[frak]

Can't seem to find...

3/8 or 5/16 inch size hose fittings?

[gregwatkins75]

i used 3/8 .. changed over the line size on radioator as it made them all one size hose.

[frak]

You mean u scewed in different size barbed fittings? Sounds like a good idea. The next question is what the thread size is?

[fordem]

I believe at least two different transmissions were used on the Vitara, a 3L30 and a 03-72LE (or AW4) - the AW4 I have uses 10mm cooler lines - 3/8" cooler hose fits the 10mm hard lines, 10mm hose is very slightly loose on 3/8" hose barb, nothing that a worm drive hose clamp can't resolve.

[frak]

Thanks. Will go with 3/8 then. Will be fitting a second cooler inline on a thermostat.

[Gwagensteve]

On a thermostat?
Have the Canadians got to you?

[britvit]

I've over cooled one on a cold night in SA.

[Gwagensteve]

what were the symptoms? what was the transmission running at?

I can start my car after camping overnight in the snow (so well below zero) and it shifts just fine from cold. I have no idea how cold the trans would have to be to be overcooled, -20˚C? Colder?

[britvit]

Poor up shift to OD, I don't run a gauge. I had the trans cooler mounted on the front of the radiator, after a long very fast road drive the cooler felt cold to the touch. It was a 5 degree night, it doesn't get to -20 here. The TC would have been mainly locked so the trans wouldn't have been generating much heat. I have remounted the cooler behind the front bar and the problem has gone. I'm guessing that's why Suzuki put a temp input to the TCM.

[Gwagensteve]

Upshift into OD is controlled by engine not transmission temperature.

[Gwagensteve]

Every time this myth of overcooling comes up I research it again to try and find a definitive answer about what overcooling does and at what temperature it becomes an issue, in case I missed something previously.

Every time the answer is the same - it can be an issue if the ambient temperature is extremely low, circa 0˚F (so that's -18˚C) and only then the transmission computer might inhibit an overdrive shift in order to slip the converter to generate heat to fix the problem. (raising the viscosity of the fluid) I do not know if the AW-4 does this because I've never been able to get one cold enough. Yes, the AW-4 computer will inhibit a shift into OD if the engine is cold (engine, not transmission) and this can be eliminated by cutting the temperature sensor wire to the transmission. This implies transmission temperature has nothing to do with the inhibited OD shift because otherwise the transmission would handle this without reference to the engine. I do not know whether this inhibited OD shift is for the benefit of transmission heat or engine heat (and therefore emissions) - I always though it was for getting the engine hotter quicker, and not related to the gearbox.

The normal minimum operating temperature of regular petroleum ATF is as low as 15˚F. (so near enough -10˚C) at which point the viscosity of the fluid will allow the transmission to work properly. Borderline operating range is -30˚F (so that's -34˚C) that data came from Amsoil, whose synthetic ATF has a minimum normal operating temp of -45˚C

So, is it possible to overcool an auto? Yes. if ambient temperature is under about -20˚C, you have a very efficient cooler, and you're driving with the converter locked without substantial speed changes from a cold start. So conditions I don't anticipate any of us are likely to see locally.

I'll add one acceleration run is enough to out about 30˚ of heat into an auto from cold. I can get my auto to 40˚C in literally seconds if I left foot brake it up to around 1500rpm.

Autos are spectacular at turning torque into heat.

[fordem]

Someone please correct me if I'm wrong - this temp input to the TCU that is being discussed is from the engine, not the transmission?

As far as I know, Suzuki does not monitor the transmission temp on their implementation of the AW4 - some implementations (Mitsubishi & Toyota) do have a fluid temp sensor on the transmission, and I've never seen that used for anything other than an "over heat" warning.

[britvit]

Gwagensteve wrote:

Every time this myth of overcooling.

Nope. You're wrong. Under certain conditions it is possible. You would have thought that after a 4.5 hour drive at 100 kph + the transmission cooler would be hot. It was cold to the touch. On the return journey the shifting was fine, the only thing that had changed was the ambient temperature, it was a 20 degree+ day.
I drive that car everyday I know when it is shifting correctly and when it is not. Suzuki put the coolant temp input there for a reason.

[Gwagensteve]

That's my understanding Fordem. as I said, engine coolant temperature is what controls OD shift, not trans temp.

Please don't take this as any sort of attack BritVit - I'm just trying to work though this issue as I genuinely don't believe you were seeing transmission overcooling.

No, I would not expect a cooler to be hot after a 4.5 hour drive at 100km/h. The converter should be locked and the transmission running quite cool. But not below ambient and not below the temperature where the fluid has inadequate viscosity to work properly. Remember the trans will run normally down to at least -10˚C. I would expect, on cold night and even with a large cooler, cruising trans temp to be around 40-60˚C. Coasting to a stop and getting out of the car and putting your hand on the fins of the cooler might mean they feel cold.

However, that's only true if the converter is staying locked and the transmission stays in OD. As soon as the converter unlocks the trans starts to heat markedly. If you're complaining of poor shifts, then at some point the trans was cycling from OD Locked- OD Unlocked - 3rd. and back at ~3000rpm. This will heat the transmission quickly and was obviously happening during your drive.

I suspect a more plausible reason is that the engine was running just around thermostat opening temperature due to the low ambient and high road speed and inhibited fan function*, and this low coolant temp was intermittently inhibiting the OD shift.

Moving the cooler behind the front bar may have improved airflow to the viscous fan clutch which is now working correctly and engine temp was more stable (and as you pointed out you have had higher ambients) . Clutch fans only work with air flowing over them. The more coolers stacked in front of them the worse they work. You'll note, I'm sure, than on cold start the fan will run until it's been freed up - they can also run when they're cold in use. Not normally a problem, but if it's holding engine temperature low the thermostat will close to bring the engine up to temperature. When it opens, engine temp will dip and OD won't engage. I know this from experience as I have a very large radiator and fan combination, with lots of coolant in the system and this can happen to me (rarely, and at <10˚C ambient from cold start) and it will lock out OD operation even once it has already engaged. Under these circumstances transmission temperature is generally >40˚C, it's coolant temperature that's causing the issue.

There may be other reasons unrelated to temperature, but I can see no reason for the cause of poor OD shifting on a (by world standards) warm night to be "overcooling"

My research, as I've pointed out, comes from the northern US and Canada where even at extremely low ambient temperatures "overcooling" is almost unknown and not clearly defined. Some people have occasionally complained of poor shifting performance when the ambient was ~0˚F. So, someone writes that on the internet, and it's then used to support opinions on the hottest continent on Earth.

A trans temp gauge would have resolved this issue before it even arose.

[Gwagensteve]

Bit of a dig, but I was reading the Allison Automatic users guide for their truck autos. (because that's how I like to spend my time

They are specific about the temperature that preheat is required before operation in cold climates. For Dexron III fluid, it's -25 centigrade.

Minimum continuous temperature of the fluid in the sump is specified at 40˚C, although no consequences are specified for what happens between startup at -25˚C and operation at 40˚C

Maximum sump fluid temperature: 121˚C, which is in line with everything else I've read for the threshold of rapid damage.

[frak]

Well i accidentally touched off a really interesting discussion.

And i even now get the Canadian reference a. (hence it took me a while to reply)

For me the thermostat is that given its cheap anyway, why bother cooling the fluid down when not needed. A thermo will keep it closer to an optimal temperature regardless of how i am flogging (or not) the tranny.

I am pretty sure i won't overcool it down to 40 let alone -40 though driving home through the ice. I had guessed a bit below 0 would function not guessed that low a minimum temp.

[Gwagensteve]

In my mind the simplest /lowest risk solution is always the best. The likelihood of overcooling is extremely low. The consequence of overcooling is insignificant. However the consequence of thermostatically controlled valve failure is potentially high- (overheating) so I wouldn't add another system which places the auto at theoretical risk to prevent something which isn't likely or harmful.

[gregwatkins75]

frak wrote:

You mean u scewed in different size barbed fittings? Sounds like a good idea. The next question is what the thread size is?

I had pissy small little pipes in my radiator, the B&M trans cooler had the larger 3/8 inch lines, so yes I unscrewed the pipes from radiator and screwed in the larger.. was easiest..most factory approach to the problem....

Interestingly I'm not sure if the previous owner butchered it, but back about 30 0or 40cm along the cooler line I found a step down pipe thing going from larger hose to smaller hose... the smaller hose ran to the radiator...
Car was a one owner when I got it.. My only thought is he did some dodgy work or a quick fix..