OTGW with USB doesn't work.

This Forum is about the Opentherm gateway (OTGW) from Schelte

Moderator: hvxl

roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

hvxl wrote:To make sure we're not chasing ghosts, can you measure the voltage on IC1 pin 18 with a voltmeter in the same situation where you run diagnostics test #5? So, IC1 installed, OK1 removed and power applied. The results should be similar to what test #5 reports for the boiler, so around 3.3V.
Correct, what i measure with my multimeter is equal to the boiler result of test 5 (both are now around 4V :?: ).
hvxl wrote: If that is the case, let's narrow down where it's coming from: Disconnect the gateway power, remove IC1, and reapply power. Now measure the voltage on IC1 pin 18 again. If you still find 3.3V, it's coming from somewhere outside IC1.
I measure still around 4V.
hvxl wrote: If you now measure 0V, the short-circuit must be to some IC1 pin. In that case, remove the power from the gateway and measure the resistance from every IC1 pin to pin 18. You should see infinite on most pins, although a few may give a reading. Any reading below 330 ohm needs further scrutiny.
I don't know if it is still relevant (because i measure 4V and not 0 V) but did this test also.
*between pin 18 and pin 5 of IC1 I measure 316 Ohm (R4 also measures 316 Ohm but it's in the 5% range so I assume this is ok).
*between pin 18 and pin 14 of IC1 I measure 407 Ohm but it looks like the value isn't stable... I checked the soldering around pin 14 and resoldered it on both sides but it is not always stable......When I measure between pin 14 of IC1 and pin 6 of OK1 there is no resistance.
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTGW with USB doesn't work.

Post by hvxl »

The image below highlights the pcb traces connected to IC1 pin 18. One trace on the bottom side of the pcb (blue) between IC1 pin 18 and R4, and one trace on the top side of the pcb (red) between R4 and OK1 pin 5.

Image

With IC1 and OK1 still removed and the gateway powered up, please check any pins and leads close to these traces to see if they show the same voltage as IC1 pin 18. If you find any, you probably have a short-circuit between the trace and that pin/lead.

The only problem is, in theory none of them can really cause the effects you are seeing:
  • A short to R5/R6 would have produced the same voltage for the thermostat and boiler measurements of diagnostics firmware test #5, but it didn't.
  • Either side of R3 or OK1 pin 3 would not have affected the reading when the boiler is not connected because that part if the circuitry would be unpowered and floating.
  • Far side of R4 or OK1 pin 2 would have pulled the voltage hard to ground.
  • OK1 pin 6 would have put 5V on the pin, not 3.3 - 4V.
Actually the last one is the only remotely possible candidate. If there is a semi-short between OK1 pin 5 and 6, that could pull up the voltage on IC1 pin 18. But then you should have found a lower resistance in your measurements between IC1 pin 14 and pin 18. But at least have a close look at OK1 pin 5 and 6 and make sure the soldering on those two pins isn't touching.
Schelte
roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

Actually the last one is the only remotely possible candidate. If there is a semi-short between OK1 pin 5 and 6, that could pull up the voltage on IC1 pin 18. But then you should have found a lower resistance in your measurements between IC1 pin 14 and pin 18. But at least have a close look at OK1 pin 5 and 6 and make sure the soldering on those two pins isn't touching.
Like I told, the resistance seems unstable. When I measure today it was about 170 Ohm but still fluctuating.
I removed the OK1 socket to be able to checked for short below it. I wasnt able to find any but with also the socket removed the voltage on pin 18 of IC1 was gone. Although I was not satisfied in finding anything I soldered the socket in place again. Now with IC1 and OK1 in place I get the following readings for test #5.
Thermostat: 3.07
Boiler: 0.07
Reference: 1.21

Thermostat: 3.05
Boiler: 0.03
Reference: 1.21

Thermostat: 3.06
Boiler: 0.03
Reference: 1.21

Thermostat: 3.08
Boiler: 0.07
Reference: 1.21

Thermostat: 3.07
Boiler: 0.07
Reference: 1.21
With the thermostat and boiler connected I get.
Thermostat: 0.64, 1.88
Boiler: 0.23, 2.59
Reference: 1.21

Thermostat: 0.64, 1.88
Boiler: 0.26, 2.56
Reference: 1.21

Thermostat: 0.65, 1.88
Boiler: 0.27, 0.31, 2.60
Reference: 1.21

Thermostat: 0.65, 1.88
Boiler: 0.26, 2.57
Reference: 1.21

Thermostat: 0.65, 1.88
Boiler: 0.24, 2.57
Reference: 1.21

Thermostat: 0.65, 1.88
Boiler: 0.23, 2.56
Reference: 1.21
But when I load the gateway firmware and connect the boiler and thermostat i only get the messages below over and over again.
23:08:49.870469 T1002010B Write-Data Master configuration: 00000001 11
23:08:49.977611 BD002010B Write-Ack Master configuration: 00000001 11
When I disconnect the thermostat I receive data from the boiler.
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTGW with USB doesn't work.

Post by hvxl »

170 Ohm is exactly what I had hoped to see. A voltage divider of 170 and 330 Ohm makes 3.3V out of 5V, exactly the symptom you had. So that's one problem fixed, but I already had the impression there was going to be more than one problem.

Interestingly the thermostat measurements of test #5 look a lot better now too. But given the fact that the thermostat continuously repeats the same message, it doesn't seem to get the messages from the gateway.

The components involved in sending messages to the thermostat (apart from IC1) are: R7, R9, R11, R12, Q2, Q3, Q4, and Q5.

First check: See if anything changes when you switch to monitor mode (GW=0). I don't really expect it will. Then check all components listed above. Make sure Q4 is a BC548A and all the other transistors are BC558A's. Also check that the mentioned resistors have the correct value.

Next, disconnect the boiler and thermostat and measure the voltages on both sides of each resistor compared to ground (gateway still/again in monitor mode). I expect the following results:
R7: 0V / 23V
R9: 5V / 5V (0V difference)
R11: 24V / 23V (0.6V difference)
R12: 23V / 23V (0V difference)

If that's all OK, short OK1 pin 5 and 6 (with or without OK1 installed, doesn't matter) and measure the voltages again. You should see the following changes:
R9: 0V / 4.5V
R12: 23V / 23.5V (0.4V difference)
Then remove the short between OK1 pin 5 and 6 again and install OK1 if it was removed.

If you still haven't found any issues, please load the diagnostics firmware again, loop X1 to X2 using two wires and run diagnostics test #4.
Schelte
roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

Change to monitor mode didn't change anything, I already tried that yesterday.
BUT Q4 is a BC548B and Q1, Q2, Q3 and Q5 are BC557B..... feel really stupid that I missed it but naive as I am I assumed the kit contained the right parts so I only checked which of the 5 was different and placed that in Q4...

As nobody responds on the emails I send to Info@opentherm-gateway.com I probably have to buy them myself....

Although the transistor numbers are wrong, I did the requested test with the following results.
R7: 0V / 22,9V
R9: 0V / 4,8V
R11: 24V / 23,4V
R12: 23,4 / 23,4V
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTGW with USB doesn't work.

Post by hvxl »

The transistors may be reasonable replacement parts. The important thing is that Q4 is the NPN one, while the others are PNPs.

The R9 measurements are more suspicious. Without a boiler connected and the gateway in monitor mode, IC1 pin3 should be high (5V). Q4 will not be conducting and no current flows through R9. So the other side of R9 will be at the same voltage.

Please check the voltage on IC1 pin 3 and specify which sides of R9 are at 0V and 4.8V.
Schelte
roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

With the above measures i'm not sure it was in monitor mode. When I remove the power is it possible it switches back to gateway?
Anyway, the above measures again but then double checked in monitor mode.
R7: 0V / 24,2V
R9: 5V / 5V
R11: 25V / 25,1V
R12: 24,9V / 24,9V
Where the 1st value is the one closest to IC1

Voltage on pin 3 of IC1 is 5V.
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTGW with USB doesn't work.

Post by hvxl »

Yes, after a reset/poweron the gateway is in gateway mode.

Those values look OK. The measurements for R11 are not what I said they should be, but that's because I forgot to take into consideration that no current is drawn from the thermostat interface. If you put a 100 Ohm resistor or even a short on X1, you should get closer to the values I mentioned. Please verify that.

Then, with the resistor or wire on X1 still in place, continue with the next test that I described on Jan 14 (shorting OK1 pin 5 and 6).
Schelte
roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

With X1 short I get the following:
R7: 0V / 22,5V
R9: 5V / 5V
R11: 23V / 23,7V
R12: 23,1V / 23,1V
Where the 1st value is the one closest to IC1

With both X1 and OK1 pin 5 and 6 short (without OK1 installed) and in monitor mode I measure:
R7: 0V / 20,6V
R9: 5V / 5V
R11: 21,3V / 21,9V
R12: 21,3V / 21,3V
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTGW with USB doesn't work.

Post by hvxl »

OK1 pin 6 is supposed to be connected to VCC, so shorting it to pin 5 would put pin 5 at 5V too. OK1 pin 5 is connected to IC1 pin 18. A high level on IC1 pin 18 in monitor mode should result in a low level on pin 3. This happens in hardware and doesn't require any program to achieve that. R9 is connected to IC1 pin 3, so a low level on that IC1 pin should result in 0V at one side of R9.

Please check the mentioned points with your voltmeter to determine at what point the story stops matching reality.
Schelte
roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

Finally it seems to work now. I also got it working on my Raspberry Pi wit Domoticz.
Thanks for you patient and your help. And sorry if I was a bit annoying sometimes.

Now I hope someone can help me with the WIFI interface. The company (another one that the current Kiwi electronics) I bought the OTGW and this interface doesn't respond anymore. On this forum also nobody seems to be able to help.
roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

Short question. Is there a possibility to make use of the so called " OpenTherm Smart Power"? That saves me some batteries.
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTGW with USB doesn't work.

Post by hvxl »

While I'm glad things are working now, it would be nice to know what had been causing the final problem. Did you find a bad contact somewhere?
Schelte
hvxl
Senior Member
Senior Member
Posts: 1965
Joined: Sat Jun 05, 2010 11:59 am
Contact:

Re: OTGW with USB doesn't work.

Post by hvxl »

Yes, you can remove the batteries from your iSense.
Schelte
roblom
Starting Member
Starting Member
Posts: 40
Joined: Wed Apr 16, 2014 9:18 pm

Re: OTGW with USB doesn't work.

Post by roblom »

That's the strangest, I didn't find a real cause. During the tests it seems quite unstable because the next day the results where different. I think it is mostly caused due to a bad soldering around ok1 but am not sure.
You don't know who's made the WiFi interface?
Post Reply

Return to “Opentherm Gateway Forum”