze on it (nothing
out of the ordinary).

All the 4 thermistors on the plates read some 60kohm and reduce in value when heated with hair dryer and return to this value
after some time. All of the plates read a fraction of an ohm.

So, I guess I'm "lucky" not to have an "easy" fault to find.

[vjekobalas]

sorry something weird is going on with the saving and removing of the message, so there is a duplicate message/thread- I tried
to delete the older message/thread (without attachments) but nothing happens. I will write only in this thread.

[vjekobalas]

One other bit of information which may be of interest - the fan seems to be on for a long time
although there is no overheating (at least not from the plates as they cause the fault quickly)

[Gyro]

I can't see anything in the photos but is there a thermistor sensing the heatsink temperature? Just a thought.

Also maybe check the soldering / integrity of all high current joints.

[vjekobalas]

Here's an initial search for possible fault sources which are directly related to overheating
(my initial thinking is that it could be a fault with the IGBT drive circuit and there is an
actual overheating or the circuitry doing the temperature measurement is not working
or both).

(a)Two blue thermistors  at line/power input next to the toroids and emi suppression caps.
My first question is -does one do any testing with power connected and how since
we're dealing with the induction plates/can't have them in place and glass in place ?
How should I test these thermistors ?

(b)There are four 20N60A4D IGBTs mounted on pcb top ,  under each of them on pcb bottom surface are 8 pin chips
(see photo) A 3126 presume drive chips
for which I cannot find datasheet. Anyone know what they are ?

 

[vjekobalas]

There's a gazillion points marked as TSPnumber- I presume test points, anyone know
whether it is possible to find any documentation for this pcb?

[fzabkar]

I think that 8-pin part is an Avago HCPL-3126 IGBT gate driver.

Here is the datasheet for the HCPL-3120:

HCPL-3120, Avago, 2.5 Amp Output Current IGBT Gate Drive Optocoupler, DIP-8:
https://www.mouser.com/datasheet/2/678/V02-0161EN_DS_HCPL-3120_2016-03-21-908803.pdf

... and an image ...

http://www.dnatechindia.com/image/cache/catalog/a3120-500x500.jpg

I wasn't able to find the 3126 part on the manufacturer's web site:

https://www.broadcom.com/site-search?q="HCPL-3120"

This supplier ...

https://ksp-electronics.com/en/products/1374693-integrated-circuit/a3126-dip-8

... claims it is compatible with Toshiba's TLP350:

https://www.glynshop.com/erp/owweb/Daten/Datenblaetter/Toshiba/TLP350_datasheet_en_20140922.pdf

[vjekobalas]

I found this doc on scribd which may give some more info in case someone
experienced is willing to give a pointer on how to proceed. The doc is not for
exactly the same pcb but looks very similar.

[fzabkar]

A temperature sensor under the glass is used in order to protect the inductor, and it moreover allows detecting that an empty container is being over-heated and stopping the heating process.

The STC allows the user to keep the temperature constant as from the instant that the user turns power on, without using additional temperature sensors

The STC system works by evaluating the variation of the inductor-container coupling with the temperature. This variation is due, fundamentally, to the variation of the electrical resistivity of the container base with the temperature

The STC system evaluates the coupling by always using a frequency of 35 kHz. This evaluation is made during 600 ms, and it is repeated every 6 seconds. For the remaining 5.4 seconds, the system applies the power necessary to keep the temperature at the level at which the STC function was activated.

I wonder if the reference voltage for the "temperature sensor under the glass" is missing??

[vjekobalas]

A duplicate of this thread exists as something weird was going on when I created the thread.
I have tried a number of times to delete the duplicate thread but nothing happens - I will try to delete
it again but anyway, this will be the thread that I will use.

Here is a comment from the duplicate thread from fcb:
Our AEG CIB6740 threw a hissy-fit literally a year and day after we bought it.

Dismantled the hob and removed the two large PCB's.  Turned out that the vents at the back of the top push out steam when using the oven and OH had placed a large tall pot on a back ring with cold water in it (unsurprising!), the steam 'diverted' was basically ended up condensing inside through a badly fitted bit of plastic, alumimium hose  - ended up under one of the PCB's via a fan.  This triggered a fault condition.

Boards were carefully dried and cleaned, and it worked perfectly when reassembled.

Not sure if this helps, but if you do have an induction hob/cooker that throws a hissy-fit, don't leave it as it might be moisture inside.

[vjekobalas]

I tried again to find documentation for my exact induction hob but to no avail. I came across two more
docs in case it helps for my case/anyone having similar probs in future (attached).

Using the docs, in addition to the two thermistors in the power input stage (which I believe are used to limit
inrush current) I see there are two SMD NTC's which measure the temperature of the IGBT's.

With the information I have been given till now, I think I will try to check the NTC's, recheck if the
temperature sensors/holders which touch the glass are seating properly / touching glass (I am not very happy
with the metal holder spring legs) , generally try to re-seat everything again in case I am lucky  and this is all
due to excessive temperature/humidity/liquid from pots. After this, I will look into the reference voltage for
temperature sensor.

A question ahead of time - how would I be able to do any testing/measurement on the live circuit eg
to check a reference voltage as I would need to remove the glass and coils to get to the pcb's ?

[vjekobalas]

I'd like to thank all that helped with input. I will need to let this wait
as I have more serious things to solve and my wife purchased
another hob.

[benneton]

With induction hobs:

1. Check IGBT/S
2. Check Graetz diode bridge
3. Check NTC and its pair resistor (you could have multiple of those: plate heat sensor, IGBT heat sensor, even diode bridge heat sensor) always check their pair resistors too. You can test it while device is on, however I recommend having isolating transformer for your safety and your equipment's sake. Just a friendly warning. NTC/PTC are almost always used in voltage divider setup. You don't even need to turn the device on. Set your multimeter to Ohm and check the resistance between one end of pair R to other end of NTC. Notice the resistance then warm it up with finger for e.g. - it should be changed resistance. Usually it is okay even if there is any resistance/conductivity. However, characterics could be changed due to overheating. Middle point should be in the around half of reference by default (if 5V => 2-3V). Ohm check could be enough, however, sometimes parts are behaving differently while device is on.
4. Check buffer caps / all caps. Usually filter and electrolyte ones. If you have oscilloscope, check DC pins. No AC should be there.
5. Check traces - very often discovering these types of errors - Trace the sensor path to MCU...

[vjekobalas]

Thanks a lot! Will try all the points you mentioned and more seriously go through it all as soon as I have time

[benneton]

You are welcome!

"If you turn them on to a higher value, they will work for a few seconds and start flashing the fault code" smells like igbt/graetz overheat protection. Tolerance could be greater on a higher power setting (software side) - device on boost will heat more, therefore some grace period. Look for a small ntc next to positive or negative pad of diode bridge or IGBT. Maybe, just maybe... Good luck and take care