Scan tool doesn't agree with gauge

Status
Not open for further replies.
Joined
Jan 3, 2006
Messages
1,965
Location
Ohio
I picked up an Actron scan tool at a Sears store closing sale. This one reads codes, clears codes, and shows live data. Coincidentally, I had to dig into the Traverse's instrument cluster to see if I could find the flaky connection that was making my driver information display go out. In doing so, I had to take the dial needles off to get to the circuit board (danger, Will Robinson!). Didn't find the flaky connection, but apparently it's no longer going blank.

Checking the gauges after I got it back together, the speedometer agrees with the GPS, and everything agrees with the scan tool except the coolant temperature. It appears though, that it's always been off since the gauge reads the same. It normally hangs around 210F, but the scan tool says it's really only about 180F. Anyone know what temperature thermostat would be in a Traverse? I'll believe 180 and the gauge is just supposed to tell you if it's stone cold, normal temperature, or overheating.
 
The dash gauge is most likely a "dummy" gauge meant to tell when the engine's cold, normal, or overheating. Trust the scan tool on this one.
 
Originally Posted By: Dave Sherman
I picked up an Actron scan tool at a Sears store closing sale. This one reads codes, clears codes, and shows live data. Coincidentally, I had to dig into the Traverse's instrument cluster to see if I could find the flaky connection that was making my driver information display go out. In doing so, I had to take the dial needles off to get to the circuit board (danger, Will Robinson!). Didn't find the flaky connection, but apparently it's no longer going blank.

Checking the gauges after I got it back together, the speedometer agrees with the GPS, and everything agrees with the scan tool except the coolant temperature. It appears though, that it's always been off since the gauge reads the same. It normally hangs around 210F, but the scan tool says it's really only about 180F. Anyone know what temperature thermostat would be in a Traverse? I'll believe 180 and the gauge is just supposed to tell you if it's stone cold, normal temperature, or overheating.


Well, maybe not

I do enough instrumentation and calibration with SCADA and DCS systems to say that it is hardly unusual to have a sending unit disagree with a physical measurement. ( which is why we calibrate in the first place) A 30 Delta is cause for concern though.

That gauge is supposed to tell you exactly what temperature probe is saying- if its not then one or both is malfunctioning. (You say you removed the needles? We do that sometimes and you need to be real careful putting them back with the gauge in a known temp source so the gauge is recalibrated and sometimes a dab of super glue helps keep it there)

On the meter- you need to make sure the meter is pulling its data from the same source or they both could be right ( or wrong) because they are pulling from different senders and/or locations.

If it were my vehicle, I would check and be sure
 
Originally Posted By: Dave Sherman
I picked up an Actron scan tool at a Sears store closing sale. This one reads codes, clears codes, and shows live data. Coincidentally, I had to dig into the Traverse's instrument cluster to see if I could find the flaky connection that was making my driver information display go out. In doing so, I had to take the dial needles off to get to the circuit board (danger, Will Robinson!). Didn't find the flaky connection, but apparently it's no longer going blank.

Checking the gauges after I got it back together, the speedometer agrees with the GPS, and everything agrees with the scan tool except the coolant temperature. It appears though, that it's always been off since the gauge reads the same. It normally hangs around 210F, but the scan tool says it's really only about 180F. Anyone know what temperature thermostat would be in a Traverse? I'll believe 180 and the gauge is just supposed to tell you if it's stone cold, normal temperature, or overheating.


the gauge might have a "buffer zone" were it will show 210 while the water temp fluctuates up and down depending on how hard you drive it, I would assume this would be done to keep people from thinking there is something wrong with their car because the temp won't stay consistant. But if you think there is something wrong, try a new temp sensor and see what happens
 
Ive noticed that OBD scanners don't always report the correct voltage as compared to a multimeter.
 
It's a good question...does the sensor which drives the "three position" gauge provide an output which can be used to drive an accurate by the degree reading as well, or the the FI system rely on the same unit.
 
You might have more than one water temperature sensor in your engine. OBD2 might simply be reading a different sensor.
 
Sometime around 25 years ago, car manufacturers started having the ECU intercept the temp from the sensor, and feed a modified reading to the gauge. This way, owners do not complain about a gauge going up a bit when they climb a hill.

So......When the sensor sees "Normal", the gauge notches in at a predetermined spot on the gauge. It will not move, unless the sensor read really high or low.

If you want to verify what is really happening, try using a infrared temp gun.......Point it right where the sensor is mounted. That reading will give you some idea of whether the scan tool is telling the truth. You can bet the gauge is filtering the truth, to save you from yourself.
 
Originally Posted By: 4wheeldog
Sometime around 25 years ago, car manufacturers started having the ECU intercept the temp from the sensor, and feed a modified reading to the gauge. This way, owners do not complain about a gauge going up a bit when they climb a hill.

So......When the sensor sees "Normal", the gauge notches in at a predetermined spot on the gauge. It will not move, unless the sensor read really high or low.

If you want to verify what is really happening, try using a infrared temp gun.......Point it right where the sensor is mounted. That reading will give you some idea of whether the scan tool is telling the truth. You can bet the gauge is filtering the truth, to save you from yourself.


Technically that's system hysteresis which is pretty much built into every type of instrumentation except when there is an absolutely critical need for all the instantaneous changes in real time. ( even they have it because there is always a lag from event to send)

Point is that hysteresis is limited and temporary ( usually a few seconds max) so if the normalized reading after that doesn't change then the loop needs to be checked.
 
Not on car temp gauges. The vast majority are made to stay "centered" through a wide range of "normal" temperatures. My temp gauge in my truck stays centered from 160F on up. Never waivers right up to the highest temp i've seen on my scangauge of 221F. Thermostat is 191F. It's exactly like 4wheeldog describes.
 
Originally Posted By: tcp71
Not on car temp gauges. The vast majority are made to stay "centered" through a wide range of "normal" temperatures. My temp gauge in my truck stays centered from 160F on up. Never waivers right up to the highest temp i've seen on my scangauge of 221F. Thermostat is 191F. It's exactly like 4wheeldog describes.


With all due respect, no that is not correct for any gauge made by anyone in any application in this universe.

The gauge is an output device that either by potentiometer, bourdon tube, magnetism, diaphragm or whatever is going to read in real time whatever the input signal tells it to.

Hysteresis is either built in at the RTD or hard/soft coded in the EPROM, PLC or whatever device is interpreting it because the time lag between affecting the sensor until it reaches the meter is a fact of physics.

Just like on HVAC/R or other process gauges, there is ( by design) a period of "fuzzy logic" ( google that in the reference to process controls) to keep things in a steady state until things stabilize.

In most applications they are also range triggered such as long as the signal is between setpoint a and b it will average at c and alarm if over and then cascade ( go all the way one way or the other) Thus why they call it "fuzzy logic".
 
Just because you don't believe this is correct doesn't change the fact that most automotive manufacturers do what they do.

Mazda does this with just about any vehicle that has a coolant gauge (RX-8, Miata, Mazda 3, Mazda 6, etc).

Go out there, and take a look for yourself, and educate yourself.

BC.
 
This gauge doesn't seem to buffer out fluctuations, it actually does climb up slightly going up hills, and falls slightly going down hills. I was just surprised that it had temperature markings but the scangauge showed something significantly different.

I think there is some truth to having gauges filter out these fluctuations though, my CR-V would remain rock solid once it was warmed up, but word was that there was a dead zone where it would not go higher unless the temperature got significantly above normal, as somebody found out with a scangauge. Even if it it sat idling to the point where it was hot enough to turn the cooling fans on, the gauge never twitched. My old Saturn I could watch the gauge creep up slowly, then once the fans kicked in it would drop down.
 
Originally Posted By: Bladecutter
Just because you don't believe this is correct doesn't change the fact that most automotive manufacturers do what they do.

Mazda does this with just about any vehicle that has a coolant gauge (RX-8, Miata, Mazda 3, Mazda 6, etc).

Go out there, and take a look for yourself, and educate yourself.

BC.


Again, with all due respect.

I did not post a "belief" I posted a fact. Popular opinion, voting, lack of understanding et al does not alter that.

automotive manufacturers do not have their own laws of physics, secret techniques or anything else.

I don't have to go "out there" because "out there" comes to firms like mine to compute those values that go in those PFD's and P&IDs and calculate those parameters and we work out the logic based on what they want.

Hopefully I educated you now. Feel free to subject everything I have posted to any qualified authority you wish and see what they say. Then you will learn.
 
Dummy gauges, like on Fords are well known that as long as the actual temp is between two values it reads on temp as perfectly normal so that the user is not questioning all the time the value. It is not averaging the value but outputting a set value as long as input is within a certain range.
 
Originally Posted By: Nate1979
Dummy gauges, like on Fords are well known that as long as the actual temp is between two values it reads on temp as perfectly normal so that the user is not questioning all the time the value. It is not averaging the value but outputting a set value as long as input is within a certain range.


Absolutely correct, whether it is averaging a value or has a timing effect depends on a lot of variables such as whether it is analog or digital or even worse having an analog input going through a transducer to a digital device. ( literally a million ways to do that)

Then it depends on if it is a direct read instrument ( gauge) or is correlated in the EPROM to take readings from multiple inputs.

This is why without the logic ladder and loop the system works on you cannot really say
 
On both of my Infinitis, the Consult will report accurate temperatures, but the temp gauge on the dashboard hits center at 160f and stays there until 225f. Above 226F the gauge moves again.
 
Status
Not open for further replies.
Back
Top