My colleagues and I are working on a similar project with climatetalk. I am not a controls guy so i am trying my best to find some solutions.
We are currently having problems with the thermostat not being able to detect the heat pump or furnace. We have a board that intercepts the signals from the hvac system to the thermostat and vise versa. The signal is supposed to continue in its original route. The thermostat says searching when the board is connected but finds the heatpump and furnace if the board is bypassed.
Bill, It was either my good timing or my actions about a year ago; i was looking into making my heat pump communicate with my Armana communicating furnace. Synch right? Well, i was able to register and download the open protocol information too ā¦ it is open right? Then website went offline and has failed to return. This industry is āfunnyā. Lets put it that way. I can see now that in order to reach initial goal, must code a thermostat controller and then make a heat pump controller. eesh! a little daunting. May be a good project for learning swift programming though. Code it on a Raspberry Pi with a tft display.
So far, have made good progress on decoding my heat pumps communication abilities ā¦ it iwasnāt as pretty as Mitsibuchiās serial communication.
@kevin_peck,
I would be interested in keeping up with your progress on your Amana heat pump. Do you have any Amana/Goodman-specific protocol details, or just the generic protocol docs? (Iām also wondering if Emerson is actually the keeper of any model-specific details, since itās an Emerson board in my unitā¦ or if that just means itās a generic Emerson board that each mfgr can customize the firmware of).
I have a unit that keeps throwing intermittent ālocked rotorā codes, and service techs canāt find anything wrong. Itās always, āEverything measures normal. Call me if it does it againāā¦ which doesnāt help if you get the same response the next time.
Iām starting to suspect transients on our power lines, but wonāt know for sure until I can gather data. Iām going to build a logger to monitor the voltage and current, and also tap into the communication to gather any data available, but especially a/c cycle status and error codes relative to the timing of any voltage transients.
rfindley : Never said what kind of heat pump I have, it isnāt armana. If it were, I would have no trouble making it work with the furnace.
As for intermittent code errors, first suspect would be the RS485 wires and interference. Terminations are important and such. Google physical layer of rs485 and such and there is lots of info on what is needed for this network to work reliably. Sometimes you just have to change the wire but donāt do that as your first step! Oscilloscope would help here.
I have been trying to get ahold of the spec for some time - the Dropbox links were handy. Seems sensible to maybe start a Github project for this. Iām not much of a Python guy, I work in C# and have done my share of protocol bit twiddling over the years, including some time on BACnet. I might be doing some dabbling in this if someone would be interested in helping out testing and/or reviewing code, etc.
Thanks for the info on the Goodman/Carrier lawsuit - unfortunately the U.S. District Court of Delaware canāt seem to keep itās documents online either. Fortunately the wayback machine was able to retrieve it.
I just installed an Amana heat pump, and the documentation emphasized that the indoor and outdoor units must be used with the CTK04 thermostat and that CTK01, 02, 03 wonāt work. That apparently also has something to do with the lawsuit:
āAt trial, Carrier asserted that certain of Goodmanās products - the ComfortNet indoor and outdoor units with one of a CTK01 thermostat, CTK02 thermostat, or CTK03 thermostat (collectively, āthe ComfortNet systemā) - infringe claims 6, 8, and 13 of the
'004 patent.ā
Since this is my first experience with a communicating system (it replaced a 24 year old Trane HP), I was toying with the idea of building a passive bus sniffer and event logger. Thanks to all for posting the information here; if I ever write some Python message parsing code Iāll be sure to send a pull request.
@David_K The only way to get out of this legalese logjam is for us to take over this protocol. Starting with the python sniffer program, Iāve re-written it in Swift 4.2 and also added support for ALL messages on the bus. Check out https://github.com/kpishere/Net485 and specifically the sub-project in diag/net485Listen. As you can see, Iāve got the fundamentals of the protocol working but I must re-tool to make completion more efficient. So, please do try out the tool and please do upload some data! I need a live system to validate against. Then we make a database for different systems etc. and damn well make our own open source thermostat that supports all manufacturers as it is supposed to!
