Interesting transmission temp findings.

[booster]

I have been continuously messing with trans, coolers, thermostats, etc to get a better idea of what is really going on with the Chevy transmissions, which do seem to get hot in the mountains for many people.

After our last trip in the mountains last spring, that showed we were still a bit susceptible to having the trans get hot per the Scangauge, I took the standalone oil cooler for the engine and used it in parallel with standalone trans cooler we had. Neither of them had been into the radiator any more, so I put the engine oil back into the radiator.

We ran the setup without a thermostat on the trans coolers for most of this summer, and found that the trans would run very cold, showing 105* at 50* ambient, of the Scangauge. The 4L80e didn't like the low temps and would modify the shift points to be as if just started, with much quicker downshifts and longer time in gear before upshifts.

I put a 185* thermostat on the coolers yesterday and took it out for a ride today in 81* ambient. We got a fairly wide range of temps ranging from 170*-195*, but that was expected because the thermostat is remote and controlling the temp from the oil it sees, and the Scangauge is reading the temp in the trans, with the sensor appearing to be mounted "in the pan per GM" which really means on the bottom of the valve body per the drawings in the manual.

When we got home I crawled under and took the temp of the trans pan and the engine oil pan to see what they were compared to the temp readings I was getting. The engine pan was within 10* at about 194*, but the trans pan was about 50* different than the Scangauge reading at 135*, which is also the temp the computer uses to determine shiftpoints, etc. The trans pan temp is also essentially the temp of the trans oil in the pan, and the reading most of us are familiar with because that is where all trans temp sensors used to mount, right in the pan oil through the pan.

Needless to say, this is a might big difference in trans temps, especially when all the temp charts say that life of a transmission starts to go south at 170* (on non synthetic) or 20+ degrees more for synthetic.

The 50* seen is also a very big coincidence in the actual readings. 135+50 is my 185* thermostat reading. It would also put you right at the 170* old school limit for long life from a pan sensor when your Scangauge read 220*, which we hear of all the time. GM sets a too hot malfunction code at 260* for 10 minutes or more, which would be 210* at the old way of measuring temp and probably about right for synthetic fluids.

Back in the day, everyone said never go over 200* or you will toast your tranny, so if you make that 210* now for synthetic, you are right at where GM sets a code, and I think takes power off the engine to some degree.

We have heard from several users that they had gotten to 240* repeatedly on the Scangauge and showed no signs of burned fluid, which never really made sense, but if you really had 190* pan oil temp, it makes plenty of sense and wasn't too hot for synthetic oil.

It would be interesting if anyone has a Scangauge and IR temp gun to see if they get similar readings, with a stock radiator cooler setup and/or with the GM add on cooler. If the 50* difference stays relatively constant, it would give a better feel for how hot is too hot off the Scangauge. EG, when you get to 240* on the Scangauge, will you really get 190* on the pan?

I had read a few places that people felt the Scangauge temps were about 20* higher than pan temps, but never anywhere near 50* higher.

[markopolo]

It's unlikely to have changed but might be worth checking to see if the Scangauge X-Gauge code has been updated: https://www.scangauge.com/support/x-gauge-commands/gm/

[markopolo]

Originally Posted by booster

................. The engine pan was within 10* at about 194*, but the trans pan was about 50* different than the Scangauge reading at 135*, .......................

So the Scangauge on the Chevy van was showing around 185* & the thinking is that it is reading high?

Maybe folks with other OBD gauge devices could also share some temp reading comparisons between pan temp and gauge display temp. Comparisons of Scangauge and other OBD devices would be useful also.

[booster]

Originally Posted by markopolo

So the Scangauge on the Chevy van was showing around 185* & the thinking is that it is reading high?

Maybe folks with other OBD gauge devices could also share some temp reading comparisons between pan temp and gauge display temp. Comparisons of Scangauge and other OBD devices would be useful also.

I don't really think the Scangauge/internal temp sensor is reading higher than it should, as it is in line with so many others we have heard about and correct at all the ambient temps we have seen. Mostly just different than the pan temperature is, so all our old rules and experience for temperatures don't really apply.

Absolutely, that would be very good information if anyone can get it, as I mentioned at the end of the original post.

I was back reading the factory service manual again after the test run and did see some interesting things about faults they will set based on rate of change of the fluid temperature, both from too fast a change and from too slow a change at startup. Those temps seemed to indicate that the temp sensor sees changes in load generated internal temps faster than a pan sensor would. This does make sense as the sensor is not in the bottom or middle of the pan fluid, but right at the top of the pan/bottom of the valve body, so it likely sees the hot fluid coming draining back out of the trans before it gets mixed in with the cooler fluid in the rest of the pan.

We also see some funny temp swings that are quick to happen and then slow to balance out again, mostly to the cooler fluid side, but sometimes to the hotter side, which may be because some of the fluid going through the transmission is return oil from the cooler so if there was a batch of warmer fluid from a quick acceleration or short steep hill, it would likely show up right away as the oil to the cooler comes right from the torque converter and that is the hottest oil in the system especially when not in lockup. Not having the radiator fluid cooler hooked up makes for more and faster swings in temperature, it appears, because the thermostat can't react extremely quickly, while the water in the radiator changes temp pretty slowly and moderates the transmission fluid temp swings.

Another thing I realized (finally) is that the fact that the fluid coolers seemed to get quite hot nearly all the time in the past when only one cooler was being used (they still do now with two coolers, but not quite as hot). I was seeing cooler temps nearly to the temp that would be considered hot for the transmission. I think what goes on is that they pull out the very hottest fluid in the system at the torque converter, which apparently can be quite a bit hotter than the actual transmission temperature, so the coolers could run at surface temps that are quite high.

I think it will be interesting to watch the temps with a thermostat in the system again to see if some of the way it works starts to make more sense.

[markopolo]

Assuming it's a voltage from the sensor on your Chevy being converted to appear as degrees on the OBD device would this post: https://www.ar15.com/forums/general/...ge=1#i45401449

help explain the too fast, too slow & swings you noted?

I've been researching how to know if the formulas in apps are actually accurate and came across that post.

[booster]

The factory service manual does have a voltage vs temp chart, so I would be able to get that information, although I would still need to remove the sensor from the trans so I could control, and know, what the actual temperature is.

I think that the faster swings are because of the location of the sensor, right at the bottom of the valve body where hot oil from the various relief ports and bearing oil return holes will dump back into the pan.

The more I think about it, there probably isn't a hard and fast conversion that we could do from the Scangauge temp to the pan temp we used to use with a gauge through the midpoint of the side of the pan. Steady state output would probably be pretty consistent for a conversion, but transitions where you were on and off the power, particularly if the converter locked and unlocked would probably not be as reliable a conversion, with larger swings in temps from the Scangauge.

Of course the big question is if the Scangauge temp is better to go by, which is likely, or if the old pan temps were OK even though missing high, short, spikes and delivering more of an average temp. The fact that GM says you can have high temp for 10 minutes before they do anything to protect anything might lean towards average temp not being all that bad, though.

Next time I have the pan off for a filter change, I think I will just braze in a gauge sensor fitting the old way, so I can compare the temps if I chose to in the future.

[Hondo]

Back when I had my Chevy 3500 based Jamboree with the 4L80E & 6.0, I had both the Scanguage 2 & a separate temp sensor hooked to the internal transmission port, read by a temp gauge-

Here is the gauge I bought- a Glowshift Maxtow Digital/Analog Transmission Temperature Gauge-





It came with the sensor, a cable to run between the sensor and the gauge, and a cable to run between the gauge & the truck wiring-



The sensor can be mounted in a few places but I chose to install the sensor in the test port in the side of my 4L80E transmission-



I stuck my cell phone up there to get a real world image & to find the plug to be removed-



After I removed the plug, I screwed the sensor into the transmission and ran it's cable into the truck & under the dash.

The gauge has 4 power wires to connect in the power harness-

1- Black - to ground
2- Red - to constant 12v
3 -Orange - to 12 volts connected to the headlights (for dimming)
4- White - to switched 12v (12v with the key ON)

I had no problems with the red, black & white wires- finding those sources was easy enough.

But the orange wire needed to be connected to 12v that is present when the headlights are ON in order to auto dim the gauge. I didn't feel like searching for that (and with daytime running lights, and 2 headlight circuits) I went another route.

I decided to add another switch to the panel that would allow me to select the 12v & manually dim the gauge.

I installed the push button ON/OFF switch into the lighting control panel. It popped out of the removed dash easily-



I flipped it over and drilled out a hole with a holesaw-



Then I drilled the specified sized hole into the panel, and inserted the switch (with the wires attached) into the panel-



After the switch was installed & the wires through, I sealed up the hole with a small circle of plastic glued in place.

Now the switch will apply/remove 12v to the orange wire & enable a manual dimming mode. Once in the dim mode, you can hold the button on the gauge and then select 3 brightness level lower than the bright daytime setting.

I mounted the gauge in an Autometer single steel pod like this-



I mounted it on the top of the dashboard right at the left pillar. I'll post up a day time pic of it tomorrow.

After I buttoned it all up, I took it out for a 20 min drive to see what I could see. I found that the max temp to be 150F, and the outside air temp was about 40F.

The lowest brightness setting is still a too high for my taste, so I'm calling Glowshift tomorrow to see if they have any stick on plastic filters to tone it down some.

Here's a pic I took with the brightness all the way down-

[Hondo]

The difference between the Scanguage 2 & the remote were just a few degrees apart, but both told me that I needed to add a thermostat between the rdiator outlet and the external cooler that I mounted- big temp swings.

i installed the Derale thermostat and the temps are much more stable-

from my build thread-

Today I installed a Derale Fluid Thermostat to help control the transmission fluid temp. The thermostat will bypass the transmission cooler until the fluid reaches 180F at which point it will fully open allowing maximum cooling.

This allows for a faster warmup and more stable temps. After installation the temps were much more consistent, varying between 175F & 185F.

Before -



And after -

[booster]

I am actually not too surprised that the gauge and Scangauge read close to the same, as everything I have read on the Chevy temp sensing said that if you use the gauge port, you will get high readings, which would more closely match the internal sensor than an in the side of the pan sensor.

My guess is that the more steady state the load is, like long highway speed, relatively flat driving, the closer the internal sesnor, cooler output from trans, and actual pan oil temps get to each other. Where we saw big swings in temp were on quick accelerations, short steep hills, etc, where the swings were way too fast to have heated the 10+ quarts of oil in the transmission.

We also don't have the radiator cooler in the system, so our thermostat does not see the much more stable temp coming out of the radiator cooler like yours does as long as the water temp is holding relatively constant on the outlet side of the radiator.

It will be interesting to see how we now work in the mountains, where we were OK most places with the single add on cooler last big trip there. The only time we got over 200* was when we went nearly 1/2 hour in third gear of steep climb and lots of curves. We hope that the extra about 40% cooler capacity will take care of situations like that, and that having a thermostat won't mess it up by not sealing off tight.

I certainly will be adding a gauge port in the pan next time it is out, but I doubt I will permanently mount a dedicated gauge in the cab. Once I determine the correlation of temps with Scangauge, no need for the second gauge.