Radiator modification

[Jbird]

I needed a temporary radiator for a 66 Chevelle SS396 while the original was being re-cored so I bought this bolt in replacement Chinese made alooneyum one. It has one, 1 1/2" row and barely kept the 325HP 396 from boiling over.

Now that the Chevelle has gone home I was thinking, if I modified this Chinese baby into a serpentine flow unit it might be capable of cooling a supercharged 98 Pontiac V6 for an upcoming project.

My plan was to slice into the top and bottom tanks where I marked in blue. Then slipping plates of alooneyum into the slices all the way to the core and tigging the plates in place, segregating the columns into thirds and resealing the tanks.

Then the coolant would flow, down from the inlet, up towards the filler and back down to the outlet. Am I looney or what??

   Jbird

[sirstude]

I have seen that done, and it is suposed to help, but can you weld on your radiator.  many of them are epoxied and cannot be welded on.  Do a search here, we discussed that a couple of weeks ago.

Doug

[phat rat]

I run a triple pass aluminum in my cpe with excellent results. But it is a 2 core

[Bruce Dorsi]

I was thinking, if I modified this Chinese baby into a serpentine flow unit it might be capable of cooling a supercharged 98 Pontiac V6 for an upcoming project.  

I may be missing something obvious here, so I'll be following this thread to learn more.

I have always been led to believe that the amount of surface area (of the radiator tubes) in contact with the coolant was a primary factor in a radiators's ability to cool.

Altering the configuration of the radiator does not change the internal volume, nor does it increase the surface area exposed to coolant.

Modifying the radiator as you describe will, in effect, triple the LENGTH of the coolant path, BUT, you will reduce the QUANTITY of tubes connecting the upper and lower tanks to 1/3 of the number of tubes currently handling flow.

I don't think there is any gain to be made, but I'm here to learn!

If the radiator, in its current configuration, is incapable of cooling a given engine, how does re-configuring make it more effective?  

I can understand the benefit of adding MORE, or LARGER, tubes, but that is not the case here.

[Jbird]

Doug, I see no evidence of epoxy and the tanks are welded to the core.

Bruce,  I'm not reducing the number of tubes connecting the tanks, so the flow volumn should be the same. My thought is, the coolant being channeled up and down the coolant tubes will increase the amount of time the coolant is being cooled and the coolant will be forced down a path instead of being allowed to meander around in those long upper and lower tanks.

I once used an alooneyum cross flow tradiator with the inlet and outlet in the same tank. It had a plate welded into that tank halfway between the inlet and outlet. The coolant was channeled in a sideways U path through the radiator. It ran really cool.

This radiator is probably fine for cooling the Poncho V6, but I just can't seem to leave well enough alone.
             Jbird

[enjenjo]

In theory, since the amount of coolant moving through the radiator, will for the most part, remain constant in GPM, this would force the coolant to travel a path nearly three times as long, and about three times the speed. I am ignoring the friction losses, and so forth. I really don't see a great advantage either way, but then I have never tested it in the real world.

I have noticed that newer cars have much smaller cooling systems than in the past, and they still seem to cool ok, but I have no idea how to apply that technology to out applications.

[Bruce Dorsi]

Bruce,  I'm not reducing the number of tubes connecting the tanks, so the flow volumn should be the same. My thought is, the coolant being channeled up and down the coolant tubes will increase the amount of time the coolant is being cooled and the coolant will be forced down a path instead of being allowed to meander around in those long upper and lower tanks.

Perhaps I didn't explain my thoughts clearly.

As an example, let's say there are now 30 tubes connecting the upper tank to the lower tank.   ......After modification, there will still be 30 tubes connecting the tanks, but the flow will originate thru only 10 tubes in the section where the inlet is located.  .......The flow can then only pass thru 10 tubes to get to the final section of 10 tubes where the outlet is located.

Another way of thinking may compare it to the difference between a SERIES circuit versus a PARALLEL circuit.  

For any given flow rate of coolant to be the same at the inlet and outlet of the radiator (Example: 20gpm IN, 20gpm OUT), the VELOCITY of the coolant flow thru the core will have to increase, if the coolant is entering only 10 tubes instead of 30.   ......This will offset the perceived gain that the coolant will remain in the radiator tubes for a longer time.

If the coolant entering the top tank can exit via 30 tubes, as opposed to only 10 tubes, the coolant shouldn't be lingering in the top tank.  .....Any coolant in the lower tank has already passed thru the core, so any possible cooling has already occurred.  

We need to remember that the volume exiting the radiator must equal the volume entering the radiator!

Maybe my thinking is flawed, so any comments/explanations/clarifications are indeed welcome by me!

[river1]

in this months car craft (july 07) they talk about cooling systems. paraphrased below are the couple of paragraphs :?  that deal with a dual pass radiator. i would expect that it holds true for a triple pas rad also.

they mention that the reason it works is the velocity of the coolant doubles when forced thru half as many tubes. this creates turbulence in the tubes exposing more coolant to the radiator tube walls and improving heat transfer. this also presents an increased load on the water pump which means using a double pass rad demands a better water pump to take advantage of the dual pass concept.

don't know the accuracy of it, just putting it out there.

later jim

[Jbird]

Bruce,  I've basically designed a radiator three times taller and 1/3 as wide as it was originally built, but with two direction changes in the flow path. I like Jims discovery of the velocity and turbulence theory. When I get my engine start stand finished I'll have to set up a control engine and test this radiator before and after modification. If it don't work I can always cut the plates out and return it to it's original configuration.  Interesting discussion though?   Jbird

[Bruce Dorsi]

I like Jims discovery of the velocity and turbulence theory.

For several years, Mercedes used aluminum radiators having round tubes (as opposed to flat or oval shape) approx 5/16" diameter.  

The radiator was a single-pass design, but there were spiral-twisted strips of aluminum  inside each tube, which ran the full length of the tube.

The purpose of the strips was to promote "swirl" in the coolant, thus disrupting laminar flow.  

I too, agree with the theory that some turbulence is good.        

When I get my engine start stand finished I'll have to set up a control engine and test this radiator before and after modification.

I agree that is the only way to prove or dis-prove the merits of the modifications.  

Interesting discussion though?

It is interesting to me!  .....My mind often ponders such thoughts, but without definitive proof.

Have you ever made more progress on Ugbugly?  ....That vehicle always appealed to me!

[Jbird]

I'm quite the ponderer too, sometimes thats all I seem to accomplish.

 Ugbugly moved to Florida. After I built the new hood two years ago everytime I'd start to work on her I'd end up wanting to re-engineer what I was trying to finish. I figured if kept redoing every little thing, Ugbugly would never get finished.....   I've had this idea for a new project, involving a 27 T body and a frontwheel drive, drive train out of a late GM or maybe even Toyota V6 (Supercharged Bonneville V6?)...... Ugbugly went on the block. I sold her to a VW collector in Fla., A lot of folks were interested, but he came up with the cash. I really miss her, she was a blast to drive, but my new idea has me pretty excited. This time I'm gonna be completely linear... start here, finish there. No rethinking, no exceptions. I post some pics when I've gathered the parts into a pile.   Jbird

[Crosley.In.AZ]

many years ago we had a custom radiator built by Ron Davis rads in Glendale AZ......  A T-bucket /sprint car style vehicle.

It ended up as a triple pass unit.  I asked them if the 3 pass style would cool slightly better.  I received a blank stare & a shoulder shrug from the shop foreman at the front counter.

So I asked him IF any body in the building would know..... blank stare.  I turned around to leave and the car owner stopped me.  The car owner had them build the radiator and it barely cooled the car in mild heat (under 90*F)

[tomslik]

US radiator built/builds copper/brass ones like that...
i was unimpressed.....