AC/Heater Climate Control experiment

[Pottsy]

Since some people have thought that the Climate Control wasn't working properly, I thought I'd take some measurements to see if I could:

• Work out if the system maintains a constant cabin temperature.
• Get a picture of how it distributes air.
• Find out if the cabin sensor is aspirated enough at low fan speeds.
• Find out if the footwell vent influences the cabin sensor.

So I went on a 2 hour journey and made some measurements. The conditions were:

• 22C set.
• Auto mode.
• All vents open.
• Night time.
• No dual zone.
• Two people in the car.
• All A roads.
• 10C outside.

The measurements taken were:

• Foot vent.
• Face vent (one of the side ones, since the centres only come on for cooling and dual zone).
• Inside the cabin sensor cavity.
• Two points in the cabin, averaged out.
• Several parameters reported by the car, including cabin temp and vent output temp.

I used electronic thermometers, which I calibrated to each other (though there wasn't much difference between them).

Initial findings were:

1. The CC's own cabin sensor (dotted green line) is reading temperature correctly (solid green line is the temperature measured by my probe inside the sensor area).
2. The system responds to the cabin sensor and warms the cabin to the set temperature, and then maintains it to +/- 1C.
3. There is always a tenancy to send warmer air to your feet, and cooler air to your face.
4. The system works fine at low fan settings
5. After about an hour stability is attained, with the control system cycling nicely and responding with warming and cooling air.
6. The cabin sensor reads about 7C too high, but appears to be compensated

Important observation:

Between 30 minutes and 60 minutes, the output air temperature gradually falls in order to maintain the cabin temperature (as the interior structure of the car warms up). This may be perceived as a fall in cabin temperature, although it can clearly be seen that it isn't - and I wonder if this is what some people are detecting and complaining about.

cc_1.gif

I then tried 4 variations by manually selecting certain parts of the system (have a look at the four sections of the graph below):

1. Turn the fan up to full, to see if the extra flow past the sensor made it read closer to actual cabin temperature. It didn't. If anything the extra blow from the foot area warmed the sensor up a bit.

2. Then I sent air to the face only, still on full fan. The face vent temperature dropped to demanded temperature, and a small amount of hot air from the footwell warmed the sensor yet further.

3. Next a reduction in fan speed to half (still face only) made very little difference.

4. Finally, sending all the air to the feet with the fan on full caused the foot temperature to return to 36, but the cabin sensor dropped by 5 degrees! Is the cabin sensor aspiration powered by a venturi in the footwell air system? It looks like it might be, but it still reads too high, so the sensor is still clearly influenced by the temperature of the footwell air.

It appears that the sensor is reading a mixture of cabin and footwell air, the system knows the temperature of the footwell air and is compensating to arrive back at a cabin temperature. Very odd (and amazing that it works at all)!

cc_2.gif

So I then decided to see the effect of footwell air on the cabin sensor reading.

At the bottom of the graph is the demanded temperature (as set on the CC temperature control). The fan is on full and the air is directed to feet only.

You can see that as the demanded temperature is increased, the foot output temperature increases and so does the sensor. These measurements were taken pretty quickly and the cabin remained at 22C the whole time. I can only assume that this increase is what's subtracted by the programmer to get to the cabin temperature.

cc_3.gif

Conclusions:

1. In this test, the system worked well. Of course, this is only for one set of circumstances (10 C outside, 22 C demanded inside).
2. The system is very odd, in so much that the cabin sensor is hugely affected by footwell air temperature and speed, but these effects have been programmed out by the system software.
3. There is no simple fault and workround to be found, so if you are getting a bit chilly, turn the temperature up a bit.

Here's a reminder of the cabin sensor aspiration pipe location:

ccpipe.gif

Here are some pics of the experiment:

cc1.jpg

cc2.jpg

Calibration:

cc3.jpg

[steeeveee]

Great write up, sort of explains the perceived fluctuation in cabin temperature. However this does not occur in my partner's Mercedes CLK where the air temperator is kept totally constant

Looks as though it's the sensor positioning and software control that could be improved on the Civic.

[Bog]

Excellent work Potsy, thanks for taking the time to do this

Originally Posted by Pottsy

Between 30 minutes and 60 minutes, the output air temperature gradually falls in order to maintain the cabin temperature (as the interior structure of the car warms up). This may be perceived as a fall in cabin temperature, although it can clearly be seen that it isn't - and I wonder if this is what some people are detecting and complaining about.

I think this IS the real answer to the issues with the CC - it basicly comes down to perception! As I and many others have said in other threads, some people are used to other systems and have different expectations - again, not really Honda's fault (or problem)

As you say, there is no fault and no work around (Unless Honda re-programme the system).

[surferjr1999]

Wow - very impressive Pottsy - however I do feel the start of a headace. I have noticed a drop in temperature after about 30mins and have just turned up control to compensate - not a big issue.
Where is the cabin sensor? I hadn't realised that the centre vents only cooled when not on dual - need to have a play.

[Bog]

Originally Posted by surferjr1999

Where is the cabin sensor?

First photo, next to steering column with the black probe with red wire sticking out of it.

[Pottsy]

Originally Posted by steeeveee

However this does not occur in my partner's Mercedes CLK where the air temperator is kept totally constant

Have another look at the graph - the cabin temperature is kept totally constant (to the nearest degree).

The CC does this by the only way it can, which is by varying the temperature of the air it expels.

[Charley Farley]

Hi Pottsy

Many thanks for the explanation. You've done alot of work. I have had my car for few days now and the first thing I noticed is that the fan is at the lowest setting initially when set to 21 degrees and auto which was a bit to low for me. I also found on cold mornings that it was best to set the system manually to get the cabin temperature up quicker. So overall I think the system works fine. At the end of the day its personal preference as to how you like the system set up.

[Pottsy]

CF, the fan should be slow when you first start up, since the coolant is cold and if the fan sped up it would just blow lots of cold (not what you need!). When the coolant warms up it blows lots of hot, then it reduces both the fan speed and temperature until your selected temperature is met. This should be the fastest way to get the cabin up to temperature.

You'll also find it does other clever things, like making the first blow go to your feet, to avoid sending residual wet (and misty) air onto the windscreen.

[elfa7]

Excellent write-up Pottsy, thank you.

As we've said before, stick it in auto, and let it do its thing.

[Parky]

Well done with the tests, but it does not reflect my experience.

I did my own test today on a 2 hour motorway journey from Warrington to Barrow, mostly Motorway or fast A roads.

The CC was set to Full Auto (i.e. everything on Auto, temp 22 degrees), I set this using the voice command "Climate Control Full Automatic".

I had a digital thermometer in the car with me on the passenger seat.
Here are the results :-

1. Starting temp in the car was 9C, all vents open
2. Started the run to Barrow, fan speed was low initially but increased to high as engine temp increased.
3. Temp of air was warm after 2 miles, cabin temp was now 11C
3. After 10 mins on the Motorway, fan speed started to decrease and cabin temp was 19C
4. A further 10 mins later fan speed was down to low and cabin temp reached 22C.
5. After 1 hour 15 mins, the cabin temp started to drop, indicated by 21C on thermometer, and it continued to fall to 18C over the next 10 mins.
6. The CC was still set to 22C, but cabin temp was now 18C with very chilly air being output, fan speed had not increased (as far as I was aware)
7. A further 10 mins and the temp had reduced to 17C and I gave in and increased the CC to 23C, which only upped the cabin temp to 18C.
8. On reaching Barrow the cabin temp was only 18C even though the CC was set to 23C with everything else on Auto.

This has been my experience on all of my long journeys (although not measured with thermometer) and it is not acceptable for a CC system.

While I can see that others have not had this problem, many other do see this and would like it to be resolved.

Ian

[Pottsy]

Ian - interesting. Have you checked that the pipe is connected to the sensor (paper test) and it would be interesting to see what your cabin temp sensor is reading when it does this.

This morning my car was blowing 13C for 22C set and 14 outside, but that's because the sun was out and the reading was high (72).

Maybe I ought to repeat my tests on a sunny day - I wonder if the system is over-reacting to light.

It would be useful to be able to record the error in detail, to make a case to Honda.

[pcr]

Pottsy,
Thankyou for this most useful information. It's a bit too much to take in at the moment but just a couple of comments.
For the same conditions as yours but in bright sun, I measure a peak of 25C from 10 to 20 mins before it settles to 22C. This maybe the sun or maybe the different engine. Anyway, this peak will increase any perceived temp drop.
Also, from your graphs it appears that the cabin sensor isn't controlling the temp from 60 mins onward.

[Pottsy]

Originally Posted by pcr

Also, from your graphs it appears that the cabin sensor isn't controlling the temp from 60 mins onward.

It looks that way - I wonder if it was the only other variable in the equation, the outside air temperature.

[pnx]

Great work Pottsy!
Don't you think that moving the sensor to another position could fix the problem?


FOR PARKY
I've exactly the same problem you described.

[richierich]

Great work Pottsy! I have noticed the problem mostly when at 20C on Auto and it is either completely dark or getting light, doesn't seem to do it in day light, will go and try the sensor though.

UPDATE!!

I never thought I would say this in a positive way, but my car sucks!!

[civicsport]

Just out of interest what influence does the light sensor have on the cc??, in the summer with the sun shining 22c seems a lot hotter than 22c in the winter, so there must be some kind of compensation factor when there is bright sunshine.

[apemonkeyman]

Just another example of why this site is so cool. Well done for all the excellent work on this & other threads!!

[pcr]

Parky,

At the risk of you throwing something at me, it sounds like yours is under control just like Pottsy's, it's just 5C too cool. What I mean is, once settled, it more or less maintains a constant temperature and when you demand another 1C you get another 1C. I would have thought the dealer could pick up on such a large error.

Just a thought, with a mind to spotting an unusual sensor reading, at the end of one of these cool trips, it might be worth jotting down the CC diagnostic readout. Here's mine after 40 mins at 22C in bright sunlight.

1 27 Cabin sensor
2 12 Outside temp sensor (trip computer says 11C)
3 26 Solar radiation
4 07 Air mix
5 34
6 33
7 F0
8 00
9 23

Note: I drove straight into garage, hence low solar reading.

[pnx]

Originally Posted by richierich

Great work Pottsy! I have noticed the problem mostly when at 20C on Auto and it is either completely dark or getting light, doesn't seem to do it in day light, will go and try the sensor though.

This could be true.
The first time the problem occurred I started driving at 6.00am when it was dark, and arrived 1 hour later with daylight.

This thing get complicated...


PS.: I believed the temperature standard unit of measurement in UK was Fahrenheit. So, do you use Celsius?

[DUNCAN]

Originally Posted by Pottsy

Ian - interesting. Have you checked that the pipe is connected to the sensor (paper test) and it would be interesting to see what your cabin temp sensor is reading when it does this.

This morning my car was blowing 13C for 22C set and 14 outside, but that's because the sun was out and the reading was high (72).

Maybe I ought to repeat my tests on a sunny day - I wonder if the system is over-reacting to light.

It would be useful to be able to record the error in detail, to make a case to Honda.

what is the paper test??