Disclaimer: The following is provided as a GUIDE ONLY, and neither myself, nor FITFREAK.NET endorse, recommend, encourage nor take any responsibility for the outcome of someone else doing the following. This is a precision calibrated system that balances airflow and temperature to maintain refrigerant evaporation before it reaches the compressor. A frozen evaporator can interfere with this process and if air is blocked for an extended period of time allow liquid refrigerant to reach the compressor, causing hydrolock and irreversible damage. You follow these steps at your own risk!

***

This intent of this modification is to lower the cycling temperatures of the stock A/C system when needed, but still allow immediate reversion to the original calibration when not. As the Fit's air system was designed for minimum fuel economy loss when in operation, the typical temperature maintained by the system is higher than most with vent temperatures kept above 50 degrees F via rapid compressor cycling on days exceeding 85 degrees. To modify the temperature range, the signal that the computer relies on to control the compressor must be altered: the resistance of the thermistor on the evaporator. However, the proper resistance value must be chosen that does not let the computer push the evaporator temperature below 32 degrees F, or ice could form and block it, potentially causing damage as the refrigerant will boil more and more slowly and may reach the compressor itself still in liquid form.

On my GD3, the thermistor Honda used is a 2-wire high resistance unit that has a somewhat linear change in resistance with temperature. This modification is not suitable for use with the 3-wire switched type, which completely blocks or passes current on a 36F to 41F temperature range, and changed resistance values will have no effect.

To calculate the resistance value we need to aim for, I consulted a service manual which listed the following thermistor resistance values on a graph: 30k ohm @ 32F, 18k ohm @ 50F, 12k ohm @ 68F, and 9k ohm @ 86F. According to the graph, if on a cool day the system shuts off at 42F (vent) as I measured with fan speed at 2, then the evaporator is likely closer to 38F, with the resistance about 24k ohm. To lower the temperature, the resistance must be lowered slightly, so that means the new resistor(s) go in parallel.

Based on the calculations on this website, as I was too lazy to do the math myself, I chose a 200k ohm setup to start as this would lower resistance to 21.4k ohm. At this value the computer would assume the evaporator is about 6 degrees warmer than it actually is, if the graph is correct, bringing the evaporator into the 32F-34F range. To provide adjustability, I added a 100k ohm micro potentiometer to allow fine tuning.

***

Items Required:

-pack of 100k ohm resistors, 1/2 watt
-100k ohm micro potentiometer, linear type
-22 ga. wire spool
-high-temperature square adhesive zip tie mounts, plastic, and zip ties
-soft touch on-off pushbutton switch
-quick-disconnect spade or bullet connectors, insulated
-wire splice taps for 22 ga. wire
-multimeter to test your new connections and the potentiometer
-soldering iron and electrical solder
-shrink wrap of various sizes
-electrical connector lubricant to prevent corrosion

cost, minus the multimeter, about $36







Assembly:

This is simple so step-by-step isn't really needed. Solder the resistor(s) you're going to be using to the potentiometer on the two leads that make the most resistance with the knob turned clockwise fully. That way, like a volume knob, up has a higher temperature and down a lower one. If you wish to reverse that, then use the other lead, but that's up to you.

Once you get it soldered up, start shrink wrapping. Keep all metal parts apart so nothing can short out. Also a layer or two of larger wrap will strengthen the resistor wires and keep it from bending, as well as cover up the leads on the potentiometer.

On my setup I have one 100k ohm resistor on the line, and the other 100k ohm provided by the potentiometer. If I need it, I will add another 100k ohm resistor on another wire between spade connectors and move the range to 200k-300k adjustable, as right now at the max of 200k ohm it's close to too cold. Can always add or take away with the spades, but if soldered it's permanent.







Installation:

This part is all up to you. I chose this location as it is nearly invisible, easy to reach when driving, and require no holes to be drilled.



The zip tie mounts holding the potentiometer in place. Notice it can be easily adjusted by just popping this cover off. Need the removable connectors if you want to be able to remove this panel (highly recommended). Remember to put some electrical lubricant into these to prevent corrosion.



This is where it's tapped, behind the glove box. Don't even have to remove the glove box as it's totally accessible from the bottom once you pop the panel off (two snaps, much like the driver's side). You know it's the right wire if it's got that grey insulation and it follows around the back of the evaporator to right next to the accelerator pedal, where it goes into the evaporator assembly. This is the wire you test for resistance to see if your thermistor is the linear resistance or switched type. Also remember to use some electrical lubricant here too. If you want additional protection, use electrical tape.



Don't forget to test your resistances to make sure you don't have a short or broken wire. With the button not pressed, it was 9.8k ohm; pressed, 9.3k ohm. Perfect, no shorts.

Observations:

First test, ambient air 77F. Temperatures taken with my old wall thermometer at center vent, HVAC sent to vent only, window vents are open, recirculate off. I have my cabin filter removed for now as it clogged with dirt, to be replaced soon. I may have to add another 100k ohm resistor and crank down the potentiometer to compensate.

At fan speed 2 and button not pressed (OEM resistance), the compressor cycles:
-at 55 mph, on at 50F, off at 42F.
-at idle, on at 41F, off at 38F. This was weird how it went down.

At fan speed 2 and button pressed (modified resistance), the compressor cycles:
-at 55 mph, on at 36F, off at 32F.
-at idle, steady at 33F.

At fan speed 1 with button pressed (modified resistance), the compressor will cycle at idle with fan speed at 1 when it drops to 32F, about one complete cycle every 90 seconds. At fan speed 4 and engine idle, it stays steady at 36F.

As this is likely cold enough to form ice, I may add another resistor if it becomes an issue. It does form a small amount of condensation on the vents. However if it does not drop that far in hot weather, I'll leave it as is because 90% of the time I'll just have it in normal mode for the fuel economy and just use the cold mode for initial cabin cooling.

Picture taken instant compressor turned off with button on and fan speed on 1:

 

I think I'll wait for a few others to try this one out before I do it myself. Props for having the guts to mess with it..

 

I've lowered the temp on other cars (not the Fit) and on muggy, overcast days when you want to de-humidify, the evaporator would freeze up.

The switch is a good idea, instantly changing back to factory specs would take care of that.

 

Thermometer calibration in ice water.



I didn't think it was that far off. That's 3 degrees low.

So take all numbers and add 3. That means it's cycling off at 35-36F with the mod (and off at 45F at the stock setting), which means with the new setup the evaporator should barely touch the freezing point when it shuts off. That would put my recommendation for 200k ohm to be the lowest accumulative fixed resistor value (2x 100k) to use and then add the potentiometer to that.

Assuming that translates across the entire measured range it also means I had a 53 degree measured vent temp when it kicked on without the mod on an 80 degree day, and that doesn't include the unusually slow response time for this thermometer, so it was actually higher than that. I knew my temps sucked.

edit: I'm going to leave the temperatures in the first post unchanged for consistency, so remember to add 3 degrees when you read it.

 

technically you are in ice water there, not ice.. so that might change it by a degree or 2.

 

Can I just put a resistor inline here, to permanently reduce
the a/c cycling?

 

Will someone add to polaski's rep?.... I have to spread some elsewhere before I can rep him again and he deserves it.

 

I dont understand how this mod would work,
Its 110F outside, my vent temps are in the 50-60's.
My a/c still cycles on and off like crazy!
So how would making the temp sensor warmer really
help in the fit??? I understand how it would in a normal a/c system. Also in winter, it cycles on and off just the same, no matter what the temperature is!

 

bookmarked for intrest

 

I would presume it works because the car isn't taking the temperature from the outside air, but from somewhere inside the A/C system, where the temperatures actually get that low.

 

Hi Jimmy101.... polaski explains the answer to your question in the 1st paragraph after the disclaimer..... I had to read it and think about it a few times before it made sense to me.... Anything having to do with wiring I have to go over many times and I still mess up a little bit and a whole lot otherwise.

 

The sensor itself doesn't get warmer. By changing the perceived resistance value, the computer that controls the AC thinks it's warmer than it really is, keeping the system going until it gets just a bit cooler. It still cycles because the thermistor still reports a reading, we're just offsetting it a bit. Most cars don't need this because they keep the system calibrated for cool temps but this one was set for higher ones giving less cooling but at the same time better fuel economy. The older Fit/Jazz overseas with the drive-by-wire option actually followed about the same temp range created by this mod (essentially the same 3-wire switching 36F-39F thermistor used by the 6th gen civic, I know that for sure).

It takes the temp on the evaporator itself inside the HVAC, the "radiator" of the AC system inside the cabin where the refrigerant absorbs the heat from the air.

I would not recommend making the change permanent (without a switch) as we're playing with an untested calibration and everyone's sensor is going to be a little bit different. If the outside temp gets low enough the climate control computer may decide to bring the air temp way down, potentially putting it below freezing which would form ice on the evaporator, and if humidity is high could block it rather rapidly. You need to be able to revert to the stock setting for those conditions if they occur. (Also in the winter time the stock setting would be best, both for that reason and the fact that it only needs to dehumidify.)

As for thermometer in the ice water... it thought it was in 29 degree water that was at a temp just above freezing, so yes it may be off as far as 5 degrees instead of the 3 pictured assuming that. I left that water with ice in the refrigerator for an hour to make it as close as possible before taking the reading.

Today's drive was nice, it was good and cold with the outside temp about 82 on the news. It didn't freeze up on the longest continuous stretch of about 35 minutes.

 

Wow, props for a great writeup.

Maybe it's where I live or maybe the thermistors vary considerably from car to car, but my Fit has never felt as though it needed more AC oomph. I haven't measured vent temps but I almost always run it at fan speed of 2, and usually I have the temperature control a couple of clicks off the coldest setting.

I think I'll count my blessings. My Fit may not get the MPGs I see some folks hitting, but it sure keeps me cool.

 

With the humidity here as high as it almost always is, my A/C has frozen up in the open circulation setting with the blower set on 3.... It has been a few months since it has been as cool outside as 83 degrees during the day but at night when it is in the 80s I turn off the A/C and roll down the windows... 2 days ago condensation formed on my wind shield with the A/C on.

 

So the Fit's sensor is not as accurate so they give it more head room? Tell me if I'm wrong, that's kind of what I'm getting so far.

 

I don't quite understand your question, but I'll try to answer.... There is almost always variations in resistance reading in resistors that are supposed to be the of same ohm rating unless you are to pay a higher price for matched ones that have been checked before packaging.... Because of this some Fits appear to have A/Cs that have higher ohm resistors than others and lower temperature air.... The modification allows a variable adjustment that will allow the compressor to stay on longer and before cycling off and lower the temperature of the air, but allow an adjustment back to the stock setting by adjusting the knob.... I am sorry if I told you something that you already know.

 

Thanks for your answer, coyote. I think I understand it a little better now. I am interested in this mod, but I am concerned about the freezing issue. Do you guys think there is a safe middle ground that would make it colder but have no chance of freezing?

 

The potentiometer is the "middle ground." It allows continuously adjustable calibration, and once you get it where you want it, you pretty much set it and forget it. If it tries to freeze, you'll know it because you'll have water all over your vents, fog on the windows, etc., and airspeed at the vent will go down dramatically. Kinda like tuning a motor, want ignition as close as safely possible to detonation to get power but don't want to cross over that threshold.

Button off for days under 75F, button on for days over 75F is my new rule.

I guarantee if you use just a 100k ohm resistor you'll freeze up some time or other. That drops the resistance 25% or so when it's cold in there. That's a huge change. My 200k ohm setup dropped resistance about 12% and temps from my abnormally high stock value down to what I believe is safely near freezing, though I don't think I'll have to add another resistor if I follow my new "rule" for cooler or really humid days. That's what the switch was put in there for.

If you're one of the lucky ones that always has cold A/C, you shouldn't need to worry about this (unless you want cold air fast when the car's been sitting in the sun all day, today it took only about 45 sec before the air got nice and cold because it waited so long to cycle). However mine in stock form feels like it's out of breath at any temperature over 75F, and doesn't even feel like it's on for almost 10 minutes if the inside of the car is hot because it's cycling at so high at temp. It only gets cold when I don't need to use it.

I must have been spoiled with the old civic. It had a naturally cool sensor just the way I like it. If the humidity was high enough it would blow fog with every cycle all day long but never freeze up, even after it was discharged and recharged with an engine R&R (clutch, timing belt) and a leaking heater core was ruled out.

 

polaski's solution here is about as good as it gets. Just turn the little knob like you would a volume control, as you wish. If it is too much the evaporator ices up and air quits flowing, so you will know that it is too cold and that you are going to have to increase the fan speed or turn the little knob back a little bit to prevent icing from occurring again...In the off position it is working like the modification wasn't there at all.... I have a hard time getting under my dash but I am thinking that it may be easier if I remove the drivers seat.... I had to read and re read everything polaski posted and know that it will all come together once I am under the dash.... This is crazy but it is to hot here to do it right now.

 

Right now your potentiometer isn't doing much, right? Because you're at the threshold of being too cold and turning the knob down only would make it colder? So if I were going to do this mod I should really use 2 of those resistors in addition to the potentiometer, right?

EDIT: I'm a little confused how adding another resistor would bring it closer to stock when no resistance added is stock.... I think I hurt something in my brain?