New RC2000 acting strange

After having my first one destroyed by lightning and having it's replacement come in with only a RC1000 firmware, the 3rd RC2000 I have is functioning but acting strangely.

Came home from work today and noticed it's back light was green instead of blue and it was set to 72 but the arrow next to cool was blinking denoting it was waiting a period to restart, but the house was up to 75 degrees. I asked the wife why it was set to 72 and she said it was getting warm unable to get down to the usual 74. I waited and it eventually started up after time out period, started on first stage, then started second stage as it should with such a large temp difference. I reset the back light to blue and thought everything was ok and went about my business. About 10-15 minutes later I noticed the A/C was off with the stage arrow blinking again and the back light back to green, WTF?

So my questions are:

1. Why would the Omnistat shutdown before it reached the cool set point and go to a blinking stage arrow?
2. Why does the backlight set itself to green? Some kind of energy level thing? Where is this coming from?
3. Could this have anything to do with Zigbee or a smart meter? My original RC2000 didn't have all the zigbee or energy level stuff in the menu.

Seems to be running ok now at night, but it's not having to keep up with daytime heat. really hope this unit isn't defective too, this is getting tiresome.

Sounds like it's not getting enough power.
When it starts a cycle it's overloading the 24VAC and is powering off, which stops the cycle and repowers the Tstat in the powerup default mode - back to green.
You may need a power module.

Do you have a common at the Tstat?

The RC-2000 thermostat is designed to work with most single state conventional 4-wire HVAC systems
(without a transformer common). However, if the RC-2000 “resets” when calling for heat or cool, or if the
heat, cool, or fan relay cannot supply 15mA to power thermostat without the relay activating, the transformer
common wire or the HAI Thermostat Power Supply Module (Part Number: 30A00-1) is required.

Click to expand...

My newer (replaced) Omnistat 2 RC1000 always defaults to green if the power goes out. HAI said that is normal. The older RC1000 though defaulted to whatever it was left at (blue or green). I've never had the resets; but I have it set up on an autonomous power supply after having the reset issues with the original RC80. I did have communications problems with the first gen RC1000. They were odd; it was relating to it talking back to the panel and or Omnitouch screens. The newer Omnistat 2 RC1000 (with the Zigbee software) doesn't do that. I wish though I could do something to make it default to the blue screen instead of the green screen; the RC80 had "registry" style entries for the defaults; and I wonder if there are key press entries for similiar on the RC1000? I'm also concerned about the temperature that it shows; it seems high; I adjust it to an adjacent temperature sensor; and if I reset it; it goes back the the original setup (looking to be maybe 2 degrees off).

I have a common, I had it rewired with a common a few years ago when I had my old A/C and just an RC80.

When I got the new 2-stage A/C put in back in 2008, I got a RC2000 and made sure they rewired with a proper common. The old RC2000 functioned perfectly for 3 years until the lightning strike.

It does sound like it is resetting, perhaps based on some power issue though. I will have to go over everything, maybe a loose connection after swapping things out, maybe one of my float switches for detecting improper draining is acting up, I think they break the common to shut the system down.

Thanks for the input!

Perhaps there is a break in the line, maybe the lightning fried something in the wall and it is shorting out?

Definitely a power issue, I happen to catch it in the act today while working in the garage. I have my old RC80 hooked up in there as an aux temp sensor and it is power off the A/C as well. I happen to look over at it and noticed it was blank, as in no power. I have seen this before when the drain line clogged and the pan filled tripping one of the floats.

I think it is actually the float switch in the condensate pump. I have three safety float switches that will all break common if they trip. One right off the drain in the unit, one off the pan, and one in the pump. The pan was dry so I check the pump, jiggled it a little, and it came to life draining and shortly after the thermostats got their power back. Took apart the pump and it was fouled up with nastiness all the way to the float switches. I actuated the float by hand and it shutdown and restarted the tstats again. Cleaned it out ran some bleach through it, and everything is running again.

Will keep an eye on it ,either way something is breaking the power to both tstats. At least I know what to look for now, if the tstat resets to green back light I know it power reset at some point. Kind of annoying it doesn't remember a the back light color between power cycles, I would have preferred a message on the tstat that you just need to hit ok too instead to notify of a power cycle, probably would let people know they specifically have a power problem more directly.

I did think about wiring my float switches into zones on my Omni instead and have automation logic to shutdown the A/C if one tripped, That way I would know more specifically what occurred, and I could even get an email or something if it did. I didn't do it because I wanted to have a self contained safety system on the A/C, perhaps instead I will get a 24v relay driven by the A/C power and tie it into a zone, leaving the circuit alone for the most part. Or perhaps I can detect the lost contact to the tstats in Omni logic? I think they read -40f when disconnected?

Well the final verdict is the the float switch in the air handler. It seem the short drain line from the air handler to the condensate pump was clogged and draining very slowly but not enough to overflow into the pan. It was barely tripping the internal float switch intermittently.

Normally I would check the float switches before anything else, but with all the RC2000 issues I have been having I focused prematurely on the tstat.

I did write a simple automation block that seems to work detecting a powered down Omnistat and alerting me, here is a sample, not sure if there is a better way to do this:


1. EVERY 5 SECONDS
AND IF HVAC CURRENT TEMPERATURE IS 0
AND IF HVAC ALERT IS NOT DISPLAYED
THEN SHOW HVAC ALERT WITH BEEP
THEN LOG HVAC ALERT
THEN RUN Email Status

It nice because Email Status button causes Web-Link to send me an email and it shows the last few log entries and current tstats status along with the message. Didn't want to go all out and have it set off a water alarm or anything.

Well that automation block seems to not be 100% accurate, but I have no idea how, the logic is sound.

Today however I just got a HVAC alert, but I looked at the tstat and it had power and hand't reset to green backlight. Also the email I got showed a valid temp instead of 0, which makes no sense.

Maybe HAI could speak to why the every trigger with a temp comparison could be inaccurate? Perhaps the temp can read 0 if it's read at just the right time while it's being updated? Is there a problem doing an every 5 seconds interval? Maybe the tstat reports 0 temp for a short time once in a while?

I could probably make it more accurate by incrementing a flag and having it trigger if it reads two consecutive 0's, but the best way is probably to put a relay on the tstat power circuit and hook it to a zone.

Also strange the Omni doesn't have a temp trigger, instead I have to use an Every with a condition.

jharrell said:

... the best way is probably to put a relay on the tstat power circuit and hook it to a zone.

Click to expand...

That's sure the first solution that comes to mind for me - that way you know 100% if there's current or not.

Just not sure why "every x if tstat current temp = 0" is reading 0 randomly during the day, to me this says the current temp condition in the Omni cannot be relied upon, which would make it useless, am I wrong?

I have implemented your program on 2 test systems, actually 1 test system and my own home. Hopefully if the controller is periodically reporting a bogus 0 temp the testing will uncover it. Just to make sure I am doing a good test can you give me the model and firmware version of the controller and firmware version of the tstat.

My first thought was an intermittent thermostat communications problem, but I could be wrong, hence the test...

EDIT: The fact that Web-Link e-mailed a temperature should be ignored. Web-Link periodically polls for temperature so temperature changes may not (usually don't) show up immediately. Since the e-mail notification should happen very quickly it is likely that you will almost always be e-mailed the last known temperature.

Fred,

My controller is a non-flash upgrade-able which I recently swapped the chip out to firmware 3.7. The Omnistat is a 1.04D firmware. I also have a old RC80 on the same bus at address 2.

I figured Web-Link polled, I did get a email status with a 0 temp once, so it must poll pretty often.

Could very well be a comm issue but it must not happen very often as it can go for hours with no alert then seemingly randomly it will trigger and it's definetly not 0 degrees in the house

Perhaps this is telling me something is wrong with the comm bus, that would be kind cool.

Out of curiosity does the Omni stat push it status to the Omni, or does the Omni poll it?

This would be very useful if I could get to work reliably, the most common issue I have with my A/C is one of the float switches getting tripped roughly every year because of some clog somewhere. I still might just wire the floats into a zone and have the Omni shut down the A/C, after all I trust it with my fire alarms. It would still be nice to know if anything is wrong with the tstat from digtal comms down and have it alert.

Thanks for looking into this.

Well my systems have been running the "test" for about 24 hours and no problems have been reported.

My home has only 1 Rc2000, but the test system in my office has an RC2000 and an RC80 so that is pretty close to your system.

I am thinking about putting this code on a light commercial installation that has 14 RC2000s just to see how often, if ever, they have a zero reading. Perhaps this weekend.

The Omni polls each tstat on the buss.

Wiring the float to a zone is probably better because you could log/send a more specific message.

Mine definitely can't go a whole day without this triggering. It's a RS-485 bus for the tstats right? Anything I should look for beside obvious wiring issues? I think my RC80 is branched off from the bus inside the air handler, could the hub-spoke like wiring not play well the bus?

jharrell said:

but the best way is probably to put a relay on the tstat power circuit and hook it to a zone.

Click to expand...

I've done similar with a condensate sump pump. I have a relay with its coil wired to the float switch (in my case, this is a 120V switch), and the contacts are wired to a zone. I use it to log when the pump runs, so I can estimate how much humidity is being removed.