Inlet side thermostat install.

[Gary Allan]

On my never ending quest for complexity for no good darn reason, I acquired an inlet thermostat housing from Shannow. The price was right (shipping from Oz).

I like the idea of this. He was paraphrasing an engineering friend of his when he said that the outlet thermostat is the greatest way to warm your radiator to a desired temp.

I got lucky when I bought an aftermarket lower rad hose. You don't know how hard it is to cheaply change sizes on rad hoses. The OEM was tapered at the rad end and big from there to the suction side of the pump. The after market, luckily, just expanded at the pump and was perfect for inserting the thermostat housing by cutting out a bit of the middle. I wasn't going to spend the (near) $100 to make this work.



http://yfrog.com/6finletstatj (direct link)

Since the new housing takes the normal bypass flow, I capped off the normal return to the water pump.


http://yfrog.com/4oinletreturnstubbyj

I ran everything in line as it was before. The flow goes from the OEM thermostat housing (about 3 o'clock in lower image) to the heater core ..through the sandwich oil:coolant heat exchanger and ends up at the inlet thermostat housing. Currently, the OEM thermostat is still in place. It's the typical 195F. The inlet thermostat is a 160F. I didn't have the courage to start off with 180F or 170F while leaving the OEM in place. I don't always get the time to spend all day on R&D, so to speak, so I like to do things in a manner that allows them to function "as is" and fine tune later.

Stupid Is As Stupid Does

Well, I got the thermostat housing installed ..refilled the rad ..and started the engine. The thing is, I now had two closed thermostats at either end of the radiator. I didn't drill a weep hole in the lower thermostat ..and I surely had air in the engine. No bypass flow ..no thermal changes at the inlet stat ..even when the outlet stat opened and was begging for release.

I had thought of tee'ing the return initially and also wanted to be able to put this thing back to normal without too many hassles, so I just left the current hoses intact and capped off the tee like I did the normal return line. I initially broke this connection and filled the engine with a funnel on one end ..and just wait for it to seek its own level with the other. I ended up draining it anyway to put the weep hole in the lower stat.

[IMG=http://img231.imageshack.us/img231/2816/inletstat.jpg][/IMG]

[/URL]


Anyway, the 160F inlet stat is transparent at idle. My electric fan cycles normally. On the highway, if anything, this runs a tad cooler.

Shannow suggested going to a common supply and parallel returns instead of my inline setup. He feels that raw engine temp should control the inlet thermostat. I (up until he brought it up) merely looked at my heater core and oil cooler as one cooling circuit, which when taxed to a certain level, opened another cooling circuit.

Right now I've got a relatively long path to the return/inlet housing. Probably 15 feet. With using a common supply and parallel return, I'm uncertain how the flows will divide, but it would surely give the inlet thermostat unaltered temp to react to. Shannow recommended freeze plugs with holes drilled in them as a restrictor to bias the flow. I'm trying to pull a more elegant (that is, complex) way out of my behind for doing this.

Again, the setup works fine as is, but that may not be the case when I up the inlet temp to 170F and/or 180F.

I chose a NAPA Superstat for this initial run. I think I'll opt for a Robertshaw stat for the final temp once that's determined. It allegedly has more proportional flow than out run of the mill types.

Ultimately I would like to eliminate the OEM stat. It adds one more point for failure without assured quick diagnosis. OTOH, some have suggested that newer thermostats are designed to fail open. If that's truly the case (coincidentally, all of my stat failures for the past decade have been of the stuck open nature), then leaving the OEM stat in place can add some fail safe redundancy.

Yes. I never had to do this. I did it for the fun of it.

 

[440Magnum]

Originally Posted By: Gary Allan


Yes. I never had to do this. I did it for the fun of it.



Well heck, here I was all primed and ready for the story to end with "And now my 4.0 finally gets 29 mpg..."

One question... by leaving the outlet Tstat in place aren't you pretty much defeating the inlet T-stat altogether? I mean, sure, the inlet stat opens up when the return from the heater core hits 160...but coolant *still* doesn't flow through the radiator and thence through the inlet-side stat until the outlet stat hits 180 degrees, right? So you can still get the condition where there's a big temperature gradient across the engine (160 at the inlet, 180 at the outlet) which IIRC is part of what an inlet-side T-stat is supposed to help with. Isn't it supposed to work so that the inlet-side stat blends bypass water with radiator-cooled water to maintain a constant input temperature to the engine, with the flow rate accellerating rapidly with increasing temp to create a minimum gradient across the engine?

Or am I all wet (with coolant, probably)?

 

[Warstud]

Autozone sells the fail safe t-stat your looking for.

 

[Gary Allan]

Quote:
but coolant *still* doesn't flow through the radiator and thence through the inlet-side stat until the outlet stat hits 180 degrees, right?


Correct, but whatever fluid enters the engine should be very close to 160F. Without the inlet stat, it could be as low a 100F ..or heck, if it's 17F out ..probably much lower.


I get what you're saying, though. What this should do is make the outlet stat cycling much less frequent. It won't be gulping coolant that may be 50-150F colder and then waiting until the outlet stat decides to close.

The thermal shock should be neutralized ..not that it's any apparent issue for anyone outside if -35F without a winter front when the thermostat finally opens

Then again, at that temp, I imagine the heat core does a good enough job at cooling the entire engine.

So far, I haven't exceeded design setpoint at any unexpected time. I have my electric fan programed above 195F (starts around 200-205f at half speed and is at full speed at 210F on the gauge. I didn't want them coming on at any speed where natural air flow would cool the rad enough.

 

[Gary Allan]

Originally Posted By: Warstud
Autozone sells the fail safe t-stat your looking for.


Thanks. I just now put "fail safe thermostat" in google

Motorad Fail-Safe® thermostats as well as other Motorad products are available at Auto Value Parts Store, Autozone, Canadian Tire, Fisher Auto Parts Stores, O'Reilly, and Parts Source

 

[TurboLuver]

I've been using MotoRad Fail Safe T-stats for several years, very good quality.

 

[Shannow]

Gary,
from a control system's point of view, I'd dump the outlet thermostat ASAP.

I'm not sure what the response time of the inlet thermostat is, but I can see where if the outlet is markedly hotter than the inlet, the inlet could be wide open when the outlet finally pops, allowing a gutfull of cool water in until the inlet either controls it, or it reaches the outlet.

Your guess that the heater could be big enough in winter is spot on.

When I used that housing (on a 4.2 litre V-8), a zero C morning, the top tank would still be cold after a 15 mile drive.

Would love to see the behaviour of your idle air control valve before and after installation... on the Q-jet, I ran out of idle speed adjustment, and resorted to leaning out to get control back.

 

[Gary Allan]

My line of thought was that at whatever process variable exists within either thermostat, it's keeping the other from drifting too far out of setpoint. At least on the outlet end at this time.

I'll make the jump to 180F on the next evolution and tighten up the span between the two before dumping the outlet stat.

What's your opinion on this setup to retain the in line bypass flow configuration? I reason that it should limit any "time to temp" alterations due to the laminar cooler, yet still should maintain the 195F blending flow ..albeit at reduced (probably) volume.



I guess I could make it a three port like we discussed over this diagram.

 

[mechtech2]

What was the point of the installation?
What was the purpose of the endeavor?

It seems to me that since no auto maker does this, I would consider that fact.

 

[Gary Allan]

Well, no automaker uses an inlet and outlet thermostat ..at least that I know of. I think the LS engines use an inlet thermostat ..which I think is misnomer(ed) as "reverse cooling". I think the pathway is the same. Patman was telling me his 180F thermostat yields 215f+/- at the sensor.


The purpose? Entertainment in building different, if not better, mouse traps. Even if some other way isn't better, sometimes you just need to go there for a change from convention.

 

[440Magnum]

Originally Posted By: Gary Allan
Quote:
but coolant *still* doesn't flow through the radiator and thence through the inlet-side stat until the outlet stat hits 180 degrees, right?


Correct, but whatever fluid enters the engine should be very close to 160F. Without the inlet stat, it could be as low a 100F ..or heck, if it's 17F out ..probably much lower.


Gotcha... that's the other half of the system that I wasn't thinking of. Older very heavy duty engines (I'm thinking in particular of the General Motors 305 v6 and its V12 twin as well as the industrial Mopars of the 50s-70s) attacked that problem by using massively oversized waterpumps and bypass systems. Even with the thermostat *fully* open, only 50% of the water being fed back to the engine was allowed to be radiator water. IOW, at full-tilt cooling, the return water was a 50/50 mix of radiator-cooled and bypass water, and at anything less than max cooling it was more than 50% bypass water, creating a very uniform temperature profile through the engine.

Of course today we have engines with reverse-flow cooling that go even further toward getting the cylinder skirts hotter and the heads cooler.

I was only half joking about gas mileage since its such a common gripe with the 4.0 anyway. I wonder if you will see a little improvement. My guess is not enough to notice, but maybe a little bit.

 

[Gary Allan]

I think the 4.0 has a high bypass flow. That's what the guru at Evans said when he had to take it out of his Cherokee. It just wants to send the more viscous stuff in a circle instead of pushing it through the rad. They used to be down the street ..at least the tech support end of it and some warehousing.

My wife's 4.0 warms up within a very short distance. I'll challenge most aluminum engines for time to temp. The 2.5 doesn't quite warm as fast ..but still pretty good for a cast iron engine.

I was hoping for a .5-1mpg out of my electric fan. It removed several pounds of rotating mass, but I bought some Dunlop M&S in load range D ..and it sapped any gains and more. I was peaking over 21mpg on a longer run. Now I'm back to 17.5+/-.

My wife likes the more "manly" tread

 

[440Magnum]

Originally Posted By: Gary Allan
Well, no automaker uses an inlet and outlet thermostat ..at least that I know of. I think the LS engines use an inlet thermostat ..which I think is misnomer(ed) as "reverse cooling".




Hmmm... my understanding is that starting with the LT-1 and continuing with the LS series, the GM corporate pushrod v8s (can't call them "Chevies" anymore... thank God) really do use true reverse-flow cooling (return water into the heads, then down and out the block skirt) AND inlet-side T-stats. Maybe I'm wrong on that... I haven't ever actually messed around with an LT or LS, I just have a lot of co-workers that own them.

Most Chrysler v6 engines now use inlet side T-stats, but they're still the conventional flow (in through the block skirt, out through the heads). Not 100% sure about the Hemi and 4.7, but I think they're inlet-side and conventional flow direction too.

 

[440Magnum]

Originally Posted By: Gary Allan
I think the 4.0 has a high bypass flow. That's what the guru at Evans said when he had to take it out of his Cherokee. It just wants to send the more viscous stuff in a circle instead of pushing it through the rad. They used to be down the street ..at least the tech support end of it and some warehousing.


Interesting- that actually would not have been my guess, just looking at the heater hose sizes and the fact that it runs the bypass through the fairly restrictive heater core itself. If you look at the bypass on, say, a 5.2, that's a 2" diameter dedicated hose that's only about 4" long (pre-1992, even shorter post-1992 'Magnum' version)- LOTS of flow potential there.

Originally Posted By: Gary Allan

My wife's 4.0 warms up within a very short distance. I'll challenge most aluminum engines for time to temp.


Mine too, and so did my wife's iron-block 3.5 that we had before she got the PT. My 4.0 has the factory 195 T-stat and runs pretty much nailded on 210 degrees indicated (probably 200-205 allowing for factory gauge inaccuracy). That took me a while to get my head around, but it works great and never overheats- even with the summer-from-[censored] we've just been through. It does ping just a tad if you lug it (5-speed) but not enough to worry about.

 

[onion]

For what it's worth- Detroit Diesel's new heavy truck engine, the DD15*, has a thermostat only on the coolant inlet.






*the DD15 is actually a re-badged overhead cam Mercedes engine. But don't tell anybody.

 

[Shannow]

Originally Posted By: mechtech2
What was the point of the installation?
What was the purpose of the endeavor?

It seems to me that since no auto maker does this, I would consider that fact.


That's where you are wrong.

Lots of cars control inlet temperature, oft in addition to outlet temperature.

My old J-car did, 4Runner V-55 did, and me 1984 E30.

 

[Shannow]

Originally Posted By: Gary Allan
What's your opinion on this setup to retain the in line bypass flow configuration? I reason that it should limit any "time to temp" alterations due to the laminar cooler, yet still should maintain the 195F blending flow ..albeit at reduced (probably) volume.


I still think that having coolers in series will lead to a less controlled coolant temp dud to the variables.

That's a purist thought 'though

 

[Gary Allan]

Quote:
Interesting- that actually would not have been my guess, just looking at the heater hose sizes and the fact that it runs the bypass through the fairly restrictive heater core itself. If you look at the bypass on, say, a 5.2, that's a 2" diameter dedicated hose that's only about 4" long (pre-1992, even shorter post-1992 'Magnum' version)- LOTS of flow potential there.


It's just speculation with some semi-authoritative speculation thrown in. I have to figure that there's a reason why a cast engine can possibly warm so fast. I'm sure that they did something to help cope with the emissions issues in its latter years. I've never had any engine warm that fast.

This whole little brain tickling session has given me cause to figure how adding a sandwich laminar cooler reduces both temps. I've been trying to explain the reason. I can now speculate that it just forces more coolant through the "also ran" rad once the thermostat opens.

The wife's 4.0 ran about 205F stock. Went to 210F when the winch went on. Went back down with one of those expensive hollowed out thermostat housings from some outfit. I threw in a bigger rad (I forget if it was 2 or 3 core) and now her temp cycles +/- 5f when she's on the highway. The hose is too short to fit this housing in there and they don't make a Robertshaw stat for the later model jeeps.

 

[Gary Allan]

Originally Posted By: 440Magnum
Originally Posted By: Gary Allan
Well, no automaker uses an inlet and outlet thermostat ..at least that I know of. I think the LS engines use an inlet thermostat ..which I think is misnomer(ed) as "reverse cooling".




Hmmm... my understanding is that starting with the LT-1 and continuing with the LS series, the GM corporate pushrod v8s (can't call them "Chevies" anymore... thank God) really do use true reverse-flow cooling (return water into the heads, then down and out the block skirt) AND inlet-side T-stats. Maybe I'm wrong on that... I haven't ever actually messed around with an LT or LS, I just have a lot of co-workers that own them.



I don't know either way, personally, but a respected source told me that it's not reverse cooling ..just backwards temp control. That's from memory ..so he could have just been talking about one series of engine. I surely wouldn't argue the point since I never researched it.

I just did some reading. The LT engines had true reverse cooling. The LS engines ...who some allege to be reverse cooled ..are not. Apparently GM abandoned the reverse cooling not only due to some issues, but for patent problems with Evans Cooling.

 

[Gary Allan]

Originally Posted By: Shannow
Originally Posted By: Gary Allan
What's your opinion on this setup to retain the in line bypass flow configuration? I reason that it should limit any "time to temp" alterations due to the laminar cooler, yet still should maintain the 195F blending flow ..albeit at reduced (probably) volume.


I still think that having coolers in series will lead to a less controlled coolant temp dud to the variables.

That's a purist thought 'though


Remember ..you had to refine yourself to become a purist. I'm still (I don't know if I want to say this) "in the rough" so to speak.