tjbreen

I have a single 22x8x2 Setrab in front of the radiator. I didn't like the idea of series or parallel and got as much surface area with the one cooler as most do with two. I work on the KISS principle and the plumbing for this setup is simple and direct. Stainless mesh in front has kept it relatively undamaged from track debris.

On hot track days with the stock AC in place I was getting hot (water 240F and oil 270F) on extended sessions (30min +) but have since removed the condenser and all is well. Water temps measured at the inlet to the radiator never go above 220F and oil temps measured at the turbo banjo bolt never go above 250F. I run water wetter and 16oz of antifreeze per refill. Radiator is original but clean.

kurt M

Straight weight changes viscosity more over a temp range than a multi weight. 5-40 acts like a 5 at low temps and a 40 at high temps for example. If your concern is keeping a stable viscosity across a wide range of temps you would want a multi weight oil.

There are massive brain hurting formulas for calculating flow through heat exchangers to design out tunneling or backwaters in exchangers. While the formula are beyond my little shade tree wrench pea brain I have dealt with the concept and outcomes in real world apps many times. The described effect also happens inside a single exchanger when the conditions are right (wrong). Flow rates, fluid viscosity, system design all play a part. I suspect that for a properly setup system the flow of the system would overcome a viscosity based backwater in a single exchanger or in a dual system. Depending on the air temp delta and temp drop from single pass you are not going from high temp oil to cold in a pass. 250 to 190 would be a huge delta. 230 to 200 is a good range to work on. That amount of a thermal drop should not make a huge change in the thickness of modern motor oils.
Yes a gear based pump is not as bothered with back pressure than a centrifugal type pump but high back pressure is what blows filters, lines and exchangers apart, often on startup. It also is what starves motors of heat absorbing oil when it is needed most at high loads and RPMs.

I am building a 993 motor based track car with dual cooler system using two 964 oil coolers plumbed in parallel. Very easy for me to record temps on both coolers and report back when it is up and running. I am hopefully about a month or so from initial testing. The coolers are plumbed in a way that is easy to switch from parallel to series so I could test and tinker if needed. I will post pix later.

67King

Arg, crap, you are right. Not sure what I was thinking. Or maybe I wasn't.

I'm with you on the formulas. And I sucked in heat transfer. A somewhat decent analogy would be how to mount intercoolers, if anyone here deals with that. Multiple short tubes are preferred to few long tubes for the same types of reasons that parallel coolers are better than series (in theory, anyway).

But, I would expect that it can be summarized as deltaQ = Cp*density*heat gradient, for measuring the ability of the coolers to pull out heat. In other words, the ability to do work all boils down to the heat gradient, as everything else - heat capacity and density (of the cooler, which is what we are interested in) - is the same. This heat gradient is the difference between the oil temperature and the ambient air temperature. So if the first oil cooler is 50% efficient (which it won't be, but just for clarification), then the second cooler's delta T would be half the size of the first, and the total deltaQ would be 1.5 (1 + 0.5), versus 2 (1 + 1). This is a drastic oversimplification because flow rates aren't factored in, but for a qualitative look, it should be appropriate.

Of course, since you've got the ability to switch back and forth, you have an ideal situation. For anyone who would have to made a decision and go with it, this is the approach I would take.

Jeff Lamb

Yes, this is correct. I can't believe I forgot about this during my reply last night. Too many other things going on . . .

That sounds great. You will be testing your setup before I finish my car so I look forward to hearing about your findings. As long as the parallel coolers don't suffer from the viscosity induced flow imbalance that BAT described, I do agree that running the coolers in parallel is the better way to go. I look forward to seeing some pictures when you have the chance to post them.

How many and what type of temperature senders are you going to run? And, where are you going to place them?

One other thing -> Do you happen to know how much oil that various 911 pumps move at various RPMs on the intake and exit sides? The BAT guys said that max flow should be around 10 gpm, however, I don't know what RPM they were referring to, nor do I know which 911 oil pump and whether or not this is the intake or the exit side of the system. I guess I could always give them another call to ask some more questions . . .

Jeff

Jeff Lamb

Wow. Those sound like some HIGH temperatures both before and after the removal of the AC condensor. I was thinking that we were supposed to avoid going above 240 degrees for oil temps for any length of time. However, I haven't really looked into this in enough detail. Maybe there are several others out there (like yourself) who run oil temps up to 250 degrees (or even hotter) with no problems?

Jeff

Van

I have my oil temp sender in the drain plug - which is obviously at the bottom of the engine and right next to the exhaust. It used to read over 250 pretty regularly, now it's in the 240-245 range with the 2nd cooler.

I change the oil every few events - and, knock on wood, haven't had a problem yet.

Oddjob

I see upto approx 260F temps in the oil pan in a 951 during long sessions or sprint races. I have dual oil coolers, the KISS mocal kit in series with the stock factory cooler.

Coolant temps run normal, approx 1/3 to 1/2 way on the gage.

I would like lower oil temps and am considering going with larger coolers, as I start losing oil pressure above 250F (approx 1/2+ bar).

A930Rocket

I've got mine in series and it seems to be working well. On a 90* day at the track, oil temps get to about 210-215* after 25 minutes. This is with 400 rwhp.

I gotta think 240-250* is high and is starting to break down. But what do I know....

tjbreen

The 250F is measured in the hottest place possible, the return oil from the turbo to the pan. I have not measured the oil pan temps but would expect those to be much lower. I do run 20-50 synthetic and when changed, I always send an oil sample to Blackstone labs for analysis since I rebuilt the head and bottom end about four years ago. So far, they have reported no issues with either the engine wear or oil life.

The water temp is also measured at the hottest place possible, at the radiator inlet. I went for worst case possible when placing sensors for my aftermarket gauges.

joonas

I do not know if BAT is right or not. Main point is that you get better cooling with parallel connection.

"When using multiple heat exchangers remember that the greather the difference in temperature between the liquid to be cooled and the air that is doing the cooling, the greater will be the temperature drop across the cooler. First, plumb multiple coolers in parallel rather than in series."

If I understood something from physics book then if you use parallel connection.
1. Input pressure is equal
2. If flow speed changes then pressure drops
3. At the output pressure wants to become equal again. So it should get more oil from high pressure side or something
I am probably totally wrong

kurt M

I think BAT does have a good point. The issue they point out can happen. It can even happen inside a single cooler if conditions are poor. Only good calulations, design, build and testing can tell one way or the other. IF the coolers get same flow then a parallel system is better. Lower flow rates and higher thermal delta = more heat moved per pass per quart. Just have to make it right and test that it works.

looping coolers makes the restriction add up. If each are OK by themselves linking them might make for high pressures/lower flow. Side by side reduces one after the other increases. It is not hard to test the systems.

kurt M

Pictures as previously promised. I am using two 964 coolers mounted mirror to each other. The lines are equal length and fittings and coolers are the same as well. I plan to mount two digital thermometers that show output temps from each cooler taken in the return portion of the cooler manifold. If the oil backwaters or tunnels I can disconnect one line from each cooler at the thermostat, connect them together with a splice fitting and change the configuration from parallel to series. I hope to have some data in a few months. (spare time and $ build) The air intake is 1/2 the exhaust vent size and the intake in the bumper is the same size as the intake hole in the body. I am fabricating a fan that can be mounted in the exhaust vent to pull air through the coolers. It will have flappers that open with air pressure so the fan will not restrict free air flow when the car is at speed. Motor is a stock for now 95 993.

Jeff Lamb

Wow. You have a VERY nice looking project!! I am looking forward to seeing it on the track. The interesting thing is that the oil cooler setup I am putting on my car is nearly identical to your configuration, except I am using two large Setrab coolers. The guys at Zuffenhaus in Charlotte are working on it right now and I will attach some pics when some further progress is made.

What thermostat are you running? And, are you running -12 oil lines? Or, are they the -16 lines?

One other question -> What are those large air scoop looking things in front of the front wheels?

Jeff

Jeff Lamb

Kurt, what are the orientation of the end tanks on your oil coolers? I am setting up my Setrab coolers so that the oil enters the bottom of each cooler and then exits the top rather than flowing from left to right or right to left.

Jeff

kurt M

Oil moves in lower fitting and out upper fitting. -12 on the pressure side -16 on the tank to motor side. Scoops are brake cooling ducts that line up with the turn/running light holes in the fenders. They make slightly more sense with the fenders in place.

you are in good hands with the guys at Z haus.