DIY otgw (usb) 2,5V on pins 5-18 of IC1 when X1 and X2 wired

[2mv]

When I troubleshoot my otgw I measure 2,5V on pin 18 IC1 when X1 and X2 are wired.

First this first: I have assembled the otwg myself and I'm no experienced solder, so it is possible it could be prone to all sorts of errors.
I have connected the owtg on the boiler and thermostat (Remeha Tzerra 28c and Remeha iSense) and do not get readings from the boiler. (I did so from the termostat)
So I've started the troubleshoot actions.

This is what I have done so far:
Opentherm interfaces:
Resistance X3 (without 230V connected): 1 kOhm
Between pins 5 and 14 of IC1 and pin 15 and 16 of IC2 (230V connected): 5V, polarity also ok:

Opentherm interfaces:
Voltage on X1: 25V DC
Current of X1: 7mA
Current of X1 when pin 3 of the IC1 connected to pin 5: 25mA
Installed octo-coupler,voltage on X2/X2: 16V
Connected pin 2 of the IC1 socket to VDD (e.g. pin 14), the voltage on X2: 5V

Voltage on pin 18 of the IC1 socket with pin 3 left open: 2,5V

(and again when it is connected to ground: 4V)

Why could I have 2,5V on pin 18? Well above the 1,25V limit. What do I have to check or change?

[hvxl]

With pin 3 left open you measured 7mA on X1. With X1 looped to X2, the opto-isolated part of the OTGW boiler interface has no other power source than this 7mA. Even if all of that current goes through R2, that wouldn't cause more than 0.7V across the OK1B LED. That's not enough to make it light up. And R2 looks like it's the right value on your picture. So this is a bit of a mystery.

Let's first make sure that the situation happens due to activation of OK1B. Remove the power, pull OK1, reapply the power and check that pin 18 then shows 0V. And while you're at it, measure the voltage between pin 3 and 4 of the OK1 socket. Assuming you found 0V on pin 18, remove the power again and measure the resistance between pin 3 and 4 of the OK1 socket (OK1 still removed). You should get around 100 Ohm.

If that didn't provide any unusual readings, put OK1 back in and power the gateway up again. Now measure the voltages (expected values in parentheses) across R1 (0.5V), R2 (1.2V), R3 (0.6V), D9 (1.7V), D10 (15V).

[2mv]

Your help is highly appreciated!

Ok, there we go:
X1 and X2 still wired, OK1 removed, IC1 pin 18-5: 0V
pin 3 and 4 of the OK1: 3,5-4V
remove power, resistance between pin 3 and 4 of the OK1: 4kOhm

OK1 back in, power back on:
R1 (0.5V) 0,65V
R2 (1.2V) 0,0V
R3 (0.6V) 0,75V
D9 (1.7V) 1,85V
D10 (15V) 14,5V

I did not get any reading on R2.

So I removed R2 and solder a new one and now get the measurements:
X1 and X2 still wired, OK1 removed, IC1 pin 18-5: 0V
pin 3 and 4 of the OK1: 0,3V
remove power, resistance between pin 3 and 4 of the OK1: 150 Ohm
( I don't believe I have a highly accurate resistance meter, but it gives a ballpark figure, if I measure other 100 Ohm resistors I also get a reading of 150 Ohm on my meter)

OK1 back in, power back on:
R1 (0.5V) 0,6V
R2 (1.2V) 0,3V
R3 (0.6V) 0,75V
D9 (1.7V) 1,1V
D10 (15V) 14,0V

So there was a problem with R2, but what problems are remaining? The voltage is still not correct?
(I already did solder D9 again, but to no avail.)
Any suggestions I could try?

[hvxl]

Sorry, I mixed up R1, R2 and R3. The expected values should have been R1: 0.6V, R2: 0.5V, R3: 1.2V. To my defense: it was late already?

Some of your measurements are a bit lower than I expected, but at least you should not find much more than 0V on IC1 pin 18 anymore in the trouble shooting step where you ran into problems last time. Continue with the trouble shooting instructions (or just give it a shot and see if communication with boiler happens to work now).

[2mv]

Of course I immediately did hook up the device to the boiler/thermostat.
It works, I get readings of the roomtemperature and boiler water temperature. So both appliances give info. Great!

The resistor (R2) I replaced had the resistance value as advertised, so I suspect it malfunctioned due to bad soldering. With my magnifying glass it looked ok, but maybe the solder didn't flow through properly of some sort.

Thanks again for your great help,
Best regards, Herbert