Is this consistent with a failing coolant temperature sensor?

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Matt - 02 Jul 2008 05:02 GMT

My car has an analog temperature gauge. The reading is below 100F when
the engine is cold, then climbs as I drive, but doesn't go much above
160F.

Bad thermostat? But I have a receipt from the previous owner for
thermostat replacement five months ago. I just got the car.

Is the behavior of the gauge as I describe it above consistent with a
failing coolant temperature sensor?---namely that it works partially but
is inaccurate---reads 35 degrees F lower than the actual temp or tops
out at 160F?

I am seeing other symptoms that seem to be consistent with a too-cool
engine though.

1995 Park Avenue Ultra
3800 supercharged V6

Reply to this Message

Erik - 02 Jul 2008 07:54 GMT

> My car has an analog temperature gauge. The reading is below 100F when
> the engine is cold, then climbs as I drive, but doesn't go much above
[quoted text clipped - 13 lines]
> 1995 Park Avenue Ultra
> 3800 supercharged V6

I bet the past owner put in a 160°F thermostat. It's probably supposed
to be a 192°F.

A quick IR thermometer reading off the thermostat housing (once the
engine is fully warm) will tell the story... the receipt from the past
owner may even have info on the thermostat installed.

The computer will never go closed loop running that cold...

Erik

Reply to this Message

Matt - 02 Jul 2008 17:23 GMT

>> My car has an analog temperature gauge. The reading is below 100F when
>> the engine is cold, then climbs as I drive, but doesn't go much above
[quoted text clipped - 16 lines]
> I bet the past owner put in a 160°F thermostat. It's probably supposed
> to be a 192°F.

I was thinking that myself. But the past owner was not a technical guy,
and 160 F seems rather extreme. When I look for thermostats for my make
and model at advanceautoparts.com, they offer only 180 F and 195 F. The
owner's manual lists 195 F for the stock thermostat.

Maybe the shop (Car-X) had a 160 F thermostat laying around and didn't
notice or just wanted to get rid of it. Or would there be some reason
to put in such a cool one? The repair was done in Minnesota. The
problem was that the heater wasn't working. Could the choice of 160 F
be related to the coolant leaks that this engine is prone to? There was
a class action settlement lately for that problem. Maybe they promote
cool thermostats so the customer's engine will be cool sooner when they
need to work on it so they can get you in and out sooner.

> A quick IR thermometer reading off the thermostat housing (once the
> engine is fully warm) will tell the story...

Aha. Maybe I'll drop by some service station and ask them to check it.
I would buy an IR thermometer if they're $20.

> the receipt from the past
> owner may even have info on the thermostat installed.

> The computer will never go closed loop running that cold...

I don't know which loop you mean. Explain?

The other running-too-cool symptoms I hinted at were:

1) the oil looks brown and not too clear, even though it was changed
only about 300 miles ago, and

2) the oil level is about 2/3 qt. too high on the dipstick.

I expect most of the high level is simply from overfilling, but could
the oil pick up that much blow-by in 300 miles at 160 F? Maybe I should
check the PCV valve too?

My impulse is to put in a stock (195 F) thermostat, but if there is a
reason to go with 180 F instead, I would be happy to listen. I don't
know much of the pros and cons about using a non-stock thermostat. The
car has 160K miles in case that matters.

> Erik

Reply to this Message

jim - 02 Jul 2008 19:51 GMT

> > The computer will never go closed loop running that cold...
>
> I don't know which loop you mean. Explain?

The computer relies on the info from the coolant temperature sensor. That is
different from the sending unit for the dash gauge. Does the coolant fan come on
in stop and go traffic on a hot day? If it does that would be a good indication
your gauge is off.

> The other running-too-cool symptoms I hinted at were:
>
> 1) the oil looks brown and not too clear, even though it was changed
> only about 300 miles ago, and

This could be related to the engine not being hot enough but it is much more
likely you need to change the oil a little more frequently to get rid of
accumulated grime that is probably simply due to not changing the oil often.
After a couple of frequent changes it will probably not get dirty so fast.

> 2) the oil level is about 2/3 qt. too high on the dipstick.
>
> I expect most of the high level is simply from overfilling, but could
> the oil pick up that much blow-by in 300 miles at 160 F? Maybe I should
> check the PCV valve too?

If the oil is overfilled and stays overfilled I don't understand how you
conclude that means there is excess blow-by?

Checking the PCV and the hose is certainly worth doing. Is the oil milky brown
or just brown? The oil will pick up whatever excess grime is inside the engine.
In theory the oil keeps the grime suspended and gets eliminated with an oil
change, but if you don't change the oil often enough then it will start to
deposit the grime inside the engine and that can be why fresh oil gets dirty
fast.

> My impulse is to put in a stock (195 F) thermostat, but if there is a
> reason to go with 180 F instead, I would be happy to listen. I don't
> know much of the pros and cons about using a non-stock thermostat. The
> car has 160K miles in case that matters.

If the reason the t-stat was replaced was to get the heater working it probably
already is 195° t-stat. If that is the case, that would suggest the temp gauge
just reads low. If your cooling fan comes on, it is unlikely that it is running
too cold.
Get the temperature checked. If the temperature gauge is not reading
accurately I wouldn't worry too much. It should still do it's job. It will tell
you how long it takes to warm up and if the engine starts to overheat you should
see the needle move upwards.

-jim

> > Erik

----== Posted via Pronews.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.pronews.com The #1 Newsgroup Service in the World! >100,000 Newsgroups
---= - Total Privacy via Encryption =---

Reply to this Message

Matt - 02 Jul 2008 22:20 GMT

>> 2) the oil level is about 2/3 qt. too high on the dipstick.
>>
[quoted text clipped - 4 lines]
> If the oil is overfilled and stays overfilled I don't understand how you
> conclude that means there is excess blow-by?

I wasn't clear. I mean maybe the full condition is all or partly due to
blow-by condensing in the oil rather than being overfilled at the last
oil change. Could the oil level rise 2/3 qt in 300 miles assuming 160 F
t-stat and a PCV problem?

Reply to this Message

Matt - 02 Jul 2008 22:24 GMT

>>> 2) the oil level is about 2/3 qt. too high on the dipstick.
>>>
[quoted text clipped - 10 lines]
> oil change. Could the oil level rise 2/3 qt in 300 miles assuming 160 F
> t-stat and a PCV problem?

I'll try again.

Could the overfull condition be all or partly due to blow-by condensing
in the oil rather than due to being overfilled at the last oil change?
Could the oil level rise 2/3 qt in 300 miles assuming 160 F t-stat and a
PCV problem?

Reply to this Message

Matt - 06 Jul 2008 16:26 GMT

>>> The computer will never go closed loop running that cold...
>> I don't know which loop you mean. Explain?
>
> The computer relies on the info from the coolant temperature sensor. That is
> different from the sending unit for the dash gauge.

That is surprising---not that I would know.

In the intake manifold I found two devices other than the injectors.
One is toward the rear of the car, (on the side with the even-numbered
bank of cylinders). When I unplug it, the gauge stops working. The
other is on the end of the manifold beneath the throttle body. When I
unplug that one, the gauge is unaffected. Your post seems to imply that
that one gives input to the computer. Which one controls the warning light?

I'm thinking that if one sending unit is bad, maybe the other is too.
Mileage is 160K.

Maybe I should get somebody to hook the car to a computer for general
diagnostics. Presumably a bad temperature-sending unit would be
detectable. Should I go to the GM dealer for the diagnostic---or would
a large repair shop have the right equipment?

Reply to this Message

N8N - 23 Jul 2008 20:12 GMT

> > In article <x5Dak.64213$t44.1...@fe105.usenetserver.com>,
>
[quoted text clipped - 38 lines]
> Aha. Maybe I'll drop by some service station and ask them to check it.
> I would buy an IR thermometer if they're $20.

is $30 too much?

http://www.amazon.com/gp/product/B000FJR0SG

it's a cool tool, I was just looking for mine this past weekend
(trying to calibrate temp gauge on my '55 Stude)

nate

Reply to this Message

Scott Dorsey - 02 Jul 2008 14:20 GMT

>My car has an analog temperature gauge. The reading is below 100F when
>the engine is cold, then climbs as I drive, but doesn't go much above
>160F.
>
>Bad thermostat? But I have a receipt from the previous owner for
>thermostat replacement five months ago. I just got the car.

Maybe. Maybe they used a cheap one that broke open. Maybe they used
the wrong one, that is set for too low a tempeature. The first thing I
would do is to measure the temperature at the radiator and compare it
to what the shop manual says it should be. The second thing I would do
would be to get a known-good thermostat (NOT a cheapie from a chain store)
and replace it.
--scott

Signature

"C'est un Nagra. C'est suisse, et tres, tres precis."

Reply to this Message