what is wrong with my car?

June 19, 2009 12:12 PM

Why is my car overheating in traffic jams?

My 96 Honda Civic (you're jealous, right?) overheats in traffic jams after ten, fifteen minutes.

The fans kick on for a while: the temperature will crawl upward as I sit there, fans kick in, temperature drops.That keeps happening until the upward creep exceeds the capacity of the fan to cool it. The temperature indicator crawls upward until it gets about 3/4 up and then I crank the heater to diffuse the problem.

I took it in last week and the mechanic could find nothing wrong. He ran it for a couple of hours, he said, checked all the fluids, monitored the temperature, and gave it a once over to be sure it would pass inspection (which is coming up.)

He kind of convinced me that I was imagining it, that it wasn't really unusually high, just maybe that I didn't notice it before. So yesterday I let it creep up to just a tick beneath the red zone, took a cell phone picture, and then cranked the heat. The picture I think will convey the urgency to him a little better, when I take it back to him.

Other maybe relevant details: as soon as the traffic jam ends and I speed off the engine cools down. I commute long distances routinely and have problems only sitting idle in traffic. Since he said he ran it for a couple of hours, I'm assuming he ran it in park. That's the main difference between my traffic jam experience -- in traffic, my foot's on the brake. We have a lot of construction going on around here and creep along at five miles an hour.

He's quite competent and a good guy who we know personally. Is there something I can tell him to look for?

Please note that I am really, really, really ignorant about cars and engines.

posted by A Terrible Llama to travel & transportation (29 comments total)

Is the fluid level in the catch tank going down any?
Does anything smell like french toast?

I would normally guess a stuck thermostat, but if it can idle for an hour, then that isn't it. I would say that the insides of your radiator are coated with scale. It forms after a long time of use and acts like an internal insulating blanket. What that does is keep the heat on the inside from getting to the cool on the outside.

The scale makes the radiator muck less efficient, but doesn't prevent it from working at all. Thus sitting still it had plenty of cooling capacity. Creeping in stop and go traffic, you exceeded its limited capacity and then it heated up.

At this point flushing (washing the system out) won't help the radiator much, but it may clean out the engine itself some.

My recommendation would be a new radiator. With how common your car is, that shouldn't cost much over $100.
posted by Antidisestablishmentarianist at 12:19 PM on June 19

muck=much
posted by Antidisestablishmentarianist at 12:20 PM on June 19

As my dad explained to me...

(1) Because the lack of movement means your engine block isn't getting the additional cooling power of moving air, despite being water-cooled.

(2) If you're too close to the car in front of you, your car will suck up their hot exhaust.

In cases where the car isn't moving and we can tell the car (which is as old as yours) is in imminent overheating danger, we open all the windows and turn up the heater. Another option is to just turn off the car.
posted by Ky at 12:21 PM on June 19

Well, if your fans are kicking on, then your fans and any related temp sensors are probably fine.
There are two possibilities that come to mind:
A faulty thermostat or other blockage in the cooling system.
A faulty water pump that isn't circulating coolant well enough.

The thermostat normally acts as a blockage when the car is cold, to help it heat up by restricting coolant flow through the radiator. As the car heats up, the thermostat opens and allows coolant flow. If it's sticking shut, you've got no circulation and the hot coolant can't pass through the radiator.

Some water pumps are comprised of a metal gear, driven by the timing belt in your case, that drives a plastic impeller in the engine block. It's not too uncommon for the plastic impeller to crack and frewheel on its shaft, allowing the drive gear to spin just fine, while not circulating any coolant.

How many miles do you have on this car, and do you know when the last timing belt service was?
Also, is your coolant all brown and crusty? You could have restricted flow in your cooling system.
posted by Jon-o at 12:21 PM on June 19

How many miles on your car? When was the last time your had your coolant system, timing belt, thermostat inspected?
posted by special-k at 12:25 PM on June 19

At idle, your car doesn't generate enough heat to get fully in the red zone. If its already hot from driving it around, it may generate enough heat at idle to keep it from cooling. I suspect if you drove the car around until it got totally hot and then put it in park, you'd be able to reproduce this.
posted by bensherman at 12:27 PM on June 19

Just to follow up: the mechanic idled the car for one hour and it did not overheat. To my mind, that rules out a cracked impeller completely, and thinking again, the thermostat could be stuck half open, and that would act the same as a bad radiator.

So, I would have the thermostat replaced, $20 part, $25-50 labor.

If it still overheats I would do the radiator $100 part, $100-200 labor.

That is my revised opinion.
posted by Antidisestablishmentarianist at 12:28 PM on June 19

Thanks for the responses so far....

Some more info:

Like 160K something miles. Maybe 170. Lots.
Timing belts -- I can't remember when, but recently--within the last couple of years
Thermostat inspected-don't know off the top of my head

I do keep every piece of paper every mechanic ever gives me at home in a folder, I'll see if I have anything when I got home re. thermostat inspection -- but wouldn't a thermostat work or not work, not work partially?

Here's something on paperwork from the dealer (they looked at it before the mechanic):"Car is running hot found both upper and lower rad hoses are about the same temp. Start with replacing thermostat and coolant flush."

I felt like they wanted to go on a fishing expedition so I bailed--I'm just icked out on the idea of throwing money at things if I don't *know* they're broken, but maybe that's how it works for cars. I just felt like, isn't there some kind of test they could do so I wasn't "starting with" replacing one thing and then moving on to replacing a bunch of other things when it doesn't work?
posted by A Terrible Llama at 12:41 PM on June 19

There is a way to test this without "throwing parts" at your car.

I would carry out the following tests:
First, I would open the radiator cap and run the car until it's at operating temperature. I would then observe the flow of coolant and see if I could gain any information about if the thermostat is opening or not. Once the car was hot, I would turn the heat on full blast. If the climate system doesn't deliver enough heat, then I suspect that the water pump isn't circulating coolant through the system. At that point, I would take the timing cover off, run the car, and use a stethescope to listen for any irregular water pump noise and maybe get lucky and hear something.
I would then road test the car until it starts to reproduce the overheating, bring it into the shop and use a non-contact thermometer to measure the radiator tempertature accross the unit. If it's hot, and there's no deviation, then it's clogged and not cooling properly.
posted by Jon-o at 12:50 PM on June 19

Coolant doesn't last forever. If fans are working I suggest flushing radiator, replacing coolant & thermostat. Another point as motors get older they produce more heat than new.
posted by patnok at 12:54 PM on June 19

If it were me, I'd drain the radiator, fill the system with water, add a jug of radiator flush, run for the official 10-15 minutes, drain it again and fill it with water and see what happens. I'm leaving the part about things working fine and then forgetting to replace the pure water with water and anti-freeze off this list (This is worth doing every few years anyway.)

If that didn't fix things I'd replace the thermostat and make sure the belt driving the water pump was good and tight. While I was getting covered in grease and my hands all scratched up I might replace the radiator and heater hoses if they'd never been replaced before and I was planning on keeping the car for a few more years.

So, when all is said and done, I think the dealer's "start" list was not too far off base.

If those two don't fix the problem then this could get spendy fast. In most cases the radiator flush takes care of it though.
posted by Kid Charlemagne at 1:00 PM on June 19

"The fans kick on for a while: the temperature will crawl upward as I sit there, fans kick in, temperature drops.That keeps happening until the upward creep exceeds the capacity of the fan to cool it."

Are you seeing a temperature drop on the indicator? Because if the fan can reduce the temperature early in the cycle, I'm wondering why they don't later on.
BTW, one, perhaps counter-intuitive, thing you can try during these episodes is turn on the A/C, possibly with heat as well, if you need more effectiveness. Counter-intuitive because A/C increases the engine load slightly; effective because, in most cars, this forces the radiator fan to keep running. Do watch your temp indicator and listen for the fan the first time you do this, and it actually reduces your engine temperature. Obviously this is not a fix; just a band-aid.
posted by TruncatedTiller at 1:02 PM on June 19

The Cooling System Story by Art's Automotive in Berkeley, CA.

I was going to suggest rodding the radiator but, as is detailed on the above webpage replacement is probably a better way to go. In any case, I think that you'll find an answer there.
posted by bz at 1:10 PM on June 19

Another "quick" temporary fix could be to add something like this to your existing antifreeze. While there's probably a root mechanical cause to your issue (or old coolant), usually these radiator additives can help in the short term while you figure out your issue and are typically less than $10. Granted, most modern radiator fluids usually contain some sort of surfactant/surface tension reducer for water, if your fluid is old it may need a refresh.

I've been through a similar situation with a '99 Isuzu Rodeo I had, but I was even more upset because the vehicle was only a few months old when it started happening to me - and I did the "roll-down-the-windows-turn-heat-to-high" trick mentioned above and it sucked on the hot summer day I was stuck in traffic.
posted by JibberJabber at 1:21 PM on June 19

Here's something on paperwork from the dealer (they looked at it before the mechanic):"Car is running hot found both upper and lower rad hoses are about the same temp. Start with replacing thermostat and coolant flush."

If it is true that the upper and lower rad hoses are about the same temperature, then I think you can eliminate the stuck halfway closed thermostat option.

If the thermostat were stuck halfway closed, then you would have a low flow rate through the radiator, which, yes, would mean insufficient cooling overall, but you would expect that the coolant that did make it through the radiator would be significantly cooler coming out than going in (it would have plenty of time to cool off). In other words, if the coolant going into and coming out of the radiator is approximately the same temperature, then you know that the radiator isn't doing its job properly. If you know that there is something wrong with the radiator, there is no point in messing around with the thermostat right now.

Flush the coolant first, especially if it is old, and see if that fixes your problem. If it doesn't, then it is time to replace your radiator.
posted by ssg at 2:28 PM on June 19

Is the fan staying on? I had a similar problem years ago with a VW Golf and a friend of mine installed a manual fan switch for me. Totally solved the problem.
posted by miss tea at 2:57 PM on June 19

I'd bet on the water pump. The key diagnostic element here is that the temp creeps up in cycles. That is, the fans cool it down, then the heat cycle builds up, and eventually overwhelm the fans. To me that indicates inefficient distribution of coolant due to the water pump (or slipping belts) - not enough coolant circulates to actually be effective, and eventually the fans can't do all the work.
posted by VikingSword at 3:05 PM on June 19

I had something like this happen with my car. It did turn out to be the water pump, if I remember correctly.
posted by AnnaRat at 3:49 PM on June 19

I do not have any mechanical advice to add, only sympathy if you live in a warm climate and have to crank the heat in summertime traffic jams. I've been there, and it's close to my idea of what hell must be like.
posted by 8dot3 at 4:00 PM on June 19

I had this happen with an old car I had. Tell your mechanic to replace the PCV valve.
posted by aninom at 11:45 PM on June 19

The rubber hoses that connect your radiator to the engine block can soften over time, particularly the lower hose, which feeds cooled water to your water pump; some lower hoses are even made with a coiled spring inside to resist collapse under the sucking action of the water pump, as they age. If you haven't had your radiator hoses and engine accessory belts changed in more than 3 years, it's a good idea to do so, and fairly inexpensive. This is usually combined with a cooling system "flush and fill" service, where the cooling system is flushed with a caustic chemical to remove scale from the engine block and radiator, then further flushed with clean water, and finally re-filled with the proper protective ratio of fresh coolant and water, which will inhibit further corrosion for a couple of years, and provide some anti-oxidation protection to the plastic and metal parts of the cooling system, like the thermostat, water pump and hoses.

The symptoms of a collapsing lower hose closely mimic your description of the temperature tending to run away during stop and go driving; at low idle revs, the hose can resist collapse barely enough to allow some coolant flow, but with each rise in engine revs beyond idle during "acceleration," (even creeping forward), the coolant flow is choked off a bit, resulting in climbing temps. Restoring air flow through the engine compartment cools the softened hose, too, temporarily restoring its ability to dribble a little coolant through to the pump, so that the car seem to return to normal temps at normal driving speeds.

An experienced mechanic checks this in about 2 minutes, simply by squeezing the hoses a bit when the engine is warm and running. A softened hose is obvious.
posted by paulsc at 2:40 AM on June 20

"Car is running hot found both upper and lower rad hoses are about the same temp. Start with replacing thermostat and coolant flush."

If this is true (the part about the temperatures being the same), I would proceed as suggested. That probably won't work and you'll need to replace the radiator.

The fact that the mechanic idled it for an hour isn't a great test. Unless he idled it in gear, on a freeway, in traffic, sitting in the sun.

Car cooling systems are all about equilibriums. With an on-the-edge cooling system, sometimes the slightest thing can make the difference between the system rejecting enough heat to not overheat, and not. Airflow, humidity, ambient temperature, in the sun or in the shade, etc.
posted by gjc at 3:16 AM on June 20

Thanks everyone for responding and for educating me. I'm going to compile a list of items for the mechanic to look at. I kind of think the water pump might be the deal, it turned up on some Google searches I was doing, but I'll just offer up a bunch of things.

I did specifically ask him if a car in park simulated the experience of me sitting there with my foot on the brake, and he assured me it did...so maybe I'll take it to someone else next time. Even as ignorant as I am about cars, that didn't make any sense to me.

Thanks for the sympathy, 8dot3. It gets even better--there's a baby in the back seat. So far it hasn't happened in a big way on a hot day with her, and I've been leaving early to try to avoid the traffic, but I've got to get it resolved because I have this horrible vision of myself stuck on the interstate during a half mile stretch where three or four converge.

I'll see if I can get someone to look at it Monday.

Thanks all.
posted by A Terrible Llama at 9:09 AM on June 20

Update: If it chokes and dies as I'm accelerating to merge on the interstate, that's bad, right?

Because that's what happened this morning.

I had it towed to the garage. Well, that's one way to force a resolution. Except now I don't even know how many problems I have.
posted by A Terrible Llama at 6:43 AM on June 22

Did it overheat when you were getting on the highway?
posted by Jon-o at 10:59 AM on June 22

No, I'd driven forty miles and had dropped off the baby. I was accelerating to merge back on when it happened. The temperature gauge wasn't running high, either.

I managed to pull over to the shoulder where it died with a shudder. A few minutes later I could re-start it, but let them tow it as I had no idea what had happened in the first place.

The mechanic thinks water pump, off the top of his head, but can't look at it until Wednesday.

It doesn't seem too great.
posted by A Terrible Llama at 2:27 PM on June 22

Don't worry too much, even though it's looking like you might have more than one problem with this car. A cooling system failure will rarely cause the car to shut off and only if there's some serious overheating going on, so your choking and dying experience is the result of a failure in another area. Since you're car is a 96, it will be OBD-II compliant, and I'd be really surprised if the Check Engine Light didn't come on with a code related to whatever cause your car to die. It could wind up being something really simple like a clogged fuel filter or a bad ignition coil that acts up after extended driving.
If you're wary about your mechanics diagnosis, just write back and I'm pretty sure that we'll be able to sort you out.
posted by Jon-o at 3:39 PM on June 22

Late to the party, but I thought I'd chime in with my very similar experience.

I have a 1990 Acura. A couple years ago, I noticed the temp gauge would "climb" occasionally. Never overheated, but got close. Coolant level was OK. So, I took it to the mechanic. Long story short, he replaced/modified everything having to do with the cooling system. The following Thanksgiving, stuck in traffic, the car overheated. Took it to a mechanic who pointed out some minor flaw in the work the original mechanic had done. So, I thought, problem solved.

But no. Just a few weeks ago, driving into Manhattan, the gauge started to creep up, until I noticed that coolant smell, and some steam coming out from the hood. Pulled over, and let things cool. After letting it sit, and replacing a little coolant, I continued on with the heater blasting (fun!). Shortly after, I had my mechanic install an override to the fan. His final analysis was that there was a problem with a controller chip. Apparently, the chip was not telling the fan to turn on, even though the engine was hot. And so far, the gauge has not creeped at all.
posted by ObscureReferenceMan at 12:55 PM on June 23

Update: Something something head gasket something something


Eight hundred dollars is the best case scenario.

For pete's sake.
posted by A Terrible Llama at 1:16 PM on June 25