Home Automation Domotica Forum Europe, Bwired Forum

Hi,

I got a OTGW from someone who couldn't get it to work. I received it today, and first had a look at the troubleshooting page: http://otgw.tclcode.com/debugging.html#power

Things i checked:

Power:
Resistance between X3 (230V): Yup, few kOhm, so OK
5V +/- 1% between pins 5 and 14 of IC1 and pin 15 and 16 of IC2 > Yup, it is there

OT
Next measure the voltage on X1. This should produce a value somewhere between 18V and 34.5V > Yup, 23V DC or so
measure the current between the two terminals of X1. It should be between 5mA and 9mA. > Yup, 7mA or so
connect pin 3 of the IC1 socket to ground (e.g. pin 5), the current between the two terminals of X1 should be between 17mA and 23mA > Yup, 19mA or so
find a voltage between 15V and 18V on X2 (and X1). Yup
If you connect pin 2 of the IC1 socket to VDD (e.g. pin 14), the voltage on X2 should drop to between 5V and 7V. Yup

Measure the voltage on pin 18 of the IC1 socket with pin 3 left open: 0V
and again when it is connected to ground. Fail! > 0V (instead of 5V)

So there it fails, the voltage on pin 18 stays 0V, even when i connect pin 3 to GND (pin 5).

What part do i need to replace now? Q4 (BC548) ?

I also see the person i got the OTGW from has used BC557B for Q1, Q2, Q3, Q5 instead of BC558A ... do i also need to change this, or is 557 also sufficient?

The BC557 is for higher voltages. For the rest it should be identical to the BC558 series.

https://www.fairchildsemi.com/datasheets/BC/BC559.pdf

Ok, so that shouldn't be the problem i guess?

No. And as far as I can see on an old picture, I've also used the BC557B's for my OTGW...

ThinkPad wrote:What part do i need to replace now? Q4 (BC548) ?

Unlikely. All the measurements on the slave interface side produced the expected results, so the problem is probably not on that side. Also a number of components on the master interface must be OK, judging by some other measurements that produced the correct results. The problem must be somewhere near D9, R2, R3, and OK1B.

With X1 and X2 looped together and pin 3 of the IC1 socket connected to ground (IC1 removed), what voltages do you measure across D9 and R2?

hvxl wrote:

ThinkPad wrote:What part do i need to replace now? Q4 (BC548) ?

[...]
With X1 and X2 looped together and pin 3 of the IC1 socket connected to ground (IC1 removed), what voltages do you measure across D9 and R2?

On D9 i measure: 4.85 - 4.90V (depends on which side (left/right) of the diode i put the black/red probe of my multimeter)
On R2 i measure: 0.37V

I expected 4.7V across D9, but your figures are in the ballpark (although they shouldn't differ depending on which way around you measure). In theory that same voltage is divided over R2 and R3 relative to their values. So R3 sould get 220/320th (3.3V) and R2 100/320th (1.5V). At around 1.2 OK1B starts conducting, preventing the voltage over R2 to rise any higher. So I would have expected around 1.2V across R2. You're only seeing 0.37V, which means R2 only gets about 1/13th of the voltage.

Possible causes:

• R3 has a higher resistance than the 220 Ohm it is supposed to be, or there's a bad contact on one of its wires.
• R2 has a lower resistance than the 100 Ohm it is supposed to be, or there's a partial short across R2.
• OK1B is damaged, or there's a partial short between pin 3 and 4.

So, check the colour bands on R2 (brown-black-brown-gold) and R3 (red-red-brown-gold) to make sure they are the correct values. If that's OK, pull OK1 out of its socket and measure the voltage across R2 again. If that's now around the expected 1.5V, there's something wrong with OK1. If you still measure a voltage below 1V across R2, check the soldering of R2, R3 and OK1 pin 3 and 4.

Thank you for your detailed explanation! I will check it this afternoon.

First thing i did when i received (the already assembled) OTGW was reheating all the solder joints, to exclude a bad solder joint. So that can't be the problem.

Ok, i think i found it.

R2 colors are correct: brown - black - brown - gold
But R3 is not, it is: brown - red - red - gold (1,2kOhm) and it should be 220 Ohm according to the parts list & your message.

I also checked the colors of the others resistors, and R8 is also not correct. I think R3 & R8 are switched. They have the same colors but in a different order (R3 should be: red - red - brown - gold, R8 should be: brown - red - red - gold).

I am now going to swap them, and check the voltages again

This fits your earlier measurements exactly. An R3 of 1.2 kOhm is 12 times as big as R2, so R3 gets 12/13th of the voltage and R2 only gets 1/13th. The value of R8 is nowhere near as critical as that of R3. That's why the checks that tested that part still worked in the static case. It would probably have caused problems when actually sending opentherm messages to the boiler.

Ok, swapped the resistors around.

But i still get the same readings unfortunately:

GND - pin18: 0V
GND - pin 18 (pin 3 connected to pin 5): 0V (where it should be 5V)

In above situation, X1 is bridged to X2 with two wires. IC removed, OK1 installed.

Optocoupler defective?
Or maybe reversed in his socket? When i have the PCB in front of me, X1 and X2 on top, 230V bottom, the dot on the OK is in the down right corner, pointing towards R7, R8 etc

That's the correct orientation for the opto-coupler. Which voltages do you now measure across D9 and R2?

To check the opto-coupler, you can move it up two pins. That will put OK1A in OK1B's place. Your earlier measurements suggest that OK1A is fine.

D9: 4,48V
R2: 1,15V

(X1 & X2 connected, pin3 IC to GND).

When i move OK1 up towards X1 X2 (so bottom 2 rows of socket are empty), measurements stay the same, 0V on pin18 in both situations.
When i move OK1 down (two empty rows near X2) same measurements...

OK, with OK1 back in its original place and the rest still the same, what voltage do you measure between OK1 pin 5 and 6? Can you also measure (on a powered down gateway) the resistance between IC1 pin 18 and pin 5?

Ok, i was measuring on pin 1 instead of 18 (no pin numbers in your schematics, so i had to Google for the chip layout, confused myself there i think).
pin 3 open: between GND and pin 18: 0V
pin 3 to GND: between GND an pin 18: 2,25V


Pin 5 & 6 are this one: ?

I measure 13,6V DC on these two pins.

IC pin 18 and pin 5 is 0,32kOhm. But my multimeter probes aren't that accurate when measuring ohms, they have a fairly high resistance by them selves.

So the voltages are OK now i think?