Mazda 6 Forums banner

1 - 19 of 19 Posts

·
Registered
Joined
·
240 Posts
Discussion Starter #1
How would one do it?

My car (a ‘15) runs ATF temps of 200 - 208F when I’m driving down the highway in warm to hot weather. That’s very approximately 5 - 10F above engine coolant temp. 4 ATF drain-and-fill cycles has not impacted ATF temp, though the fluid looks much better than it did when I got the car.

As far as I can tell, ATF doesn’t leave the tranny. Rather, engine coolant goes to the tranny to a small heat exchanger. Couldn’t one rig up a completely separate engine coolant circuit to run from that heat exchanger to a small radiator such as an ATF radiator, up front? An electric pump could circulate this coolant. One could also include a thermostatic control to either cycle the pump on and off, or to bypass the radiator in order to not over-cool the ATF (My car’s ATF heats very slowly in winter on the highway).

Am I missing anything here?
 

·
Registered
Joined
·
1,077 Posts
What temp does the ATF ultimately get to (steady-state) in winter- and during the shoulder seasons?
 

·
Registered
Joined
·
1,077 Posts
See post no. 27 and thereabout?

 

·
Registered
Joined
·
240 Posts
Discussion Starter #4
I may not originally have understood what that post implied. Will the tranny pump its ATF out of the case and to an external cooler if one uses that block you mentioned?

If it would, that would certainly make things simpler than an electric motor circulating coolant through a totally separate cooling system.

I would of course monitor it if I changed the ATF heat management system, but I’m not particularly concerned about wintertime ATF temps. The tranny eventually warms up. Remember that I’m in the southern AZ desert, so winter has a different meaning here than for most of the continent. :) A simple thermostat that shunts to bypass rather than opening & closing would prevent over-cooling.
 

·
Registered
Joined
·
1,077 Posts
Will the tranny pump its ATF out of the case and to an external cooler if one uses that block you mentioned?

If it would, that would certainly make things simpler than an electric motor circulating coolant through a totally separate cooling system.

....

A simple thermostat that shunts to bypass rather than opening & closing would prevent over-cooling.
That hockey puck-like heat exchanger is likely a furnace-brazed stacked plate unit... and the transmission pumps the ATF thru lubricant-side passageways inside it... with one of the "captured o-ring" ports as an inlet and the other port as an outlet. Meanwhile engine coolant is taken from near the coolant pump discharge to supply the liquid-to-liquid HEX and the coolant outlet from the HEX is delivered to somewhere near engine coolant pump suction-side. That pressure differential drives flow on the coolant side of the HEX.

The A/T pumps ATF which often is en-route to a lubricant supply point... thru the HEX and then on to that lube-requiring supply point. You do not want to restrict that flow much by adding restrictive additional components. "Restrictive" is the key word. Generously sized pipes / tubes / fittings are NP. Also, lifting the flow 'way-up in elevation should not be attempted (either). Putting a moderately-sized "stacked-plate" type liquid-to-air ATF cooler in front of the A/C condenser... as close in elevation (or centred by way of elev. with the OEM liquid-to-liquid HEX would be ideal).

How to achieve this? Make a sandwich plate. Take a 1-1/2" thick aluminum plate (say 6061 T6 mtl) and recreate the captured O-ring ports on either side of the sandwich plate's thru-drilling. Do this on the INLET side to the liquid-to-liquid HEX only. On the outlet side recreate the o-ring recess to mate-up with the HEX but instead of thru-drilling, go only say 0.600" deep (i.e with a square-bottomed blind hole). Then cross-drill the edge of the plate (to intercept the blind hole) and tap the edge-drilling with 3/8" NPT. Do the same on the other side of the aluminum plate... but "phase" the cross-drillings to prevent fitting interferences with the two 3/8" NPTF tappings. Go on your merry way with 3/8" SS tubing and quality hose product to get to your new Aux. ATF cooler. Use a "beading" process when attaching hose product to SS tubing so that hoses / hose-clamps don't slide off as they get hot. You can use Swagelock ferels (that have been made up to torque and have been "swaged" then soldered to make your hose-beads).

I have written too much. Comments re adding block and bypass ball valves for bypassing this aux cooler in winter apply.

Re your idea for a T-stat... I can comment in a PM if you're interested.
 

·
Registered
Joined
·
1,077 Posts
To be clear, here, I am saying add the (disconnectable / bypass-able) aux. ATF cooler downstream of the OEM HEX.
 

·
Registered
Joined
·
240 Posts
Discussion Starter #8
I’m not much of a fan of factory ATF ‘heaters’ in my climate. About all they can realistically do here is assure that the ATF is overheated nearly all the time, IME. Combine that with ‘lifetime’ ATF, and you’re working very hard to assure your tranny won’t last as long as it should, IMO.

I can easily make all the calcs necessary to size the lines and the radiator correctly (I still have those texts and field manuals) if it could be discovered what the design parameters of that ATF circuit are. Same goes for the engine coolant side, though that seems like a super-easy hack: Just leave that side all hooked up.

Yes, a simple cover for the ATF radiator would be a simple and much cheaper solution. I’d prefer thermostatic control, but it isn’t necessary.

Cdn, PM away with other thoughts.
 
  • Like
Reactions: DrFeelGood

·
Registered
Joined
·
240 Posts
Discussion Starter #9
To be clear, here, I am saying add the (disconnectable / bypass-able) aux. ATF cooler downstream of the OEM HEX.
This is how Honda designed the ATF cooling system on my pickup.
 

·
Registered
Joined
·
1,077 Posts
I’m not much of a fan of factory ATF ‘heaters’ in my climate. About all they can realistically do here is assure that the ATF is overheated nearly all the time, IME. Combine that with ‘lifetime’ ATF, and you’re working very hard to assure your tranny won’t last as long as it should, IMO.

I can easily make all the calcs necessary to size the lines and the radiator correctly (I still have those texts and field manuals) if it could be discovered what the design parameters of that ATF circuit are. Same goes for the engine coolant side, though that seems like a super-easy hack: Just leave that side all hooked up.

Yes, a simple cover for the ATF radiator would be a simple and much cheaper solution. I’d prefer thermostatic control, but it isn’t necessary.

Cdn, PM away with other thoughts.
I had indicated that I would PM you, Bulwnkl, as I gauged that the level of interest in this thread was not 'real high... but what the h*ll...

What I wanted to say is that thermostatic control of the ATF temperature, it seems to me, is not that super-easy (in a practical sense). Now I read on BITOG (Bob is the Oil Guy) that if a person were to sense the ATF temp in the transmission outlet piping (the outlet towards the ATF cooler)... which one would think is pretty close to the A/T sump temperature... well, that can be not accurate at all, and results in poor ATF temp control. Apparently when, say, climbing a high-load highway hill, the ATF-out temp can seemingly be quite high... notably higher than the sump temp. I really don't know why this would be the case, as the ATF-out probably constitutes sump-ATF that is drawn-in to the A/T main pump, possibly pressure-regulated down to the desired pressure (for the needs of the lubricant supply point that it is routed-to)... and then delivered to the cooling system. So, if you DO subscribe to this notion that temp-sensing of the outlet line is inferior to temp-sensing at the sump... then I THINK you are left with a control system design challenge... You would generally then want to sense at the sump, but control ATF cooling system flow / BYPASS of cooling system flow... based on a remote signal... Not based on the actual fluid you are diverting... I guess pilot-operated control.

Sorry, control system design is not my forte.

So, if you instead consider controlling the rad-fan, to have MORE or LESS fan-action over the auxiliary aerial ATF cooler... based on a bung / temperature switch stuck directly on A/T pan... that might be the better way to do this.

As to the need for block- and bypass (with low-restriction ball valves) on the aux. cooler (versus simply shielding the aerial cooler) in winter - I worry about the actual lines themselves... and ATF "gelling" in those lines. No, it does not gel...but it does become viscous... and it therefore can / does impede ATF flow. I feel better to simply take the aux. ATF cooler totally out of the circuit to avert those possible problems.

Comments invited...
 

·
Registered
Joined
·
1,077 Posts
I should also add that in summertime I think you can throttle the coolant-out side of the HEX say by interposing a ball valve in the outlet line and adjusting the restriction. I don't think valve errosion would occur at these low pressures. Depends on just where said outlet line is tied-in to. If there is either an upsteam-of-HEX or downstream-of-HEX user of that coolant... then mebe this ISN'T a good idea. This action would ultimately reduce the heat-up action of the ATF... in your Arizona desert situation.

I think optimal ATF temp is 175-185F.
 

·
Registered
Joined
·
240 Posts
Discussion Starter #12
Thanks for the additional posts.

I agree with your control system challenge point. I had considered trying to control flow (or I might say bypass of the external cooler) based on ECU-reported ATF temp. That is doable, but excessively complex and I have no interest in either designing or building that controller for a mechanical bypass valve in the aux cooler circuit.

That said, I am more willing to use an inline thermostatic bypass than some on BITOG. A member there from long ago had very good results with them for just this kind of thing.

At this point it seems that monitoring ATF temp with the coolant flow shut completely off would be instructive. I’m away from the car for a while now, so I’ll look at it after I get back to see whether I can simply clamp a hose to shut off the flow.
 
  • Like
Reactions: Cdn17Sport6MT

·
Registered
2016 Mazda 6 Touring AT, coilovers 1.2" drop, OVTuned, Injen CAI, Thrush Exhaust
Joined
·
119 Posts
This is an interesting discussion, I'm also concerned about ATF temps (though not because I'm in the desert, but when I'm autoX-ing or track days or aggressive uphill canyon driving on a warm day).

@bulwnkl - How are you measuring the ATF temperature? (I'm assuming a bluetooth OBD reader, what app are you using).
Also, for synthetic ATF is 210 degrees really a problem?
I thought for synthetic oils <230 should be fine, or is that only for engine oils? (sorry if this seems like a noob question, this is an area I do not know much about).

Also, I have a few friends with track cars that have lowered crank case oil temperatures drop by 5-10 degrees after adding LubeGard, and the LubeGard Platinum ATF additive is rated for Mazda FZ specs/compatibility, may be worth a try, though unlikely to make more than a few degrees difference.

Also, probably not worth the risk/impact elsewhere in the system, but what about running a slightly lower temp thermostat (if the current one is 195 degrees, maybe run 190?)

Or, again mabye this is noob-ish or wouldn't do enough, but could there be a way to affix some sort of heatsink/fins to the AT pan to just radiate more heat?
 

·
Registered
Joined
·
1,077 Posts
Re ATF temp... oxidation / degradation of the fluid isn't so much the issue (I believe) but rather degradation of valve body (especially)- and seal soft parts. Like i said above, 175F to 185F ATF temp is the desirable range for most automatics.
 

·
Registered
Joined
·
240 Posts
Discussion Starter #15
@bulwnkl - How are you measuring the ATF temperature?
I use a ScanGauge, so I’m looking at the factory ATF temp sensor data. I had a connecting cable for a laptop when I had my Subaru Baja, but that software didn’t work with the Honda I got later, so I just got the ScanGauge. I’m actually super happy with it. It’s a great tool for me.
 

·
Registered
2016 Mazda 6 Touring AT, coilovers 1.2" drop, OVTuned, Injen CAI, Thrush Exhaust
Joined
·
119 Posts
All due respect @Cdn17Sport6MT - it seems your assertion of 175F-185F is referring to older transmission fluids, as is most articles talking about ideal transmission temperature from 10+ years ago.
First, if our engine coolant regularly operates above 190F why would Mazda engineer it to have that flood pump through the transmission since that would mean 'overheating' the transmission by design.
Second, synthetic low viscosity transmission fluids (like Mazda FZ or Dex VI) have an optimal operating range between 175F-225F.
See article here TCI Automotive Automatic Transmission Tips - Classic Trucks Magazine
"At approximately 240 degrees, important additives in automatic transmission fluid (ATF) begin to cook. The result is the formation of varnish inside the transmission. At approximately 260 degrees, internal transmission seals (which are typically manufactured from a polyacrylate material) begin to harden..."

@bulwnkl - I'm not sure there is any need for ATF cooler as long as it's staying under 225F, if want to figure out if the fluid is degrading consider getting it tested/analyzed. (back in the days of non-synthetic ATF then going above 200F would be concerning).
 

·
Registered
Joined
·
1,077 Posts
Fair enough comment re it being synthetic... but as I mentioned it's not fluid degradation that I'm worried about. It's seal / elastomer damage... in various locations... like in the valve body, the solenoid operated valves, and in the transmission-proper... due to high temps.

Mazda is far from the only mfr which uses engine coolant to heat up and ultimately to cool the ATF. Not sure it's cool as would be optimal though.
 

·
Registered
2016 Mazda 6 Touring AT, coilovers 1.2" drop, OVTuned, Injen CAI, Thrush Exhaust
Joined
·
119 Posts
So I finally got my bluetooth OBD-reader with Torque app going. So far the transmission temp and coolant temp seem to stay in lock-step (though I haven't done a lot of driving with it yet plus the weather has been fairly mild).
Thus far, it's been only 185-195 and have not seen any deviation between the two temperatures of more than 1 degree (no matter how hard I hit the gas).
@bulwnkl Have you considered flushing your coolant and running higher percentage of water? (as it seems the coolant dictates the transmission temp)
 

·
Registered
Joined
·
240 Posts
Discussion Starter #19
No, I have seen nothing to indicate that the cooling system lacks adequate heat transfer capability in my car’s easy service. Engine coolant temp is where it would be expected to be given a thermostat temp of ~195F.
 
1 - 19 of 19 Posts
Top