Repair guides and support for the 5th generation Honda Accord coupe, sedan, and wagon.

59 Questions View all

temperature gauge pinned hot

when starting the car the temp. gauge jumps to red hot ....I've been driving the car for 15 minutes max. one way thinking that the car may overheat but so far the car seems normal at these short distances.

Answered! View the answer I have this problem too

Is this a good question?

Score 1


Hi, Have you checked the coolant level in the coolant reservoir?


Yes, coolant level is fine! The temp. gauge goes to red hot or pinned as soon as I start the car cold. And...after approx. 15 min. driving locally, the engine doesn't seem to be overheated. so I'm assuming that the thermostat is working fine. I'm just afraid to drive it longer or farther when the temp. gauge says red hot.


Add a comment

2 Answers

Chosen Solution

Hi @jamesjew,

Since you are experiencing no overheating problems the most likely cause is that the engine coolant temperature sending unit is faulty. This sends a signal directly to the dashboard gauge. There is another coolant temperature sensor close by which sends a signal to the ECU to help in engine management. You shouldn't get confused between the two as the one you want only has one wire and the other has two.

Here is an image showing the two sensors. The one on the left, where the single wire black connector has been disconnected is the sensor sending unit to the dashboard gauge, the one on the right with the grey connector (and being held by fingers) is the ECU sensor.

Block Image

Not quite certain if my figures are correct , but to test the sending unit, disconnect the single wire black connector and using a DMM (Digital Multimeter - Ohmmeter function) measure the resistance between the sending unit terminal and earth (or ground - the engine block will be earthed). With a cold engine the resistance should be approx 140 Ohms. With a hot engine it should measure 30-50 Ohms. If you measure around this value (or definitely not near 140 Ohms) with a cold engine then the unit is faulty and needs to be replaced.

A quick test is to disconnect the lead from the sending unit, turn on the ignition and observe the temp. gauge. If it registers 'zero' then most probably the sending unit is faulty. If however it still registers a 'full scale' reading then either you have removed the wrong lead, there is a fault in the wiring or there is a fault in the instrument cluster/temp gauge

Note: Be safety aware - always switch OFF the engine when you are going to work on it

Was this answer helpful?

Score 4


Thank you Jayeff! Your diagnosis including the photo was correct. I installed a new temperature sending unit this morning and the temperature gauge is now working properly. Thanks again for your help! James


Add a comment

I have similar problem:

upon cold start gauge on dash goes to full red in 1 or 2 seconds.

When disconnecting single wire coolant sensor, and starting car gauge on dash stays at zero.

However, when replacing with new sensor, gauge on dash goes to full red at startup, very surprised by this.

(not possible the car is red hot yet at cold start)

This is the new sensor:

Block Image

Wondering what else can check for, seems this coolant temp sensor is at the end of the line of things. There is nothing else feeding into that dash gauge right?

Was this answer helpful?

Score 0


@xyzabc if the sensor is not the problem the the gauge is your next likely culprit . if however the sensor has a ground wire check the ground first as a faulty ground can cause this but since the sensor is metal by the picture its probably self grounding


Hi @xyzabc ,

If the gauge stays at zero with the lead disconnected then check the new sensor with an Ohmmeter.

The resistance of the sensor between the body of the sensor and the sensor connector tag should be ~140 Ohm when cold. (you can do this out of the engine)

If it reads ~30-50 Ohms or even s/c (short circuit) when cold then the sensor is faulty.

You can also check that there is no visible connection between the brass body of the sensor and its' silver connector tag, due to a manufacturing defect From your picture there is an insulator (brown ring) there. Use a magnifying glass if you want to. A small shard of metal is all that it takes ;-)

If the readings aren't 140 Ohms or s/c or 30-50 Ohms when cold, check that you have the correct part.


It looks like the reading is .57 for the new sensor while still installed, which I think would not affect the reading. I'll take it out tomorrow for up close examination for burs. Also perhaps I am reading the meter incorrectly, this is when I set it to 2k, when I set it to 20k it still the same. So this .57 means that it is roughly half an ohm? Or does the .57 mean 57ohms or 570 ohms because the range is set to 2k, I've been trying to google how to read the ohm meters, but I don't get why it's not 1k 10k instead of 2k 20k and whether the number on the screen is still ohms or some ratio of the current setting. When I set the meter to be just 200, then it just reads 1 when connecting the leads to the sensor, my guess is the meter is not sensitive enough at that setting.

When I test the old part, it reads .610 when the ohm meter is set at 2k and 0.61 when set to 20k so I believe this means I should read it at less than one ohm.

Perhaps I got a bogus part. I'll bring my ohm meter to napa tomorrow too.


Hi @xyzabc ,

.570 Ohms is 570 Ohms on the 2 K ohm scale but it should read 0.05 (possibly 0.06) on the 20 K Ohm scale, depending on the type of Ohmmeter. (see link below)

Your readings for the old part of 610 Ohms is way too high according to what I've found out. The readings are as I posted in the Chosen solution. This is if it is for the same vehicle make and year model as shown above that is. So check that measurement again.

Also check what your meter reads when you just s/c the meter leads together it should read 00.0 Ohms on the 200, .000 on the 2K and 0.00 on the 20K Ohm ranges. It may possibly read 00.3 up to 00.9 and varying when on the 200 scale depending on how good a s/c connection is between the leads. Oxidization of the metal probes may introduce a bit a resistance into the circuit. You'll notice the reading change as you rub the lead's test probes together until it stabilizes.

Changing the scale only means that you have to interpret the readings a bit differently. If the reading is showing 1 when the meter is connected to whatever you're testing, its' resistance is either above the scale set or it is o/c (open circuit) and it will display 1 on every scale. (It should also show 1 when the meter is set to Ohms and the test leads aren't connected to anything i.e. an o/c.)

If the resistance was 570 Ohms it would read 1 which represents that it is above the chosen scale maximum reading, if the meter was on the 200 Ohm scale but .570 on the meter if it was on the 2K Ohm and 0.57 on the 20 K Ohm scale. Whereas if it was 57 Ohms it would show 57.0 on the 200 Ohm scale and .057 on the 2 K Ohm and 0.05 (possibly varying to 0.06) on the 20 K Ohm scale.

Conversely if it was 5.7 Ohms it would read 5.70 on the 200 scale, .005 on the 2K Ohm and possibly 0.00 on the 20 K ohm scale and if it was 0.57 Ohms it would most probably read 00.5 (and varying) on the 200 Ohm scale and most probably .000 on the 2K Ohms and 0.00 on the 20 K Ohms scales as the meter isn't accurate enough on these scales for the low measurements.

The best way to read resistance is to start on the highest scale and switch down until you get the most accurate reading, which is usually the one above the one that shows "1" unless you have a good idea of what it should be e.g. .574 on the 2K scale is more accurate than 0.57, which is all that would show on the 20K scale ;-)

Hope this makes a bit of sense.

The above the usual way for most low end but adequate DMMs (Digital Multimeters) which have what is known as a 3.5 digit display to show the readings, but it may be different for meters with different display resolutions. If you have the user instructions for the meter it should state how the individual scales are displayed for Ohm readings.

Also check that the lead to the sensor isn't being pushed on too far and bridging over the connector tab onto the body of the sensor so that it is shorting to earth and causing the gauge to read high temp.


Add a comment

Add your answer

James will be eternally grateful.
View Statistics:

Past 24 Hours: 38

Past 7 Days: 300

Past 30 Days: 1,612

All Time: 12,413