[BO185]

Knowing a resistance of the sedner is worthless unless you know the actual coolant temp hitting that sender to verify it. Shooting the sender with an IR gun will give you its body temp that's it not coolant temp, granted yes they are close. An IR gun on rubber hose close. but won't be 100%.

So again till you get an actual COOLANT temp number, with a know good gauge, you're pissn in the wind.

Based on your IR gun readings alone,doubt there is wrong with your setup is most likley your sender as people here have posted over and over.

 

[55 Tony]

Sending Unit

Here is what I'm using on my 56 temp gauge. It works well with no added resistors or anything. Goes up to almost 3/4 normally and higher with temps over 200. It's a GM part. It's a big fat sucker, no adapters needed.

 

[johnonetrillion]

From your photo, it's clear you've got the wrong sending unit. It should have the pin-type connector.

Yes, that is certainly true. It's mostly likely for a '57 and has had the connector swapped.

And that means the spare I have is also not correct – it too would be for a '57.

I have just gone to the source of the units supplied by Zip (Corvettes) and ordered one direct ... from Lectric Limited.

https://www.lectriclimited.com/elec...sending-units/temperature-sending-unit-108007

The fellow I spoke says the GM service replacements have been made for a range of years and do not match up with the resistance figures for '56.

Thanks for all the info (and also the PMs).

Cheers

 

[johnonetrillion]

Temperature sender resistance figures ohms

When I initially searched before posting this thread I missed this because the term 'sender' was not in the title.

It has, as you point out, all the resistance figures required to test these items.

Digging through the archives: https://www.trifive.com/forums/showthread.php?t=9213&page=2 has some useful info. If the specs shown are correct, you won't see a difference when the engine is cold, but you'll definitely see it when warm. The resistance goes down when the temperature goes up. At 220 degrees, 100 ohms +/- 10% for the '56 but 80 ohms +/- 10% for a '57 would explain why a '56 gauge would read hot with a '57 sending unit.

10% either way is quite an error margin, right?!

For future search reference, here is the post and the data:

https://www.trifive.com/forums/showpost.php?p=1268447&postcount=14

56 Chevy Electrical Sender for Temp Gauge (NOT the same as a 57)
½”-14 NPT Thread
100°F Ohm Value 469.20Ω - 312.80Ω
220°F Ohm Value 111.65Ω – 91.35Ω

57 Chevy Electrical Sender for Temp Gauge (NOT the same as a 56)
½”-14 NPT Thread
100°F Ohm Value 441.60Ω – 294.40Ω
220°F Ohm Value 88Ω – 72Ω

 

[johnonetrillion]

Knowing a resistance of the sedner is worthless unless you know the actual coolant temp hitting that sender to verify it. .

Right you are.

Except there is value in knowing if the sender is correct. (With the wrong one, the reading at the top of the gauge is meaningless – the needle just past the highest dot can easily mean 210˚F or 240˚F ... there's no relativity left.)

 

[johnonetrillion]

Also, here's a pic of the gauge reading when the temp sender unit pin is earthed.

(I never got that high when driving, btw.)

 

[Churchkey]

Just speculation on my part I don't have an early car.

Are you supposed to have an instrument voltage regulator on this model?

High gauge readings on some vehicles can be caused by a failing IVR.

The IVR's are normally mounted on the back of the cluster.

 

[Rick_L]

Are you supposed to have an instrument voltage regulator on this model?

High gauge readings on some vehicles can be caused by a failing IVR.

The IVR's are normally mounted on the back of the cluster.

I thought the regulator was built in to the gauges themselves - but I can't verify that.

 

[Pops @ Tri-5.com]

Are you supposed to have an instrument voltage regulator on this model?

High gauge readings on some vehicles can be caused by a failing IVR.

The IVR's are normally mounted on the back of the cluster.

No IVR on a TriFive

 

[BO185]

Right you are.

Except there is value in knowing if the sender is correct. (With the wrong one, the reading at the top of the gauge is meaningless – the needle just past the highest dot can easily mean 210˚F or 240˚F ... there's no relativity left.)

And only way to know if the sender is correct is knowing what the coolant temp actually is which you have no clue what it is hence know resistance is point less if since you don’t know if coolant is 200* or 100* without a Thermometer in the coolant. Or remove the sender and placing in hot water with a thermometer and reading Reaistance. But if doing that just install a new correct one at that point.

 

[angs1957]

You can place a resistor in series with the sender you have, it will adjust your gauge to read approximately from about 1/2 to just over and increase to hot as the temp rises, may be an easy fix for you.

 

[stoveboltgeek]

And only way to know if the sender is correct is knowing what the coolant temp actually is which you have no clue what it is hence know resistance is point less if since you don’t know if coolant is 200* or 100* without a Thermometer in the coolant. Or remove the sender and placing in hot water with a thermometer and reading Reaistance.

...which is exactly what I plan to do.

I just bought a NOS '56 temperature sending unit. I have an infrared thermometer and a professional-grade voltmeter.

My plan is to put the sending unit in a pot of water on the stove and measure the resistance at 10 degree increments. Then, we'll settle once and for all the difference between the '56 and '57 sending unit calibrations.

 

[55 Tony]

...which is exactly what I plan to do.

I just bought a NOS '56 temperature sending unit. I have an infrared thermometer and a professional-grade voltmeter.

My plan is to put the sending unit in a pot of water on the stove and measure the resistance at 10 degree increments. Then, we'll settle once and for all the difference between the '56 and '57 sending unit calibrations.

Only problem is the infrared red thermometer is probably only good for + or - 10°F each measurement. Just set an old fashioned thermometer in there. Clear water and shiny stuff makes for even less accurate readings. The closer you get the better it is, but I don't know how steam on the lens will affect it. Again, a real thermometer will do a much better job.

 

[stoveboltgeek]

Only problem is the infrared red thermometer is probably only good for + or - 10°F each measurement. Just set an old fashioned thermometer in there. Clear water and shiny stuff makes for even less accurate readings. The closer you get the better it is, but I don't know how steam on the lens will affect it. Again, a real thermometer will do a much better job.

Okay - I have stovetop style thermometer (sometimes called a candy thermometer) I can use as well. After all, I'm all about accuracy. And, I call myself "stoveboltgeek" for a reason.

 

[BO185]

Depends on the IR gun. My fluke IR gun is perfectly on, passed calibration every year when I use to have to get it calibrated for work. Place sender tip in pot of hot water to test resistance. Then you can verify temp with IR gun, temp prop on volt meter, and a drop in thermometer now you can verify all 4 easily.

The temp probe on a volt meter like a fluke is most accurate typically.

 

[stoveboltgeek]

Depends on the IR gun. My fluke IR gun is perfectly on, passed calibration every year when I use to have to get it calibrated for work. Place sender tip in pot of hot water to test resistance. Then you can verify temp with IR gun, temp prop on volt meter, and a drop in thermometer now you can verify all 4 easily.

The temp probe on a volt meter like a fluke is most accurate typically.

On further thought - it also depends on whether the aiming light on my IR thermometer is a LED or a laser. If it's a laser, it's too dangerous to use on reflective surfaces like water or stainless steel pots.

I don't have a laboratory grade thermometer like we used back in chemistry class.

Interesting you should mention a temperature probe for a Fluke voltmeter. I have a higher-end Fluke meter, but didn't buy the temperature probe. Hmm...and my wife just asked me for ideas for Christmas...

 

[55 Tony]

Depends on the IR gun. My fluke IR gun is perfectly on, passed calibration every year when I use to have to get it calibrated for work. Place sender tip in pot of hot water to test resistance. Then you can verify temp with IR gun, temp prop on volt meter, and a drop in thermometer now you can verify all 4 easily.

The temp probe on a volt meter like a fluke is most accurate typically.

I didn't know they could work that well. A Fluke huh. How does it do on shiny or reflective surfaces? I admit, I have one from horror fright, I don't use it to measure temperature so to say, more to compare one temp to another, taking multiple readings of each location.

I used to tease my buddy that it's a fluke that his Fluke DMM works. I just don't like it because it's autoranging and it takes so long to take a measurement. It may sound stupid but when you are used to a quarter second and then it takes 1.5 seconds, it seems like an eternity.

 

[Pops @ Tri-5.com]

You can place a resistor in series with the sender you have, it will adjust your gauge to read approximately from about 1/2 to just over and increase to hot as the temp rises, may be an easy fix for you.

I believe the gauge won't be accurate Angelo....The resistance of the sender is not linear as the coolant temperature changes.

 

[stoveboltgeek]

I believe the gauge won't be accurate Angelo....The resistance of the sender is not linear as the coolant temperature changes.

Pops, let me add a bit more detail.

The engineering term for these devices is a "thermistor" - a resistor whose resistance changes predictably over a range of temperatures. They're typically specified as change in ohms per one degree change in temperature.

For the moment, let's assume that spec is linear over the range of 100 to 220 degrees Fahrenheit. Looking at the specs we dug up the other day:

56 Chevy Electrical Sender for Temp Gauge (NOT the same as a 57)
½”-14 NPT Thread
100°F Ohm Value 469.20Ω - 312.80Ω
220°F Ohm Value 111.65Ω – 91.35Ω

57 Chevy Electrical Sender for Temp Gauge (NOT the same as a 56)
½”-14 NPT Thread
100°F Ohm Value 441.60Ω – 294.40Ω
220°F Ohm Value 88Ω – 72Ω

If you're a geek like me, you'll notice that the 57 Chevy 220 degree spec is actually 80 ohms +/- 10%. Likewise, the 56 220 degree spec is 101.5 ohms +/- 10%. The lower spec is much looser (+/- 20%).

So, taking the nominal values for each, we find:

(101.5-391 ohms) / (220-100 degrees) = -2.41 ohms/degree F (for '56)
(80-368 ohms) / (220-100 degrees) = -2.40 ohms/degree F (for '57)

Now if we plot this on a traditional x-y chart (x=temperature, y= resistance), we'd see a) the slope of the two lines are only slightly different (-2.41 vs. -2.40), and b) there's an offset between them. (21.5 ohms and 23 ohms at either end)

Now, if the measurements I'm proposing reveal the relationship is indeed linear (constant -2.41 or -2.40 Ω/°F over the range), the lines will be straight. Then, you ought to be able to fudge it by putting a 22 ohm resistor in series with a '57 sending unit to make a '56 gauge read correctly.

But, if we plot the measurements on the above-mentioned chart, and the lines aren't straight, a 22 ohm resistor would still work, as long as the difference between the lines remains constant (21-23 ohms). If not, then all bets are off.

Anyway - I'm going to check with a buddy of mine to see if he has a NOS '57 sending unit I can borrow for this test. If not, I'll use an aftermarket I've found to work well in my '57 (BWD WT203). And, we'll see what's what.

Yeah - living up to the name "stoveboltgeek"...

John - if there's an electronics supply store in your area, get a 22 ohm resistor (1 watt, just to be on the safe side) and see if your gauge reads correctly when warm. I'd go with a 5% or tighter tolerance, but the composition (carbon, wire-wound, film) doesn't really matter.

Also, this conversation does make me wonder if the guy who sold you this car got rid of it because the temperature gauge said the engine always ran too hot, and he could couldn't figure out why.

 

[BO185]

On further thought - it also depends on whether the aiming light on my IR thermometer is a LED or a laser. If it's a laser, it's too dangerous to use on reflective surfaces like water or stainless steel pots.

I don't have a laboratory grade thermometer like we used back in chemistry class.

The laser in pointers on IR guns won't damage your eyes persay unless you getting it right in your eye for mins on end, most people are smart not to do so. Biggest issue is the IR reading off a reflective surface will if false readings. Chrome for example in powder coating I can shoot it in the oven and its IR reading will be off.

"Class 2 visible-light lasers are considered safe for unintentional eye exposure, because a person will normally turn away or blink to avoid the bright light. Do NOT deliberately stare into the beam -- this can cause injury to the retina in the back of the eye. Be aware of beam reflections off glass and shiny surfaces."

nteresting you should mention a temperature probe for a Fluke voltmeter. I have a higher-end Fluke meter, but didn't buy the temperature probe. Hmm...and my wife just asked me for ideas for Christmas...

Yeah the temp probes are handy. Can use in water or air.

I didn't know they could work that well. A Fluke huh. How does it do on shiny or reflective surfaces? I admit, I have one from horror fright, I don't use it to measure temperature so to say, more to compare one temp to another, taking multiple readings of each location.

Any reflective surface will not be accurate IR gun wise as the sensor will get false readings from reflected IR. So have to watch that.

I used to tease my buddy that it's a fluke that his Fluke DMM works. I just don't like it because it's autoranging and it takes so long to take a measurement. It may sound stupid but when you are used to a quarter second and then it takes 1.5 seconds, it seems like an eternity.

Most flukes you can it to whatever you want reading wise they have alot of options on them.