My latest coolant reroute
 

So after my high coolant temperatures at the track I have been on a quest to bring them down. I am using a multi-pronged approach, one of which is to rethink the coolant reroute. I have incorporated several elements from previous discussions/implementations posted here as there are a lot of great ideas.

I have already done the spacerless reroute, and this is in fact what I was running at the track. The coolant flow at the front of the head is completely plugged (freeze plug) and all flow exits the back of the head.

One of the reasons I think I may have been having problems with temperature is the heater core recirculating back into the 'mixing manifold' or in my case the lower radiator hose. Because of this route, there is a significant portion of the flow that is not circulating through the radiator. The pro is that it is good for a street car; quick heatup, no overcooling (been there, done that). The con is that it is not great for a track car for the reason already stated. This has been discussed at length in other threads so I will not rehash it all here. But what am implementing is the idea of recirculating the heater core loop into the upper radiator hose, but regulating it with an in-line thermostat.

Another change is that I am redoing the way the small diameter lines circulate through the heater core and turbo; I am running them in parallel. I am also eliminating the circulation through the throttle body. There are two points at which flow is tapped for the two parallel loops. One is the existing, stock point at the back of the head. The other is from the pre-thermostat point in the in-line housing for the heater core flow. The former will be the one that feeds the oil cooler. The latter will be the one that feeds the turbo. The flows are teed together just before the bung located at the inlet manifold to the water pump.

I'll post up some pictures of the actual parts. But for now here is a diagram of the reroute. I am looking for some input to see if I overlooked anything or if anyone sees any potential problems. The car is offline right now so this is one of several things I am working on while I can.

A quick comment on the dual thermostats. Right now they are both 180 deg F units. I think I may switch the one at the back of the head to a 200 deg F unit. That way the flow will be biased through the heater core. If that flow is not enough to keep the engine cool, then the one at the back of the head will open and provide additional flow.

Also, note that there is continuous flow through the motor. The flow through the parallel lines for the turbo and oil cooler are always free to flow. Plus I drilled a small hole, about 3/16" in the flange of the front t-stat so it always flows as well. Hopefully those three paths will be enough to keep a sufficient amount of coolant moving through the motor during warm-up, without flowing too much and keeping the motor from warming up completely.

I am also trying out my previously posted double oil cooler sandwich idea using two stock water/oil heat exchangers. I already installed it and routed the coolant lines and it all fits. I'll be installing a temp sending unit in the oil drain plug to see what the resulting delta-T is between the oil and the coolant.

The only thing I see is the possible lack of flow before the t-stats open.
__________________
Best Car Insurance | Auto Protection Today | FREE Trade-In Quote

Yep, I agree. There is flow before the t-stats open, just not sure whether it is enough. If it is not enough, I can always drill more holes in the front t-stat flange to allow more flow to bypass the t-stat.

But, how do I know if there is enough? I am not sure of a good way to tell if I am getting enough flow through the head before the t-stats open, short of something like instrumenting the head with a bunch of thermocouples to look for hot spots, or checking the flow rates against a stock setup.

So you already have holes drilled? How many, how big? If the holes combined are as big or close to the inner diameter of the heater hoses I would say your good then.
__________________
Best Car Insurance | Auto Protection Today | FREE Trade-In Quote

Well I am not sure I want them to be that large. If they are, I think I will be back to having the same problem I had before with the unregulated upper radiator heater core hose return. There is too much flow through the radiator when warming up and cruising, and the car takes too long to get up to temperature, or never gets there.

Why do you guys keep re-inventing the re-route?

With this routing, the heater will see very little flow.

I have done 6 re routes on '01 - '05
 

This works! Wheel b/4 reinvention! These kits are for '01 - '05 cold side S/C or Naturally aspirated. $175 if any one wants one. I am working on a kit for the intake side. I will say that stock coilpacks are a problem, of course any one concidering a reroute is a "track dog" and has changed to COPS anyway.

Well, I think I answered that already, but maybe not, so I'll try again. If the heater core return goes into the lower hose, like stock, then (I think) there is too much flow that is not circulating through the radiator and the car runs hot. If the heater core return goes into the upper radiator hose, unregulated, then the car runs too cool for the street. The winters here in South Texas are not harsh so little heater core flow is not a big deal.

I have seen three-way regulators, though not big enough for a car. Instead of blocking/opening flow based on inlet temperature, they divert flow between two different outlets. The application I worked on was for a heat exchanger for a high-powered radar system; that is all I will say about that. But basically it either took the incoming coolant and diverted it to the heat exchanger, or flowed it directly back into the coolant loop. That way there was always full flow through the system, but temperature was still regulated.

If I could find one of these regulators with 5/8" inlet-outlets (or adaptable to that) it would be ideal. One output goes to the upper, one goes to the lower. Done.

I would think that the exhaust manifold would absolutely dump heat into the coolant pipe, I'd at least wrap it.

I used a cold-side reroute. No way I wanted to route mine hot-side with the S4 manifold.

BTW noob if you want to sell your reroute kit, which really has nothing to do with mine, please start a new thread.

What's wrong with the standard reroute then add a valve or t-stat to stop flow through the heater when hot?

Yeah that could work too, but then you are giving up some flow. On the street, that is no problem. But I think it is better when the engine is shedding a LOT of heat, like on the track (where I had the problem), to use all available flow through the engine and radiator to help with cooling.

Where does one get a standard packaged t-stat that closes with heat instead of opens? I would be interested in that.

Someone previously suggested an oil cooler thermostat

IIRC they talked to an oil thermostat supplier who said it would not work correctly for some reason, but I could be wrong.

Yes, the use of an oil T-stat was discussed with one of the companies that makes them, and it was concluded that due to the much lower viscosity of water, the coolant would pretty much bypass the thermostat all the time. (Oil thermostats never close completely- they have a bypass path designed to allow 10-15% "leakage" in order to maintain circulation through the whole system. With water instead of oil, you'd get closer to 100% leakage.)

Why not just restrict flow between the heater core and the lower rad hose? Many Hondas have a little ball valve that only opens when the climate control is set to 'hot'. Nab one out a junkyard and find a reasonable setting that gets some flow, but not enough to hurt cooling. Or just close it entirely before a session.

I have suggested a tap in that heater return line a long time ago, would still work actually. Just find a tap.

That would cause water to stop flowing past the rear mounted thermostat and cause it to not open as readily when necessary.



One possible fix would be simply to reduce flow volume through the heater core enough that it wouldn't be as much a detriment to the temp at the mixing manifold, but would still have a little flow for the benefit of the thermostat and for offering a little heat from the heater. Like 1/3 or 1/4 the flow?

A normally open 12v solenoid valve,wired n parallel w/ the rad fan should work.

Zxtex, the problem I see with your original sketch is that before the rear thermostat opens, the only coolant flow through the engine is through that heater hose, which you've blocked with the second thermostat. I will be doing this same reroute without the second thermostat, soon. It is encouagimg to hear that it is too cold for the streets.

And to save me five minutes of searching, do you remember the freeze plug diameter for the front of the block?

Actually there will always be flow through the two loops that flow through the turbo and the oil cooler, plus a little flow through the heater core thermostat as well due to the hole I drilled in it. That is important, since both thermostats need to see flow across the paraffin actuators so that they will 'know' when the engine is warm and open up.

In the diagram I show those lines connected to the 'upstream' side of the thermostat, probably not clear though.

On that freeze plug, dang it I wish I could remember. It is metric, I want to say 34mm. I do remember I had to go to O'Reilly to find one. Auto Zone and Advance did not have it. Cannot remember if NAPA does.

30mm, usually labeled as 30.05 or so. Autozone had 'em down where I live.

Thanks Joe.

You know, back to my question about determining if there is enough flow through the engine while the thermostats are closed... The first time I start up the engine with this hooked up, I could log/watch the coolant temperature closely; the temperature sending units (stock and aftermarket gauge) are right at the back of the head, before the t-stat.

If the temperature suddenly jumped (or oscillated) at or near the thermostat opening temperature, this could be an indication that there were some pockets of hot coolant in the engine that were not circulating well with the rest of the constant-flow circulation (while the thermostat was closed).

Alright, ill ask for a 30 (30.05), thanks.

Do you really trust two 3/8" lines and a little coolant going through some (1/8" or so I'm assuming) holes to supply enough coolant to your head for warm up? That doesn't seem like much.

Yeah that is what I am not sure about. How much is really needed?

You're thinking of using the wrong pair of ports. You can choose a pair of ports that have near zero leakage when hot.

Why don't you run the Turbo and Oil Cooler outlets to the radiator inlet?

Chris

Good question. I think they will flow more if they are connected at the pump inlet since the pressure should be a little lower there. Plus, though I doubt they will flow enough to significantly effect the engine warm up time, they will be circulating warm flow like the heater core does in the stock configuration.

So, how do we solve this? I am about to route the heater return to the radiator inlet, but understand that the car may take ages to warm up on the street...

What do you consider too cool for the street? My reroute is kinda similar, just without the t-stat in the heater hose but I'm getting 195-200° temps in normal driving. Have you tried turning the heater on full blast when on track?

The car would take 15 minutes or more to get above 160-170F. And, the temperature would fluctuate greatly from 160-200 depending on if I was sitting at a light or on the highway. It is just not very stable.

I run the hard top so I really do not want a setup that relies on running the heater to keep the engine cooled off. Even with the windows down it would heat up the cockpit too much for me on a hot day.

I'm in the process of installing this reroute so I'll report back on how well it does once the car is running again. I am making several other cooling related changes as well so it will be hard to look at it as a back-to-back comparison. But, I think the logs of the coolant temps will still be useful data. I fabricated a thermostat housing for the heater core circuit using a spare thermostat cover and the thermostat housing I removed from the front of the engine. It looks to me like it will work well.

I just put a valve in the heater core return that goes to the mixing manifold. Put it at 'almost closed'.

That, and I remolded the hood vent. Flushed all coolant with pure water.

Thursday's track day will tell if my temps stay lower now.

Spookyfish, is this somehow switching between returning the heater outlet to the mixing manifold and to the radiator inlet? Or are you just choking off the return to the radiator to speed up the warm up process, wouldn't that limit flow?

I think what he is doing is restricting flow through the heater core (and into the water pump via a stock mixing manifold type configuration) to a bare minimum. This will force most of the flow through the radiator return hose (and through the radiator) once the t-stat opens.

Do you have a coolant reroute with the thermostat at the back of the head? Seems like this would be a must for this setup, otherwise there will not be much flow through the back cylinders.

Correct. Just enough flow for warmup. After that, maximum flow through the radiator.

In fact, I have a front AND rear exit. I want to cap off the front still, but haven't done that yet. Both exits mix at the top radiator inlet.

None of that. Just limiting flow through the heater core return that goes to the water pump/mixing manifold.

Freeze plug that bitch already! :)

Nah, too much trouble. My turbo sees coolant from that spot. I will use a block off plate. Someone needs to make me one.

http://www.dny.se/poker-table3/img/poker-03.gif
+
http://www.cordlessdrills.us/cordless-drill-374.jpg
+
http://www.china-saws.com/productsim...gsaw_42200.jpg
+
http://ecx.images-amazon.com/images/...KL._SS500_.jpg

I know, just haven't found thick enough Al plate around here that isn't an acre in size.

6061 aluminum bar, 0.25" thick, 3" wide, 10-12" long, $2.88

6061 aluminum bar, 0.3125" thick, 3" wide, 10-12" long, $3.60

6061 aluminum bar, 0.375" thick, 3" wide, 10-12" long, $4.31

6061 aluminum plate, 0.25" thick, 8"x8", $14.11

6061 aluminum plate, 0.375" thick, 8"x8", $21.01

LOL you guys just are not going to let him off the hook, eh? :)

Can't he feed the turbo coolant off the side of the intake manifold? Or is that an inlet.

Yeah you can skip that connection on the front entirely and connect the turbo to the outlet of the coolant line for the throttle body. That is how I had it connected after I got rid of the front water neck.
Back of head->oil cooler->throttle body->turbo->mixing manifold

LOL. I found a local machinst, a guy I already knew but I didnt know he runs a full machine shop... he's making a plate for me. I will probably have him make soms stainless studs for my turbo to mf connection and go that route.

flyin miata has a kit now, not sure if that helps or not.

That is just the M-Tuned kit. I have a full custom reroute, just need to block off the front.

Baller front block off plate
 

Behold the most baller front block off plate:

I still say put a freeze plug in and feed the turbo from the throttle body. If you want to pretty that up a bit, just use a thermostat gasket instead of silicone. That'll avoid all the orange goop.


Oh, and holy overkill batman!

And Red Line Water Wetter SuperCoolant to go with it:

my phone is more baller than yours, but if you know how to use yours, then you win.

I believe your design in the original post is a good one but not the most efficient as it will split the thermal load of the engine, oil cooler and turbo and so to trigger each TS you must provide sufficeint temperature to both which would occur at best a few seconds apart however, more than likely the spacerless reoute TS will open first and because that has the larger fittings (and therefore flow) will be cooled and the coolant in the other section will not reach the critical temperature and so not flow round for quite some time as the heat within the system increases as a whole with a single loop open. Not being a very hands on Turbo person although the theory is fine means I am cannot be sure as to what this will do to your turbo.
You could swap your turbo and oilcooler couplings round which would mean the turbo gets cooling instantly once opened whereas the oil cooler will only come into function once at operating temperature properly.
I say again I do not have the experience but have more than enough theoretical science knowledge. I would believe my diagram below would be the optimal reroute for your situation given parallel looping and thermal load onto a single point.
http://img199.imageshack.us/img199/3851/rerouteu.jpg

I'm sooo tired of seeing reroute diagrams. I also don't like the idea of sending all the coolant through the heater core and not directly to the radiator. You guys can't stop coming up with variations of the same thing can you?

also, the coolant out of the oil cooler and turbo would not be magically cooled :P

There are many variations yes but there is only one that is most efficient. TBH I think someone should do some datalogging and use the different kits available on the same car to give genuine comparison. Then we would know the best reroute although autox cooling is different from drift cooling (what with the air not going straight in :D ).

Where is any magic cooling or suggestion of it?
Twin thermostat would create flow variation and differing thermal loads in sections of the original design posted.

It's a SPA962 from Linksys. If you have the right supply channel, it's only marginally more expensive than the less fancy models. Oh, and I had it tftp our company logo, so I win :)

- All flow through the heater core is no good. The lines are not nearly large enough and will restrict the flow through the radiator. On the track, probable death by overheating.
- I'll say it again... There is always flow through the oil cooler and the turbo. Always. The branches for those happen BEFORE the t-stat plates in both cases.
- I have a cooler thermostat in the circuit for the heater core. That means it will open first, before the larger main branch. The idea is to bias the flow through the heater core so that it will warm up the cabin for a DD. You know, so the driver is kept comfortable during these harsh South Texas winters :bowrofl:

Using a lower TS at the rear is a very good idea kinda wish I'd thought of it.

Here is my question though. Apparently you've said it twice now but the flow through the oil cooler and turbo which is always there? Isn't it stopped by the two TSs you are using until they are open because otherwise the TS are actually doing nothing. The TS opens allowing flow from white to blue and the joining hose from heater core to oil cooler TS will only flow once both TSs are open. I know this seems silly to you but I can't wrap my head around it.

As I see it you have three stages of cooling in the system with your design (shown in the picture below). Arrows indicate flow therefore no arrow means no flow. The line between heatercore will be presurised but wont flow anything until the TS opens so the turbo won't get flow as the turbo feeds the stat and so does the heater core (2 ins 0 out = no flow, 2 ins 1 out = flow)
http://img141.imageshack.us/img141/2263/coolthis.jpg

I don't know anything about reroutes other than what I have read here on MT.net. In an ideal world, there would be no heater core and all water would go through the head and back to the rad right?

If so, what is the option for a two postion valve on the infeed line to the heater core? Switch it to bypass on the track days so coolant still flows through the lines to keep the thermostat happy but doesn't see the heater core. Reposition it to the other position and it would be like a normal reroute but still flows through the heater core for a DD car. If you want to be really baller, you could rig it to be cable actuated in the cockpit. (Or forgetful and can switch it on track when you forget to do it in the paddock like I would.)

The rest of the reroute would be the "standard" reroute.

Or, tell the NOOB to shut up and stick with things that he has a clue about.:noob:

Sorry man but the flow through the oil cooler and turbo on your diagrams is all wrong.

http://www.threadbombing.com/data/me...acepalming.gif

they are beyond wrong. He learned fluid dynamics from the Hyper School of Fail. And furthermore, stop illustrating the lines out of the cooler and turbo as blue, they are not cold lines, they are even hotter than before.