Ok, one thing I don't understand here is that I've seen over and over that the optimum range is 160-180 but I notice on mine, on the highway it typically runs at 120. Even around town, usually I can keep it to 150-165. And I have no cooler whatsoever except for the radiator. I only see 180-200 when I'm really trying ie. I went to find a steep hill that I had to climb in third gear for about 2-3 miles, TC slip all the way up. And the cold warmup takes FOREVER...1.5 miles of around town driving and the gauge hasn't budged off 100 yet (to be fair it's mostly downhill)

 

I can't tell what the problem is without first knowing how the sensor on your gauge is installed / mounted. Auto Meter makes decent gauges.

First, the basic ATF cooler is in the radiator. Until the engine coolant temp hits 78-deg C (172-deg F), the thermostat is not going to open and send the coolant to the radiator to be, well, cooled. In short, your radiator is a big liquid heat sink for the ATF cooler, which is a big brass cylinder with brazed fittings sitting in the bottom of the radiator.

That's why the long warm-up times...especially running downhill (your motor isn't working that hard).

Now, knowing that a running and fully warmed up motor will have the thermostat open, the coolant is going to be at least 170-deg F or higher until it has completed a pass through the radiator. The hot ATF fluid running into this "cooler" is not going to exit any cooler than the lowest coolant temp in the radiator. That said, how your ATF temp is hanging at 120-deg F doesn't sound possible, because of how the stock ATF "cooling" loop through the radiator is constructed. I don't believe you are seeing the true ATF temperature.

watt-man is an instrumentation guru. See his treatise on P/S fluid temps. He includes a JPEG that shows both ATF and P/S temps. Note his lowest temp while going downhill. It is not 120-deg F, but just under 160-deg F on the graph. Here is the link:

P/S & ATF Temp Graph on Hillclimb and Downhill Travel

Here is the graph by itself:

The graph

To sum it up, either your ATF temp gauging system is not delivering true data, or there is more I just don't know about.

Regards, OF

 

My sender is located before the fluid gets cooled and I tested it with hot water prior to installing it. It was reading 5-10 degrees cool at most, if even that. Also, seeing as how my temperatures spike up to 195, if my 120 degree fluid is really actually at 170, then my 195 reading must mean the fluid is at 250? I think that is a bit too high with nothing inside the van and no towing. I have also found other posts where people are seeing 120 degree temps on the highway.

http://www.odyclub.com/forums/showthread.php?s=&threadid=5461&perpage=15

Maybe a possibility that the fluid is not really flowing while on the highway? I did not cut any of my lines so some of them are routed kind of weird. There are no kinks that the eye can see, and since they are secured further along the line, I don't think there are any kinks period. But at one point, the line coming out of the tranny and into the radiator does get higher than where it was originally. I thought this might have killed tranny flow at low stress (like on the highway), but since there are others with the same readings (3 in the thread I posted), I think this shouldn't be the case?

Does fluid only flow when the TC is unlocked? I was thinking there is a built in thermostat that cuts off flow at low temps but that makes no sense

 

Your A/T has a pump to move fluid where it needs to be. If the engine is running, it is pumping, period. Even under "low stress" it will pump fluid if it is operating correctly.

No possibility that fluid is not flowing on the highway. Otherwise, your tranny would not work. Pump pressure and fluid movement is required for an A/T to work.

One of the posters you reference has the stock Honda ATF cooler. Think about this. His observation, based on his gauge sensor location, is that his ATF enters the radiator/ATF heat exchanger at 120F...well, if the thermostat is open on his warmed up engine, how hot is it going to be upon exiting? At least 170F, then it runs through the Honda finned tube ATF cooler located in front of the AC condensor. I have one on my 2002. It's not a lot of added cooling area, though the pipe diameter does increase quite a bit to increase dwell time prior to returning back to the A/T. Will that Honda finned tube ATF cooler dump enough heat at an ambient temp of 60F so that once that ATF gets run through the A/T once more that it only heats back up to 120F at the A/T's ATF outlet? I doubt it.

However, I can believe an aftermarket ATF cooler, with no thermostat, installed after the radiator/ATF heat exchanger could perform that well.

In short, based on how the Honda system is built, your readings on your particular Ody don't make a lot of sense. They are way low. I'm not commenting on what the actual readings should be, for obvious reasons. watt-man's numbers seem sound, because I assume he does instrumentation for a living and explains possible error sources in his sensor data loop. His gear is not made for everyday people like us, though.

Think about it. If you heat water to 120F on the stove and put your hand in it, well, you'll be able to tolerate it, but it will be very uncomfortable. That just doesn't sound like the temp of ATF out of a warmed up and operating car.

Regards, OF

 

One of the posters you reference has the stock Honda ATF cooler. Think about this. His observation, based on his gauge sensor location, is that his ATF enters the radiator/ATF heat exchanger at 120F...well, if the thermostat is open on his warmed up engine, how hot is it going to be upon exiting? At least 170F, then it runs through the Honda finned tube ATF cooler located in front of the AC condenser. I have one on my 2002. It's not a lot of added cooling area, though the pipe diameter does increase quite a bit to increase dwell time prior to returning back to the A/T. Will that Honda finned tube ATF cooler dump enough heat at an ambient temp of 60F so that once that ATF gets run through the A/T once more that it only heats back up to 120F at the A/T's ATF outlet? I doubt it.

However, I can believe an aftermarket ATF cooler, with no thermostat, installed after the radiator/ATF heat exchanger could perform that well.

Nice thread. I am confused as to whether the thermostat to which you are referring is the engine coolant water thermostat, or another thermostat.

Are the power steering fluid cooling, and/or transmission fluid cooling, circuits thermostatically controlled?

 

No, no thermostatic regulation of ATF or P/S fluid in stock form. Really, the only way to get a great picture of what the A/T and the respective coolers are doing is to place at least 3 temp sensors (calibrated properly, of course) at these positions:

1.) ATF output from transmission / ATF inlet into internal radiator cooler.

2.) ATF output from internal radiator cooler / ATF inlet into aftermarket cooler.

3.) ATF output from aftermarket cooler.

Of course, this is more info that most/all of us need, but boy it would satisfy my curiosity!

v/r, OF

 

So there's still no explanation as to why four people's temp gauge are reading the same. watt-man might do instrumentation for a living, but I have yet to see anyone point out something wrong with my set up? I don't think I am reading 50 degrees low either as I already explained...

If I put my hand on my tranny hoses after a quick drive, it is not warm at all, nothing even close to the radiator hoses.

 

You just answered your own question in part...at least two of them (yours and the guy with the Honda finned tube air/fluid cooler) should have different temps. Why they are the same isn't the point...why they aren't different when they should be is. You don't have the secondary cooler.

Also, you answered your other question...a quick drive won't warm up the tranny. It has over 2 gallons of fluid, and the radiator acts like a heat sink until the engine is fully warmed up for a while. A quick drive won't cause your ATF temp to rise appreciably. Even the owners' manual gives you a minimum amount of run time to get it warmed up for a drain/refill.

v/r, OF

 

Even after a quick drive where the engine gets up to temp for quite a while?

And uh so...still no suggestions on why I'm reading so low...

 

A quick drive for quite a while? That's an oxymoron...if it's for quite a while, then it's not a quick drive. Seriously, putting your hand on the hose is nowhere near the same as immersing your hand in a bucket of the stuff. You've been there...tranny doesn't feel hot, you pull the drain bolt, the initial surge of draining fluid splashes on your hand...damn, that's hot. I probably should not have introduced that analogy, as your hand is not a good temperature determination device.

There's nothing wrong with your setup. It's just not a well-calibrated gauge setup, nor can it be, because it was manufactured to fit in any car. You have no control over that. All that means is that it's just a generic gauge and sensor, and it's not going to be even close to near as accurate as a gauge that can be trimmed out with its own tightly regulated power supply. Think about it...the other Ody owner has an extra cooler with more fluid line volume, all of which helps cooling, with a different gauge and sensor setup and he doesn't get a different value. All things equal, it just points to really wide tolerances in what the automotive gauge manufacturers expect when they build these things. They're not Swiss watches, for sure.

Also, to make your comparison tougher, ODYTRB, who also registers 120-deg F, is using a Nordskog gauge set which has a linear gauge face scale (100 bottom peg, 190 midpoint, 280 top peg). The currently marketed generic Autometer trans temp kit is non-linear (100 bottom peg, 180 midpoint, 250 top peg). In fact, the Autometer scale is more appropriate for a radiator coolant temperature range, not an automatic transmission, which under duress can shoot higher than 250-deg F. So, we have different gauge manufacturers, different installation hardware, different sensor gear, different power delivery (no two cars are alike), and different ATF cooling loops (one w/ and one w/o an extra cooler)...and they read the same! That points to large errors in the manufacture and calibration of these mass-produced gauge kits made to work on every car with a 12V (ahem) electrical system. It's acceptable, though, because we're not building a nice time-stamped graph like watt-man did with multiple aligned data inputs. What's good enough? If I get the needle to read the same place on the gauge face during the same driving conditions every time, even if it's not accurate, that's good enough. If the needle starts to rise a lot, then I'll worry.

When I flew for a test squadron, we had these difficulties all the time with gauges, and we had a unit (PMEL) dedicated to making sure they read correctly prior to and after installation in an aircraft.

If you're expecting great accuracy from these auto gauges, it's a crap shoot. The important point is not the supposed indicated value, but the trend data. If it says 120F, even though that's not the actual temp value, but the gauge does it every time, and the car is operating well, then that's a good place for the gauge needle to be. Your gauge may not be giving you a true reading ofthe actual temp, but it's a repeatable value. That's called a good degree of precision if it gives that same value all the time under the same conditions, even if it's the wrong value (i.e., inaccurate). You can still use the data presented by that gauge.

Even if it were a digital gauge, it would still be subject to the same manufacturing errors...the only difference is the display because it's still an analog feed.

Automotive gauges rely on a variable power supply...your alternator. Is system voltage 13.2V? 14V? 14.5V? None of the above? It's not perfect power delivery in the least, but good enough for the job. That's one reason (among many) why many standard auto temp gauges on vehicles on the sales lot don't have numbers. Just knowing where the needle ends up is good enough. If it moves to a very different part of the gauge face, it alerts you to a problem. How fast it does it is your trend data. To get numbers on an automotive temp gauge face, often you have to buy aftermarket gauges for that.

Regards, OF

 

They're precise, but Autometer does not guarantee accuracy

My sender is located before the fluid gets cooled and I tested it with hot water prior to installing it. It was reading 5-10 degrees cool at most, if even that. Also, seeing as how my temperatures spike up to 195, if my 120 degree fluid is really actually at 170, then my 195 reading must mean the fluid is at 250? I think that is a bit too high with nothing inside the van and no towing.

You did your own home calibration check at only one temperature point, using boiling water, right? If you're near sea level, we'll say that 212-deg F is a faithful comparison point.

That said, you can't use that to determine error at different needle displacements (temp indications) on that Autometer gauge. In short, you can't assume that your 120-deg F reading is 170...it could be far different from either 120 or 170. Same for your 195 reading. It could be higher or lower than that. You calibrated at only one data point, and the inherent equipment error at that point is not necessarily indicative of errors at different gauge indications. The only way you can interpolate is if you calibrate at multiple points and use a data scatter.

Believe it or not, that's how we build performance tables in aircraft manuals. You can't test for every single point on a performance graph...there's not enough jet fuel in the world for that. Therefore, you test "enough" and then build your data curves from there. watt-man uses the same test and eval methodology in building his graphs. Actually, his is better because he uses a continuous recording device with multiple channels because he can in this application.

One thing I noted on the Autometer home page...they only claim their gauges are capable of a good degree of "precision". They make no accuracy claims, mainly for liability reasons.

v/r, OF

 

I also checked it at 105F and 195F and it looked ok at those temps too.

It depends on how you define quick drive. I have places that I can go that I would consider a quick drive but others wouldn't. But I would always get it up to operating temps for a good while, either due to warm up or lots of hills.

The gauge is specifically a transmission temperature gauge. It is not a generic "temp" gauge. It has TRANS printed on the face. The readings I am taking off them can be assumed to be linear since there are also intermediate markings of 150 and 210, and these regions are also split in half again. On the highway, it's not even hitting halfway to 150. I think the 100-250 range is not bad for a temp gauge - if you look at Cyberdyne gauge it is also a 100-250 range. Consider the small sweep of the analog gauges, it's a better setup than a the B&M one, which goes from what 100-360? You can't tell much from that...

 

Only 3 points don't make a good data scatter

They all work the same. Voltage goes in, heat changes the resistance value of the sensor, and this gives you your needle deflection. Autometer puts the "Trans" silkscreen label on the gauge face so that you, as the driver, will know that you are reading a sensor feed from your A/T.

A while ago I helped a friend purchase gauges for that 3-gauge binnacle Autometer sells to put on top of the dash. It's a nice setup. The coolant and trans gauge faces were identical except for the gauge face labels.

If you are expecting great accuracy, I wouldn't for all the reasons mentioned. You can't get there from here w/ generic gauges, or even high-end stuff like Nordskog. There are too many things that will negatively affect the reading you see in a running car. What you're interested in is repeatability. I'll take an inaccurate gauge kit that gives the same value everytime for given conditions (precision) any day over a sometimes-accurate gauge w/ low precision.

If they were interested in accuracy, Autometer would provide user-adjustable trimpots in the back of their gauges to do final calibration against reference equipment. This is too much for the average Joe (hey, I wouldn't use it, either), so what we all get is a one-size-fits-all gauge kit. Accurate or not, it's a precise kit, and that's what you really need.

The B&M is a race setup. You can, under race conditions, see over 300-deg F. There are synthetic ATF formulations made to function, if only for one race, up to 400-deg F. Of course, all of the seals, clutch packs and other items will take a big hit. Yes, the Autometer gauge scale is perfectly usable for everyday normal conditions for us average Joes.

v/r, OF

 

I have one of those digital cyberdyne gauges fitted and I was wondering what the accuracy was as well - having just got back from a camping trip towing my Coleman Sun Valley (around 2800lbs) the transmission peaked at 241 F after a 7 mile slightly uphill grade to the campsite,

is 241 F high especially since I have a B&M cooler fitted as well?

 

I have no idea how accurate your installation will be. It depends upon a lot of things, like voltage draw at the time of the reading (i.e., how many electrical accessories were running, did the radiator fans activate, etc.) and how you positioned the sensor.

The most accurate temp reading devices I've seen automakers use are thermocouples that plug into catalytic converters...but in these applications, they don't connect to a gauge. A thermocouple generates its own voltage based on temperature of the probe. The engine management system (I don't know if our Odys use this) can tell if the catalytic converter is overheating when the thermocouple exceeds a certain voltage threshold. It gives an accurate indication of how hot the cat is getting.

They are way more accurate than the gauges we get for our cars from AutoZone or Kragen, but not really practical for us to use.

The only way to see what the actual temperature you are reading is to build a correction card. You build one by using a temperature measuring device that is independent of the vehicle's electrical system (like an electronic thermometer that employs a thermocouple probe), and note what the thermocouple temp is as well as the auto gauge's displayed temp at the same time. You do this over the entire range of temps. Bear in mind that you also have to know what the electrical load is on your car. The correction card is only valid for the conditions under which you made the measurements. If you change the electrical load on your car (i.e., on/off status of high-draw items changes), then the card is not valid.

My understanding is that most generic auto temperature gauges use a thermistor, which simply changes resistance with temperature, so changes in supply voltage will change the output voltage, thus changing displayed temperature even though the actual temp may not be changing.

Knowing that voltage changes can destroy accuracy in these applications (but not precision...if the conditions are repeated, you'll get the same reading for a given temp with a precise setup), Auto Meter also sells mechanical gauges. These use a bulb that is heated and changes pressure in a capillary tube that connects to a mechanical gauge. I think the gauge uses a sealed aneroid, like an altimeter without a static port. The only drawback is that underhood temps can heat the media in the capillary tube and affect accuracy negatively.

Nobody said getting an accurate temperature reading was going to be easy.

I got my 4-channel K-thermocouple data logger not too long ago, and I'm ordering 2 more thermocouples and the appropriate cabling and plugs/jacks to start getting data (A/T torque converter fluid output temp, ATF temp out of OE radiator heat exchanger, ATF return temp from air/fluid cooler returning to A/T).

Once I get some data, I'll post the results. Don't look for it too soon, though...I've got a ton of projects, like a boat repair, on the burners right now!

v/r, OF

 

Well, I hooked up a pair of K thermocouples directly to the ATF output and return lines (metal tubing) on top of the A/T today, securing them with a piece of 3/8" rubber ATF hose and a hose clamp.

OAT was 70F, no wind. T1 was my ATF output temp, T2 was the ATF return line. This particular 2003 EX has no additional cooler, just the OEM radiator tank cooler. Gas tank was almost empty, no passengers, no cargo. In short, I was operating at near the lightest gross weight possible, with no additional loads from towing or winds or hilly terrain.

I drove about 16 miles north on flat terrain at 65mph to run an errand, and ran 16 miles back at the same speed to get home.

Steady state T1 was 147F to 151F. I selected T1-T2, and the cooling delta T ranged from 8.7F to 12.2F. That's not much cooling at all from that in-radiator tank cooler. Also, bear in mind that I did not insulate the outside of the small length of rubber hose placed on top fo the thermocouple ends, so there are some minor heat losses...the actual ATF temps may be a couple degrees higher.

So, that's my baseline, approximately 150F ATF output for steady cruise at 65mph with only a driver in an empty Ody on a 70F no-wind day on flat terrain.

Once I install the coolers and tow Moby Dick (my bass boat), I'll run the data logging function from a cold start and post more data.

Stay tuned.

Regards, OF

 

150F is what my gauge reads under similar conditions, so I feel like my 125F reading on a cool day is pretty accurate, as I've said all along...

 

Ok, one thing I don't understand here is that I've seen over and over that the optimum range is 160-180 but I notice on mine, on the highway it typically runs at 120.

I do believe that 160F-180F is a healthy range as well...but your 120F gauge indication should be ducky.

And why not? Given that our tranny starts a "warm/normal" shift schedule at temps well below what either of us is indicating on our respective temp indicating apparati (how's that for a plural form of apparatus? ), I'd say we're better than fine. I don't think the 120F reading is accurate, but I think it's precise. I mean, if you get that as your typical highway reading, then that's your normal reading. Kind of like the difference between using Kelvin, Fahrenheit, Celsius or Rankine scales...you get the number you get, it's repeatable, and that's good enough.

I'm beginning to think that getting the warmest temps you can get without degrading fluid or tranny life is a bigger benefit to MPG than anything else. I was operating at the lightest loading possible without removing my seats and carpet or clothing . We'll see what happens when I climb a light grade with the family on board.

This little device I've got is amazing. I've got 6 wire lead K-T/C's of two different types from two different manufacturers, and 4 inconel solid K-T/C probes of two different lengths and OD's from yet another manufacturer...and all of them end up within 0.1 degree celsius of each other whenever I do an equilibrated test in still air. Although it's a really accurate way to measure temps (relative to other means in my financial grasp), my indicated ATF temps are probably a little on the low side since I didn't put any appreciable insulation around my clamping devices.

I'm going to do a cold start tonight when I leave work (in exactly 3 minutes and 47 seconds ), and tell y'all when it allows a normal 3-4 upshift on my T1 thermocouple (ATF output line temp). I'm guessing that will happen at 100F by design.

Regards, OF

 

I don't understand your point...did I not say it usually runs around 120-125? It is usually high 50's low 60's around here.

 

Well, I always try to address questions on an information thread:

I don't understand your point...did I not say it usually runs around 120-125? It is usually high 50's low 60's around here.

The title of the thread is "ATF Coolers...TEMPERATURES...some more info". Just presenting info about what I've seen from a few different sources and my attempts at both accurate and precise measurements. I promised some data once I figured out how this stuff works, and there it is.

Well, on the way into town, then to work I got some surprises.

First, insulate your gear from external factors you don't want to measure. It was 56F this morning, and during my run into work I found that holding an even 54mph with a 12mph tailwind, no passengers, no cargo, flat terrain, gave 133F. Now, that's nice and cool...but it was an incorrect reading, as I soon found out.

Well I switched off the T1-T2 delta indicator to get a reading of the underhood temp (my T3 channel), and it gave a whopping 62F. That's amazing. I always thought underhood temps would be warmer. Were my quickie T/C mountings experiencing heat losses from turbulent cooler underhood air? Well, yes they were! I stopped just long enough to jam a piece of bubble wrap from an envelope in front of the T/C's to act as a windshield. I got back up to the same speed, let things settle down, and now the ATF temp under the same driving conditions was 140F.

With a 14 degree OAT drop, finding the optimum speed for coolest ATF temp (driving at a snail's pace on the freeway), and getting a nice tailwind to help, I was expecting less than 140F, but it's possible my 150F reading under the previous conditions was unrealistically low as well because I hadn't insulated the thermocouples in the same way watt-man did. All this proves is that I can't get 120F in my Ody's ATF output line on a flat highway, no matter how gently I drive, even with the wind pushing my car. Others' cars may be different.

Second tidbit: bb5 noted that our tranny goes into "normal/hot" mode at around 100F. Last night and this morning I got the 3-4 upshift to happen at 28mph just as my T1 temp passed 98F and 95F respectively. He definitely got that one right.

BenjiBoy, I don't think either of us is getting perfectly true temperatures. I mean, my data from this morning is probably still a degree or two lower than actual. Just the nature of my current T/C setup, but I know I'm pretty close. Also, my data from yesterday is probably a bit more cooler than what the actual ATF temp really was because I didn't shield my T/C's from all that wind. Also, if yours is only 10 to 15 degrees off from true, that is actually pretty good since it relies on an NTC thermistor-based device that needs a fixed power supply to be accurate, but it has to put up with the variable voltages of automotive power supplies.

As well, that OEM radiator in-tank cooler just doesn't do much cooling. My T1-T2 delta stayed at around 9 degrees today. Also, once warmed up, this tranny doesn't lose heat quickly when shut down. I got home near midnight, then got going today at around 6am. Even after 6 hours of sitting last night with temps dropping from the mid-60's to mid-50's, the ATF temp was about 85F right after startup in the morning. Amazing. Also, when stopped at a light, my delta figure went negative...the "cooler" was actually acting as a heater.

Regards, OF

edited for grammar and to finish up the post.

 

Damn, this thing gets HOT

I thought I would bump this for most current data.

My last posts above were based on strictly controlled conditions: early morning, no traffic, perfectly flat roadways, constant speed, no winds or tailwinds only. Gas tank almost empty, no cargo, carrying only me, driving as carefully as possible, accelerating as gently as possible up to a snail's pace for the given roadway. I know, I can hear it already: Those are unrealistic conditions.

So, my wife's 2002 EX-RES received the 2-channel instrumentation setup this time for a roadtrip from SW OK to Deltona, FL and back. The conditions are different for this bit of data sampling, since our 2002 EX has the Honda OEM ATF cooler, but my 2003 EX used for previous data sampling does not.

If other vehicles were around on a highway, requiring me to actually change our speed just a little to maintain safe spacing or ease back up to cruise speed, I couldn't replicate the 140-deg F or 150-deg F ATF temps. Even with the finned tube cooler providing an extra 7-10 degrees on my T1-T2 readout (7-10 degrees extra cooling), I still couldn't get temps that low.

Differences? First, it was warmer out. 72-deg F at sunrise (it's summer). Second, I had my children and wife with me, but we packed light. I did bump the tire pressures up to 40 psi on the 2002 EX for this trip. Third, this Ody has an added ATF cooler, but the 2003 did not.

Results: ATF pump output temp would quickly get up to 150-deg F, then would not climb so quickly after that. However, it wouldn't stop climbing, either. It just slowly kept marching up as we drove. Once stabilized, driving as carefully as possible in the slow lane during the coolest part of the day, it liked to stay in the 177-deg to 185-deg F range. Drafting a big rig would help drop it a couple degrees, though.

It got hotter. If I stopped to let everybody stretch their legs, the ATF temp would not really change. No surprise here. You can shut down for hours, and the ATF temp will still be pretty warm. The big surprise was getting back onto the freeway. I had to use a good bit of throttle, jockeyed a little to get around a big rig so that I could get back into the slow lane, and the ATF temp marched right up to 205-deg F. It took many miles for it to ease back down into the 180's. Again, this 2002 has the extra ATF cooler.

Honda OEM finned tube cooler doesn't do much. In short, the combination of the radiator in-tank cooler and the Honda OEM finned tube ATF cooler for the 1999-2004 Odys is insufficient to stabilize ATF temperatures even with the most careful driving. ATF temps exceeding 200-deg F just by getting back onto the freeway and establishing position in traffic...that's just wrong.

I drove as carefully as possible, when I could. Since I had real-time data accurate to within a few tenths of a degree Fahrenheit, if I saw the temp march up, I would immediately do something about it...like ease off the gas, or maintain throttle position going downhill (slowing down or using the engine as a brake going down a gentle slope can raise ATF temps, too, if you were already carefully monitoring them and driving carefully to being with).

Things to do If you have a 1999-2004 Ody, you need to do something to enhance transmission longevity. That finned tube Honda OEM ATF cooler is completely unsatisfactory and not big enough to stabilize ATF temps even under the most favorable operator inputs. Things to do for this year range (1999-2004):

1.) Get an ATF cooler. Any type seems better than that Honda OEM finned drinking straw.
2.) Extra filtration is not a bad idea (Magnefine or other remote filter w/ a FilterMag on it.)
3.) More frequent ATF drain/refill schedule

I already have Magnefines on both and do a single drain/refill with each oil change. Real coolers are next on the list.

Regards, OF

 

Great reading and very interesting. Not sure how many others out there are reading this, but I definitely am. I'm glad I've installed my cooler and Magnefine.

ODYFAMILY, did you get your coolers yet? Time is ticking brother.....

Keep us posted.

 

Brother, I am immersed in multiple projects...Just finished an 80-foot-long fence, finishing an outbuilding (trimming out, plus have to fix the new roof I put on it...it's leaking! ), and finishing the re-fit on my boat (have to locate my V6 outboard down one notch on the jackplate; had prop 'blowout' during hard turns during its maiden voyage out of my "garage drydock" last weekend ).

I'm purchasing a very used Toyota MR2 for $700 to be my personal transpo while I take down our Odys (one at a time, natch) for the planned work. Once I get that beater car, the Tru-Cool coolers are next on the purchase list.

My 2003 EX needs a new A/C compressor (buying the compressor this week, got the filter-desiccant kit already), so I'll just pull everything off the front end (bumper cover, splash guards, condensor, radiator, fans) to make life easier. As well, I'll change the little ATF filters located under the solenoids that reside on top of the tranny. When I re-install, on will go my coolers.

I promise I'll take pics and post these.

Wife's 2002 EX is next. I'm just going to replace the OEM coolers with the Tru-Cool items. I have to purge the A/C so I can replace the filter-desiccant, as its little plug is leaking refrigerant oil, so I've got a bit of work to do.

I attached my data logger again to the 2002 EX when I pulled our boat to the lake last weekend. Flat roads, only one hill, driving very carefully over the course of only about 21 miles, pulling about 2,600 pounds.

224-deg F. Yikes.

 

FWIW, I didn't use my manual condensor switch yesterday on a 15 mile run in city traffic and the input and output lines of the stacked plate tranny cooler felt identical meaning it wasn't doing a thing to cool the fluid....

This morning did the same trip and left the condensor fans on....what a difference in temperature. I really don't need a gauge to tell me so. My oh so accurate back side sides of the fingers felt the results....

 

Brother, I am immersed in multiple projects...Just finished an 80-foot-long fence, finishing an outbuilding (trimming out, plus have to fix the new roof I put on it...it's leaking! ),
Hmmm, let me guess, flashing around a vent or chminey?
and finishing the re-fit on my boat (have to locate my V6 outboard down one notch on the jackplate; had prop 'blowout' during hard turns during its maiden voyage out of my "garage drydock" last weekend ).
Oooo, that sucks. Stainless?
I'm purchasing a very used Toyota MR2 for $700 to be my personal transpo while I take down our Odys (one at a time, natch) for the planned work. Once I get that beater car, the Tru-Cool coolers are next on the purchase list.
How big a coolers are you putting on each of the ODY's? Can't remember if you've already posted that in a different thread or not.
My 2003 EX needs a new A/C compressor (buying the compressor this week, got the filter-desiccant kit already), so I'll just pull everything off the front end (bumper cover, splash guards, condensor, radiator, fans) to make life easier. As well, I'll change the little ATF filters located under the solenoids that reside on top of the tranny. When I re-install, on will go my coolers.

I promise I'll take pics and post these.
Good deal.

Wife's 2002 EX is next. I'm just going to replace the OEM coolers with the Tru-Cool items. I have to purge the A/C so I can replace the filter-desiccant, as its little plug is leaking refrigerant oil, so I've got a bit of work to do.

I attached my data logger again to the 2002 EX when I pulled our boat to the lake last weekend. Flat roads, only one hill, driving very carefully over the course of only about 21 miles, pulling about 2,600 pounds.

224-deg F. Yikes.

Sounds likes lot's of fun....makes me wonder why we get ourselves into this maintenance nightmare. The more stuff you got...the more you got to maintain!