Electric fence energizer repair

[Bubaaak]

Hello,
 I am trying to repair energizer for an electric fence. The model is Horizont ranger an50, no schematic available. As it's mcu controlled it should be switched on by a press of a button, but it's completely dead. No signs of water damage, however as used outside it's possible.
The picture of the board is in attachment. I have tried to measure some of the components and haven't found anything wrong so far. It's taking about 20mA at 12V, with the power button pressed it goes up very slightly (by about 0,5mA).
I have identified  buck converter for mcu at the top left corner, with both input and output caps having the correct voltages (3,37v at the output). The power switch is connected to the "SW" terminals near this circuit, where one being ground and the second one going to pin 46 of the mcu, as well as the "Wake" terminal near the mcu itself. (3.3V at this pin as well )
Although I already doubt it is repairable, I would appreciate any suggestions on what to check /try

[fzabkar]

In the middle of the bottom edge of the PCB is a transformer, a diode and a large capacitor. Do you measure any voltage on this capacitor? I presume that the SW terminals will need to be shorted to switch on the unit.

Warning: This area is a high voltage supply.

[Bubaaak]

No, there is no voltage on that capacitor, with or without SW terminals shorted. I assume(d) this boost converter for the actual fence voltage supply is controlled by the mcu and it's switched on only when the whole thing is turned on. But your post gave me idea it might be an issue with this boost converter and the unit doesn't want to turn on because the mcu doesn't see any voltage on it's output, so I will try to focus on that area.

[fzabkar]

The SOT23-5 IC appears to be the PWM controller for that boost supply. Can you identify its markings? Perhaps its Enable pin is inactive?

[Bubaaak]

Had to scrape the coating, it's marked FFUR. Pins 3&4 are common and connected to Gate of near mosfet, 2 is gnd, 5 is +12v and pin 1 goes all the way to the little board behind the grey capacitor, which on both sides contains HC4538 multivibrator, divider to sense the high voltage, comparator and one/two and gates.  Pin 1 goes specifically to the output of both comparators.

Edit :looks like I got the pins mixed up, hopefully it's correct now.

[fzabkar]

I haven't yet been able to track down the marking code, but I suspect that the IC is a PWM controller. I think pin#1 is probably an Enable pin. If this pin is low, then the boost converter is probably being switched off by that daughter board.

That said, these ICs normally have a feedback pin. I was expecting that the divider on the main PCB (5 x 620K + 1 x 10K) would connect back to this IC. Now I'm confused. :-?

[Bubaaak]

I don't think it's pwm controller, as the exact same component is under the comm socket, with its pin 3 connected to the output thyristor. Pin 1 of this one is connected to output of one of the and gates.

[fzabkar]

I think it's a MOSFET gate driver.

Something like this (IRS44273L):

https://www.infineon.com/dgdl/Infineon-IRS44273L-DS-v01_00-EN.pdf?fileId=5546d46269e1c019016a4e9c18ca0cff

Pin #1 is the PWM input.

The marking code is really weird. The top of the chip is marked as YWLC (YW = year/week, LC = lot code). The bottom marking is A + IR logo.

Examples here:

https://pl.aliexpress.com/w/wholesale-IRS44273L.html?initiative_id=

Is the MOSFET also an IR/Infineon part?

The daughter board appears to have two identical redundant circuits. Is this a safety feature?

[Bubaaak]

Mosfet is STM part. Looks like you have nailed the driver, but since nothing is on pin 1, it's probably not the culprit.

Daughter board looks like two redundant circuits indeed, except for one more AND gate on one side. For the record SOT-23-5 ICs on this board are marked K503 and C085. I will try to draw some sort of schematic, although it's not easy to track the connections with all the vias and coating. I will also try to figure out the rest of connections with the main board . It has 7 connections - high voltage, gnd, 3,3v, outputs to mosfet and thyristor driver and two, probably inputs which I have yet to find.
On a side note - when connecting to the lab bench power supply initially the connection sparks and something buzzes very little. Unfortunately I have no way to measure the inrush current, but I am not sure whether it's normal just for charging capacitors or the boost converter is trying to start but fails to do so.

[fzabkar]

C085 is SN74LVC1G08:

https://www.ti.com/packaging/en/docs/partlookup.tsp?partmarking=C085
https://www.ti.com/product/SN74LVC1G08

K503 is LMV331ILT:

http://markingcodes.com/search/c/k503
https://www.st.com/resource/en/datasheet/lmv331.pdf

FOD4108:

https://www.mouser.com/datasheet/2/149/FOD4118-191117.pdf

[fzabkar]

I wonder if the dual multivibrator works this way. One half of the multivibrator generates fixed width pulses which drive the MOSFET, and this then charges up the boost capacitor in discrete steps. When the boost voltage reaches its preset level, the comparator stops the multivibrator. The other half of the multivibrator is then triggered by the MCU, and this then sends a single pulse to the thyristor which energises the fence. Then the cycle repeats. The MCU senses the boost voltage via a separate potential divider on the main board.

Would it make sense to disable the thyristor part of the circuit and then see if the boost voltage comes up? If so, then this could point to some fault protection issue.


Another thread with a similar device:

https://www.elektronik.si/phpBB2/viewtopic.php?p=566285

[Bubaaak]

Great find of the similar thread, although he had it way easier with the component being visually burned I will however try to measure/test this opto-triac to rule it out.

I am for a few days out now, but my next step will be drawing a schematic of the daughter board and at least part of the boost converter to better understand its function. I will report soon

[Bubaaak]

So I was able to create a crude schematic of the daughter board and part of the HV supply, but it didnt really help me with understanding the circuit so far.

Only a part of this board is duplicated, and that is the comparator part with HV sensing divider. Second input of the comparator is another voltage divider of 6k8 and 10k resistors, creating almost exactly 2 volts out of the 3.37V supply. This should mean the comparator will turn off the output once there is about 620V at the capacitor. This part of circuit doesnt have any connection to the rest of the board, or atleast i wasnt able to find any. The comparators and AND gates does have some conections to the MCU, but it looks like that part is driving only the output thyristor.

While measuring I found out the 10k resistor of the divider at main board was meauring about 5k, so i tried to desolder it, but apparently the double diode nearby was skewing the measurement. Then I was unable to solder it back and lost it later, so i have to find another one somewhere first The double diode is marked LD3, so it should be BAT54S. It measures 300mV drop / 5,5kohm in forward , 1,2V / 8,2k in reverse. This looks kinda weird to me, but all the other LD3s on the board measure similiar values, so i guess it is a feature?
In same way the 10k resistors of the low voltage comparators measures 4,7k, but once again thats due to the comparators connected and since both measure the same, it is probably not a fault.

Edit: I have just noticed that those comparator parts are not exactly the same either, one comparator have HV sense at pin 1, the other at pin 3. Therefore when one comparator outputs high, the second is low and vice versa, however both outputs are just connected together.

[Bubaaak]

Took a beter look and I got that wrong during creating the schematic. The comparator parts are the same, with 2V from the 3,3V reference connected to the non-inverting input and the HV-sensing divider connected to the inverting input. Also found out the pin1 of mosfet driver is also connected to MCU via 1k resistor, so thats where PWM for driving the mosfet comes from, comparators are probably just a fail-safe? I will try to solder back the resistor and look around with a scope to see if there is any activity on the MCU

Pin 41 of mcu, which should control the opto-triac is 0.1v (floating?) and gets pulled to zero as I bridge the SW contacts. Could this be sign of shorted diode inside the opto triac?

[fzabkar]

Pin 41 of mcu, which should control the opto-triac is 0.1v (floating?) and gets pulled to zero as I bridge the SW contacts. Could this be sign of shorted diode inside the opto triac?

I would think that the 270 ohm resistor would prevent the MCU pin from being pulled down to ground. In any case, if the output is being pulsed, you won't see this on a multimeter.