GM coolant sensor discrepancy

[nibblet787]

Hi all, I've spent countless hours/weeks/days/months building and installing my MS-II v3 into my 68 Camaro with a mildly built 350. Things seem to be coming along nicely, except for the coolant sensor. I bought the MS-II as a DIY kit from DIYAutoTune, and they included a GM coolant sensor. I have an Edelbrock Air Gap manifold, so I installed the GM sensor at the back (next to the distributor). During the whole MS-II installation, I also took the opportunity to install all new dash gauges from Autometer. This included an electric coolant temp sensor, which I then also installed in the intake manifold, but at the very front (next to the water neck).

Here's the problem: I'm getting MAJOR differences in temp readings between the two sensors. I'm using the newest beta of TunerStudio and, according to that, I'm getting differences of up to 40 degrees F or more, as compared to my Autometer gauge. It's not only differences in temp readings, but also in the rate at which the temp increases. For example, my dash gauge says I'm at 190, but TunerStudio reports that I'm at only 155, or so (I forget the exact numbers, but I'm not far off). I've calibrated the thermistor tables at least a dozen times (with the engine off, of course), hoping that something might magically change. I've used the GM numbers, and even experimented a little by trying the numbers from the other common sensor tables included with TunerStudio, but no luck getting the two sensors to agree. I've also just tried messing with the GM numbers until I get TunerStudio to report what the dash gauge says. That has worked, except that it doesn't work for the rate of temp change, which is still majorly skewed.

As of a few minutes ago, I just tested the GM sensor in a pot of water, going through the steps of ice water --> almost-boiling water. The numbers are pretty damn close to stock; here's what I got:

32.5 F ----- 8,970 ohms
111 F ------- 1270 ohms
186 F ------- 285 ohms
203 F --------200 ohms

So, can anyone shed some light on this? Could there possibly be that much of a difference between the front of the manifold and the back? My biggest problem is that my MS-II is thinking that my engine is still too cold (and therefore all the enrichments are still active), but the dash gauge says it's heading towards the boiling point because MS is keeping the idle way too high (trying to still warm up the engine).

[kjones6039]

Just a thought............

I am not particularly surprised to find 2 different temps at different points on the engine, although 40+ degrees is rather large.

I have a similar issue on 383 SBC. The electric gauge reads higher than the GM sensor connected to my MS. I just ignore the gauge and use the temp reported in TS and keep on tuning.

If it was me I would swap the location of the 2 sensors and see if the issue remains the same or follows the sensors.

Ken

[nibblet787]

I'm actually considering moving the front sensor (next to the water neck) back to the other rear coolant port. The intake has an identical port to the one that the GM sensor is currently installed in, but on the opposite cylinder head. Unfortunately, the rear ports are a different sized diameter than the front port, so having to go to the hardware store to buy yet another pipe fitting is not something I'd like to do. But, seeing as how I'm changing the water pump right now, this would be the time to move it.

Also, can anyone tell me if the resistor bias could be at fault? I'm 99% certain that I have the correct resistor installed (well, I'm certain that I installed the one that DIYAutoTune packaged for me), but just in case, I'd like to know if changing the resistor bias value would fix it. Checking the actual board would be a huge pain, because of the lengths I went to in tucking the MS box away to make things look clean, so I'm hoping more for a software fix.

[trakkies]

One easy check with the MS CTS and ATS is what do they read at ambient when the engine is left overnight? They should agree.

[904svo]

Did you copy the matfactor,thermfactor,throttlefactor.inc files over to My Documents\TunerStudioProjects\MyCar\inc after you used easytherm? If you didn't TunerStudio and Mega Tune will read different.

[nibblet787]

Unfortunately, my Autometer gauge only goes down to 100 F, and the daily temps around here (Detroit) haven't been hot enough to register on it (not complaining, mind you, lol). The MS does, however, report a fairly approximate ambient temp. My main concern is that, if the temps really are that much hotter at the front of the engine, then the thermostat is going to be opening a lot sooner than MS says it should be, which is the main cause of my problems with MS, since it's so dependent on getting accurate coolant and air temps.

I haven't touched EasyTherm or Megatune. I've considered it, but since TunerStudio does everything I want (especially the paid version), I figured I didn't need to deal with anything else. On top of that, the GM default values in TunerStudio come pretty close to agreeing with what I measured in the cooking pot, yesterday. When I was just messing with the thermistor numbers, I had to bump some of the Ohms values by well over a thousand points just to come close to what I needed. Even then, I could get the gauges to agree on a temp at that exact moment, but as soon as one of them changed, the other wouldn't. Or if they did, it would be something weird like, for example, my MS temp reading would very slowly creep up to 220 F, while my dash gauge still says 190 F.

I also forgot to mention that I'm using the lastest MS-II code version I could find, which is 2.905, if that helps.

[trakkies]

Unfortunately, my Autometer gauge only goes down to 100 F, and the daily temps around here (Detroit) haven't been hot enough to register on it (not complaining, mind you, lol). The MS does, however, report a fairly approximate ambient temp. My main concern is that, if the temps really are that much hotter at the front of the engine, then the thermostat is going to be opening a lot sooner than MS says it should be, which is the main cause of my problems with MS, since it's so dependent on getting accurate coolant and air temps.
[snip]

It was more just a rough and ready check they are correctly set in MS. Both the CTS and ATS agree with ambient on mine to+/- 1C at 20C.

Are you sure you have no airlocks in the cooling system? A pocket of steam might give a false reading.

[nibblet787]

Are you sure you have no airlocks in the cooling system? A pocket of steam might give a false reading.

Well, now this comment has me wondering if the ports at the back of the intake are just no good to use. As far as I know, those rear ports are not flowing ports. In other words, the sensor is just dipped in coolant that doesn't really go anywhere, except back the way it came (stagnant coolant?). If a sensor wasn't there, that coolant would bump up against the underside of the intake and just stop.

Meanwhile, the front of the intake has the water neck. So, the front sensor (the dash gauge, as well as the thermostat) is exposed to a constant flow of coolant via the water neck. Grrr... if that's the case then I need to find a spare port at the front. Damn heater hose, why must you always be in my way?

[kjones6039]

There are 4 plugs (2 per side) in 350 sbc heads. You can install a sensor in one of those that will be exposed to circulating coolant.

A word of caution though.......

Those plugs are 1/2 NPT and the GM sensor is 3/8 requiring a bushing to install it. Bushings tend to shield the sensor bulb from the coolant flow. Just curious..... Are any of your sensors installed with bushings now? If so, it could be a factor in your problem.

Ken

[nibblet787]

Actually, the Autometer gauge is using a bushing. But, A) it came with a bushing from the factory, and B) the actual sensor is 1/8 NPT (seriously, Autometer? 1/8 NPT?? Where are you expecting me to install it??)

Anyway, tomorrow is my day to finish the water pump installation, so I guess I'm headed to the hardware store to find a 1/2 to 3/8 bushing. Man, I've probably blown over $30 just on pipe fittings for this project that either didn't fit or didn't work out the way I thought it would. Those little bastards are expensive! Of course, if I hadn't ripped the labels off each time, I might've gotten my money back...

So as of right now, I'm gonna try the GM sensor in the cylinder head port, just under the exhaust manifold flange, and remove/reinstall the Autometer sensor to make sure the bulb isn't all "turtled up" (that's my new adjective) inside the bushing.

(side note: I realize there's a 1/8 NPT port just above the oil pan, next to the motor mount. But with thermal dynamics being what it is, I think that'll just end up reading colder than the sensor up top, thus defeating my purpose.)

[kjones6039]

"turtled up" works for me!

I realize there's a 1/8 NPT port just above the oil pan, next to the motor mount

I think you will find that plug is actually 1/4 NPT, generally used for the knock sensor. It will not accommodate the bulb on a coolant sensor anyway.

FWIW, I modify my 1/2 to 3/8 bushings to ensure that the bulb is always exposed, but that is a lot of work that you probably don't want to do.

Good luck....

Ken

[trakkies]

I'd be inclined to fit the MS sensor where the maker fitted the temp gauge one, if this is possible. And see if you get sensible readings on the MS by reference to the temperature the thermostat opens. If this works, it might make more sense to make an electronic interface to share that sensor with your dash gauge - perhaps someone has already done this and could supply the circuit?

[nibblet787]

Success!! I moved the GM sensor to the location next to the water neck, and relocated the Autometer sensor to the cylinder head. There's still a little bit of difference, but we're talking maybe 5 degrees or so. Thanks to everyone who chimed in and helped solve this annoying little problem!!

EDIT: I forgot to mention that I cut down both sensor bushings, so that the bulb would stick out a little more. Not sure how much it helped, but I feel better about their accuracy, at least.

[kjones6039]

I forgot to mention that I cut down both sensor bushings, so that the bulb would stick out a little more. Not sure how much it helped, but I feel better about their accuracy, at least.

That's how I do mine and it certainly seems to help.

Glad it worked out for you!

Ken

[convertables_10]

Is the original intake in question a professional products? If so It is setup for rear coolant return lines to aid in circulating water from the back of the engine. I run my sensors off the water neck area of the intake. It has a constant coolant flow. Glad to hear you got it figured out as well.