Radiator Flow Rate Test

[Bambusiero]

So, I've been having overheating problems. It's a familiar story if you read here much, and other forums, too. I could recite the same litany of "did you check this, did you check that" from Isuzu diesel conversions (Wheelingnoob - sheesh! It was like reading "Lord of The Rings") to Subarus to other regular 80 series trucks.

I won't go through all the gory details, but it's load / speed dependent, not airflow dependent, and EVERYTHING in the cooling system has been replaced / serviced / checked.
Going faster runs hotter! It's almost linear, and pretty darn repeatable.

One last nagging doubt before I have the head gasket replaced :
Are you SURE that brand new CSF 2517 old school all metal radiator is OK? Maybe it's just a piece of crap. It's made in Thailand or somewhere. Maybe the tubes are all full of solder or flux or elephant dung or something. It's only brazed copper after all, and nobody uses THAT anymore, you know!

OK, let's just see about that!

Radiator flow rate test setup:

Stopwatch.
1 willing helper.
5 gallon plastic bucket.
1 1/4" PVC pipe.
1 1/4" PVC elbow.
1 1/4" PVC male thread adapter.
1 1/4" to 1 1/2" PVC thread bushing - as a retainer nut.
1 1/4" slip plug.
Heater pipe insulator foam - gaskets.
1 1/2" radiator hose - InFlow and OutFlow adapters.

Head pressure : Bottom of bucket is roughly 14" above radiator inlet.
Test volume: A bit less than 5 gallons.

Step by step:
1) Plug Outflow pipe.
2) Fill plastic bucket with water to rim.
3) Get stopwatch ready, watching inside the bucket.
3) Have helper pull plug (start timer).
5) Water level down to top edge of pipe (stop timer).

Measurements:
Old OEM radiator, 2 row aluminum core, plastic tanks, no visible sign of build-up or blockage.
10.8 +/- 0.1 Seconds.
New CSF 2517 radiator, 3 row, all copper/brass.
6.7 +/- 0.1 Seconds.

Yeah - you heard that right! Seconds!
You have to be fast or you'll miss it. Like flushing a toilet - no, faster!

Well, you can argue about high or slow flow rates having better cooling capacity, but really, that's for heat exchanger engineers. They'll both work just fine, thank you.

Oh, one last thing: FOR SALE - CHEAP - Calibrated Radiator Flow Test Set. (Phoenix area preferred)
Anyone want to add their data?

Here's the test setup assembled and ready to go with the dismounted old radiator.

I love the juxtaposition of "DANGER" and "KEEP OUT OF REACH OF CHILDREN" and that lovely happy child. Nice.

 

[Bambusiero]

And on the truck:

 

[richardlillard1]

 

[inkpot]

I always thought it wAs more heat transfer related than flow rates. I recumbent it IS related to some degree?

 

[Output Shaft]

All that's been proven is that one radiator flows faster. That's not very useful information. The important metric is cooling efficiency.

Hot water wizzing around the engine at high velocity doesn't cool very well.

 

[e9999]

True, you can't compare 2 radiators of different design just based on water flow rate. Both inner and outer tube and fin configurations matter a lot too.
However, it might be a useful test to see if a given radiator is starting to get plugged up over time, assuming you could reset the setup exactly the same way after some time.

 

[Bambusiero]

I know, I know...It doesn't really PROVE anything about cooling capacity.

What I think I did prove, to myself, is that there is nothing wrong (as in defective, plugged) with my new radiator.
If it has hot water flowing through tubes, and air blowing around those tubes, well, it's a working radiator. The rest is fine print.

Also, I wanted to put some quantified numbers down so that anyone else who is fighting a cooling issue will have something to compare to.
This was really a pretty easy DIY project, and cheap. The most expensive part was the radiator hose, long enough and with 2 right angle bends. About $20 I think. The most time spent is draining coolant.

What I know about heat transfer is that there are more variables than you can shake a stick at, and that there are textbooks full of mind boggling equations for it, written by Thermodynamics professors for PhD students to read, and at the end of the day, the all the math and theories only get in the ballpark, and then real fun begins with fully instrumented test rigs. All of that is way above my pay grade.

I do think it's fully possible to have 2 heat exchangers, one with a high flow rate and low delta-T, and another with a low flow rate and high delta-T, with equal cooling capacity, as in Watts/second transfer rate. Not necessarily so, but certainly possible. After that, no armchair theorizing will do.

Actually, I would think it's pretty much adjusted for by the thermostat anyway. It self regulates the flow-temperature relationship, assuming there is some margin to work with.

That's a key point. Margin. Apparently, my truck is has no cooling capacity margin. It's right at the edge, and pushing beyond the cooling capacity of the system, even though the system is designed with plenty of margin for any normal duty. Even Phoenix in the summer, loaded down, uphill, with A/C running, is normal duty. The system should handle it. Nope - it's the load. Way above normal, somehow.

 

[Hornd]

Sound Logic! Your making too much heat and can not dump it fast enough.
Have you done this yet:

I know, I know...It doesn't really PROVE anything about cooling capacity.

What I think I did prove, to myself, is that there is nothing wrong (as in defective, plugged) with my new radiator.
If it has hot water flowing through tubes, and air blowing around those tubes, well, it's a working radiator. The rest is fine print.

Also, I wanted to put some quantified numbers down so that anyone else who is fighting a cooling issue will have something to compare to.
This was really a pretty easy DIY project, and cheap. The most expensive part was the radiator hose, long enough and with 2 right angle bends. About $20 I think. The most time spent is draining coolant.

What I know about heat transfer is that there are more variables than you can shake a stick at, and that there are textbooks full of mind boggling equations for it, written by Thermodynamics professors for PhD students to read, and at the end of the day, the all the math and theories only get in the ballpark, and then real fun begins with fully instrumented test rigs. All of that is way above my pay grade.

I do think it's fully possible to have 2 heat exchangers, one with a high flow rate and low delta-T, and another with a low flow rate and high delta-T, with equal cooling capacity, as in Watts/second transfer rate. Not necessarily so, but certainly possible. After that, no armchair theorizing will do.

Actually, I would think it's pretty much adjusted for by the thermostat anyway. It self regulates the flow-temperature relationship, assuming there is some margin to work with.

That's a key point. Margin. Apparently, my truck is has no cooling capacity margin. It's right at the edge, and pushing beyond the cooling capacity of the system, even though the system is designed with plenty of margin for any normal duty. Even Phoenix in the summer, loaded down, uphill, with A/C running, is normal duty. The system should handle it. Nope - it's the load. Way above normal, somehow.

 

[e9999]

what determines the temp of your engine is the interplay between load and effective cooling. Also the ambient temp once your thermostat is fully open. Of course, it will get hotter if the load is higher, the ambient temp is higher, or the cooling is lower. There is only so much load though, like going fast on an uphill, so I would think the cooling is usually the key.

if your system is struggling as in not cooling enough, then the thermostat will play no role as it will (should) open all the way once your engine is warm. Might as well remove it then, that will help a bit on the high temp side.

 

[Bambusiero]

Hornd asks if I did the combustion gas detection test.

Why, yes I did, thank you for asking.

Bought a new kit (NAPA brand, exactly like your picture), drove around to get the system up to operating temperature, left the truck idling, (carefully) removed the radiator cap, sucked an inch or so of coolant out, docked the filled test tube, applied mild vacuum with the supplied rubber bulb.

No color change at all, and the rubber bulb held vacuum too, after a few sucks.

That there is a head scratcher, for sure.

I wanted to repeat the test at speed, but everyone in my family refuses to cooperate. I thought it would be fun...

I am convinced at this point that there are several distinct types, and degrees, of head gasket leaks, each with different symptoms.

Mine does not leak at idle / low load, but it does at combustion pressure above some threshold, and leaks correspondingly more gas volume (and heat) at higher RPM. I should call it LPM, Leaks Per Minute. That's the theory, anyway. It almost acts like a check valve. I don't usually loose coolant, but have, a little, a few times. It's not consistent.

If coolant temperature is high from driving at highway speed, I can pull over, idle for a few minutes, and watch coolant temperature drop down to 186 F or so. Have idled indefinitely with A/C running full blast, in the shade, 100 F day, 195 F coolant temperature.

Outside temperature definitely does have an influence, as do loads like A/C, driving up mild grades, etc. Everything adds directly in, because there is no margin, and the system is operating on a fairly steeply sloping portion of it's characteristic curve. If the system was operating within it's designed range, the slope of load vs temperature is pretty flat because the thermostat acts as local negative feedback. But, the thermostat correction range is used up, all the way open, no more influence. (Yes, thermostat checked on the kitchen stove. Perfect.) In other words, what e9999 said.

 

[Bambusiero]

richardlillard1, I keep watching that and laughing.

 

[bahamasair]

In my experience for a bad head gasket to overheat the engine you have to be loosing coolant, mixing oil and coolant or pressurizing the coolant system which will also cause loss of coolant or explode the radiator. Are you loosing coolant? Have you checked your timing? it seems any cooling related problems on these rigs get blamed on the head gasket pretty quickly without any real proof.

 

[baggar11]

If coolant temperature is high from driving at highway speed, I can pull over, idle for a few minutes, and watch coolant temperature drop down to 186 F or so. Have idled indefinitely with A/C running full blast, in the shade, 100 F day, 195 F coolant temperature.

This seems like classic weak fan clutch. If it's been addressed already, it has to be the radiator.

 

[Rivman1243]

Could it be plugged cats or bad O2 sensor causing it to run lean?

 

[Tools R Us]

In my experience for a bad head gasket to overheat the engine you have to be loosing coolant, ...

This and the leak often prevents the cooling system from drawing back from the recovery tank, so the radiator will be low, and the coolant tank full. The other indicator is a miss, most often #6, setting a P0306 code, the can be other reasons, but combined with above it's a pretty sure thing. Not having these symptoms and a clean combustion gas test, I bet it's not the head gasket.

The brass radiators run warmer, don't shed heat as well as the aluminum ones. Did changing the radiator make any difference? There is a difference, but doesn't sound like your whole problem, a new radiator should work?

The '80 is fan dependent, yes even at speed, especially when loaded and hot. What clutch is on it, has it ever been serviced?

 

[Bambusiero]

This seems like classic weak fan clutch. If it's been addressed already, it has to be the radiator.

The '80 is fan dependent, yes even at speed, especially when loaded and hot. What clutch is on it, has it ever been serviced?

My fan clutch is the Eaton unit, from reading what you and others have posted in the past.
Since I was replacing the radiator anyway, it was a good time for a cooling system overhaul.
I followed your service & fluid upgrade recommendations as closely as I could.

1) Crack front & back halves apart.
2) 200 F overnight reservoir drain, ports held open & facing down.
3) Wipe, wipe, scrape & wipe some more. All those little grooves full of sticky silicone fluid.
4) Checked the thermostat operation with a hot air gun (photos below).
5) Refill reservoir with custom 15000 Cst silicone fluid.
42%, 1 tube, 18 ml, Toyota p/n 08816-10001, 10000 Cst silicone oil (amber)
58%, 25 ml, Factory Team brand 20,000 Cst Silicone Diff Fluid (clear), from a local RC car hobby shop.
(That is some GOOEY SNOT, took quite a while to spatula that stuff into the small reservoir holes.)
6) Re-assemble. Check for leaks.
7) Operational checks. Seems good. Spins no more that 1 blades worth when hot.

Just for boy science project fun, checked thermostat operation with an adjustable temperature hot air gun.
It was hard to be precise because the hot air stream was not enclosed, so room air gets en-trained, depending on your aim, etc.
Also, it responds to temperature changes quite slowly, so it lags behind your aim adjustments.

Ports just beginning to open at 163 F.

Ports beginning to fully open at 181 F.

Ports more fully open at 201 F.

 

[Bambusiero]

... The other indicator is a miss, most often #6, setting a P0306 code, the can be other reasons, but combined with above it's a pretty sure thing...

I have not gotten P0306, but I am now getting consistent P0330 Knock sensor 2, bank 2 -circuit malfunction.
This has gradually changed from none during springtime when weather was cool, old radiator was in, and the overheating issue was not yet apparent, through almost every trip now.

It was only every now and then right after the cooling system rebuild, when the weather was first getting hot, A/C always running, and we really ran into the overheating issue on a mid-day trip out of town.

So, I don't know if this is just a distracting side issue or a confirming indicator. I think I might just buy 2 new aftermarket units to confirm. I understand they can be had for $45 apiece?

 

[Bambusiero]

This whole adventure started innocently enough, earlier in the spring.
Weather was cool, not using A/C, not paying any attention to coolant temperature, ScanGaugeII not plugged in.
No sign of trouble.
Fat, dumb, and happy.

I was changing out worn alternator belts and accidentally nudged the radiator bypass nipple with my elbow, and broke it right off.

Well, it is an old radiator, and the plastic seemed a bit degraded from age, for sure.
The plastic tank surface has stretch marks, like your Mom's belly.
I've had the truck 12 years. Same OEM radiator, no troubles.
It might even be original, I have no idea.

But, I needed to use the truck, it's my daily driver, so I patched it up, and was back on the road that same evening.
Found some un-named fuel fitting with a righteous looking metal nipple at the auto parts store.
Found a stainless fender washer in my junk.
Soldered it into a proper fitting using Harris "Stay Brite #8" tin/silver solder and "stay-clean" liquid flux.
FANTASTIC stuff BTW. HIGHLY RECOMMENDED for low temp repair and fitting fabrication.
Check out the solder fillet below. Perfect. Soldered with a preform of solder wire and a hot air gun. (good only for small stuff, but so nice)
Glued in with standard JB Weld, thickened with some resin filler powder for no sag, cured with 150 F hot air gun for 1 hour.
Even though the possibility of radiator replacement has now presented itself, I thought this repair could actually last, but alas, it was not to be.

One week later, there's a hairline crack (see black marker line) on the back side of the upper tank, spraying coolant.
Great! I seems like the radiator has suddenly reached it's end of life.
Normal easy duty, and it's trying to explode!
OK, shopping for a radiator, but I still need my DD on the road.
Can this be repaired? I've never tried such a goofy thing before. Sooo trailer park!
2nd patch with JB Weld and a single strip of fiberglass cloth ribbon.
Tank surface carefully sanded and alcohol cleaned.
Surface buttered with JB Weld.
Fiberglass patted & prodded into the wet epoxy with gloved fingers until it's soaking through.
Outer surface buttered with epoxy and fussed with until reasonably smooth.
Rough areas tend to smooth out with heat & some small bubbles pop.
Again, cured with 150 F hot air gun - here for a while, there for while, 2-3 hours while I got some dinner. On the road next day.

BTW, the idea has snuck in at this point that this is because of pressure... Naw, it's just OLD.

2 days later. Hairline crack on the frontside now. NOT FAIR!
It's literally exploding! What's causing this?
New radiator is already ordered, must wait.
3rd patch, DD needed, taking no prisoners this time, same MO. I'm a JB Weldin fool.
Crack location in black marker line.

By the time the new radiator arrives, the 2nd patch (1st tank patch) is showing signs of de-lamination and seepage along its bottom edge, along the crimp. You can barely see a white stain in the 2nd photo.
It looks as if the plastic tank is more elastic than the fiberglass re-enforced JB Weld, and is separating due to shear stress when hot.
More indication of an over pressure condition? Or just the natural consequence of material mismatch? You be the judge.

 

[NLXTACY]

So, I've been having overheating problems. It's a familiar story if you read here much, and other forums, too. I could recite the same litany of "did you check this, did you check that" from Isuzu diesel conversions (Wheelingnoob - sheesh! It was like reading "Lord of The Rings") to Subarus to other regular 80 series trucks.

I won't go through all the gory details, but it's load / speed dependent, not airflow dependent, and EVERYTHING in the cooling system has been replaced / serviced / checked.
Going faster runs hotter! It's almost linear, and pretty darn repeatable.

One last nagging doubt before I have the head gasket replaced :
Are you SURE that brand new CSF 2517 old school all metal radiator is OK? Maybe it's just a piece of crap. It's made in Thailand or somewhere. Maybe the tubes are all full of solder or flux or elephant dung or something. It's only brazed copper after all, and nobody uses THAT anymore, you know!

OK, let's just see about that!

Radiator flow rate test setup:

Stopwatch.
1 willing helper.
5 gallon plastic bucket.
1 1/4" PVC pipe.
1 1/4" PVC elbow.
1 1/4" PVC male thread adapter.
1 1/4" to 1 1/2" PVC thread bushing - as a retainer nut.
1 1/4" slip plug.
Heater pipe insulator foam - gaskets.
1 1/2" radiator hose - InFlow and OutFlow adapters.

Head pressure : Bottom of bucket is roughly 14" above radiator inlet.
Test volume: A bit less than 5 gallons.

Step by step:
1) Plug Outflow pipe.
2) Fill plastic bucket with water to rim.
3) Get stopwatch ready, watching inside the bucket.
3) Have helper pull plug (start timer).
5) Water level down to top edge of pipe (stop timer).

Measurements:
Old OEM radiator, 2 row aluminum core, plastic tanks, no visible sign of build-up or blockage.
10.8 +/- 0.1 Seconds.
New CSF 2517 radiator, 3 row, all copper/brass.
6.7 +/- 0.1 Seconds.

Yeah - you heard that right! Seconds!
You have to be fast or you'll miss it. Like flushing a toilet - no, faster!

Well, you can argue about high or slow flow rates having better cooling capacity, but really, that's for heat exchanger engineers. They'll both work just fine, thank you.

Oh, one last thing: FOR SALE - CHEAP - Calibrated Radiator Flow Test Set. (Phoenix area preferred)
Anyone want to add their data?

Here's the test setup assembled and ready to go with the dismounted old radiator.

View attachment 1286287

View attachment 1286288

View attachment 1286289

View attachment 1286290

I love the juxtaposition of "DANGER" and "KEEP OUT OF REACH OF CHILDREN" and that lovely happy child. Nice.
View attachment 1286291

I think I <3 you

 

[Bambusiero]

In my experience for a bad head gasket to overheat the engine you have to be loosing coolant, mixing oil and coolant or pressurizing the coolant system which will also cause loss of coolant or explode the radiator. Are you loosing coolant? Have you checked your timing? it seems any cooling related problems on these rigs get blamed on the head gasket pretty quickly without any real proof.

As for the exploding radiator, maybe so, maybe so. See above.

And for the next round - Coolant System Pressure Test
Started a new thread so it can be searched and found easier. It's a related but separate subject.

Have not checked timing. Not sure how, although a read a somewhat intimidating thread about that a while back. Do I really need to?
Long ago and far away, had a timing strobe light, but it's gone now. Does that work nowadays?

I'll be trying to make a correlation between operating coolant temperature, pressure, and radiator / overflow tank status.
Stand by.

Also, I think I need to repeat that combustion gas test.