Manual transmission cooler? or for that matter diff cooler
 

Bored and looking to waste money! ok mostly kidding
I was wondering if anyone who regularly tracks our cars runs a transmission cooler or even a diff cooler. I seem to recall a thread where hustler said a temp sticker on his tranny showed north of 250 degrees. Do you think a cooler would reduce transmission failures or do you just change the fluid more frequently?

Ive been thinking of doing a diff and tranny cooler, but its toward the bottom of the list.

Build a large box around the tranny with a small weep hole near the top of the box. Cut a moderately sized opening out just ahead of your gear shift turret (build a funnel/pipe if you need too) so that you can reach the transmission from the cockpit. Buy a large bag of "rock salt". Before you get to the track, stop by a gas station and buy a dozen "big bags" of ice - throw them on your tire trailer/passengers seat/trunk/roof/whatever. While on the track, keep a bag of ice and the bag of rock salt in the passengers seat. Everytime you cross the start/finish line of the racetrack dump 1/4 bag of ice into the hole, followed by about a cup of salt.

By the time you get done with your session, you'll have track-made ice cream!!

***as an added bonus, your cooling problems will be gone too!***

Lols didnt expect the end.

I use a Manuel transmission cooler. My transmission has never been cooler thanks to that mexican.

It's worse than that - TrackDayHookey stuck a temp probe in his and measured fluid temps of over 340*F.

I sent my oil into Blackstone when I popped my tranny last year - no insolubles or viscosity breakdown, so the fluid isn't actually "overheating", but IMO a cooler would probably still help. If we can keep bearing temps down and keep the gears meshed better, reliability/longevity goes up.

IMO, track cars over 300whp should consider diff and tranny coolers. Even SM guys see high diff temps. It's something I will get around to this year - we'll have a kit for ~$500 with a pump, cooler, filter, and the necessary fittings.

:giggle:

Did you buy it at home depot?

I think you ment in front of home depot ;)

I've been thinking about this also. I wonder what the diff temps get up to...I've had a auto trans cooler sitting on my workbench begging me to install it.

Why not use thermal epoxy to glue fins one them? You can easily double or triple the surface area to the airflow by doing this.

You could cut a heatsink like this into long narrow inch-wide strips, with 3-4 fins on a strip, and glue them between the ribs of the tranny and/or diff

http://www.aluminum-profiles.com/products/9.jpg

Simple glue would insulate and not transfer heat. There's got to be some kind of heat transfer adhesive available to use. I've used small heat sinks on RAM chips that had a tape with heat transfer ability.

Thermal tape has very poor heat conductivity compared to the various thermally conductive chemical adhesives available, there's even conductive epoxy, but last time I checked it was in the 10$ range for just a few ounces.

I used to work for an HVAC companies "special projects" department making weird heating and cooling solutions, this is the one area I might be able to provide advice in.

once you clean off and plane the surface that you want to mount said heatsink to, you'll want to spread an extremely thin layer of thermally conductive paste (not adhesive) to the heatsink and then bolt the heatsink to the machined surface using holes drilled through the housing and tapped...

Or you could do it the easy way, and use an external pump and cooler...

A thin layer of thermal epoxy to attach the heat sink will work fine. Epoxies do not conduct heat as well as aluminum but if the layer is thin, and there is a lot of surface area, the delta-T will be low enough not to matter much.

A pump and a well placed cooler is more complex to implement but will work much better.

Yeah, I wouldn't focus too much on epoxies and heat sinks. Yes you may eventually get it to work, but you could throw a tried and trusted cooler/pump system in and have much better results.

Someone does it already, beetches:

http://www.turbomagazine.com/tech/08.../photo_08.html

That is a less than stellar testing methodology used in that article combined with anecdotal data. At any rate a cooler (fluid to air heat exchanger) is going to work much better. There is no way those fins on my oil pan would lower engine oil temperatures by 70F+ like the oil cooler does.

Agreed about not being as effective as an extra cooler with pump.

The Q is, maybe this will do 10-20* improvement, with low cost and simplicity.

They will drop a 10 or so degree difference. We use them on experimental trans. tests. I would not expect the tape to hold up under road conditions for very long though.

Is this still on the to-do list? Any ETA?

I would guess based on everything else on TSE's to do list that do not expect one till at least towards the end of this year.

all CPU heatsinks are attached with a synthetic silver grease stuff. perhaps something like that, but with more adhesive power (or make some attachment points for the heatsink).

Although I agree that you will have better results with a pump/cooler setup.

Yeah, they do seem to stay busy.

End of this year is fine for me. I'm not planning the major power mods until then anyway.

Is this still in development, or has the turbo kit absorbed all of TSE's R&D time?

Still on my todo list, but not near the top. I'll get something put together this fall after NASA Nationals are over.

New Electric Oil Pump for Diff Cooler or Turbo Scavenge Conversion Unlike VDO | eBay

Some line, fittings, a cooler, relay, wiring, and a switch.... Or get fancy and run a coolant temp switch like the OEM miata fan switch.

I got to get to work on mine since the flat underbody is going to trap heat even if I litter the bottom with Naca ducts.

I coping the Porsche RSR race car with 2 side ducts for Trani and Oil cooler on each side of the radiator duct.

What about making ducts for each Oil/Trani cooler then run piping from those to the Trackspeed Engineering Brake Ducts?

Bump!

Anybody install a trans or diff cooler on their miata? It seems reasonable that cooling the transmission would improve reliability, this at least seems accepted as true in other applications (trucks towing, automatic transmissions).

I found this Boat Marine 12V Engine Oil Change Gear Pump 3 2 GPM Transfer Water Electric DC | eBay on ebay, it's rated at 3.2 GPM, and it's a gear driven oil pump. I've searched a lot, and it seems to be the biggest oil pump that's gear driven and electric and not 1000 dollars and weighs 40 lbs.

Doesn't seem like much flowrate to me? Even if you had a 100% effective oil cooler, you'd only be pumping about 3 gallons/minute of cool oil into a box that's having to handle a lot of power on a track. I would think you would want more flowrate, but maybe 3 is enough if your cooler is awesome?

At 3 gallons per minute, that's 12x the capacity of the trans, per minute. I'd almost be concerned with not getting to proper operating temps

That would be a great problem to have! I will have a temp sensor for the transmission fluid so I could easily control the pump based on temp, run it only when needed. I guess in theory it will for sure help cool the trans, and the question is "how much"?

Is it rated for 300*F?

It says it has bronze gears and a brass housing. From what I've read about other pumps rated at less flowrate, plastic = sucks, metal = high temps ok. Also if it comes on at say 220*F and keeps the oil below 230*F, then it won't "see" 300*F oil in theory.

Keeping the trans temp below 230 may be a tall order, it runs a lot hotter than most realize. All the more reason for some experimentation in this area though. Personally would prefer an electric pump rather than gear driven I think..

-Ryan

I'd be worried about parts like seals, which are often constructed of plastic.

Well so far it's the best pump I've found under 500 bucks, so I'll probably just buy it, hook it up, and report back whether it works or not and if it leaks or not. Unless someone finds a better one!

Ryan- I agree, I don't think pumping 3 GPM of cool oil into a box that's handing 300+ hp is going to be enough to keep it cool, but I'd love to be wrong. I guess either way I'll find out.

So lets talk about heat exchangers for this. Where would you put it? How big? Fan or no fan? What size lines? I mean we're going to be pumping gear oil, so I'm thinking minimum 1/2" hoses.

Probably have the pump close to the transmission, so you don't have a lot of restriction on the inlet side, then let the pump push the fluid around to a heat exchanger and back.

That pump looks fine, ryan its an gear-rotor electric pump like the same type of pump as the oil pump but driven with an electric motor. Probably only has one seal from looking at it and its probably an oring so it should be fine.

Pat, pretty spot on on pump placement. For tranny connection, were you planning on drilling and tapping or using drain and fill holes already there? A banjo fitting would probably fit on the drain hole pretty easy and have a low profile.

Heat exchange wise. Does anyone have any temperature data at all? Just knowing the mass of fluid, the mass of aluminum and the mass of steel in the tranny, and the time it took to raise the tranny and fluid to x temperature you could figure out how much heat rejection capacity you need with the heat exchanger. Or you could just guess randomly and measure temps and compensate from there.

From my couple hours of google searching, my best guess is the people that have this data, and know what works/what doesn't are either not telling or simply not on the internet posting about it. I found almost nothing.

I'm figuring that too much cooling is never bad, so build it big and see what it does. If it's too big the pump won't run all the time, so no harm their.

I will do a little calculations at some point, just haven't yet. Working on wiring design for my car right now. But the CP of oil looks like it was ~1.7 kJ/kg*K, so much less than water.

Too big just ends up being more weight and more blockage of the already precious airflow through the mouth of the car.

True, but for my application I'm thinking about cutting up the rear of the trunk and putting the a/c condenser and a couple oil coolers back there. Seen Porsche do this (a/c condenser in rear) to get clean air to intercoolers and radiator up front.

I have no idea for inlet/outlets, drain and fill plugs would be convenient, but I don't remember where they are. No way I'm doing a banjo on the drain, that would be too restrictive I would think, unless it's a custom high flow banjo I don't know of... I'd prefer a large hose that goes straight to the pump inlet, no restriction at all.

It shouldnt be that bad with the banjo. That pump as long as its the same level as the bottom of the tranny is going to have less head loss before the pump than your oil pump would so it shouldnt be a big deal. The inlet is the annoying one with it on the exhaust side.

I've pondered a tranny cooler, and was intending to use one of the Tilton pumps that Summit sells for that purpose. Dunno if it's a better choice than that ebay one, though:

http://www.summitracing.com/search/p...Cooler%20Pumps

Putting the AC condenser in the trunk seems like an odd choice. I wouldn't do it on a street car (because I'd want the trunk to carry stuff) and if it's a race car then why does it have AC? :)

Fittings and hoses should be fairly straightforward, also a thermostatic controller or perhaps just a manual switch in the cabin along with a temp gauge.

--Ian

I've never used a Tilton pump, or the ebay one for that matter. The sticker on the pump in your link, the 40-527 pump shows 8 Amps, 1.9 GPM, 50 PSI Max. Vs the ebay pump has no pressure rating but "up to 10' lift pressure" (not much) and rated at 3.2 GPM.

Without testing either pump, my guess is if there is little restriction, the ebay pump will move more oil, if there's a big pressure drop, the Tilton will be better. So then the question is, which will move more oil if it has to push 200*F+ gear oil through an oil cooler?

Tilton sounds like the safer bet with the whole, being advertised for this function and not being all that much more than the ebay one.

Finally found some info, this is specific to rear end cooling

Article: Keeping You're rear Gears Cool - Circle Track Magazine

Quote from very bottom of article:

That's for a diff, and they're saying 20 GPM is ideal flow, that's 10x more than the electric pumps we're looking at..... I would think a transmission would generate more heat than a diff, but I could certainly be wrong. I really have no idea.

If they are right, an electric pump wont' cut it at 1-2 GPM. I can say that from what I've read, NASCAR does use pumps to cool their transmissions and diffs, and they are mechanical pumps, not electric.

See. This is why I love this forum. Pertinent information being shared. I now rescind my earlier reservations!


One question though. If the 20gpm is ideal for differential oil, is it because it holds, and or transfers heat different than that of the trans?

I think it's simply that to remove enough heat to keep the diff cool, you have to pump that much cooler oil into per minute to remove all the heat.

Rough numbers, but check this out:

This site post efficiency for diff gears Comparison of Gear Efficiencies - Spur, Helical, Bevel, Worm, Hypoid, Cycloid

From there,

I believe our diff is spiral bevel gear. So if we split the range and say we have 97% efficiency, then 3% of the power the diff sees will turn into heat.

So on a 300whp car, rough numbers, 3% of 300 = 9 HP worth of heat generated. If we use the worst case and say 95% efficiency, then 5% of 300 = 15 HP worth of heat generated.

I'm not gonna do the flow rate/temp calcs right now, but if the above is correct, our cooler needs to pump out 9-15 HP worth of heat on a mere 300whp car. Anyone please feel free to confirm/deny/laugh at the above!

20 gpm for the diff sounds crazy high. It's only got a quart of oil in it, so that's pumping the entire thing through the cooler in less than a second.

Let's try some back-of-the-envelope calculations. Poking around a bit, mineral oil has a specific heat of 1.6 kJ/kg-K. 15 hp is 11.2kJ/second, and with a 90K temperature different (250F vs say 100F) that suggests that the fluid could carry that much heat with a flow of only .077 kg/second. Gear oil is pretty close to the same density as water, so a 1L capacity is giving you around 1 kg total mass.

So if I did the math and unit conversion right, and assuming a heat exchanger big enough to output 100F gear oil, 20 gpm is about 10-12x higher than it needs to be. That'd suggest the 2 gpm of the Tilton is just right. :)

--Ian

+1, and I would say this is conservative. Also keep in mind though, you are assuming that all of the heat needs to be evacuated through the heat exchanger, and it doesn't. The gearbox radiates a large amount of heat.

It's also conservative because the heat exchanger does not need to output 100F fluid, I believe that is too low. I would guess you would want the tranny oil to stay in the 180-230F range.

According to AGMA(American Gear Manufacturers Association) gear design regulations, they do not apply a temperature safety factor to gears until they are over 240F, I believe. This is from memory, so the temps may be +-20F, but everyone here can get the point.

On a side note, a 20 GPM pump seems WAY too big for me. Even a 2 GPM pump would empty the gearbox off all oil in less than 8 seconds if a hose popped or got damaged in some way.

I don't pretend to be as smart as some of you guys but I always thought I would use the power steering pump I removed from my car to circulate the oil from my transmission through a cooler. It is rated for oil and for high temperature. I thought it would likely be mounted to a bracket off of the PPF and use a belt over the driveshaft to spin it. When the car moves, oil flows. It wouldn't make any pressure except for the little bit of flow resistance from going through the cooler. It is a simple little gear pump that should last forever in that application. The belt wouldn't need to be very tight because there wouldn't be much pressure on the pump. A little spring loaded tensioner pulley like the one used under a lawn mower deck on a Snapper would work just fine. Keep it simple.

It would be good because it would involve no heavy additional electric motors or additional electrical power drain. In my car there is already a deficient alternator output when the radiator fans, headlights, or power windows are on. I need that power for spark plugs and Megasquirtseses.

This is all good post guys.

I will say though, if his numbers rely on a 150*F temp drop, and the new system only drops the temp say 50, then to remove the same amount of heat we will need more flow.

BUT, as has been mentioned, we don't need to remove ALL the heat, just some.

I don't have the data, but if someone with say, 150whp, or 200whp miata could tell use, "hey, I have 150whp and my transmission oil is 220*F after a 30 minute track session", then we could basically guess that the cooler needs to be sized to remove the difference between YOUR whp and the "limit" of what the stock trans can handle without getting too hot all by itself.

I have no idea if anyone has this data though! But does this seem like a reasonable way to estimate the system requirements?

And +1 to Sixshooter's idea, i like that, simple and cheap!

The cooler doesn't actually need to reduce the temperature of the fluid by the entire desired amount in a single pass. It's a small volume of oil and it is all going to make several trips through the cooler each lap. Less than 2 gallons of oil in the trans, 2GPM pump, 2.5 minute lap time = a pretty serious amount of cooling even at 15*F drop per pass.

The pump size gives you the maximum amount of heat you can transport from the diff to the heat exchanger. If the fluid coming out of the heat exchanger is too hot, then the problem is that the heat exchanger isn't big enough (or efficient enough, or have enough airflow, or whatever), not that the pump isn't transferring enough fluid.

And yes, the diff cools by conducting heat to the air passing over the fins as well, so the cooler doesn't have to handle the full load. My intention wasn't to do a full analysis, just a quick back-of-the-envelope calculation to look at the question of 20 gpm vs 2 gpm.

I don't think a PS pump is the right choice. For one, it's intended to make pressure, not volume. For another, you don't want the pump to run all the time, that would overcool the oil. I suspect the Tilton is lighter than a stock PS pump, as well.

--Ian

That is true. But you can calculate the amount of cooling, it's m*CP*dT, mass flow rate time heat capacity times change in temp. So that takes into account how many times it passes through.

The tilton is rated at 1.9 GPM.

This 75W 90 oil Mobil 1™ Syn Gear Lube LS 75W-90 is listed at having a density of 0.859 g/cc. 1.9 gpm is 7,192 cc/minute, so that's .859g * 7,192 = 6,178 grams/minute, or 6.178 kg/minute.

Cp of oil is 1.67 kJ/kg*K I think.... Not 100% on that. Can anyone confirm?

So if our GOAL is to keep the oil at say, 230*F max, then the oil exiting the trans will be 230*F, and we'll cool it a bit and then return it to the trans.

230*F = 110*C
So if we cool the oil say 50*F, which I would think would be pretty reasonable/good, that's 230 - 50 = 180*F = 82*C exiting the heat exchanger heading back to the trans. So temp drop is 110-82 = 28*C I didn't put numbers in Kelvin but the delta T will be the same K or C.

heat removed would be m*cp*delta T, so 6.178 kg/minute * 1.67 Kj/kg *C * 28C = 288.88 KJ/minute, which is 4.81 kJ/sec, which is basically 4,810 watts, or 6.46hp worth of heat.

I think.... That's actually a decent amount of cooling, more than I thought it would be.

Agreed! See my math, let me know if you agree or I messed something up....

FWIW, the article I linked that claimed 20 gpm was a racecar with a 9", not a miata making 300whp. No idea if there application needed more flow, just numbers I found on the internets!

NASCAR belt driven pump that runs off of the driveshaft:

https://www.miataturbo.net/attachmen...ine=1414530018

https://www.miataturbo.net/attachmen...ine=1414530018

https://www.miataturbo.net/attachmen...ine=1414530018

Hell, you can mount all kinds of interesting things back there on a race car:
http://www.jalopyjournal.com/forum/a...nt-jpg.631082/

https://encrypted-tbn0.gstatic.com/i...9cOdtCGe-GfFpb

That nascar pump, cooler, and lines all look way too small to be able to do 20gpm. I'm thinking the link that was quoted had a typo?

That is possible. They said it ran at 5 PSI, which sure sounds more like 2 GPM vs 20.

If so, then everything looks ok for 2 GPM.

And from the pic above, that oil cooler isn't that big at all. But I'm guessing if it's from a NASCAR car, it probably gets 170mph air, so it doesn't need to be huge to be effective.

Just a thought, why not call a nascar team and ask what they think! I suspect that they would stop laughing long enough to give a tip or two!!!