tacoma5050

My 2010 civic has a digital water temperature gauge. The gauge does not display numbers, rather it has a LCD bar graph. During the spring, summer and fall the gauge displays 9 bars when the engine is fully heated, which is just shy of the halfway mark on the gauge.

In temperatures below 32 degrees Fahrenheit, the gauge only rises to 4 to 7 bars, even when the car has been driven for over 15 minutes. An odd thing is, sometimes it will display a normal temperature of 9 bars, even on very cold days, say 25 degrees, but this is rare. The heater works OK, the air seems warm enough. It seems that if I turn heater off, that the temp will raise one or two bars.

Maybe these small engines just dont have enough mass to warm up in the winter?
Dealer says nothing is wrong with 2010 civic, but this does not seem right to me. Any ideas, thanks.

Heres the kicker, my old 2001 civic did the same thing in its older years, dealer said there was nothing wrong with that one either. I had replaced the thermostat in the 2001 thinking it was stuck open, but no help.

97emsedan

My 2010 Si has been run in temps as low as -36C, and my temp gauge goes up, slowly, but it goes up. I drive for 30 minutes to get to work and on the really cold days ( see temp above) it takes about 20 mins for the car to reach full operating temperature. Try shifting at 3K until the second bar shows up, then bump your shift point to between 4 and 5K. It allows the engine oil to heat up at a steady rate, and the coolant will warm up faster as well.

Did you replace your coolant with a different brand, say Water Wetter or some other race coolant? Or maybe it is not diluted to a 50/50 coolant water mix? More glycol than water? Are you running an 8 PSI radiator cap? Any of these would increase the boiling point of your coolant, and result in the temp rate to rise slower.

tacoma5050

Hi 97emsedan,

Radiator cap and coolant are original as shipped from factory.
My other two honda's, a Pilot and CRV, do not have this issue of taking 30 minutes to completely warm up on cold days. After about 5 minutes, their temp gauge is at full normal temp. While on those same cold days, my civic will never reach full temp even after 30 minutes. It just seems odd to me.

motegicivic

I would not worry as long as you have decent heat in the car. Those gauges are never 100%, just a guideline.

tacoma5050

I agree motegicivic, not much to worry about.

But my temp gauge must be fairly accurate, because for the 9 warm months of the year it is consistantly at 9 bars. It only fluctuates lower during the winter months.

dpetro1

If the car is not getting up to normal operating temp in cold weather, your thermostat is most likely stuck open. The thermostat's job is to regulate coolant flow depending on coolant temperature. If is fails open (even just the slightest bit) the car will not be able to get up to temp because coolant will flow through the radiator at all times.

if the coolant is too cold the ecu will compensate and add fuel which can decrease your fuel economy and is generally not good for the engine. the temp gauge should stay at a steady temp and not fluctuate.

you can remove the thermostat and immediately see if it is held slightly open, if so replace it with a new OEM unit. Only $20 or so from Honda

tacoma5050

I found this Honda Service News Article from Feb 2010, which states

http://dvpatel.homelinux.com/forumfiles/SN/A100200.pdf

- Temperature Gauge Reads Low When It's Cold Outside -
In very cold weather, your customer might notice that the engine temperature gauge appears to reach the normal operating level a little slower than it would in warm temperatures. This slower movement shouldnt be mistaken for a bad thermostat or any other cooling system problem, as this is typical operation of the temperature gauge. This is because the temperature gauge doesnt operate in a linear fashion like a fuel level gauge does. Check out the example below of this logic shown on a 2006-10 Civic:
Gauge segment 1 doesnt show until engine temperature reaches approximately 132°F or 56°C.
Gauge segments 2-9 are approximately 138-165°F or 59-74°C.
Gauge segments 9-11 Normal operating range is 170-222°F or 77-106°C.
Gauge segments 12-20 are approximately 228-249°F or 109-121°C.
Gauge segments 21-22 are approximately 255°F or 124°C.

On cold days my civic sometimes never rises above 6 or 7 segments, even after 15 to 20 minutes of driving. Which, according to the Honda Service Article, means that my civic is not operating within a normal range on cold days?

dpetro1

correct, all honda temp guages work this way. its just the newer civics have the digital display.

in cold weather it may take longer to get to normal temp, but once it does it should stay there. if it never reaches normal temp or gets to temp then goes back down, then there is a problem. almost certain it is the thermostat as i described above.

tacoma5050

Well the weather has finally turned cold enough to debug my civic coolant temp problem again.

Last year the Honda zone rep told me this is a common complaint of civic owners, and he explained it away by saying that it’s due to the difference between analog vs. digital temp gauges. He said that digital temp gauges respond faster than analog gauges.

But I just recently confirmed that my digital temp gauge in my civic is not responding any faster that an analog gauge.

Take a look at the two graphs below in PDF format. One graph is from my 2010 Civic and the other from my 2008 pilot. The graphs show the engine coolant temperature over 30 minute rides in 10 degree weather and from a cold start. The data was collected live via an OBD computer.

You can see from the Civic chart that the temp takes a very long time to rise to 180, then dips to almost 140, and never rises above 160 for the remainder of the trip.

For the same trip, the pilot coolant temp rose very quickly to 180 and stayed within 5 degrees of 180 for the whole minute trip.

So the temp gauge in the civic is accurate and the vehicle is not maintaining a temp near 180 like the pilot does. It would appear the thermostat is bad or sticking open, but Honda already replaced it once and it did not fix the issue. Honda says there is nothing else to check and that the car is working fine :-(

In addition to the graphs, I have also included the raw data files below:

https://www.dropbox.com/s/0jm1z5ju159g6va/08Pilot.pdf
https://www.dropbox.com/s/xhz7erdorft2c81/10civic-.pdf
https://www.dropbox.com/s/te3k7z2ib2...-09degrees.csv
https://www.dropbox.com/s/o6vx53nlwy...-10degrees.csv

97emsedan

I would trust the techs at Honda. It could be as simple as a faulty sensor in the coolant system.

In cold days here ( -27F, -33C) such as I had this week, my temperature gauge didn't rise at all after 5 minutes of driving... it went to 2 bars after 10. I had to rev to over 5K RPMs to get the gauge to get to normal and it still took 20mins to do so. T

he thermostat isn't open when you cold start the car, if it was...your gauges would never rise. It is closed when you start the car and opens when the temperature of the coolant rises to where the thermostat is triggered to open, then the coolant in the rad is run through the engine, and you might have noticed a temperature drop on the gauge. If the air temperature outside is really cold the car might not ever see normal operating temperatures, due to the efficiency of the rad, but as long as you have cabin heat and window defrost capability, then don't worry.

I have experienced what you went through, Canada is a bitch in the winter, colder than most other places on Earth...

tacoma5050

Hi 97emsedan,

The Engine Coolant Temperature Sensor ECT was checked and is good. And it works fine the rest of the year. We also took voltage readings directly at the ECT and they corroborated the data from the ECM.

Also, regardless of why it takes sooo long to reach the operating temp of 180, it should not drop below 170 once 180 is obtained.

Also you said: "...If the air temperature outside is really cold the car might not ever see normal operating temperatures, due to the efficiency of the rad..."

No matter how efficient the rad is or isn't, once the thermostat closes, water is no longer flowing thru the rad and thus should reach 180. Also why would it reach 180, then drop to 155?

This really has me baffled, it really seems like a thermostat that gets stuck open. But I can't image 2 OEM thermostats were bad. Plus the Honda zone rep said others have this same issue.

97emsedan

Well, it has me baffled then... the thermostat would be the obvious problem if the coolant temp levels drop, but I can't fathom a failure twice in a row...it sounds kind of like there is a shunt of some kind and the coolant is getting cooled off in a place other than the rad... maybe the heater core... could be a compounding of sources... I am stumped...

tacoma5050

"a shunt".... I never thought of that! It certainly would be a valid explainion of the symptom I am seing seeing. I will ask Honda if they are aware of a shunt in the system...thanks :-)

None2Slow

Before taking anything apart, try putting a piece of cardboard partially in front of radiator. This will limit the airflow across the the rad and help keep a more consistent water temp. If you ever see the big rigs, they have a cover that actually blocks the radiator and they can choose how much air they need to go through it. The thermostat allows the water to stay in the block until it reaches 180* then opens and allows the cold water in and holds the hot water in radiator until it cools down and the water in the block reaches 180* again. The colder the outside it is, the colder the water in the radiator will be. Then, when it enters the block, it takes longer to warm up. It is normal for your gauge to read lower in the winter time then in the other months. I had a 92 Thunderbird with the supercharged v6. The fan went out towards the end of summer. If I drove the car to work on the freeway, it was fine running about halfway up the gauge. In town, it would get to hot. I stopped driving it until winter. It never got more then 1/4 up the gauge. Even with an infrared temp gun, it never got more then 182* and this is with a 195* stat in it. Maybe try putting a temp gun on it nad wee what you record this time of year and what you get as the seasons change. I bet your going to get a lower reading this time of year. Also, try getting a reading with, and without the cardboard in place. See if there is a difference. Let us know what you find.

tacoma5050

None2Slow, Most of what you say makes perfect sense. My only question is: why is it then that when my Pilot(and every other car in my life) reachs 180 degrees, that it maintains that temp with 5 degrees for the rest of the trip, even in 10 degree weather? The card-board thing is a good idea for a test. I will ask Honda what they think, I dont want to void my warranty.

None2Slow

Your talking about 2 completely different vehicles with different engines and driving styles. The heavier Pilot has more load on the engine which will hold more heat in. I drive truck for a living and this was a "trick" we would use in cold places. Also, older cars have old technology. These new sensors are way faster to respond to changes. My 99F350 fuel gauge reads full 5 minutes after I fill it if I leave it running while fueling. My 2012 Civic SI reads full 30 secnds after. Look at the O2 sensors from back then. 1 wire is all they had and were heated by the exhaust. Nowdays they have 4 wires and are heated by the wires for faster reaction times. If the dealer says its normal, I would go with that. As for asking about the cardboard, I don't know where you live, but I would try it and not say anything to them. Yor not going to cover the whole thing. Maybe 25% and try that. I've been around cars since I was 14, now 40, but the basics are the same. More air in radiator, the cooler it runs. If your fan stops working, it overheats from lack of airflow. If you lived nearby, I would give you a hand. I've taken cars to the dealers for repairs and they couldn't find anything wrong. I would take them for a ride and have to point it out. I've decided to work on my own cars years ago and at 1 time almost opened my own shop.

1 more thing, if you go to the parts store and buy a scan tool that does live data, you can actually see what the engine is doing in real time. I have 1 just for this reason.

tacoma5050

I have narrowed this down to 3 possibilities that would explain this symptom:

1. The thermostat does not always close fully when needed. But unlikely since 2nd OEM thermostat did not change symptom.

2. There is a bypass of the thermostat in the cooling system by design or defect. Either in the thermostat it's self, or else where in the system.

3. The coolant temp sensor is mounted on the radiator side of the thermostat. Although, I would find this hard to believe.

A. Does any one know if there is a coolant bypass of the thermostat in the civic cooling system?

B. Does anyone know if the coolant temp sensor is mounted on the radiator side or the engine of the thermostat?

97emsedan

Don't cover the whole rad opening... I did this with my 97 Civic and my AC didn't work right... it needs some air to keep the condenser cool. Try just covering the lower half of the rad side of the front...

eddieE

This is a great thread. Any updates? My fiance has a 2010 civic that is taking about 15 minutes of driving to get up to normal operating temp. on the really cold days, while my car (2001 civic with 222k miles) was taking about 6 minutes at most on the same day. In her case, it does not drop once it gets to temp, although, she is going by the gauge in the car and we don't have actual temperature readings. The parts guy at honda said he has a 2009 civic that does the same thing when I went in to buy a thermostat. He convinced me not to replace the thermostat. Not sure how to move forward. I will be following this thread carefully.

Thanks

Eddie

tacoma5050

Honda USA finally admits my 2010 Civic may not reach "normal" operating temperature in cold weather.

Honda USA has all along been trying to “explain away” my cars symptom by saying that it was due to my misunderstanding of reading a digital temperature gauge vs. an analog temp gauge. They claimed the digital temp gauge is more responsive and thus I would see these temperature variations that I don’t see on my other cars that have analog gauges.

I knew they were wrong, because I can feel the air from the heater is just not that warm on very cold days.

So to disprove Honda, I collected data from the OBD on my Civic and my Pilot. The data shows the engine coolant temperature over 30 minute rides at 10 degrees weather and from a cold start.

Analyzing the data, it showed that the Civic coolant temperature takes a very long time to rise to 180, then dips to almost 140, and never rises above 158 for the remainder of the trip. (Note: on some trips it never got above 158)

For the same trip, the Honda Pilot coolant temp rose very quickly to 180 and stayed with 5 degrees of 180 for the whole trip. For both vehicles their temp gauge displayed accurately what the OBD was showing for current temperature.

So the data proves my temp gauge in the Civic is accurate and the vehicle is not maintaining a coolant temp near 180 like the pilot does.

From these symptoms it would appear the thermostat is bad or sticking, but the Honda dealer already replaced it twice when they confirmed this problem last year.

I gave this data to Honda USA, and now Honda finally admits that this symptom is not a result of the way that the digital temp gauge works, but in fact is a result of the cars design. Here is exactly what I received from Honda:

“The vehicle is operating within normal specifications given the conditions and climate. The best way we are able to describe it is that the efficiency of the cooling system in the car is such that when the temperature is extremely cold out, the vehicle may not reach "normal" operating temperature under the circumstances. The cooling system is simply working so efficiently, combined with the low ambient air temperatures, that the vehicle simply does not get that warm. The only way for it to reach a higher operating temperature under the circumstances would be to limit air flow into the engine compartment (for example placing a piece of cardboard in the front bumper opening in front of the radiator) which would help in increasing this operating temperature. Not that we recommend that, but just as an example. At this point, given the vehicle is operating normally and the HVAC system is operating normally, there is no further action that can be taken to change the situation that you are describing.”

BTW, I tried the “cardboard” suggestion and it did not help.

I guess I am just stuck with a Civic that does not have good heat on very cold days. Also Honda USA has refused to these questions: Do all 2010 Civics’ have this problem? Do Civics’ newer than 2010 have this same problem or has it been fixed? And if it has been fixed, how?

tacoma5050

Here are charts from the OBD data I collected from each vehicle.

disk_55FL

Thanks for sharing the data ........ my '07 runs cold in the winter also !

I guess it's designed that way ?? but , I guess that better than running to hot !

DaveSi677

I never had any of these problems on both my 06 EX and my 08 Si (when I drove it 1 winter)... They are always hitting normal temp and it get really cold here too like -40C.

tacoma5050

Hmmm...I wonder if there are any differences between my 2010 Civic LX and your 06-EX/08Si that would account for that? -40C is really cold, I wonder if the cars in your area have some design differences? Coldest it gets around here is -5F, but I start to notice the problem in early winter when temps start getting below +30F.

tacoma5050

To me the only thing that would make sense here is a thermostat that does not close all the way. But the Honda dealer claims they put a new one in and Honda USA acknowledges problem. I wish Honda would release a tech bulletin on this, but the zone rep told me that Honda would not be doing that. Maybe Honda had a hugh batch of faulty thermostats? ....seems unlikely for such a simple part.