I just got off the phone with the tech from Perma-Cool. The guy said, as we've suspected, that it will not work properly. He said instead of 5-10% passthrough you will likely see closer to 40-50% getting past. He said their unit works under the notion that there will be more resistance/restriction through the cooler in conjunction with the added resistance of the oil itself. In short, he says, it will not flow the way we are wanting it to.

ATM a simple T valve seems to be the cheepest/easiest solution and just manually switch it for Summer/Winter operations. Given it's not exactly the most elegant solution, but would be effective and very simple...

One could put a 160deg thermostat in an inline housing between the oil cooler and the upper radiator tee. That would force all flow into the mixing manifold until 160 is hit, at which point flow diverges ~50% to upper radiator hose inlet and ~50% to mixing manifold. Once the oil Tstat activates 100% goes into the upper radiator hose.

That provides automated hands-free operation, and shouldn't ever result in a complete flow blockage.

So I started looking for parts to do this, and I found this nifty Y-diverter thermostat:
AutoZone.com | Shopping | Parts | Product Detail - Thermostat

Autozone part number 25692. This would seem to provide the functionality needed but I'm unsure about hose sizes etc making it impractical to fit.

Obsidian's manual T valve has more and more merit as I think about it. Under normal usage you leave it set to the mixing manifold. For heavy duty usage you can switch it after warmup.

Apparently, Corky and Larry Bader had already done a diverter valve. (Like I know what's going on around here, eh?) They designed it for an oil cooler, but I fail to see why it would not work with the coolant. When I mentioned it Saturday to Corky and Larry, they both seemed receptive. Hustler saw the piece Saturday too. I will go take some photos of it.
Stephanie

The problem likely would come into play that even when the oil t-stat is routing most of the flow to the upper neck; when doing this it's not a 100% blockage (I don't believe... inside the t-stat is more like 5-10% when bypassed.) If that's the case then using an oil t-stat as a starting point becomes moot.

You could do basically what you suggest and eliminate parts IF you were able to locate a higher temp/normally open t-stat (say ~190f for argument's sake.) It would then allow 100% flow to the Lower hose untill 190f. Have a 2nd t-stat (lets say 160f for example) that is normally closed. Then between the temps of 160-190f the flow would be a 50/50 between upper & lower allowing for an intermediate cooling setting. Then above 190 the normally open t-stat on the lower radiator line would close sending 100% of the cooling to the upper hose. Thereby providing max cooling when needed most while maintaining a progression and avoiding a pressure spike when switching from one routing to another.

I'd thought of the above a while ago but the primary issue lies in finding a normally open t-stat. Thusfar I haven't had much luck in locating such a unit. If you find one by all means let me know. The main issue I see with most of these solutions lies in the complexity, size, and sheer points of failure.

I've always been of the philosophy that the fewer joints, adapters, & the smaller the better. Using a manual 3-way valve is by far the simplest solution, however I'm not certain I would want to switch it while the engine is on. It'd likely depend on how the valve is. Many ball type T-valves will close off all outlets from the inlet when diverting from one routing to another. Given this could be overcome by strategically drilling holes or grinding the ball's passages down to avoid a blockage which would lead to a pressure spike.

In the end for me, it's more about overall system simplicity and fitment than anything.

Here is the first version.
Stephanie

That is awesome! Looks just like something Mad Max would have on his AMX!

I wouldn't want to switch it with the engine on either. I don't think the pressure spike would be a problem ... since this wouldn't be switched until the engine's warm, the stock thermostat should be open and flowing. So the heater core would see a small spike, but the flow has somewhere else to go, so it'd be nothing like what you'd see with a full flow blockage with the water pump pushing a static pressure head.

Question: what's going to keep this locked into Position 1 or Position 2?

That's all fine and good, but why not just use something already off the shelf?
http://www.drillspot.com/pimages/2563/256366_300.jpg

That one has 1/2"NPT ports and you could simply use a 1.2"NPT-5/8" barb and problem solved.

Why does a dog lick himself? Because he can. They have the shop, the ability and the skills. One looks sexy/custom and the other is Home Depot.

Personally, for the money, home depot likely works just fine for me too.

Will brass react with glycol & other antifreeze ingredients?

hopefully.

No... If it was an issue BEGI and most others would have issues as many of the barb fittings they already use are brass. Even so the valve is available in stainless steel as well.

I mean no disrespect or any doubt of BEGI's fabrication capabilities. I just failed to see the point making something like that in house that appears as if it would require more than $18 in aluminum. Given I can totally see it is they were simply bored and was looking for something to do. I just don't see it making sense from a production standpoint was all. Plus you would have to ensure the top seals properly, fitment, and pressure rating. The brass valve has the lowest pressure rating of any of the materials it is available in and is rated to 400psi.

Given, if I had those sort of fabrication resources I'd work on an actual thermostatically controlled diverter. Instead of making a system that uses a ball to divert the flow could modify a thermostat such that when it would open (change in horizontal position) it would switch between two "Y" paths. Basically, instead of rotating something to change the flow pattern you would be actuated via height. Essentially having the two Y's sandwiched one upon the other just angled differently on each plane. This would allow for 100% diversion (or a certain % could go to the other hose with a few small holes drilled), and allow for a progressive taper from the flow from one hose to the other. Basically ramping from 100% from one hose, down to 50% to both, to 100% to the opposite hose. It should be more reliable than using 2 thermostats and a simpler overall design. That's if I had the fabrication and engineering resources they have, but that's just me. ;)

http://i6.photobucket.com/albums/y22...ta/y-route.jpg

'Cuz the photo above (home depot part) is just ugly. Granted, my engine bay is not the cleanest, but that ugly thing would never be allowed in. I am sure the photos above are not going to be a production piece. It was just meant to represent what some people have been asking about. Personally, I think a thermostat controlled unit is going way overboard. I subscribe to the Keep It Simple Stupid frame of mind. In an ideal, perfect world, yeah it would be great. But for something that needs to be changed once a year, what is the point?
Stephanie

That's a perfectly valid judgment. Personally, I think the fabricated one looks a bit bulky and pieced together. Given that it is just a quick and dirty prototype. On a personal level I think the (home depot version) is much cleaner and seemingly not as noticeable... especially considering exactly where it would be located (against the firewall down along the shelf.) I just don't see it being that noticeable to where the cost and effort become worthwhile. Again, this is simply a personal judgment and to each their own.

I too think a thermostatically controlled unit is going a bit overboard. However, it would be ideal as it would allow for fast warm up's while providing the extra cooling achieved from the reroute... best of both worlds. Unfortunately, in order to achieve that it would require either a multitude of fittings or some custom fabrication. Either way as Stephanie says... it's likely more trouble than it's worth for the sake of keeping things as simple as possible. ;)

Despite it not being the most popular I decided to opt for the manual valve. It arrived this morning and I don't believe it will be too obtrusive once it's installed. I just had to pickup some 1/2npt > 5/8 hose barbs and it's ready to go. I'm just waiting on the new pipe from BEGI before I install it.

I hate to bring back a somewhat ancient thread, but I'm in the same situation as Obsidian, planning to use the manual 3 way valve and just switching it once the car is warm with the engine off. If it works well enough for the track then I'll be overjoyed, if not the a conventional re-route will be in my future.

Did you ever successfully complete this 3 way valve setup? Has anyone???

I'm not going to give BEGi a cent of my $ or ever deal with their unacceptable delays and fuck-ups again, so I'm going to get 4 of those 1/2 npt to 5/8 hose barb fittings and drill and tap my own splice for the lower rad hose, maybe using that stock metal pipe that used to bolt to the chassis.

It looks like these are the parts that would be needed to complete this setup.
Hose Barb,Straight,1/2 MNPT,5/8 In Barb - Push On Hose Fittings - Distribution Equipment - Pneumatics & Hydraulics : Grainger Industrial Supply
Three Way Ball Valve,1/2 In NPT,Brass - Ball Valves - Valves - Plumbing : Grainger Industrial Supply

I'll try and get some pics up once I start this endeavor.