Today I got the temperature controllers installed for my warming pads. All that is left is to clean up the wiring and get the controllers secured.
I decided (for now) to not use the temperature sensor/controller that came with the kit that Ultra Heat created for me. The first controller they provided worked inconsistently. Testing it at home it worked about 50% of the time. I suspect there was an internal wiring issue, but I didn't open it to verify that. The replacement sensor they sent worked every time but the temperature range seemed like it was 30° F to 44° F, not the labelled 35° F to 45° F. I used the temperature sensor from my Fluke 325 meter to check the temperature at the sensor/controller. The Fluke sensor was taped to the Ultra Heat sensor/controller. It isn't clear how accurate it is to tape the two together, but that's all I could figure out.
The new controller I'm using I got off Amazon for $11 based on recommendations from other forum members. It's the
LM YN DC 12V Digital Thermostat Module. It has a programmable range, which I prefer to the preset range. Other benefits are that it shows you the observed temperature and the target temperature on the two displays. Each battery now has its own controller. With the Ultra Heat kit, there was only one controller, so it was warming one of the batteries without knowing what that battery's temperature was. Given that the two batteries sat side-by-side and the temp range was conservative there wasn't much risk. With the new setup, there is zero risk.
In the test today, after 40 minutes, the warming pads raised the observed temperature at the top of the battery by 8° F. From ~47 to 53° F. I had to fool the controller by putting the temperature sensor into a gel freezer pack. I could have just raised the target temperature on the controller, but didn't think of that until later.
Each time I checked the current on this circuit, it was no more than 1.68 amps. This was with two warming pads active on a single battery. In my opinion, the results are excellent given the small amount of current that was consumed. The pads must be incredibly efficient.
The temperature reported by the controller was within .75° F of the temperature reported by the BMS's remote temperature sensor. I was pleased to see the two devices report almost the exact temperature. The Victron BMV-712 reported a similar temperature, but it reports only whole numbers. Still, it was close enough to call it accurate. I have both sensors (controller and BMS) taped down together at about the middle of the battery. I used some spare rigid foam to create a hollowed out "cap" over the two sensors so they will be reading the battery temperature without (much) interference from the ambient air. The cap isn't sealed, so this isn't 100% effective.
I checked the backside of the warming pads multiple times during the test. I couldn't feel any heat at all. I used an infrared thermometer and it detected only a 5° F difference between the back of the warming pad and the top of the battery.
The new controller is pretty slick. It's a bit confusing to program at first and comes with no instructions. I found a set of user supplied instructions and cleaned them up. If you want a copy, let me know. I'm going to buy another controller and use it for running a fan to keep the component compartment cool in the summer.