There is considerable misinformation regarding what makes a radiator transfer heat (so take what you read with a grain of salt).

Ill try to make a short radiator and heat transfer for dummies post :flipoff2:

Heat is transferred from the coolant to the radiator mass, and from the radiator mass to the airflow. The temperature difference at each transfer junction (coolant to metal, and metal to air) drives the quantity of heat transferred.

Aluminum weighs less, and the lower mass allows it to transfer heat more quickly than brass. The steady state heat transfer between the two materials (aluminum and brass) is very similar, but aluminum reacts quicker to a change in the temperature difference of the coolant fluid (like when you are on the throttle) because the heat transfer takes less time heating up the mass of the radiator itself before establishing the temperature difference between the radiator and the airflow. The fact the aluminum experiences an elevated surface temperature quicker allows it to reject the heat into the airflow quicker.

When the radiator is underdesigned (to reduce weight or fit a poor location) this rapid thermal response provided by aluminum can make a difference in overheating (or not) after a brief period of WFO travel (like in racing). When the load is steady state (a fixed industrial engine or a long crawl up a grade towing), the response time benefit is insignificant because while the brass takes longer to respond, having the proper area for heat transfer is more important (both radiators eventually reach the same steady state temperatures and the same heat transfer rate).

Sizing the face area of a radiator core to have one square inch for every cubic inch of displacement has been around for a long time (400 ci. engine matched to a 20"x20" radiator). This rule of thumb does not work everywhere (dry deserts need more face area) so some add 10% or more extra face area.

Core thickness is an airflow restriction on the air side (bad for heat transfer), and more tubes (one tube per core) increases surface area (good for heat transfer) on the coolant side. Multicore radiators are great for getting the radiator metal hot, but not always great for getting that extra heat into the air, and the hot air away from the radiator.

The coolant to metal transfer is fairly efficient, because both materials (coolant and metal) have significant mass to store the heat being transferred. The heat transfer junction, the required wetted area of the tubes inside the radiator core, can be calculated with fairly good accuracy. The surface area can be accomplished with more small oval tubes (1/2" and 5/8" typical of brass radiators) or with fewer larger tubes (typical aluminum 1" and 1 1/4" tubes). Go with the most tubes (cores) for the greater surface area, if the fan drive can handle the airflow restriction.

The aluminum radiator core manufacturers already take any credit for the more rapid thermal response of aluminum when calculating the tube surface area, they use the minimum surface area they can design, so further reducing the face area of the core (to try and squeeze even more efficiency from the aluminum construction) is risky unless the design can be tested and adjusted (if the budget allows doitagain engineering). I advise against reducing the face area of the core, and any thinking that the material choice allows a discount factor to the heat transfer potential, unless you are racing to shave weight and the load is transient.

The coolant side of each core tube sees the same temperature coolant flow and the same rate of heat transfer from the coolant to the radiator core metal. The result is the radiator metal is almost always the same temperature with minimal gradient front to back. The coolant to metal heat transfer is the same for each core (what heats the radiator metal is the same), and the coolant temperature leaving each core (to be delivered to the engine) is nearly the same, but the air side of the radiator is not so simple.

The discount factor for heat transfer on subsequent cores is only on the airflow side. The face cores experience unheated air, and subsequent cores experience air at a higher temperature.

The best heat transfer occurs where the greatest difference between the air temperature and the radiator metal is found, the face or entering side of the radiator. If you want the best heat transfer, increasing the face area of the radiator metal that sees cool air will gain you the most impact for your effort (a larger radiator face will beat more cores nearly every time).

The construction of the radiator fin design is important to the increase in temperature through subsequent cores in series, and is related to a "bypass factor" that models how much air bypasses heat transfer from direct contact to the surface area of the fins. The mass of air that can squeeze between (bypass) the fins without picking up heat mixes with the mass of heated air that does make contact, and the result raises the air temperature of the downstream cores.

In reality, without getting into math or fin designs, the elevation of air temperature between cores is less than 15%. If the airflow temperature is raised from 70dF to 170dF through a four core radiator, a 100 degree increase, 45 degrees (~45%) of the temperature rise (and heat transfer) is in the air to metal contact in the first core, and a lower percent from subsequent cores (something like ~30%/17%/8%).

The aluminum radiator guy's are right that two cores are more efficient (nearly 80% of the cooling is from the two front cores), but if the total surface area of the coolant to metal, and metal to air, contact is less ... the net result is not so great of a design (just like ricers 100 ci 4-banger @ 3 hp/ci is good, but nothing like a 302 @ 1 hp/ci: there is no substitute for more surface area in a radiator unless you want to spend a lot).

The actual area of the core face that experiences the high air to metal temperature difference is more important than the calculated face area. People tend to forget that the radiator face is not working to transfer heat, unless it's moving air.

What makes air flow through a radiator core (through a restriction)? Pressure drop (static pressure) or the momentum of the air mass (velocity pressure) through the core establishes the airflow, and the resulting heat transfer.

Most people with a cooling problem try to increase the mass of air blowing through the radiator by increasing the velocity pressure acting on the core. They add a round fan (usually electric) in front of the radiator (a pusher fan).

When an unshrouded fan is used, a puller with no shroud or a pusher, the core area experiencing the airflow (and temperature difference) is limited to only that is the direct path of the high velocity air. A 20x20 (400 sq.in.) radiator face with a 16" diameter fan blade, without a shroud, is little better at heat transfer than a radiator with a 16" round (201 sq.in.) face area.

Adding a 16" pusher in front of this radiator, with the unshrouded 16" puller, gains almost nothing in air side heat transfer effect.

How do you get the entire entering face of the radiator to work transfering heat? You try to get airflow across the entire face of the radiator core.

Try many small fans (fit lots of round high velocity airstreams in a square area)? It can work, but it's looks complicated (and is probably expensive).

It's easier to make a static pressure difference across the radiator core work to move air through the entire face of the radiator, by using a puller fan and fitting a shroud on the suction side of the core. You only need one fan, and it can motivate airflow across the entire face area of the radiator core, just by adding a shroud to contain an area of negative pressure on the leaving air side of the core.

The problem with using static pressure to draw airflow through a radiator is that it takes a significant increase in power to generate pressure (research fan laws). Using a multicore radiator core that is thick and restrictive on the air side requires that it be matched with a fan and shroud that can generate a static pressure difference great enough to overcome the restriction.

Replacing a two core with a more airflow restrictive four core can sometimes work against you if the fan clutch is weak, or it is combined with a swap to an electric fan.

Most electric fans cannot develop significant airflow at higher static pressures, because the power draw must be limited to protect the wiring. Compare an electric fan to a high flow (and pressure) mechanical fan. The blades of the electric are narrow, and the blade pitch is shallow (compared to the mechanical fan), both design aspects limit the potential of an over-amp condition. Yes, electric fans are great to gain power on the end of the crankshaft, because to generate a significant negative pressure behind the radiator sometimes takes three to five horsepower (the gains we read in the electric fan advertisements). Use electric fans when you can, when the radiator is overdesigned for the power and transient heat transfer required, and use a shroud. Just do not expect a 1/4 hp electric to pull the same airflow and pressure drop of a fan drawing 3 hp off the front of the engine.

I read that fans have little effect to gain airflow at speed ("airflow is going through the radiator on the highway because you are going faster, so adding a fan or shroud to cool the engine on the highway is not going to solve the problem.") We read it all the time. It's BS.

What is difficult to understand is that most vehicles generate a high pressure area under the chassis at speed (the air above and to the sides of the vehicle is high velocity/low pressure, but underneath it's low velocity and high pressure). Yes, the pressure in front of the radiator can be higher (conversion of the velocity pressure to static pressure) but the pressure behind the radiator can increase with speed as well. Betting that the converted high velocity air in front of the vehicle radiator can overcome the pressure under the vehicle (the pressure on the leaving side of most unaided radiator cores), to motivate airflow, is almost like betting that you can piss up the inside of the airhose of your compressor with 20 psi streaming out the end.

You still need a fan and shroud to eliminate most radiator airflow problems at highway speeds, because you still need to establish a pressure drop across the radiator core (and sometimes it's more difficult at 60 mph, than when parked at the curb).

Crossflow vs. downflow design is not as important as where the radiator cap pressure relief is located. The cap should be on the low pressure side of the water pump, something that is easier to package and service with a crossflow core.

I hope this helps (I have had enough internet tech for the week)?

In summary: if the budget demands a choice between a high dollar aluminum radiator, or a brass radiator and a well fitting shroud, get the system with the shroud (and use a mechanical fan drive with a fan blade that has some pitch angle to the blades, and a clutch to save power when it's not needed).

Happy Trails!

 

[No message]

 

Somebody is after Billavista's crown as king nerd, eh Ed?

 

Ill try to make a short radiator and heat transfer for dummies post
.............................
short eh??????????

 

I don't have any "scientific" input here but I'd bet that many that see improved cooling with an aluminum radiator just replaced a smaller and/or plugged up radiator... It's always amazing what a new radiator will do for marginal cooling conditions...

 

One thing to keep in mind about aluminum radiators.

They have a useful life of 5-8 years. At which point they stop transfering heat properly and allow the engine to run hot.

The issues is corrosion on the OUTSIDE of the radiator. The radiator can be flow tested and everythign and come out ok, and still be bad.

Learned this from an old fart behind the counter of his personally owned radiator shop. I didn't believe him until he convinced me to dump the $$ on a new replacement radiator. Sure enough, the old fart was right... engine temp dropped from 240 to 130. Had to replace the thermostat after that, it turned out to be stuck open as well.

 

Maybe I need a quick refresher on heat transfer, but the below quote doesn't jive with what I remember (great post though - not knocking ya). Most brass radiators are actually a copper and brass mix, mainly being copper. Copper has twice the thermal conductivity of aluminum. I don't remember much about elemental mass having much to do with thermal transfer...

...
Aluminum weighs less, and the lower mass allows it to transfer heat more quickly than brass. The steady state heat transfer between the two materials (aluminum and brass) is very similar, but aluminum reacts quicker to a change in the temperature difference of the coolant fluid (like when you are on the throttle) because the heat transfer takes less time heating up the mass of the radiator itself before establishing the temperature difference between the radiator and the airflow. The fact the aluminum experiences an elevated surface temperature quicker allows it to reject the heat into the airflow quicker. .........

I do know that the radiator design ultimately plays almost more of a part as to how well it cools than the material it is constructed of - which is where a well designed aluminum radiator has an advantage over a traditional copper/brass one as they can use the wide and thin tubes that copper/brass can't, which makes up for most material thermal conductivity shortcomings...

 

Maybe I need a quick refresher on heat transfer, but the below quote doesn't jive with what I remember (great post though - not knocking ya). Most brass radiators are actually a copper and brass mix, mainly being copper. Copper has twice the thermal conductivity of aluminum. I don't remember much about elemental mass having much to do with thermal transfer...

No need for the refresh, you remember well.

Copper has a higher thermal condictivity than aluminum.

The greater mass of copper takes longer to heat up through conduction from the coolant into the metal (although with the thin wall material of a radiator or heat exchanger coil the difference is fairly small). The longer response time to heat the copper, adds delay time to the Conduction/Convection/Radiation heat transfer from the metal to air. The difference is minimal, but it gets advertisement copy.

Eventually, under a continous duty load, copper will reach the same temperature as the aluminum, when it will have the greater advantage of faster conduction of heat from the leaving side of the radiator (the fins with heated air) to the entering side of the radiator (where the fins see the coldest air, and greatest heat transfer).

I do know that the radiator design ultimately plays almost more of a part as to how well it cools than the material it is constructed of - which is where a well designed aluminum radiator has an advantage over a traditional copper/brass one as they can use the wide and thin tubes that copper/brass can't, which makes up for most material thermal conductivity shortcomings...

Correct.

Extrusion flexibility of aluminum shapes allow designs that are better balanced between the coolant to metal heat transfer, with the metal to air heat transfer rate. Fin bonding is also easier with aluminum tube to aluminum fin, allowing for better conduction at the bond (better performance).

Happy Trails!

 

:beer:

Thank you that was exactly the response i was looking for, I can run with that info, now that i have a better understanding.

I'll have the 17" mechanical 7 blade fan with shroud, so cool, now its just a budget issue.


Dusty

Bet your fingers are sore.


One of three places in california..... Rubicon, Fordyce, Barrett Lake

 

One thing to keep in mind about aluminum radiators.

They have a useful life of 5-8 years. At which point they stop transfering heat properly and allow the engine to run hot.

The issues is corrosion on the OUTSIDE of the radiator. The radiator can be flow tested and everythign and come out ok, and still be bad.

Learned this from an old fart behind the counter of his personally owned radiator shop. I didn't believe him until he convinced me to dump the $$ on a new replacement radiator. Sure enough, the old fart was right... engine temp dropped from 240 to 130. Had to replace the thermostat after that, it turned out to be stuck open as well.

Removing the oxide layer for aluminum takes seconds. Mag wheel cleaner or a aluminum truck trailer wash will do the job nicely !
How do you think I keep a fleet of these clean.

 

Ed or someone else, could you walk me through the actual vs advertized benefit of a dual pass radiator?

Wouldn't Hood louvering (I can't spell) be a really good way to improve air flow through the radiator and allow the hot air in the engine compartment to escape?

Thanks.

 

Ed or someone else, could you walk me through the actual vs advertized benefit of a dual pass radiator?

Wouldn't Hood louvering (I can't spell) be a really good way to improve air flow through the radiator and allow the hot air in the engine compartment to escape?

Thanks.

They have three tanks and flow like this:

 

What about mounting your rad above the motor with an electric fan drawing the heat up?? Similar to the hummers?


The only issues I see with that is you had better be running either a snorkle or a damn good cold air induction setup. The heat from your manifolds or headers is only being sucked up no towards your intake.

 

Thanks for the pic of the double pass old scout, but does it really help w/ cooling?

You haven't changed the amount of surface area at all. The distance the coolant moves through the radiator (and therefore the time in the radiator) has been ~ doubled but how does this increase the amount of cooling that can take place? In the above posts, everything seems to be surface area and cooling efficiency. The double pass doesn't effect any of that. There is no talk of 'residence time' or what ever.

Keep the info coming. This is good stuff (so far).

 

Think of it this way.
Say 190F coolant drops 50 degrees travleing the 30"s across the core. , so a single pass rad will have an input temp of 190 and an exit temp of 140. The double pass will have an exit temp of 90F!

 

Ed or someone else, could you walk me through the actual vs advertized benefit of a dual pass radiator?

Wouldn't Hood louvering (I can't spell) be a really good way to improve air flow through the radiator and allow the hot air in the engine compartment to escape?

Thanks.

Heat transfer is better when the temperature difference between the two materials is greater (coolant and airflow).

In a single pass radiator the entering coolant is usually in the 200-230dF range, and the leaving coolant is in the 150-180dF range (with air entering the radiator at 70-95dF and leaving at 130-160dF). The radiator is designed to have the 200dF coolant transfer as much heat as possible into 70dF air, at the radiator core area where the temperature difference is greatest: the entering airflow side of the radiator face.

Extra cores (more rows of tubes) help the cooling effect, but the majority of the heat is transferred in the first two tube rows (what I explained earlier). The majority of the cooling effect in the first two tube rows is also on the side of the radiator where the coolant leaves the entering radiator tank (again, where the hottest temperature difference from coolant to airflow exists). The half of the radiator core next to the entering radiator tank is doing most of the heat transfer.

As the radiator core is widened (as the core tubes get longer, or the radiator is redesigned to be a two-pass configuration) the coolant temperature in the leaving end of the tube drops, reducing the heat transfer rate. The resulting 140df coolant in the tube near the leaving radiator tank cannot transfer as much heat into the 70dF entering airflow, as the 200dF coolant can transfer. The result is the radiator face area near the leaving end of the tubes is not doing much heat transfer. The expected heat transfer efficiency, with respect to face area, can be severely compromised by long core tubes (or dual pass construction).

The wide aspect ratio of the XJ radiator is a fair example of core tube length that is not providing good heat transfer efficiency on the leaving end of the core tubes, and the factory Engineers placed the secondary electric cooling fan on the coolant entering side to balance out the heat transfer across the face of the radiator. They paired the greatest temp difference with the less effective electric fan to overcome some of the efficiency limitations of the short and wide space constraint, and also to allow the hot end of the radiator core to provide some reasonable heat transfer efficiency without the electric fan in operation. It's a compromise forced by the body space available.

The advantage of a dual pass radiator, over a wide radiator core, is the coolant is mixed in the intermediate tank. This mixing improves the quality (consistancy) of the coolant temperature entering the second pass across the radiator face. It allows the temperature of the coolant entering the core tubes to be more consistant, front to back and top to bottom, as it flows through the remaining radiator core.

This mixing is beneficial with a very tall aspect ratio radiator core (the opposite of the wide XJ core) and with deep multiple-row cores, as coolant flow will take the path of least resistance. The intermediate tank mixing helps when the core tubes cross section area exceeds the coolant volume flow rate (like when high flow custom rectangle aluminum tube extrusions are employed, or when cores rows are added).

If the top half of the radiator core tube bundle can flow the majority of the coolant, like when replacing a two row core with a four row core, the velocity of the fluid in the lower tubes of the core will be reduced. The top entering face tubes will provide great heat transfer (due to the temp difference) but the velocity of the fluid may be too fast to gain the best heat transfer effect (the fluid velocity and ample tube cross section area of the tube can establish a thermal boundary layer). Combine this with fluid in the leaving side of the core tube bundle (third and deeper tube cores) where the airflow temperature is elevated enough that there is minimal drop in coolant temperature. The result is coolant leaving the core without gaining as much heat as expected.

Lower in the core tube bundle the remaining fluid flow may also be so slow (remember, the majority is already flowing in the extra deep tube core fed by the top half of the tnak) that after the first few inches of tube the drop in coolant temperature is so great that the heat transfer efficiency is no better than at the end of a long tube. You end up with cool fluid leaving the tube, but not much fluid (less than the optimum for the tube cross section and surface area).

The dual-pass construction allows the radiator designer to better match the coolant flow to the core tube's cross section area at the entering radiator tank. The intermediate tank mixes the hot coolant that was restricted by the boundary layer in the top of the first pass tube bundle (and more effectively the hot coolant passing through the leaving airflow side of a deep radiator core bundle), with the colder coolant near the bottom of the first pass (and the coolant leaving the entering face tubes). The goal is to provide the hotter coolant from the inefficient tubes in the first pass another shot at getting closer to the coldest airflow where the best heat transfer exists.

Does it work? With a deep core radiator and with some of the wild aluminum core tube profiles I would expect it can make a significant improvement in the leaving coolant temperature. I have not read any SAE papers on the measured improvement provided by this style of radiator construction, but I would expect there is considerable heat transfer to be gained with a tall aspect ratio radiator (for the 30's vintage hot rods & other space constraints where deep cores are required).

It would be instructive (read -- interesting for tech Engineer/nerds) to build a four row XJ radiator with an intermediate tank to see what benefit result exists (maybe a potential thermo/ME grad needs a senior project)?

I Hope This Helps (HTH)?

Happy Trails!

 

Ed, I must ask, where have you received your education? You sure know a lot about everything from 4 links to radiator tech

 

Ed, I must ask, where have you received your education? You sure know a lot about everything from 4 links to radiator tech

BrettM, I grew up in a family where we worked on mechanical things to keep out of trouble. Bicycles, skateboards, go-karts, motorcycles, tractors, cars & trucks, anything with wheels to tweak and keep busy. Family hobbies include antique tractors, old motorcycles, and small stationary engines.

I worked in a lawnmower shop for a while, as a flat rate auto mechanic, and as a refrigeration tech for Pepsi (when the economy and construction work was slow). This was in addition to Apprenticeship in sheetmetal fabrication and air conditioning and refrigeration, and eventually a BSME from Cal Poly Pomona. Employment has been in the HVAC&R industry, and Digital Control and Fire/Life-Safety & Security systems. The work in construction tends to keep me busy and somewhat up to date with technology (my customers demand I know what they do, to provide what they need for an environment).

I lived for a while below the longest natural terrain closed road race course in the western USA, Willow Springs, and the management didn't seem to care if some skinny high school kid stood in the doorway while they were teaching the paying students how to road race or build and tune the cars. They also didn't seem to care if I parked my motorcycle in the pit lanes when the race teams and factory Engineers were testing their toys. Watching the Japanese and German Engineers work with a translator to decode the feedback from the British and American drivers was comical at times, but good exposure to patience. Along the way my interests found a way to qualify for the NHRA Division Six finals, and work with a low budget team that placed second in class in the Baja 500 (including prerunning/chase and sitting suicide seat for hours, thinking of ways to make the car faster and more comfortable). I have former shop partners (and good friends) who worked for Sandy Cone, Walker Evans, Spencer Lowe, Jim Russell, Keith Black, Ed Pink, Cal Wells, Dick Guildstrand, and other competitors, people who seem to do well with innovation. Bench racing can lead to enlightening fabrication techniques (what failed, as well as what worked).

There is a lot to learn if you can keep quiet long enough to let it soak in. The other aspect is my father demanded that I, "cut the College Boy crap and explain how it works in words people can understand."

I'll keep working on the last part (it's a hobby, so I can make mistakes now and then).

Happy Trails!

 

If I understood all of that right then the dual pass design has limited benefits other than mixing the coolant (homoginizing the temps some?) before a second pass and giving more of the coolant a chance to pass through the front cores of the radiator?

Assuming I've only got so much room and I'm moving air well, what is the best "bang for the buck" to increase my cooling capacity?

Thanks.

 

If I understood all of that right then the dual pass design has limited benefits other than mixing the coolant (homoginizing the temps some?) before a second pass and giving more of the coolant a chance to pass through the front cores of the radiator?

Assuming I've only got so much room and I'm moving air well, what is the best "bang for the buck" to increase my cooling capacity?

Thanks.

You have the summary correct on the dual-pass benefit.

Best Bang? Add an engine oil cooler and a trans cooler (if you run an auto). The direct oil cooling will take some of the load off the radiator and get you more efficient overall performance (as long as it's not overheating now due to age and wear, or poor design and shroud construction). Oil coolers have come way down in price in the past ten years.

Happy Trails!

 

same

You have the summary correct on the dual-pass benefit.

Best Bang? Add an engine oil cooler and a trans cooler (if you run an auto). The direct oil cooling will take some of the load off the radiator and get you more efficient overall performance (as long as it's not overheating now due to age and wear, or poor design and shroud construction). Oil coolers have come way down in price in the past ten years.

Happy Trails!

Ed I always love to read your "hobby" so keep it up.... Oil coolers, I recently changed my radiator to an 17 x 27, 2 core brass unit from an 87 legend. Fitment went well and cooling is fine, I switched because the CJ radiator was toast this fit and the price was right. This is a an auto radiator, can the engine oil be used in the trans cooler portion of the radiator or will the oil be too thick for the flow. The Radiator trans inlets are about 8mm ID.

 

The only thing I would add is that fatigue failure due to thermal gradientsin the core (at the tube/tank junctions primarily) can be an issue.

Garrett Turbo (also builds heat xfer devices) worked on alum radiators for big rigs and had lots of failures trying to meet the endurance testing requirements - I believe it was a combination of thermal and mechanical fatigue.

Brian

 

In the training I have had, all the literaure that I read, and my personal experience w/ heavy trucks and heavy equipment, the temperature differential between the inlet side of the radiator & the outlet side of the radiator is only about 10*. Any more and the coolant flow is too slow, any less & you have either a coolant capacity problem or an air flow problem.

A 30* difference in the 2 sides seems extreme to me.

 

6/10/2010 update enough people ask and don't go to the above listed link so heres the scoop on my setup that works in the summer heat of the Sacramento valley


People talk about it, people allude to it but no one ever seems to post the specifics. heres what has worked for me in pretty hot central valley CA temps 100+ in the summer

this combo keeps a warmed 401 reasonably cool in the summer.


Engine fan:
For engine fans that fit the stock shroud and suck lots of air.

1. early 1970s - late 1970s Cadillac eldorado and others running 472s and 500's came with an 18" 3-1/4" high pitch 7 blade fan
2. also there is the 1981-1986 Cadillac with the 4.1L v8 it is 7 blade an 18 inches also with similar pitch right at 3" -3-1/8" and moves a lot of air. It is unique in that it as a reverse bend on the top of the blades easy to spot

I picked up one of each for $5 a piece from pick and pull so $10 total they both seem to pull the same amount of air. pretty incredible difference.

Fan Clutch:
For the fan clutch I used a Hayden Severe duty fan clutch #2797 (buick with the diesel in the early 80's on some models like napa i had to notch the holes just a smidge to clear the bolts into the water pump but on a summit model i didn't, it was slotted deeper with universal slots). $48.79 from my local napa

For Shroud:
Stock CJ7 V8 shroud (stock no cost), if you use teh JP2006 the 81-86 HD w/ AC 6cyl fan shroud does work and it accomodates a 18" fan but i recommned you build one if you want the bigger radiator. it cleaner and you can center things up better


For Radiator:
For a bolt in radiator i so far have found a couple of options for bolt in at a decent price with a stock shroud.
but remember the surface area of a V8 radiator is only 372 cubic inches +/-, the shroud can't accomodate anymore. other options are cross flow and they require a different shroud and radiator hoses which is fine but i didnt want to spend booko bucks and changes, too much money at once and the side tanks really cut down on the amount of core surface area you can capitalize on.


For cheap and not much messing around:
For aluminium down flow design radiators they are out there, they are not overly cheap but they are pretty close price wise to a brass radiator these days, bolt in and thats the key, usually fit and cooled better than the extreme 4 core radiator i got from extreme radiators a few years ago. the one i had was 3 rows of 5/8" cores and with a stock shroud it got the job done alright but the surface area wasnt there in the core for the cubes of my 401, im convinced now that surface area dimensions are key to keeping a crawler cool as well as efficiency. on this radiator, the number of tubes and the fin count was less than the brice thomas radiators but bang for your buck its a good deal, i used mine only over the summer and decided i wanted to bite the bullet and go big on my cooling system. BTR also makes a v-8 cj radiator also is a great high quality bolt in unit, uses your stock shroud the only catch its still a little small on core surface area.



To really get the most core suface area and cooling:
If you have the know how to stamp out a 3/4" deep rectangular tray, cut a round hole and make your own fan shroud.... it takes an afternoon and is well worth it then the better option is the 6 cyl radiator with more surface area than a cj V8 radiator (408 cubes (6cyl) vs 372 cubes (v8 15.5"x24" )) and 81-86 radiator support/grille (the late model grille is deeper than the 76-80 and allows for 3" thick tanks over 2.5" tanks = the option of a thicker core) these guys at brice thomas make jeep radiators that are down flow and they'll do custom work for reasonable. I ordered the JP3006, the core is 17"x24", i had them modify it to run dual 1.25" tubes vs their configuration of dual 1" for a minimal charge you might even be able to do slightly thicker by 1mm or two. the tanks ended up 3" thick fit flush which into the late model CJ radiator support meant no issues or changes in the fan to core clearance from my old 1976-80 raditor support and stock style V-8 radiator. my advice is Do not let them tig on the brackets, make your own, also have them notch the lower tank 1" x 5" wide on the driver side to give you more PS line clearance (highly recommended), you'll need to move it up or down and set the depth yourself, nobody ever sends a raditor with the brackets set perfectly and with tanks this thick you need every 1/4" worth of proper clearance make it count ask them to send the brackets seperate if you know someone with a tig welded its even better.
http://www.cgj.com/shop/radiators/antique-radiators/jeep/jp-3006/

I wish they would make the lower tank with the notching and sell this radiator as a V8 option because it really does work much better than anything else ive tried and ive tried alot to keep my bored and cam'd 401 cool

All said and done its a little more money but a whole lot more radiator and cooling. Lets just say im more worried about an exhaust leak these days even on a 105 degree day i havent seen much over 180-185 degrees out of my jeep. Novak wants $400+ for their setup and i can tell you that this has more surfae area in the core and more capacity and is imho a better fit than their setup when it comes to running an amc 401 under the hood




For a thermostat:
Combined with a 180 mr gasket or robert shaw high flow thermostat $14.21

For coolant:
I run 30% dextron synthetic coolant and 70% Distilled water with 1 bottle of water wetter $32.87

Water pump:
Flow cooler water pump $89.95 (i already had one since my old water pump puked i did not include this in total price since most guys i know have em already on their jeeps)



Few details on the jeep. CJ 7 mildly warmed 401, t18 D300 4.5" spring lift, 35's 8274 winch in front 1" body lift.

I have a 1" body lift which put the stock 7 blade 17" fan right close to the shroud. To clear a bigger fan and to healp with geometry when lifting the tranny I moved the MORE motor mounts up and that created ample room clear the 18" caddy fan in stock shroud. Since i was going for a flat belly pan with my clocked dana 300 things needed to be shifted little any ways i re=drilled the more mounts 7/8" up and 3/8" for ward sliding the motor slightly ahead helped with valve cover clearance at the fire wall, moving it up allows you to continue using the stock radiator shroud. My flat belly pan is actually 5/8" below my frame but that is fine for what i wanted.

 

Ed, correct me if I'm wrong, but I was of the understanding that most cores for our applications were constrained to an overall depth front to rear.

If so, that would mean that you could not fit 4 tubes measuring 5/8" in the same space you could fit 2-1 1/4" tubes. You have to allow for the space between the cores and using that logic a 2 tube core with 1 1/4" tubes actually has the capacity for more tube to fin contact area that one with 3 5/8" tubes.

That would make the 2 row core more efficient with the greater tube to fin contact area all other things being equal.

Or, was I misunderstanding again?