stun_shift

[1991 4Runner V6 4WD 5-Speed]

Hey guys, long time lurker here and although I've solved a lot of prior problems just by browsing the forums, I have a question that I've yet to find an answer to.

I recently repaired a leak in the cooling system near the rear-top of the engine compartment when a hose developed a crack and sprayed coolant when prodded or put under pressure.

Having fixed *that* leak, I've since discovered another on the bottom of the radiator. This time, the leak only occurs once the engine has heated up and the cooling system has pressurized. As far as I can tell, the coolant appears to be leaking from the seal between the main radiator grill and the bottom manifold that feeds from the engine into the radiator.

I believe the radiator was replaced with an aftermarket product, as the bottom manifold has ports for both a manual and an automatic engine.

My question: is there anything else that could be causing the leak or am I doomed to re-replace the radiator?

Another site (specifics escape me) mentions that there might be an air bubble trapped in the system, but I wanted to get more information before proceeding.


This is my first post. I've read through all the FAQs, but if I've missed anything or left out necessary information, please let me know. Thanks!

rworegon

Time for a new radiator or, depending on the radiator, having it repaired by a radiator shop. Do not use any radiator stop seal products unless you want more problems later on.

A air bubble would not cause a leak like that.

The ports on the bottom of the radiator are used for automatic transmissions, not manuals. Also, coolant flows in the radiator at the top and back to the engine out the bottom on the 3.4L. I know nothing on the 3.0L at all.

millball

Not on the 3vze.

Hot coolant enters the bottom of the radiator, and cooled coolant exits the radiator top.

rworegon

Lol. Oops, my bad! I had 3.4l on the brain....that's what I get for reading too quickly. Can't say I know anything about the 3.0L.

stun_shift

Thanks for the feedback guys! I'll probably go ahead and replace the radiator as it's not a repairable model (plastic w/ aluminum core)

wyoming9

iTrader: (1)

You could have a new gasket put in unless the plastic is cracked but the cost would not be worth it at least to me.

I love my all metal radiators easy to fix if you know how to solder.

bswarm

I talked to a very good custom radiator builder recently. He said the aftermarket copper radiators are junk because they use thin copper and low lead solder. He suggested getting the plastic/aluminum type at $160 as they would last longer, or he could custom build a copper type at $300 that is better than aftermarket. My Modine radiator started falling apart after 2 years.

Marc

I think all conventional rads flow from the bottom of the rad, through the lower rad hose through the block and out the upper rad hose to the rad, otherwise they would potentially suck air from the top of the rad.

scope103

Before you replace the radiator, remove enough trim parts so you can actually see the leak. I thought I had a broken solder connection on the lower outlet tube, but it turned out to be leaking from the lower radiator hose-radiator connection. Replacing the factory spring-type clamp with a cheap standard worm-screw clamp fixed it for me.

RJR

Coolant flows into the radiator at the top, and out at the bottom, as the previous poster said.

From the FSM for the 3VZE:

RADIATOR
The radiator, mounted at the front of vehicle, consists of upper and lower tanks and a core connecting
the two tanks.
The core contains many tubes through which engine coolant flows from the upper tank to the lower
tank: Air passing over the radiator fins cools the heated engine coolant flowing through the radiator.
The upper tank has an inlet for engine coolant from the water jacket and it has a filler inlet. It also
has a hose attached through which excess engine coolant or steam can flow. The lower tank has
an outlet for the engine coolant and a drain cock. Automatic transmission models include an automatic
transmission fluid cooler.

millball

This is an example of an instance where the FSM is in error.

It is a certainty that the hot water enters the bottom of the radiator on 3vze engines.

scope103

A certainty? The drawing in the manual is very clear; if the manual is "wrong" this is not just a typo. http://personal.utulsa.edu/~nathan-b...06operatio.pdf (The '94 manual has the same drawing and similar description; for what that's worth.)

Short of pulling the upper radiator hose and running the engine, is there a way to figure this out?

millball

All you have to do is look at the position of the thermostat. When it opens, the coolant is permitted to flow out.

The 3vze thermostat is positioned at the bottom hose connection.

Conventional cooling systems that flow the hot coolant out the top have their thermostat positioned at the top hose connection.

I don't see how the 3vze thermostat could ever open if the hot flow is indeed from the top of the radiator.

RJR

I find it hard to believe that anyone would design an automotive cooling system where the coolant was pulled from the top tank of the radiator. That would only work if the radiator was always completely filled. If the coolant level was down even just to the top of the radiator cores, the water pump would be pumping huge air bubbles into the block, which as we know isn't good.

I'll take a look at my vehicle when I get home - I'm not yet convinced the manual is wrong.

millball

I'll put my IR temp gun to the top and bottom tanks later today and see what it says. I have to make a hundred mile trip later.

scope103

I wondered about that too. The thermostat would be exposed to hot coolant (from the "back" side, where the temperature-sensing slug is). But as soon as it opened, at least part of the thermostat would be exposed to cold coolant from the radiator. Would it snap shut? Or is the thermostat somehow positioned so the temperature sensing element is in "hot" coolant, and the cold coolant from the radiator somehow just flows through the "valve" part of the thermostat.

So I searched the internet. And the answer is: just about 50% are sure it's one way, 50% are sure it's the other way, and both groups think the others are idiots.

This site http://www.k20a.org/forum/showthread.php?t=78154 is about Honda/Acura K engines which also have the thermostat on the lower hose side. The Honda manual is also clear: coolant flows from top to bottom to thermostat to pump.

This site http://www.pbase.com/aw11mr2/image/109957581 has the cooling system of a MK1 disassembled; the thermostat is not right next to the pump, but it is also on the exit-from-radiator side.

Millball has the right idea; this is a job for actual experimentation. I hope an IR sensor works; I might try it myself.

RJR

I suspect this is the case. The thermostat simply acts as sort of leaky single-pole, double throw switch. When the engine is cold, coolant flows directly from the engine block through the thermostat housing to the water pump and back to the block. When the sensor opens the t-stat, water from the radiator is mixed into the coolant stream going to the water pump, but the direct path from the block is not completely closed off. If the t-stat housing is properly designed and the sensor is positioned correctly, it can still sense the temperature of the coolant coming from the block and keep the temperature regulated properly.

It's interesting that there are so many variations on such a basic system that's been in existence for over 100 years. You'd think by now that one "best" design would have completely taken over.

millball

Well, the IR thermometer readings don't lie. It is clear that I was wrong.

The hot coolant does indeed discharge into the radiator top. I measured about 170f
at the top tank and about 90f at the bottom, after engine warmup.

It must be that the thermostat heat pellet is protected from the incoming cooled coolant, as scope103 suggested.

I was wrong, wrong, wrong about this, but it was an honest mistake

RJR

Hey, thanks for making the measurement and sharing the results. Can't tell you how many times I've been proved wrong by the evidence. Part of this journey of discovery.

scope103

Millball beat me to it (hey, it was his idea!) I was curious enough to run the same test this afternoon. The truck is garaged in the winter, so I parked it on the driveway to get it to ambient temperature. As you can see, we are just getting hammered by Old Man Winter here in Northern California. ;^)

time Lower Upper time in min:sec, temps in °F, measured on hoses
0:00 67 69
1:00 60 64
2:00 63 73
3:00 65 81
4:00 68 85
5:00 77 94
--------------------- idle dropped to 850 at 5:00, so truck now "warmed up"
6:00 71 99
7:00 73 112
8:00 87 121
9:00 83 133

So I got the same result as Millball. And like Millball, I'm now convinced the FSM (on this point, anyway) is correct.

Note how the temperature at the lower radiator hose actually dropped at 6:00, possibly because the thermostat opened allowing cold water to drain out of the radiator into the hose. (But don't go nuts; it's hard to get an accurate temperature with an IR thermometer.)