Second Generation Nissan Xterra Forums banner

The Radiator ATF Cooler Doesn't Heat the ATF After a Cold Start

17K views 42 replies 20 participants last post by  DezurtRat  
#1 ·
This topic has gotten quite a bit of attention as the interest in bypassing the radiator ATF cooler has become a hot topic. This information is copied from the thread, http://www.thenewx.org/forum/showthread.php?t=30287 . I'm posting just this part of the thread for those who don't want to read the longer thread.

1. There was no evidence of the rad-atf cooler heating the atf. On three cold starts temperature out the cooler was always colder than temperature out of the trans as the engine and trans warmed up. As the trans warmed up the difference became larger. The difference is a few degrees when the temperature out of the trans is mid 90s and 13-14F when temperature is about 120F. It takes about 20 minutes for the temperature out of the trans to reach about 150F with normal highway driving when the air temp is about 75F.

2. The atf is heated in the return line to the transmission. From a cold start and during all operation the atf is absorbing heat. On one cold start at 75F ambient driving about 65mph, after 3 minutes temperature out of the trans was 108F, temperature out of the air-atf cooler was 99, and temperature into the trans was 104. At 10 minutes, Tout was 131F, Tair was 111, and Tin was 124. At 20 minutes, Tout was 147F, Tair was 121, and Tin was 137. I suspect some of the heat is from the steel outlet line. The two lines are clamped together. Without making a few additional measurements it is just a guess.

I am planning on making additional measurements in future. I have a trip planned to Ouray, CO, at the end of the month and will get some cooler ambient temperature data. I don't expect the rad-atf cooler to heat the atf under those conditions either.

Jeff
 
#10 · (Edited)
Yes.

I suspect you'll see a crossover in the data.

In extreme frigid temps, when the ATF is EXTREMELY viscous, there will be little heat generated in the TC because there will be minimal slippage due to the fluid viscosity under normal driving. Typically the transmission is programmed to shift differently under these conditions to compensate for the increased viscosity.

The engine will warm up to normal operating temperatures, the T-stat will open, and the coolant will flow through the radiator. Between the throttling of the t-stat, and the fact that the fan clutch will be slipping to minimize the airflow through the radiator, the engine cooling system will try to reach ideal coolant temperature for the coolant going back into the engine. At that time, there will be heat from the coolant added to the ATF across the radiator ATF cooler. Heat always flows to the lower temperature. Also note that even with the thermostat completely closed, there is still a 'min flow' bypass port that allows some fluid flow, because dead heading a centrifugal pump like the coolant pump is bad for it and will lead to cavitation on the impellor blades.

Yes, some heat will be lost from the secondary ATF cooler to the air (and that small amount of heated air then travels into the radiator... not enough to really make a difference, but not lost entirely from our two interacting systems either.) At this time, also note that the main fan is slipping due to the fan clutch not locking up because of the cold air temperatures coming out of the radiator... trying to minimize the airflow through the radiator.

Not a bad system of controls and balances from a completely analog system that relies on thermal expansion / bimetal springs in the thermostat and viscous fan clutch.

And BTW, I know this because I have studied it many years ago in one of my first advanced thermodynamics systems design courses.

As to the original data that Jeff collected, it is VERY interesting how much heat is transfered between the supply and return lines running parallel with counterflow. I had noted how closely they were plumbed, but I hadn't really thought it would be that significant.
 
#7 ·
As far as heating up in below freezing temperatures, I think it would be essentially the same except for taking longer.

The coolant in the radiator will cool faster than the engine and trans. The coolant in the radiator doesn’t start to flow with hot coolant from the engine until the thermostat opens. The coolant in the bottom header, where the rad-atf cooler is located, will be the lowest temperature.

The main source of heat for atf is from the converter. When is isn’t locked it is multiplying torque. It is also slipping and converting mechanical energy to heat energy. The extreme case is using the footbrake and running the engine to 2000 rpm in drive. The trans will heat up very quickly. All of engine power is being converted to heat in the converter by the atf.

Driving gently after a cold start in 0F temperatures would just seems prudent to allow the entire driveline to warm up. I would expect the trans to warm up to about 100F faster than the engine. If the trans is shifting normally when cold then the atf isn’t too viscous to hinder performance.

The atf is flowing through both the rad and the air-atf cooler. The lower the ambient temperature the more effective the air-atf cooler will be at cooling the atf on its way back to the trans. Even if the atf was heated by the rad it would be cooled in the air-atf cooler.

I obviously don’t have any data to support my opinion. It is unlikely I will ever be in the conditions where I could even make measurements in below 20F temperatures. However, I would be surprised if the system behaved differently at low temperatures. If someone chooses to bypass the rad-atf cooler and replace it with an additional air cooler, you could always block it off from airflow in cold weather. Without data all you can do is make an estimate of the performance.

Jeff
 
#12 ·
I've really enjoyed these threads for some reason, makes me really think about what's going on. Very interesting to see the real data and and understanding its behavior. Thanks for your efforts Jeff, I will be curious to see some cold temperature data.

I guess the part I'm having trouble understanding is if the ATF is intended to be warmed in the radiator on a cold start, why is it immediately sent into the air-atf cooler out front? Seems like this would be undoing whatever ATF heating had just occurred.
 
#13 · (Edited)
I ordered four additional thermocouples and extension cables to measure temperatures at a few extra points in addition to points already instrumented. I’ll be able to measure the temperature at outlet of the trans, the outlet of the steel line (inlet to rad-atf cooler), outlet of the rad-atf (inlet to the air-atf cooler), outlet of the air-atf (inlet to steel return line), inlet to the trans, and two points near the steel lines. The thermocouples come with a surprise gift in the bottom the box:

Image


This will allow 4 points to be recorded and downloaded in the evening. The existing meter will allow 2 other points to be manually monitored.

This data set should answer the question of what is the heat source for the return line. If the feed line is the source it will have to be cooler at the outlet with a delta approximately equal to the temperature rise in the return line. If the air around the lines is equal to or hotter than the feed line then the air would be the source and the feed line won't be cooler at the exit, maybe even hotter. I suspect the will be a bit of both, but without data it's my opinion at this time.

I was looking at the steel atf lines to see where one or two thermocouples could by placed to measure air temperature near the lines. The lines aren’t tightly clamped together. There is an air gap most of the way between the two lines, even at the zip tied spots and most of the clamps. Any air gap will reduce heat transfer between the two lines. The uncertainty is the reason I’m collecting the data.

Another aspect that is vehicle specific is the presence of skid plates. If they are present they will alter the air flow over the engine and transmission. Whether this improves or decreases heat transfer from the transmission is open to speculation. My opinion is that skids would tend to trap heat and increase the temperature, but not understanding all of the boundary conditions makes this a swag.

Jeff
 
#18 ·
New data

I added the full data set to the other thread, http://www.thenewx.org/forum/showthread.php?t=30287 . Just the cold start heating data is presented in this thread.

I added additional thermocouples and have the Omega data logger to record the data, one nice feature is the data is real time stamped. There are now 8 thermocouples. There is an indoor/outdoor thermometer used for measuring the temperature of the air entering the grill.

1. Trans out, start of the steel feed line.
2. Rad-ATF cooler inlet, also the end of the steel feed line.
3. Rad-ATF cooler outlet, also the Air-ATF cooler inlet.
4. Air-ATF cooler outlet, also the inlet to the steel return line.
5. Trans in, end of the steel return line.
6. Inlet to engine thermostat housing, feed from the lower radiator header where the Rad-ATF cooler is located. TC between the lower radiator hose and housing nipple.
7. Engine bay at the front of the steel ATF lines.
8. Engine bay at the rear of the engine by the steel lines.

Image


Image


The logger is only four channels so not every one can be recorded. The first time I did a cold start log with the trans out, rad-atf inlet, air-atf outlet, and trans in TCs. This was to get some feeling for the temperature change in the line during heating and hot-operation. The ambient temperature was about 85F for the short drive of about 9 minutes on city streets at about 40 mph. Nothing unusual was observed, the atf heated with temperature peaks from sitting at lights and slower heating with easy driving. The difference between the rad-atf inlet and air-atf outlet is the sum of the temperature difference for both coolers so no conclusion could be drawn about the radiator coolant heating the atf fluid from this log. It appears the there is more heating in the return line than cooling in the feed line as the engine and transmission heat up. A possible conclusion is there isn’t much transfer between the two lines.

Image


Next day log used the thermostat housing or rad coolant, rad-atf in, rad-atf out, and air-atf out TCs. Cold start at about 35 mph.. The temperature of the thermostat housing increased rapidly until the thermostat opened and the coolant started to flow into the housing with the observable temperature drop. Given the thickness of the casting I would expect this temperature measurement to be the least reliable, though it is probably higher than the coolant temperature. From the start the temperature drop in the rad-atf cooler was measurable, no evidence of any heating. Given the low speed and the low atf temperature there wasn’t much cooling in the air-atf cooler.

Image


Given the way the thermostat controls coolant flow through the radiator it is hard to image any condition where the atf would be heated by the coolant after a cold start.

Jeff
 
  • Like
Reactions: JeffInDFW
#19 ·
Judging interest in temperature data

The new data was posted two days ago and there haven't been any responses. If anyone is still interested in seeing this type of data in the future please let me know. It takes time and effort to prepare it for posting and it's time well spent if it is of value. If members aren't interested I don't have a reason to keep posting this information.

Thanks,
Jeff
 
#20 ·
The new data was posted two days ago and there haven't been any responses. If anyone is still interested in seeing this type of data in the future please let me know. It takes time and effort to prepare it for posting and it's time well spent if it is of value. If members aren't interested I don't have a reason to keep posting this information.

Thanks,
Jeff
Very cool work - not sure if everyone who is a non-engineer appreciates it.

mitt
 
#21 ·
I've been following both threads with interest ... however I can appreciate that it takes a ton of work, not just to collect, but also to format to the point where it can be interpreted. Data analysis is no joke ... I think you're doing a great job.

Know that it's appreciated, but I'd certainly not fault you for simply posting the raw information ... rather than parsing and formatting.
 
#22 ·
I'm glad that there is some interest in the data. I don't mind the time as long as it's of interest to a few. I've learned a lot about the X from others efforts and this is a way to give something back.

Jeff
 
  • Like
Reactions: JeffInDFW
#31 ·
We at the 2005+ Pathfinder forums: http://www.thenissanpath.com/viewtopic.php?t=2139&postdays=0&postorder=asc&start=210 are keeping a close eye on this as well.

We have the same problems and dilemma about doing the bypass. However, the 08+ Pathy had a slight design change in the way the atf enters the rad cooler. All tranny failures so far that I have researched seems to be from 2005-2007 models years only. Is there any anyone here with a later model year (2008+) with the tranny failures, and does the newer X (08+) have a different setup than the 05-07, like the Pathfinders?
 
#34 ·
An update on some colder weather data for the ATF heating up frpm a cold start. With the X about 30F and the ambient temperature 27-28F from thermocouple data, I did a cold start. The ATF from the trans shows a temperature rise very quickly, the fluid id being heated in the converter as soon as a drive gear is engaged. And there is also temperature drop as the fluid flows through the Rad-ATF heat exchanger. I think there is no ambiguity about the performance of the Rad-ATF heat exchanger, it cools the fluid.

The interesting observation that was made for the Air-ATF cooler. I stopped at a store after the engine and trans were warm. I changed two of the measuring thermocouples to read the temperature drop across the Air-ATF cooler. Initially there was a large temp drop as expected in the low 30F air. After driving for about 5 minutes there was no cooling from the Air-ATF cooler, the temperature out of the cooler was only about 1F lower. This was disturbing, it should be vey effective in cool air at 40-50mph.

I intentionally drove around rather than going directly home because of the odd behavior. Then I thought maybe there is a thermostat of some sort in the cooler. I pulled in a parking lot and ran the engine at about 1500 rpm against the converter to heat the ATF. Wnhen the temperature of the ATF into the Air cooler got to about 140 the temp out started to drop. I let it heat for a while longer before starting to drive normally on the road.

After driving easily between 40 and 50 for about 10 minutes, the trans cooled off. During this time there was 10-20F drop in temperature from the Air cooler. When the temperature of the ATF into the Air cooler dropped to about 120F the thermostat closed and the temp out was only about 1F cooler. It looks like the device opens flow through the cooler at about 130F and bypasses flow at about 120F.

I went back and reviewed the data from last summer and the same behavior was present. I just hadn't paid attention to it. The ATF heats up so quickly in warm weather that it wasn't obvious. I was also mainly interested in the high temperature performance and wsn't lookng closely. Even some of the data from Colorado showed the same behavior.

I don't have any idea what the actual desigh of the themostatic device is but it does work. I'll collect some additional logs of the ATF temperature in the future to get a better idea of the performance.

Jeff
 
#39 ·
Take a close look at the ATF to air cooler. Notice it is a double pass system. It looks like both fittings are on the same side of the cooler. In the middle there is a little block. In that block is a thermostat. It closes when hot forcing the ATF through the cooler. When cold it is open allowing a straight shot bypassing the cooler. This is why you had no cooling from the cooler before the trans was warmed up.
 
#36 ·
ATF is flowing through both coolers and back to the return tube and transmission. The Air cooler is a double flow cooler. It it's orientation in the X the ATF flows in the top right-hand side (passenger) nd out the lower right-hand side. The fluid flows right to left on the top half and then left to right on the bottom half and out the cooler. The left side manifold is open the entire length and the right side is open to the inlet on the top half and to the outlet on the bottom half. It there is a temperature sensitive valve in the right side manifold, it would open and allow flow to bypass the heat exchanger tubing. It may be something as simple as a bi-metallic spring.

Jeff