R8810A, binding problems
Hi all,
a year ago I bought an R8810 Open Therm boiler gateway to take advantage of my boiler's modulation capabilities.
Since the beginning I had issues binding it to my EvoHome controller (the colour, non-wifi, last year model, fw 25). It took more than a few tries to get them bound together, but I eventually succeeded and since then it kinda worked.
"Kinda" worked because there's some behaviour that doesn't look right, for example the R.F. test signals 5 red flashes but it takes several seconds to kick in (e.g. I set the test on the controller, the controller says a test signal has been sent, then it could take as long as 30 seconds before my R8810 starts flashing, that seems too long). Sometimes the controller picks up a scheduled temperature change (or a quick action) and my boiler is operated within 5 to 10 minutes, sometimes it takes even longer.
I have installed a thingy that logs OpenTherm communications between my boiler and the R8810, and, from what I see in that log, the OpenTherm handshake looks fine: as soon as the R8810 sets a parameter, my boiler reacts. The boiler itself provides feedback to the R8810 OpenTherm requests. So, as of the R8810, I'd say the OpenTherm side is good.
By converse, the R.F. part looks dodgy. It seems that the controller and the R8810 cannot reliably communicate to one another, thus making boiler control problematic. It's like some commands are received with some delay, and some are not received at all.
Two weeks ago I was away for the weekend, I turned off my heating system (used the "off" quick action on the controller when leaving my house) and when I turned it on again, the temperature rose for a while and then stopped before the setpoint was reached, and startedd decreasing. I had to turn heating system off and on several times to fire my boiler.
Last weekend it happened again. I switched to the "off" quick action then on again, the heating system fired and then, after one hour, it went off again - I had to stop/start again to resume it.
Yesterday, it happened even without having turned my heating system off: I schedule a lower temperature during the day as I'm at work, and in the evening it didn't fire the boiler as it was supposed to.
Last weekend, my controller reported comms faults to the R8810. Funny, they were not at the same time as the problems mentioned before. Two weeks ago there were no comm faults, nor there were yesterday.
So yesterday I found a cold house... I tried switching the heating system off and on again and it did not fire my boiler. I tried to rebind my R8810, but it didn't work (I was holding the controller about 50 cm away of the R8810 and still it didn't pick up the signal). I tried to reset the R8810 (holding the button for more than 10 seconds) and placing it back into bind mode (holding the button for about 5 seconds). The red led flashed as it was supposed to, but binding at the controller did't work again. I tried at least 5 times, then it worked, without having done anything different. Now it's appearently working (RF test ok, boiler firing), but I still have doubts about the reliability of my R8810.
I had the chance to play with the BDR91 on/off boiler relay in the past, and it was totally different, the binding procedure was always successful and the response was always quick.
Is it the R8810 that behaves like this or it's my unit that's dodgy?
I'm still unsure if it's the ordinary behaviour of any R8810 or if it's mine that's faulty... anyway, since monday (when I had to reset it, rebind it and resync everything), my EvoHome seems A LOT more reactive in terms of boiler control...
Seems weird to write the first answer to my own post... is there anyone else using an R8810? Could you share your experience? binding issues? slow reactions?
Thanks!
Hi all,
Today I had some other strange things going on.
I was away for the day and, before leaving my house at about 11.15, I set te "economy" quick action on the controller. Later in the afternoon I checked with my iPhone app and edited the duration, to last to 19.58. I checked again around 20.30 and schedule was back to normal and temperatures were rising from 18° (the scheduled temperature reduced by the "economy" quick action) to 21° (the scheduled temperature).
I checked a few more times with my Windows phone and everything looked fine. I arrived home at 21.45 and my house was warm.
Then I checked my OpenTherm monitor graph ad discovered it was NOT fine at all:
otggraph.jpg
I can see that my EvoHome told my boiler to stop generating heat for some time (from 21.22 to 21.41), you can see the break in the green bar (heating "on") and the red one (flame status). The thin red line is the CH water temperature. The thin gray line that has a drop at 21.22 and suddenly returns high at 21.41 is the "control setpoint" and the low level is 10°. The thin black line is the relative modulation level, that drops in the same moment to zero.
This means that at 21.22, my R8810 told my boiler to set the water temperature to 10° and stop the circulation. At 21.41 it changed its mind and started calling for heat again (control setpoint jumps from 10° to 75° that's the max. temp. I'm allowing).
At 21.45 my house was warm but not all my zones reached their setpoint. Thus there was no reason at 21.22 to stop the burner for 20 minutes.
The RF check for the heat request yelds the usual response: 5 red flashes with up to 30 seconds delay from when I send the test signal from the controller, plus sometimes the flashes would stop and maybe start again in some other 30 seconds.
No faults in the controller log today.
The OT Gateway logs seem fine, so it seems not to be responsible for a bad power problem nor the boiler is (I can provide logs if someone is interested).
Honeywell guys... is this normal? Do you think I have a faulty R8810? Or worse, a faulty controller?
I don't think there's many of us with r8810's in use in here going by the feedback when I was having problems.
I have the newer wifi controller, but the boiler tends to fire up within a minute or two of me setting the temperature. Usually straight after the valve has opened on the radiator calling the heat. In terms of temperature set point, I find that most of the time the water set point is 45 degrees (this may be the lowest my boiler allows). It only seems to raise the water set point if a zone is taking a long time to reach its set point.
I had a bdr91 initially, and the 8810 is far superior. Rooms stay at set point and barely fluctuate at all. Maybe a .5 degree deviation. Where the bdr91 was overshooting by 2 degrees or more and then oscillating the whole time.
To add to the above- my first r8810 was faulty and wouldn't even power up. So it's not beyond the realms of possibility that you could have a faulty one too.
Thanks SpeedyChuck for your feedback. That tends to reinforce the idea that my unit could be dodgy.
I really hope someone at Honeywell could tell if there's any way to have a suspicious part diagnosed, in your case it was pretty clear there was something wrong, being your unit unable to power up. Mine might be screwed up as well, but in a more subtle way... I had to try the binding procedure many times before having it correctly bound to my controller, and looks like it loses connectivity every now and then...
I'm tempted to open an RMA (if I only knew how to do it), but winter's biting and I need my heating system now, I wouldn't like to send my R8810 someplace for a week or more...
there is an OpenTherm test tool but if you think that there's an issue I would simply recommend exchanging it.Originally Posted by emmeesse68
I would also double check that it is bound correct as an OpenTherm Bridge under SYSTEM DEVICES
It is spaced 300mm away from metal objects I assume?
Last edited by top brake; 21st December 2015 at 03:01 PM.
I work for Resideo, posts are personal and my own views.
On the OpenTherm side, everything looks fine. My OpenTherm gateway detects no communication errors between my boiler and my R8810A. I think the problem is the other side, between my R8810 and my controller. Let me say I hope the culprit is the R8810 and not the controller...
As I described in another post (EvoHome acting up?) I tried to fix the problem by resetting/rebinding my R8810. I had to try the binding procedure several times, holding the controller pretty close to mt R8810, before it eventually succeeded. The impression is that the RF communications between the controller and the R8810 is unreliable, the only way to check as far as I know is the RF test, that yelds 5 red flashes but sometimes is delayed and interrupted.
I confirm it's about 30cm away of my boiler and I can't see metal objects nearby (...as a separate subject, I'd like to understand if this rule is to be applied to all RF components in the evoHome system... my HR92s would all be problematic, since they're less than 300mm away of the radiators...)
Back to my R8810, I put it close to the ceiling, above a kitchen cupboard, line-of-sight to the controller, about 6 meters away. Closest metal object being the boiler, roughly 30cm away. I can provide a photo this evening if it's useful.
Assuming my R8810 is still under warranty (legal warranty in Italy is two years, I bought my R8810 more or less one year ago, and it's always being, like, a bit dodgy), can I ask for a replacement... and how?
Thanks!
Please rebind the R8810A1018 to the evohome control. Please ensure that you're at least 1 meter away when you're doing this. If you're too close it might only bind one way.
Man, I really hope the designers of the Evohome technology are taking note of this issue for the next generation. I wonder if you searched for the words bind/rebind how many times they would show up in this forum?
Each component has different bind/clear bindings sequences. The fact that a component can have multiple bindings (what? why?), which is something we basically *never* want. Each component has different kinds of buttons (some with lights inside them, others without, some just unmarked touch-sensitive areas), and a different look and feel, and battery compartments and ways of accessing those battery compartments (trying to take the batteries out of the Evohome controller is one of the most frustrating things I've had to do in a while).
The endless restrictions on distances and placements of components (not something I have ever come across previously), which implies that the protocol is not very robust and doesn't have an acknowledge/retry aspect to it. The fact that there is no repeater, which seems pretty essential in a product such as this. The insanely complicated process for testing wireless signal quality (it's basically too hard to seriously test every component).
I am very happy with my Evohome installation (8 rooms with a total of 9 rads over 3 floors with a separate Y87 thermostat in one of the rooms), and it seems to be performing flawlessly, with no connectivity issues that I've ever noticed (and no fault logs ever). It's definitely not perfect (I seem to get a fair bit of overshoot), but compared to not having it, I love it. It has made my house much more comfortable (especially the lounge, which used to be too cold), and hopefully will save me some money in the long run. I also enjoy this forum, as I am very much a nerd and enjoy those aspects of the product.
Having said all that, as a techie I find it quite frustrating that Honeywell surely must be aware of these issues, yet doesn't think they are enough of a problem to address them (these products have been out for years).
I reckon one thing that Honeywell lacks is a bunch of opinionated young bucks that question the status quo, and are savvy to how 21st century users want to interact with products. Nest has an amazing user experience (e.g. each smoke alarm comes with a unique bar code on the back that you scan from the app to add it to your account), in a very flawed concept (the whole idea of a single thermostat with a motion sensor that adjusts itself based on people constantly fiddling with it seems really dumb to me). Imagine the Nest user interface guys matched up with the Honeywell tech guys -- it would be magic!
Anyway, rant over, please carry on. And thank you top brake and the other Honeywell folks for your continued and much appreciated help!
Posting Permissions
Reply With Quote