overriding Goodman "Comfortbridge" communicating mode

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
  • Hope everyone has a wonderful and warm Thanksgiving!
  • Super Cedar firestarters 30% discount Use code Hearth2024 Click here
So how's it running Rusty?
It's been so pleasant here the last few days that we haven't run the system at all. But hey, it's still August, and it's still North Carolina, so I'll find out soon.
 
  • Like
Reactions: Pinesmoke and EbS-P
Earlier fan speed was mentioned. Again, without reading into your air handler, fan speeds are programmed on the AH logic board on mine.
 
Earlier fan speed was mentioned. Again, without reading into your air handler, fan speeds are programmed on the AH logic board on mine.
No way to do it that I see. If you look at the air-handler install doc (that I linked in OP), under "Quick Start Guide for Non-Communicating Outdoor Units" (pages 16-17), under bullet point #4 it says "Go to the Tonnage Units Menu and select the tonnage value that corresponds to the desired airflow for the outdoor unit." That's it.

Also, I posted about this to Reddit, and a guy who definitely sounds like he's familiar with these units said "You're also unfortunately going to be stuck with the same airflow for heating and cooling for a given stage, so you're going to have to go with the higher of the two. That's going to be slightly less comfortable than it otherwise has to be. That's unfortunate." at:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.
 
No way to do it that I see. If you look at the air-handler install doc (that I linked in OP), under "Quick Start Guide for Non-Communicating Outdoor Units" (pages 16-17), under bullet point #4 it says "Go to the Tonnage Units Menu and select the tonnage value that corresponds to the desired airflow for the outdoor unit." That's it.

Also, I posted about this to Reddit, and a guy who definitely sounds like he's familiar with these units said "You're also unfortunately going to be stuck with the same airflow for heating and cooling for a given stage, so you're going to have to go with the higher of the two. That's going to be slightly less comfortable than it otherwise has to be. That's unfortunate." at:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

i find lower CFM more comfortable. Better dehumidification in the summer and warmer outlet temps in the winter so it feels less drafty. Summer capacity won’t take the hit that winter does. Blowing air that’s 8-10 degrees warmer than the room might not feel that “warm”. They reset mine to 400 cfm per ton with stage 1 running 70% of that and I turned it back to 350 cfm.

Now if you have restricted ductwork I might keep it higher. Really a good professional would verify flow and use measurements to set the correct flow. Without that measurement you can’t verify capacity.
 
i find lower CFM more comfortable. Better dehumidification in the summer and warmer outlet temps in the winter so it feels less drafty. Summer capacity won’t take the hit that winter does. Blowing air that’s 8-10 degrees warmer than the room might not feel that “warm”. They reset mine to 400 cfm per ton with stage 1 running 70% of that and I turned it back to 350 cfm.
Well, the way they do it is that whatever tonnage you select, it sets the high-stage CFM and 400x the tonnage, and the second-stage maybe 70% of that.
Now if you have restricted ductwork I might keep it higher. Really a good professional would verify flow and use measurements to set the correct flow. Without that measurement you can’t verify capacity.r gri
I'm getting some weird ESP measurements. The return is 0.35" or so, which is nice, thanks to converting my filter grill to handle a 4" filter, and streamlining the ductwork from the filter grill to the air handler. But the supply side is barely measureable, maybe 0.1" or so. That just seems weird. I've done a lot of work on sagging and not-tightly-stretched flexduct, and also sealed all the joints in the supply trunk (as advised by folks here). But that still seems oddly low.
 
So, being kinda annoyed that this fancy Honeywell thermostat won't provide statistics on percentage of time the system is running in various modes (unlike the Google Nest, much maligned by "professionals").

So I built a little monitor for my system and connected it to the thermostat wires - by mounting it on the cover of a junction box with 4 cables going into it: air handler, compressor, thermostat, and monitor.

For each thermostat wire, I use this circuit:

[Hearth.com] overriding Goodman "Comfortbridge" communicating mode


The optocoupler puts a load of only a few milliamps on the thermostat wires, so should not interfere with them. When one of the thermostat signals is asserted (so 24vac between it and the common signal 'C'), the optocoupler turns on its internal NPN transistor and the logic output signal goes low. The diodes in the opto-coupler won't conduct until there's a volt of so across them, so when the thermostat signals are not asserted (and are in fact tri-stated**) the logic output stays hi (at 3.3vdc). These 5 signals go into an Arduino in the monitor (which treats them as active-low inputs). The 2N3906 and LED just provide a visual indication of each signal's being asserted.

I have a USB cable going from a wall-plate inside my house down to the Arduino. I can use this to download a "sketch" to the Arduino. But normal interaction with the Arduino is via wifi (so I don't need my laptop connected to the wall plate unless I'm updating the sketch). I can program the Arduino to display graphs of the various signals, statistics on percentage the system is in each mode, etc.

** I finally found another person (on stackexchange) who knows all about HVAC but can also talk like a EE. He confirmed my suspicion that a thermostat will open-circuit (or tri-state) its outputs when these are de-asserted. In fact, he says one can install multiple thermostats in parallel, any one of which can activate the system. This explains a confusion I've long had about the 'W' wire. In a heat pump system, it's generally connected to three terminals - on the thermostat, the compressor, and the air handler. So either the thermostat (if backup heating is required), or the compressor (if it goes into defrost mode), can assert W and activate the backup heat; this would only work if the signal is tri-stated when de-asserted, otherwise the thermostat and the compressor could conflict, with one pulling W high and the other pulling it low.
 

Attachments

  • [Hearth.com] overriding Goodman "Comfortbridge" communicating mode
    Screenshot 2024-10-19 at 5.23.55 PM.webp
    22.7 KB · Views: 11
  • [Hearth.com] overriding Goodman "Comfortbridge" communicating mode
    Screenshot 2024-10-20 at 3.43.19 PM.webp
    23.4 KB · Views: 10
Last edited:
So, being kinda annoyed that this fancy Honeywell thermostat won't provide statistics on percentage of time the system is running in various modes (unlike the Google Nest, much maligned by "professionals").

So I built a little monitor for my system and connected it to the thermostat wires - by mounting it on the cover of a junction box with 4 cables going into it: air handler, compressor, thermostat, and monitor.

For each thermostat wire, I use this circuit:

View attachment 331019

The optocoupler puts a load of only a few milliamps on the thermostat wires, so should not interfere with them. When one of the thermostat signals is asserted (so 24vac between it and the common signal 'C'), the optocoupler turns on its internal NPN transistor and the logic output signal goes low. The diodes in the opto-coupler won't conduct until there's a volt of so across them, so when the thermostat signals are not asserted (and are in fact tri-stated**) the logic output stays hi (at 3.3vdc). These 5 signals go into an Arduino in the monitor (which treats them as active-low inputs). The 2N3906 and LED just provide a visual indication of each signal's being asserted.

I have a USB cable going from a wall-plate inside my house down to the Arduino. I can use this to download a "sketch" to the Arduino. But normal interaction with the Arduino is via wifi (so I don't need my laptop connected to the wall plate unless I'm updating the sketch). I can program the Arduino to display graphs of the various signals, statistics on percentage the system is in each mode, etc.

** I finally found another person (on stackexchange) who knows all about HVAC but can also talk like a EE. He confirmed my suspicion that a thermostat will open-circuit (or tri-state) its outputs when these are de-asserted. In fact, he says one can install multiple thermostats in parallel, any one of which can activate the system. This explains a confusion I've long had about the 'W' wire. In a heat pump system, it's generally connected to three terminals - on the thermostat, the compressor, and the air handler. So either the thermostat (if backup heating is required), or the compressor (if it goes into defrost mode), can assert W and activate the backup heat; this would only work if the signal is tri-stated when de-asserted, otherwise the thermostat and the compressor could conflict, with one pulling W high and the other pulling it low.
I gotta re-read this when I’m in the thinking space. Ecobee gives you nice graphs and can export runtime data.
 
I gotta re-read this when I’m in the thinking space. Ecobee gives you nice graphs and can export runtime data.
Would those be the ones @Pinesmoke showed in post #16 earlier in this thread ?
 
Would those be the ones @Pinesmoke showed in post #16 earlier in this thread ?
not sure what interface his came from but here is a screen shot of mine. You get stage 1,2 aux heat, and dehumidifier all in a graph. It’s not the best interface but it’s ok.
I can download all the data. As a csv.
 

Attachments

  • [Hearth.com] overriding Goodman "Comfortbridge" communicating mode
    IMG_2260.webp
    28.4 KB · Views: 11
So, being kinda annoyed that this fancy Honeywell thermostat won't provide statistics on percentage of time the system is running in various modes (unlike the Google Nest, much maligned by "professionals").

So I built a little monitor for my system and connected it to the thermostat wires - by mounting it on the cover of a junction box with 4 cables going into it: air handler, compressor, thermostat, and monitor.

Here's the little "monitor" widget I built. Bolted to the cover of my little 24vac junction box.

[Hearth.com] overriding Goodman "Comfortbridge" communicating mode


At first I had a USB cable going from the Arduino up to an outlet box in the house; I'd power the Arduino from that using a wall wart, or plug the cable into my laptop to download a new Arduino "sketch". But then the Arduino reset when I removed the cable from my computer and moved it back to the wall wart - kinda inconvenient.

So now I'm powering the Arduino from a 5vdc wall-wart in the crawlspace; it bypasses the USB jack, so the cable going up to the house is still connected to the jack, and I just connect my laptop to load a new sketch. I actually had to cut the red +5vdc wire in the USB cable though, because my new Macbook Air's USB port gets upset if it sees other 5vdc power at the far end of the cable.

Seems like it'd be a lot cleaner to just power the monitor from the 24vac from the air handler (using the 'C' and 'R' thermostat wires); a rectifier and then a high-efficiency DC-to-DC converter like https://www.mouser.com/datasheet/2/468/R_78E_1_0-1711201.pdf . I wonder if it's ok to put that much add'l load on the 24vac transformer in the air handler though. Those things are usually rated at 40VA, so good for 1-2 amps. The Arduino and the my 5 little circuits probably draw half an amp max. But since the regulator is 90% efficient (not a linear one like a LM7805) it should pull less than half that off the 24vac (less than 250 ma or so). Sound ok ?
 
Last edited:
Here's the little "monitor" widget I built. Bolted to the cover of my little 24vac junction box.

View attachment 331522

At first I had a USB cable going from the Arduino up to an outlet box in the house; I'd power the Arduino from that using a wall wart, or plug the cable into my laptop to download a new Arduino "sketch". But then the Arduino reset when I removed the cable from my computer and moved it back to the wall wart - kinda inconvenient.

So now I'm powering the Arduino from a 5vdc wall-wart in the crawlspace; it bypasses the USB jack, so the cable going up to the house is still connected to the jack, and I just connect my laptop to load a new sketch. I actually had to cut the red +5vdc wire in the USB cable though, because my new Macbook Air's USB port gets upset if it sees other 5vdc power at the far end of the cable.

Seems like it'd be a lot cleaner to just power the monitor from the 24vac from the air handler (using the 'C' and 'R' thermostat wires); a rectifier and then a high-efficiency DC-to-DC converter like https://www.mouser.com/datasheet/2/468/R_78E_1_0-1711201.pdf . I wonder if it's ok to put that much add'l load on the 24vac transformer in the air handler though. Those things are usually rated at 40VA, so good for 1-2 amps. The Arduino and the my 5 little circuits probably draw half an amp max. But since the regulator is 90% efficient (not a linear one like a LM7805) it should pull less than half that off the 24vac (less than 250 ma or so). Sound ok ?
They power large lcd screens and WiFi for the new thermostats off the 24v AC. I don’t see how this adds that much but I’d measure its startup draw.
 
  • Like
Reactions: RustyShackleford
They power large lcd screens and WiFi for the new thermostats off the 24v AC. I don’t see how this adds that much but I’d measure its startup draw.
My thermostat has those though (WIFI and LCD screen). I guess I could dummy up and connect the PS (for my monitor) hook it up, and see how much the 24vac droops. I think it's like 27v now.
 
My thermostat has those though (WIFI and LCD screen). I guess I could dummy up and connect the PS (for my monitor) hook it up, and see how much the 24vac droops. I think it's like 27v now.
Or just install a higher current rated transformer.
 
My thermostat has those though (WIFI and LCD screen). I guess I could dummy up and connect the PS (for my monitor) hook it up, and see how much the 24vac droops. I think it's like 27v now.
The transformer is usually protected by a fuse either 3A or 5A. In the days before circuit boards the thermostat would close and directly pull in a gas valve or A/C contactor which can pull an amp or two easily. I think you'll be fine adding this small load to the transformer.
 
I actually blew that fuse screwing with the thermostat wires (moron). And yeah, I think it was 5 amps.
 
  • Like
Reactions: EbS-P