PDA

View Full Version : 2 ROW vs 3 ROW Radiators - what's the REAL deal?


NeXJ
July 19th, 2010, 18:09
I have a 9 or 10 year old GDI 3-row radiator in my 92 XJ that's been sitting a lot... the autozone replacement water pump is blown and the rad hoses are leaking a bit... I'm thinking there must be some blockage especially since there was some significant rust inside the water jacket of the 97 HO that I installed in the thing when I got the engine. So anyway- after talking with a radiator shop - the guy told me I should hang onto the GDI and they would charge me $60 to rod it out and clean it up. At any rate, I started thinking about it after coming across a thread that mentioned the 3-row rads tend to use a 1/4" tubing and there are some pretty good 2-rows out there using 5/8" tubing.

So here's the question... since the 3 row is 1/4" (1/4-1/3 the throat area and 50 percent longer) wouldn't it present WAY more resistance on the water pump and also make the engine work a lot harder....? Imagine having to squirt water at high velocity through 10 feet of 5/8" tube vs 15 feet of 1/4" tube... I would think the later would require way more static pressure... thoughts??

winterbeater
July 19th, 2010, 18:40
The tubes are not all in one line, are they? One tube doesn't snake back and forth across the radiator. All the radiators I have seen run each tube directly from one tank to the other. So each tube is about 2 feet long. And you have more of the small tubes than the big ones. Where does the 10 ft vs 15 ft come in?

NeXJ
July 19th, 2010, 18:49
nope... well mine seems to snake... maybe each row snakes all the way in each layer (would seem) - the visible first row definitely snakes though.... 15 feet was an estimate - based on imagining the two row being 10 feet... well it doesn't matter... it's purely academic... IF that's the setup - then the 3 row would be 50 percent longer.. make sense?

cruiser54
July 19th, 2010, 19:44
You're really overthinking this................

NeXJ
July 19th, 2010, 19:59
Ha! well- I'm not positive I AM overthinking it - which is why I posted... perhaps it's something nobody's really researched before... though i guess I'd like to hear from a hydraulic engineer... I actually have a friend in europe who IS a hydraulic engineer for ESA (european space agency) - maybe I'll ask him... but hey - maybe it's a totally moot point.

I was just kind of riffing on a post I saw earlier where someone was saying that the modine 2 row radiators were superior to most 3 rows for that exact reason (though from the perspective of clogging).

joe_peters
July 19th, 2010, 20:58
This is a chicken and egg kind of thing.

Does the 3-core cool better than the 2-core?

Airflow--the 2-core permits higher airflow than the denser 3-core, however, the 3-core has more surface area for the airflow to exchange heat with.

Tube size--the larger tubes of the 2-core will not plug up as soon as the smaller tubes in the 3-core. Except, if the system is properly maintained, we are talking YEARS of reliable service with either one. I use the 15-minute flush AND the 3~6 hour flush in alternating years.

And you ARE way over thinking this!

bchulett
July 19th, 2010, 21:48
Here's the real deal...

http://www.naxja.org/forum/showpost.php?p=387695&postcount=7

.

carves
July 19th, 2010, 22:08
And you ARE way over thinking this!

In fact he's probably more OCD about it than me ... :laugh:

I have a 2core .... copper/brass with 5/8 tubes .... with about an, unsubstantiated 10-15 percent better cooling efficiency than the factory 2core radiator - manufacturers comment.

- Less than 30/70 anti-freeze mix,
- A 195'f thermostat that is fully open 15' sooner than the chrysler one i.e. 205' compared to 220' (approx),
- New stock equivalent waterpump & fanclutch - USMW stuff marked heavy duty ?
- Completely separate AW4 tranny cooler.

And 195' temps ...... whether at 110'f+ ..... or 30'f ambient temps.


Prior to that I had factory radiator ... 2core aluminium - maxcool / modine whatever ??
- Less than 30/70 anti-freeze mix,
- 185' non factory thermostat,
- factory AW4 oil cooling,
- always good condition, waterpump & fanclutch.

and temps of 190' - 230 depending on weather and engine load.


Biggest change is the radiator - which sorta highlights that not all 2cores are the same.


If your 3core is still serviceable ... and the servicing/rodding cost is realistic ... keep it. Airflow more than water flow will probably be more of a concern - if the fins are tightly spaced.

Its possible your tubes snake through the radiator ... bit like the cheaper oil cooler types .... but unlikely. If the repairer says he can rod it ... then the tubes most likely only go straight from tank to tank.

NeXJ
July 19th, 2010, 22:24
Well HEY... I'm just trying to take an alternative point of view... and suggest that maybe the three row rads are kind of victims of a 'specifications game' and that JUST BECAUSE it has three rows does not mean that it's necessarily BETTER... I'm trying to say that maybe there are other, more relevant issues at stake than 'how many rows' - and certainly a few have been touched on here...

but it's also occurred to me that nobody discusses (or has even thought about, it seems), the load required to push a column of water like that through the cooling system... that's all...

Yeah- the $60 is the 'rodding cost' but it will cost another $90 for removal and replacement... so that's $150 total - which is why I even considered the possibility of swapping it for something else...

NeXJ
July 19th, 2010, 22:25
BCHULETT - yes - I think that was the original post that made me consider this line of inquiry.

joe_peters
July 19th, 2010, 22:30
You are going to charge yourself for removing and replacing your own radiator?

JUST KIDDING!

No one is saying you don't have a point, but unless you have a flow test setup to make accurate measurements, well, its just too esoteric for most.

iwannadie
July 19th, 2010, 23:01
I think surface area is the key factor, with the 3 row you are increasing the surface area. It's not like the water is being moved very fast through the system either. You're not trying to get water from point A to B as fast as possible. You are trying to cool the fluid while keeping it moving.

Obviously the more cores the better logic won't work because the tubes would have to get smaller and smaller to maintain the same size radiator. Blockage becomes a problem with the smaller tubes even on a well maintained system a pin hole size tube would clog up. Having a single tube would give you more flow but drastically less surface area.

You will never get a solid answer either way as people will cheerlead for which ever one they have(Notice I have a 3 core) so there's no end all radiator choice. I think the type of driving, stock/after market associated parts, how well maintained your system is and cost all will help decide which is best for you.

bchulett
July 19th, 2010, 23:13
BCHULETT - yes - I think that was the original post that made me consider this line of inquiry.

I think I may have posted more about this issue back in 2004-05.. you're really testing my memory. I researched this extensively and found the 3 core aftermarkets were mostly hype.

If I recall correctly, there isn't enough room for a good robust 3 core. The 2 core aftermarket Modine (factory HD radiator) was only available through NAPA.. I compared both NAPA and OEM and they are the same. I went with the factory OEM unit for a few bucks more - primarily because I get discounts at the Mopar parts counter.

Personally I would avoid a recore .. radiator shops aren't what they used to be back in the old days. Essentially aluminum and plastic tank radiators are a throw away items IMO.

.

carves
July 19th, 2010, 23:50
Well HEY... I'm just trying to take an alternative point of view... and suggest that maybe the three row rads are kind of victims of a 'specifications game' and that JUST BECAUSE it has three rows does not mean that it's necessarily BETTER... I'm trying to say that maybe there are other, more relevant issues at stake than 'how many rows' - and certainly a few have been touched on here...

but it's also occurred to me that nobody discusses (or has even thought about, it seems), the load required to push a column of water like that through the cooling system... that's all...

Yeah- the $60 is the 'rodding cost' but it will cost another $90 for removal and replacement... so that's $150 total - which is why I even considered the possibility of swapping it for something else...


Its all good ... cant get answers without questions .... ;)

As joe_peters said ... Only flow tests would substantiate hi-flow pump benefits.


Too much flow can actually cause overheating ... coolant doesnt stay in the radiator long enough to cool ..... and can flow too fast through the engine to absorb enough heat.

Additionally ... With all the variations in radiator manufacture ... flow tests would need to be available from the pump maker - for a specific radiator in a specific vehicle ... along with thermostat operating specs, coolant mix ratios and ambient temps ..... to ensure the customer got the promised results.

Since all that info about something not nearly as important - as a new pair of foglights, is a bit ho hum, for most drivers ..... fast flow pump advertising generally just says it will fix cooling problems ... same as hood vents ..... :rolleyes:

And they will .... in the right circumstances .... ;)


$160 .... mehhh :)

Easy $400 for a decent aftermarket radiator down here ... and I still have nightmares about the dealership price I paid for a OEM one about 6 odd years ago ... :laugh:

cruiser54
July 20th, 2010, 05:13
Someone with all these hypothetical engineering questions should surely be able to R&R his own radiator, shouldn't he?

5-90
July 20th, 2010, 05:34
If you want the "real deal" on information, there are two essential factors to think about:

1) Flow area of coolant. Are the 2-row tubes significantly larger than the 3-row tubes, or are they the same size? If they're the same size, then the 3-row wins. Period. More available flow area (given cross-section of individual tubes vice number of tubes.) If they're not the same size, calculate the cross-sectional (internal) area of a single tube and multiply by the number of tubes.

(Yes, this is similar to figuring resistance of a charge air cooler for forced induction. Hang on...)

2) Area of tubes exposed to fresh incoming are vice heated air from passing over tubes. Recall that as you stack layers of a heat exchange element, you can actually lose efficiency. Why? Because the tubes in the back aren't facing fresh incoming air, they're getting air that is already heated by the first layer or two of heat exchange tubes. This increases the heat content of the air, thereby lowering the potential for heat rejection by the coolant/heat transfer into the cooling air.

(Look up charger air cooler math! The only differences here are that: 1) the radiator is an gas-liquid heat exchanger, and 2) the heat differential is usually greater - except in extreme cases of forced induction! The charge air cooler is typically set up as a gas-gas exchanger, although sometimes the "ambient" air is chilled by the addition of a liquid or by passing over ice before it gets to the heat exchanger. Other methods have been tried - a "cryogenic" charge air cooler using liquid CO2 from a syphon tank has been done. I've also seen dry ice placed in a duct before the charge air cooler, to bring the ambient temperature of the cooling air down and generate a sharper gradient, which increases heat rejection.)

The "ideal" would be staggered tubes. You'd probably reduce the actual count of the tubes by a third or so doing this, but that would be made up for by the third layer of tubes. However, this would give the tubes in the back a greater temperature differential between the coolant and the cooling air, which will increase the potential for heat rejection (in addition to increasing the heat rejection proper.)

Which is better? I don't have anything scientific on the subject - but I do have a direct comparison - ran a two-row in my wife's 1989 Limited, and ran a three-row in my 1987 Pioneer. Both had 242ci I6 engines, both had AW4, both were 4WD - except for accessories that had nothing to do with heat, they were functionally identical (for purposes of the test. Although I did have about 70K more miles on the 1987...)

I will definitely say that the three-row in the 1987 kept the engine about 8-10*F cooler than the two-row in the 1989, as gaged by a sampling of temperature measurements using non-contact infrared thermometers and direct-contact thermocouples at various points in the system (midway up the pax tank, midway up the driver's tank, thermostat housing, inlet tube on water pump housing where the lower hose connects, and probably a few more. I need to see if I can find those notes.)

Doesn't sound like much - but I spent a good deal more time idling on the side of the road than she did, and I needed every advantage I could get (my alternator was also good for about 90A more than hers. Had it purpose-built.)

I haven't crunched the numbers on the two radiators (at least WRT liquid flow area,) so I can't give a direct comparo there. Knowing that the extra heat exchange potential (from the increased contact area) was offset by the air over the third row being heated, I am pleased with the results - knowing that the offset still worked in my favour! And, going from an internal transmission cooler to an external (using a B&M thermal bypass valve to allow for quick heating, and a remote spin-on filter for an extra quart of fluid capacity) was good for another 5-7*F reduction. So, figure a net reduction of 13-17*F, just in the radiator and transmission cooler. (Adding Water Wetter gained me another 7-8*F - I never did like running the engine up around 215*! I'm not comfortable with the idea of losing coolant when I lose pressure, because it suddenly starts boiling...)

I had a couple of advantages:
1) RENIX. None of this OBD crap, so no kvetching from the computer!
2) I've done this sort of thing before.
3) Living here in CA, I also had to deal with the Smog Nazis on a regular basis. So, I got regular feedback on fuel efficiency as measured by combustion products. No negative effect on combustion efficiency was noted as a result of dropping the engine's operating temperature by about twenty degrees - in fact, I noted a slight drop in NOx. Even with the EGR disconnected (did it as an experiment at the Smog Check one year. I never did like that damned thing...)

Not fully scientific, but experiential. And it will serve for me.

RCP Phx
July 20th, 2010, 11:03
How about another twist.Alumrad's radiator is a single row(but its 1-3/8" wide)!
http://www.alumrad.com/alucop.jpg

NeXJ
August 10th, 2010, 21:15
Someone with all these hypothetical engineering questions should surely be able to R&R his own radiator, shouldn't he?

ahhhh.... the old theory and practice dilemma rears it's ugly head... haven't you heard the dictum "those who can do and those who can't, teach"??? :D

GRNT
August 16th, 2010, 22:14
NEXJ
You are on the right track.
The proof is in the pudding and several years ago this was the same debate that took place. It only took a few to prove that the MODINE 2 core radiator was the only radiator that fixed the typical overheating problem that all xj's suffer from. Everyone was putting the GDI 3 in and some would work well while others were minimal. All of us that put the true MODINE 2 core in never had a problem with them and cooling was not an issue.

The fact is that the origional design of the XJ was not for the straight six 4.0 engine in the first place. It was for the much smaller V6 which proved to be a bad motor for the car. Hence they put the straight six in an was not able to change anything in the engine bay to help cooling. Hence the largest movement of AIR through the radiator with the largest movement of water is what makes the difference.

Do yourself a favor and quiet thinking and guessing. Just put a true (not imitation) Modine 2 core rad in your car. You won't regret it.

NeXJ
August 16th, 2010, 22:47
Impossible unfortunately. I was attracted to the modine solution because of the corrosion in my heater core... since it had bigger pipes.. but the Modine is no longer being made as they (apparently) are no longer in business.... so I ended up putting a different 3-row that was available... people here on the forum seem to speak well of them...though the brand escapes me for the moment...

Banditt007
August 17th, 2010, 19:51
Had the CSF 3 core all metal rad. Also used a 1 row aluminum, cheap pepboy's radiator, and honestly the 1 row cools better IMO, and has much more space between the fan clutch and the radiator, so less chance of bad engine mounts/severe engine movement smashing up your ran w/ the clutch.

GRNT
August 18th, 2010, 08:47
Impossible unfortunately. I was attracted to the modine solution because of the corrosion in my heater core... since it had bigger pipes.. but the Modine is no longer being made as they (apparently) are no longer in business.... so I ended up putting a different 3-row that was available... people here on the forum seem to speak well of them...though the brand escapes me for the moment...

Someone must be makeing a 2 core with the same specs as the modine. I'm not sure exactly what the measurements where for the modine. I'm thinking the tubes were five eights inches or more each which gave a good unobstructed flow. If you can find a radiator with larger than 5/8 then you should be ok.

Perhaps a good radiator shop can build one.

Check with Goatman on this cause he and I were the ones doing all the research. He may have a good replacement suggestion to the modine.

joe_peters
August 18th, 2010, 12:10
Isn't Modine/GDI/Proliance Ready Rad all one in the same now?