OTGW stops when burner starts

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

Moderator: hvxl

OTGW stops when burner starts

Postby thunderbird » Sat Sep 12, 2020 7:26 pm

I have a OTGW module from Nodosho with a USR-TCP232-T2. I noticed that the module stops sending serial messages when my central heating device starts the burner for hot water. The heating device is an Intergas HR 28/24. I read the serial messages via a telnet session on the TCP module.

Typical output:
Code: Select all
R00000000
BC0000000
R00110000
BC0110000
R001B0000
B601B0000
R801C0000
BC01C2900
R900E6400
B500E6400
R00000000
B40000004
R00000000
B40000004
R80190000
BC0192A00
R10010A00
BD0010A00
R80380000
B40383C00
R00390000
B40393200
R00000000
BC0000000
R00000000
BC0000000
R00000000
BC0000000
R80190000
BC0192A00
R10010A00
BD0010A00



I tried several power supplies without luck (24v, 2.5A). The OTGW unit is used standalone, so without thermostat.
The TCP module is communicating fine all the way. A parallel ping shows no communication problem or outages. When I connect the reset jumper on the board, the OTGW starts sending messages again. This happens in the same telnet session.

So my conclusion is that the problem is in the OTGW module. When no hot water is used, the serial messages are ok for hours (tries a whole night).

Any ideas what the problem could be and/or how can I debug this?
thunderbird
Starting Member
Starting Member
 
Posts: 3
Joined: September 2020

Re: OTGW stops when burner starts

Postby hvxl » Thu Sep 17, 2020 4:25 pm

The OTGW doesn't do anything special based on the DHW mode flag. So if it consistently hangs at that moment, something else must also be changing. Some people reported that their boiler was unable to produce a good Opentherm signal when the flame was on for central heating. Maybe your boiler exhibits a similar problem with hot water. However, if this is the case, I expect that you would get error messages. The OTGW should not get into such a state that a reset is needed.

You can debug this by running test #3 of the diagnostic firmware and then generating a hot water request (iow open the hot water tap). This test requires an Opentherm thermostat to be connected to the OTGW. I hope you have one.
Schelte
hvxl
Senior Member
Senior Member
 
Posts: 1322
Joined: June 2010

Re: OTGW stops when burner starts

Postby thunderbird » Mon Sep 28, 2020 8:56 pm

Thank you for the reply. I do not have an Opentherm thermostat. I looked better at the timings, it is a little bit different.

When I start using hot water, the OTGW reports ok. But then when I stop using water, the boiler has a 3 in the service display. As soon as the 3 disappears, OTGW stops. After a minute, OTGW comes up again.

I have measured the voltage on the Opentherm connectors.
When OTGW is:
  • not connected: 42V (before and after hot water tap)
  • connected and running: 13V
  • 'crashed' after the hot water tap: 37V
Direct after a hot water tap when OTGW is 'crashed', I tried the following:
  • disconnect and connect the Opentherm connector: OTGW keeps silent
  • power off and on OTGW: OTGW immediately started sending reports

I will try to get an Opentherm thermostat...
thunderbird
Starting Member
Starting Member
 
Posts: 3
Joined: September 2020

Re: OTGW stops when burner starts

Postby hvxl » Tue Sep 29, 2020 12:47 pm

There seems to be a hardware fault in your OTGW. You should never have 37V on connector X2. The voltage is limited by D10, Q1 and either D5 & D8, or D6 & D7, So you can only have a maximum of 15V + 3 p-n junctions, which amounts to something between 17.0 and 17.5 volts. No amount of bugs in the firmware can alter that.

42V from the boiler is very high, although allowed by the Opentherm specification. But it exceeds the maximum allowed Vceo of a BC558. So it is possible that Q1 was blown. If that's the case, you may want to replace it with a BC557 or BC556. I appreciate replacing Q1 may be difficult on an OTGW from Nodoshop because they used SMD components.
Schelte
hvxl
Senior Member
Senior Member
 
Posts: 1322
Joined: June 2010

Re: OTGW stops when burner starts

Postby thunderbird » Tue Sep 29, 2020 7:07 pm

Hmmm, I made a mistake. My multimeter was on AC while measuring voltage. Now the DC voltages:
  • Disconnected: 19.5V, steady before, while and after hot water tap. My multimeter doesn't measure any spikes or drops.
  • Connected: 6-7V
  • Connected and after hot water tap: 17-17.5V

Today I saw several times - just before the 'crash' - an orange led blinks one time (RB3, pin 9). Normally RB4 blinks roughly every second. RB4 stops blinking when in 'crash' state.

It takes exact 60 seconds before OTGW 'resets' itself. Is there a fail safe mechanism which resets the PIC after 60 seconds?

I agree that the changing voltages couldn't be triggered from the software of OTGW as the boiler-OT schema is isolated. Main question is, how can the voltage changing (remember the voltage does not change when OTGW is not connected).

So bottom line the OTGW comes in a crash or freeze state. Reconnecting did not help, only a reset of OTGW. While in crash state the voltages goes from 6 to 13 and drops back to 6 after reset.
thunderbird
Starting Member
Starting Member
 
Posts: 3
Joined: September 2020

Re: OTGW stops when burner starts

Postby hvxl » Tue Sep 29, 2020 11:59 pm

Those voltages look much more like I would expect.

The LED connected to RB3 (pin 9) is normally referred to as LED A. Unless changed from the defaults, it indicates that the boiler's flame is on.

There is a fail safe mechanism that restarts the program when 64 messages have gone unanswered. That would be after about 64 seconds. But then you should still see LED B (RB4, pin 10) flashing until the reset, which you don't.

The voltage can change because the OTGW drives the opto-coupler OK1A. That's how the OTGW communicates to the boiler.


It would be interesting to see what the OTGW reports on the serial interface when it recovers on its own after the 60 seconds freeze. If it says "OpenTherm Gateway 4.2.5.3", then it rebooted. In that case the output of the PR=Q command indicates the reason for the reboot.

Another intriguing remark you made is that LED A flashes just before the problem starts. The fact that the boiler's flame is on is not immediately reported on the OT interface. It only becomes known when the OTGW requests the status (MsgID 0). So the initial suspicion that something changes on the boiler side when the flame is on sounds less likely. Because in that case the problem would start before the OTGW is aware that the flame is on.

The problem now seems to happen when the OTGW is informed that the flame is on. As you observed, that causes LED A to be switched on. If there is some short circuit in the device, or if the extra power consumption is a problem for the power supply, the LED turning on could actually be the trigger for the problem. You can easily check if this is the case by assigning a different function to LED A. There is no function Z, so by issuing the command LA=Z, the LED will never switch on. If the problem still occurs after that, we have at least ruled out a hardware issue.

However, the log you provided in your first post does not contain any status messages where the flame is on. So that does not match with the flashing of LED A before the problem. Or was your log not showing the last messages before a problem occurrence? When collecting a log in the future, can you please use the otmonitor program? That tags messages with a time stamp, and it translates the messages so I don't have to decode the hex numbers by hand.
Schelte
hvxl
Senior Member
Senior Member
 
Posts: 1322
Joined: June 2010


Return to Opentherm Gateway Forum

Who is online

Users browsing this forum: No registered users and 1 guest