Page 1 of 1

Strange jumps in measured room temperature

Posted: Sun Dec 17, 2017 6:07 pm
by HarmOtten
Strange temperature jumps_.png
Strange temperature jumps_.png (50.03 KiB) Viewed 10670 times
As you can see in the chart, it jumps up +/- 1,5°C at a certain temperature, and jumps down later during the night.
I have no idea what is causing this behaviour. But as you can see it is quite reproducable. I'm quite sure this wasn't the case without the OTGW.
I checked with a second measuring device to find out that the 1,5°C higher temperature is incorrect.
The wrong temperature is also visible on the thermostat itself: Honeywell Chronotherm Modulation T8851M1000.
Does anybody have an idea what is causing this? Any help is much appreciated!

Re: Strange jumps in measured room temperature

Posted: Wed Dec 27, 2017 5:31 pm
by hvxl
It is unclear what the source is of the data that you display in your graph. However, it is unlikely that this is caused by the OTGW on its own. Also, my T8851M1000 never behaved like this. Either something is sending some strange command to the OTGW, or something is otherwise interfering with the thermostat. Do you have anything connected to the RT terminals of the thermostat? Is there something near the thermostat that switches on between 16:00 and 6:00 (guessed from your graph) and produces heat?

The occurrences appear to happen at the same time every day. (The OTGW doesn't have a clock, so it wouldn't be able to do that even if I wanted to.) So it should be relatively easy to capture a log of such an event.

Re: Strange jumps in measured room temperature

Posted: Fri Dec 29, 2017 4:24 pm
by HarmOtten
Thanks for your reply!
The source of the data of the chart are the Room Setpoint (0x10) and Room Temperature (0x18) messages received from esp-link.
I have nothing connected to the RT terminals of the thermostat, only A and B are connected to the OTGW.
The temperature around the thermostat does not change so rapidly like the chart displays, and there is no other equipment nearby.
It looks like it happens every day at the same time, but that is just a coincidence.
I can actually see the same (wrong) temperature on the display of the thermostat. So I am getting more and more in doubt, that this issue might have been the case before the OTGW was installed, but back than I could not measure the temperature that accurately.
I will first try to figure out if the thermostat is wrong.

Re: Strange jumps in measured room temperature

Posted: Mon Jan 15, 2018 6:33 pm
by HarmOtten
I did a simple test, just removing the gateway from between the Central heating and the thermostat. Then the measured temperature on the thermostat is correct! in this case 18°C. After connecting the OTGW, the measured temperature is immediately incorrect, in this case 19°C, which is too high.
As you can see in the chart above, this issue starts every time around 17.7°C or higher, and stops when the temperature falls below 16.8°C.
I do not know the cause of this, but I suspect there is something wrong in the OTGW firmware.
Hope you can help!!

Re: Strange jumps in measured room temperature

Posted: Sat Feb 03, 2018 1:33 am
by HarmOtten
@hvxl,
Is this something you can look into, or is there somebody else who can do that?

Re: Strange jumps in measured room temperature

Posted: Sat Feb 03, 2018 11:16 am
by hvxl
I have no more suggestions to offer based on your scanty graph. And I haven't seen the log I suggested you collect of such an event.

Re: Strange jumps in measured room temperature

Posted: Sun Feb 04, 2018 4:11 pm
by HarmOtten
Strange temp jumps.png
Strange temp jumps.png (26.63 KiB) Viewed 9467 times
In more detail:
Jump down:
Strange temp jump down.png
Strange temp jump down.png (11.98 KiB) Viewed 9467 times
Jump up:
Strange temp jump up.png
Strange temp jump up.png (15.13 KiB) Viewed 9467 times
Logs:
Strange temp jump down - log.pdf
(188.95 KiB) Downloaded 512 times
Strange temp jump up - log.pdf
(241.64 KiB) Downloaded 460 times
Please let me know if you need more information.
Thanks!

Re: Strange jumps in measured room temperature

Posted: Mon Feb 19, 2018 1:38 pm
by HarmOtten
Did you have a chance to look at the log I provided?

Re: Strange jumps in measured room temperature

Posted: Sun Feb 25, 2018 2:33 pm
by hvxl
You really are intent on making it as hard as possible for people to help you, aren't you? Along with the OTGW I made the otmonitor application, primarily to get a well-defined log when people have problems. But you decided to use some other method to create a log, so that it would be unclear to me what is and isn't logged, and it doesn't decode the messages. With your logging method you thought a nice additional complication would be to use an inconsistent number of digits for the parts of your timestamp. Then finally you considered what would be the least convenient format to provide the data in and came up with PDF, which indeed is much more difficult to automatically process than a plain text file.

After conquering all the obstacles you put up, it turns out there really aren't any jumps. The temperature is steadily changing over several minutes. In the log of what you call "jump up" you didn't include much history to really see what's going on, but you can just see that the thermostat has instructed the boiler to produce heating. As a result the temperature is going up. Nothing unusual. You only seem to have quite a bit of overshoot. But the OTGW is basically just passing on the messages between the thermostat and the boiler, without making any modifications. The only exceptions are messages with ID 16 and 18, so the OTGW does provide the current time and an override setpoint to the thermostat. I don't see the SC commands, which clearly must have been issued, so I guess your logging tool fails to record commands.

Your "jump down" log shows absolutely nothing strange. Sure, the temperature is going down over a period of time. But that's exactly what the thermostat reports. Again your log is too short to see what led up to this. But it shows no evidence that the OTGW is interfering with the communication between the thermostat and the boiler in any way.

Re: Strange jumps in measured room temperature

Posted: Wed Feb 28, 2018 1:41 pm
by HarmOtten
Thank you very much for your extensive help

Re: Strange jumps in measured room temperature

Posted: Thu Mar 01, 2018 6:17 pm
by hvxl
Looking at your "Living Room" graph from around 08:45 to 14:50, it looks like your system is struggling to get the house up to the desired temperature when the setpoint changes to 18 degrees at 12:10. Maybe the thermostat is first trying to get there while running the boiler at its minimum load (this is the most energy efficient mode of the boiler and is usually used to maintain a temperature). Then at 14:34 it switches to a more energy hungry method and overshoots the desired temperature. Or maybe there are requests for hot water between 12:15 and 14:30 that hamper the system in its attempts to heat the house.

To prove or disprove those theories a complete log from 12:00 to 15:00 would be needed.