Page 1 of 2

lost OT communication after restart

Posted: Fri Oct 12, 2018 9:41 am
by Philip_Inc
I use OTG in stand-alone working mode. Without thermostat with boiler of Thermona.
Everything worked well for a few months but few days ago after some hours of work or after power outage OTG stops communication with boiler. The same with GW=R command.

Code: Select all

10:17:35.314614	OpenTherm Gateway 4.2.5
10:17:35.400497	Command: SC=10:17/5
10:17:35.470303	Command: SR=21:10,12
10:17:35.495786	Command: SR=22:7,226
10:17:35.520152	Thermostat disconnected
10:17:35.556328	SC: 10:17/5
10:17:35.569668	SR: 21:10/12
10:17:35.592700	SR: 22:7/226
10:17:36.253090	R00000000	Read-Data 	Status: 00000000 00000000
10:17:37.269427	R00000000	Read-Data 	Status: 00000000 00000000
10:17:38.284816	R00000000	Read-Data 	Status: 00000000 00000000
10:17:43.661204	R00000000	Read-Data 	Status: 00000000 00000000
10:18:00.000881	Command: SC=10:18/5
10:18:00.161051	SC: 10:18/5
10:19:00.001411	Command: SC=10:19/5
10:19:00.214855	SC: 10:19/5
10:20:00.000968	Command: SC=10:20/5
10:20:00.150603	SC: 10:20/5
The only thing which helps is placing a wire on X1 interface and it starts works well for a few hours.

I already tried to reflash PIC, but nothing changed. I have very strange test results with diagnosis firmware:

Code: Select all

Opentherm gateway diagnostics - Version 1.1

1. LED test
2. Bit timing thermostat
3. Bit timing boiler
4. Delay symmetry
5. Voltage levels
6. Idle times

Enter test number: 3
4403.
3910.
4583.

Opentherm gateway diagnostics - Version 1.1

1. LED test
2. Bit timing thermostat
3. Bit timing boiler
4. Delay symmetry
5. Voltage levels
6. Idle times

Enter test number: 5
Thermostat: 2.43, 2.50
Boiler: 2.56
Reference: 1.22

Opentherm gateway diagnostics - Version 1.1

1. LED test
2. Bit timing thermostat
3. Bit timing boiler
4. Delay symmetry
5. Voltage levels
6. Idle times

Enter test number: 6

Opentherm gateway diagnostics - Version 1.1

1. LED test
2. Bit timing thermostat
3. Bit timing boiler
4. Delay symmetry
5. Voltage levels
6. Idle times

Enter test number: 6
B: 9223.107
T: 14326.677
B: 19429.080
Please be noted that there is nothing connected to X1.

Re: lost OT communication after restart

Posted: Wed Oct 17, 2018 9:26 pm
by hvxl
You are running diagnostic tests that are intended to be executed with a thermostat and boiler connected, without a thermostat. Then you shouldn't be surprised that you get meaningless results.

I'm not sure why the OTGW stops sending messages to the boiler, but in the log you provided it never received (or recognized) any response from the boiler for the ones it did send. So there's some problem there.

If you put a wire across X1, the OTGW will instruct the boiler to start heating. That's not a situation you would normally want for a few hours.

Re: lost OT communication after restart

Posted: Thu Oct 18, 2018 3:30 am
by Philip_Inc
The boiler stops answer without any reason, the OTGW stops sending without any reason. But closing of X1 they start normal communication.
I need to understand what to replace: OTWG or boiler PCB.

Re: lost OT communication after restart

Posted: Thu Oct 18, 2018 9:22 pm
by hvxl
If you suspect hardware issues with the OTGW, I suggest you go through the troubleshooting procedure.

Re: lost OT communication after restart

Posted: Fri Nov 16, 2018 10:41 pm
by Philip_Inc
I've checked everything from the troubleshooting section you mentioned and didn't a problem there. I talked with other user of OTGW and he confirmed that it not works without shorted X1.
My symptoms: after a few days of normal functioning it lost any communication with boiler. When I short X1 for a second it start working normally.

Could you please specify what functions are available when X1 is shorted? I see that I can switch heating off by CH=0. Is it possible to change the temperature by CS or SH? I see one of them is works but still not figured which one. I don't have a way to use your software because OTGW is connected to PLC by serial port.

Re: lost OT communication after restart

Posted: Mon Dec 03, 2018 6:46 pm
by hvxl
As indicated before, the OTGW is not intended to be used that way. So I don't know what functions are available. And I'm not interested to find out.

What I don't understand is why you keep going in this direction. So, you have a problem with the OTGW when used in the normal configuration? Then you try to use it in some unusual configuration and you have a different problem. Why do you then focus on the problem with the unusual configuration rather than the problem in the configuration that is supposed to work?

If you want to investigate the original problem, I can try to help. If you want to get things working with a shorted X1, you're on your own.

Re: lost OT communication after restart

Posted: Wed Dec 05, 2018 11:15 pm
by Philip_Inc
As I reported previously it losts a communication over OT after some hours of working or restart. If the X1 is shorted there is no such behavior.
I will be very glad if you can help me to resolve this issue and make it works without X1 shorting. I already reimaged it for a few times with the latest firmware.

Re: lost OT communication after restart

Posted: Sat Dec 08, 2018 7:09 pm
by hvxl
In your last post you indicate that the OTGW keeps working when X1 is shorted. Earlier you said that it "works well for a few hours" in that situation. Which is it? Because if it also stops working after a while with X1 shorted, the problem is likely in a completely different area than when the problem only happens with nothing connected to X1.

In any case, the first thing to check is the A/D value when the OTGW stops communicating. You do this by running the command DP=57. Repeat the command a couple of times to get a good picture of the measured values. Also run the commands DP=7e and DP=61. These last two only need to be run once. Finally use DP=00 to disable the debugging again.

Re: lost OT communication after restart

Posted: Mon Dec 10, 2018 11:47 am
by Philip_Inc
Thanks a lot for your assistance!
OTGW works fine for a long period of time if X1 is shorted. And it works for a few hours or until the first power reset when X1 is opened.

I had removed a jumper over X1 and made a test. Before reboot (power disconnect) registers had following data:
7E = 20
57 = 70-78 (values in range)
61 = 07

After reset of power:

Code: Select all

Thâ–’OpenTherm Gateway 4.2.5
Thermostat disconnected
R00000000
R00000000
R00000000
DP: 57
57=81
DP: 7E
7E=40
DP: 61
61=04
DP: 57
57=81
DP: 57
57=81
DP: 57
57=81
DP: 57
57=81
DP: 7E
7E=40
DP: 61
61=04

Re: lost OT communication after restart

Posted: Wed Dec 12, 2018 6:58 pm
by hvxl
Yeah, that's not clear at all. Please clarify "OTGW works fine for a long period of time if X1 is shorted". Does it stop working after that, or do you just end the test then?

How do you perform the reset? Power reset sounds to me like you remove the power and then reapply it. But in that case, how do you get it to work initially? You always have to start by applying power. That results in a power reset. Do not you get the problem if you do a reset using GW=R?

The reported values look OK. In fact the only suspicious values are the ones before the reset. 70-78 for address 57 are too low to consider the thermostat disconnected. But the gateway has bit 61:2 set anyway. So nevertheless, it should operate in NoThermostat mode.

I have done a test with an OTGW without any connections. After a reset it reports "Thermostat disconnected" and starts spitting out R00000000 messages at a rate of around one per second. Do you have the boiler attached? Do you get the same behavior when you disconnect it?

Can you dump a few more values? Please give the following arguments to the DP command: 5E, 55, 5A, 54, 67, 6D, 05, 9C. Repeat the sequence after a minute or so. Then do the same after you reproduced the problem.

Edit: More than 6 hours later, my test setup is still sending out its messages on the boiler interface.

Re: lost OT communication after restart

Posted: Thu Dec 13, 2018 1:08 am
by Philip_Inc
If I place a wire across contacts of X1 terminal OTGW communicate with boiler without any issues. At least for 14 days. It is not a solution for my task because I cannot manage it in such mode. But technically communication works well.
I do reset by removing of power and then reapplying it. It really doesn't start after restart. In order to start it I have to short X1 for a second or wait for hours (after about 8-24 hours it starts by itself). I noticed same problem after GW=R also but not always.

When the boiler is connected after power reset OTGW sends R000... 3 times only. OT communication LED is not blinking after that. If I disconnect the boiler and do power reset or GW=R it sends R0000... and OT activity LED blinks for while. It looks like if the OTGW doesn't receive any communication from boiler within 3 seconds after restart it cease any OT communication attempts.

Code: Select all

Boiler disconnected:
7E=10
57=7C (7F)
61=04
5E=10
55=00
5A=80
54=3D
67=00
6D=01
05=10
9C=83

Boiler connected, no OT communication (problem):
5E=10
55=00
5A=80
54=39 (05,06,12,27,07,26,0F,3C...)
67=00
6D=3B (FF)
05=10
9C=03 (83)
7E=40
57=7D
61=04

All OK:
5E=00 (10)
55=00-0F
5A=98
54=3D (38,39)
67=00-01
6D=00-01
05=10 (18)
9C=83
7E=40
57=7A-80
61=07

Re: lost OT communication after restart

Posted: Sat Dec 15, 2018 1:10 am
by hvxl
This shows that the OTGW is aware that it hasn't sent a message to the boiler for a long time (6D increasing until it reaches FF). It also has determined that it is time to send a message (bit 4 set in 5E). It would do that as soon as it is idle. But apparently that never happens. This conclusion is reinforced by the fact that the value for 57 doesn't change when you have the problem. Measuring the line voltage is also only done when the device is idle. If it never goes idle, no new measurement is ever done, and you keep seeing the same value.

To check that the OTGW feels it is busy, you can dump address 12. Bit 2 indicates the busy state. So, 04 means busy, 00 is idle. The only reason I can think of for the device to not become idle is if there is constant chatter from the boiler (5E bit 3 indicates that it's not the thermostat side). This chatter doesn't match Opentherm communications, so it would result in an error. But also errors are not reported until the device becomes idle, which is why you don't see an error report. But you can check the stored error by dumping address 65. If you configure a LED for transmission errors, I expect it will light up constantly. In case you don't have any LED attached, you can dump address 66 (anything other than 00 means an error was detected) or address 06 (bit 3 is LED A, bit 4 is LED B, bit 6 is LED C, bit 7 is LED D).

If dumping this latest set of addresses confirms my theory, we have an idea of what is happening. But resolving it is another story. At the moment I have no ideas how to fix it, while still conforming to the Opentherm specs.

Re: lost OT communication after restart

Posted: Mon Dec 17, 2018 1:50 am
by Philip_Inc
Thanks a lot for your efforts!
12=04
65=00
66=00
06=DE

OK, if your theory confirms why does it recover from this state by shorting of X1?

Re: lost OT communication after restart

Posted: Sun Dec 30, 2018 6:54 pm
by hvxl
No, that doesn't confirm my theory. As I explained, I expected the values on address 65 and 66 to be something other than 00. The value of address 06 doesn't have any meaning if you don't reveal which LED you configured to show transmission errors.

Can you dump the values of addresses 38 and 5D a dozen times or so?

(I hope it goes without saying that you need to dump all of the values I ask for when you have reproduced the problem.)

Re: lost OT communication after restart

Posted: Fri Jan 25, 2019 1:47 am
by Sjoerd
Hello Schelte,
After a long time (firmware PIC was corrupt after update) I connected the OTGW again.
I think I see the same behaviour
The OT monitor only got messages like
R00000000 Read-Data Status: 00000000 00000000
after shorting X1 I suddenly saw data coming in.
In the meantime I tested to I-sense thermostat to see if there is an OT signal (normally I have a floorheating CF2 ON-OFF) thermostat.
Connecting the I sense I noticed a voltage on 5-6 volt fluctuating as mentioned in the trouble shooting guide
When initial measuring the X2 voltage I have a stead 18.41
After shorting the X1 I noticed the voltage start fluctuating between the 5-6v
Now I got the data coming in as well.

So it looks a bit like above, only data is coming in after an X1 input.

While writing this I got within 10 min's an Error 03 and data stops agian.

Code: Select all

00:01:45.168790	R801C0000	Read-Data 	Return water temperature: 0.00
00:01:45.330643	BC01C2599	Read-Ack  	Return water temperature: 37.60
00:01:46.230838	R900E6400	Write-Data	Maximum relative modulation level: 100.00
00:01:46.330691	B500E6400	Write-Ack 	Maximum relative modulation level: 100.00
00:01:47.297044	R80750000	Read-Data 	CH pump starts: 0
00:01:47.627017	B407550A0	Read-Ack  	CH pump starts: 20640
00:01:48.367409	R80000100	Read-Data 	Status: 00000001 00000000
00:01:48.528578	Error 03
00:01:49.417484	R80000100	Read-Data 	Status: 00000001 00000000
00:01:50.418409	R80000100	Read-Data 	Status: 00000001 00000000
00:01:51.422911	R80000100	Read-Data 	Status: 00000001 00000000
00:01:52.428689	R80000100	Read-Data 	Status: 00000001 00000000
Switching X1 on - off - on does not help
Voltage is still 5-6 v
Connecting the Isense to X1 display Themostat dat, but boiler does not response

Tried several options, reset, reconnect etc, no communicatio with the boiler again...

Firmware is 4.2.5
As far as I know I did not had this with the prev firmware (2 year ago?)
I can not find the prev firmware to test

boiler Remeha Calenta