Hey all, software engineer here feeling over my head on the hardware side of a project. My whole home humidifier has never worked, and I finally got around to checking it out. I am fairly certain I have discovered that the issue is that the outside thermometer probe reads 0 resistance, indicating a short of some sort. The problem is, I can only follow that wire so long before it runs behind walls in my finished basement. Also, I haven't got a clue where it runs from there - I suppose outside somewhere, but I can't find it.
So, the idea for a project was to sort of hack the thing by using a digital potentiometer wired to a WiFi arduino board, and the board would get outside temperature data from the internet and adjust the digipot resistance accordingly (I have a chart of temp vs resistance). I bought a MCP41HV51 digipot because it supports single supply operation of 36V, and my HVAC wires running to the existing humidifier controller (and thus to the thermo probe) are 24V. The problem I have suddenly realized this is AC power. I don't believe this chip will support AC as a source, so I started looking at other chips, like the AD5293. These do support AC, however, it only allows up to 33V single supply or +/- 16.5V dual supply, which my understanding is that AC is dual supply due to the nature of the polarity switching (I could be totally off base in my wording here).
I was also planning to power the Arduino itself on the R and C wires (live when the furnace kicks on) since the new Uno R4 supports 24V on the VIN, but again, since it's AC, I think this won't work. For this part of the circuit though, I can just use USB-C or barrel jack instead, a little unfortunate but not a deal breaker.
My question really is, I suppose, am I correct in feeling like neither of these chips can help me accomplish my goal? Any other approaches, chips, etc. to consider, or perhaps this is just a sort of helpless project? Worst case, the humidifier controller did come with a resistor that can be used instead of a thermometer probe, and then can be operated in a "manual" mode, where I'll need to adjust a dial based on outside temperature.
Almost all "electronics" run on DC. Your computer, TV, stereo, etc., all have internal or external power supplies that convert the power-line AC voltage to lower voltage DC.
There are some circuits here showing how to use one or more diodes to "rectify" the AC, and this page shows how to add a capacitor to convert filter the rectified AC to "smooth" DC.
Note that the capacitor charges to the AC peak which is 1.4 times the RMS so 24VAC will give you about 33VDC. (There is a slight voltage drop across the diode.)
So you may need a voltage regulator. Look for a "switching" or "switch-mode" regulator. Linear (non-switching) regulators heat-up a LOT when "dropping" from 33 to 5V or 12V.
If you need to switch/control 24VAC you can use a relay. (A relay is an electrically-operated and electrically-isolated switch.) The Arduino needs a driver circuit to drive a regular electro-mechanical relay, but you can get a "relay board" that includes a relay and driver. Or there are solid state relays that can be directly-controlled with the Arduino, but make sure to get one rated to switch AC. (AC & DC solid state relays are not generally interchangeable.)
Have you contacted the distributor or manufacturer about your system and enquired about spare parts?
This page may help.
The controller may be AC, but the temperature sensor will most possibly be running on a DC control supply.
If you have disconnected the sensor and measured a short in the sensor, then with your DMM on DC Volts, measure the voltage on the two wires you have disconnected and go into the controller.
Caution, connect the DMM leads with clips and the controller OFF, do not use bare hands as the sensor may be at a high AC above gnd.
Tom...
PS, If you sensor has failed, then it might be best to get it functioning correctly BEFORE doing any modifications.
Thank you for responses. It does appear that the controller circuit itself is indeed VDC, by testing the sensor leads with a multimeter. So, I believe the digipot I have will work - I will double check the actual voltage through this part of the circuit to make sure it satisfies the minimum V supply on the digipot.
To answer whether I have verified proper operation prior to tinkering, yes, the humidistat does work correctly when using their provided "manual" resistor to override the outside temp probe, and putting it into test mode does successfully run the system.
@mikb55, appreciate the idea to use an opto isolator. Had I not incorrectly assumed or misinterpreted that this part of the circuit is DC, I am fairly certain that would have made for an excellent solution. It still would, I suspect, but feel as though the digipot may make for simpler coding and less parts.
