use of two radiators

[neeraj]

We were having trouble with the width and the height of our radiator it was too big.so we were thinking about using 2 smaller radiators one in each sidepod with dimension of 9" * 9" * 1.5".can this system work?i know that there will be an added load on the batterey because of the two fans but the electronics team say that they can easily make it work.
please advice.

[neeraj]

We were having trouble with the width and the height of our radiator it was too big.so we were thinking about using 2 smaller radiators one in each sidepod with dimension of 9" * 9" * 1.5".can this system work?i know that there will be an added load on the batterey because of the two fans but the electronics team say that they can easily make it work.
please advice.

[KU_Racing]

what engine are you using?
Does the car currently run?
What cooling loads did you calculate?

our radiators are typically about as big as one of the 2 you are talking about, and we dont have any issues.

[Dave M]

and from experience, plumb them in series not parallel too.

[MalcolmG]

why would you plumb them in series? Doing so increases the flow resistance of the radiators (much the same as electrical stuff, eg resistors in parallel vs series), also because the water entering the second radiator has already been cooled, you get a lower temperature difference between water and ambient air on the radiator so you get less cooling from it.

We regretfully had to add a second radiator to our system before the Australasian comp last year because we found that we just weren't getting enough cooling from our existing radiator to be confident about running a full enduro in Melbourne summer heat. We plumbed them in parallel, didn't run a fan on the second radiator, and the system worked perfectly. Our data logs showed that it had no trouble maintaining a steady temp through both enduros. Our second radiator was about 50% the surface area of original one, and about half the core thickness. The biggest hassle was bleeding air from the second radiator, and this was solved by tapping a m4 screw into the top tank that we could pull out to let air out.

[Chris Allbee]

if you are going to use a dual system plumb them in parallel, being careful to keep the hoses shielded from heat sources such as the exhaust. To solve the bleeding issues implement the use of a swirl pot or header tank and place it so that it is the highest point on the system.

[mtg]

One radiator with good airflow is sufficient, if you have to add a second, you're probably not getting any air through the radiator you already have.

If you do run two radiators, heat transfer is proportional to delta T. A parallel dual radiator setup has a higher delta T on each radiator than a series setup, thus is the more efficient method for getting heat out, all other variables held constant.

[SpdRcr]

be careful...we did this 2 years ago and the idea was 'two is better than one..hopefully'....the car overheated worse than one with a single radiator. it was also heavier. make sure you run through the equations (HT, fluid systems,...)

[Marshall Grice]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">If you do run two radiators, heat transfer is proportional to delta T. A parallel dual radiator setup has a higher delta T on each radiator than a series setup, thus is the more efficient method for getting heat out, all other variables held constant. </div></BLOCKQUOTE>

holding all other variables constant is the part we had trouble with. Poor pipe design for a parallel twin radiator setup cause a serious flow imbalance. The result was much less 2 times the cooling capacity.

[Scorpio]

If you have to use two radiators, and if they are mounted on each side pods, I would mount them in a way so that the hot air that is getting pulled by the fan is targeting the rear tires. This will help you to heat up the tires before acceleration, skid pad, and autocross.

[Maverik]

Shhhhh, don't be giving away tire warming secrets!

[Gasperini]

I like how there's a bunch of CPP guys posting in here.

...And that's because we've tried this without success on the 2004 car. It didn't work for the reason Marshall mentioned. Another drawback is fan efficiency. Two small fans will almost pull twice the current as a large fan for the same amount of flow.

Two radiators is easy to package in sidepods, but has serious drawbacks, even if you can prevent flow problems. Like most of the stuff on these cars, it's a compromise.

[Chris Allbee]

I'm curious as to what the "serious drawbacks" are? I know there are compromises, but serious drawbacks implies that there should be serious gains to justify the use of such a system. Marshall also admitted that it was "poor" pipe design that caused the flow imbalance, not that it was impossible to obtain.

done correctly you can increase your surface area without increasing capacity (if that is your desire) and therefore incur only a modest weight penalty.

besides, who says you need to run two fans?

I agree that decisions on these things are compromises, but I wonder if your bias is based on experience with a bad design....

[Gasperini]

I was referring to the excessive current draw and extra weight from two fans. I think the water problem could be solved, but the other compromises add up to "serious drawbacks" in my book.

[neeraj]

we are using honda cbr600 f4i as our engine(sorry for not mentioning it in the post).During initial tests with the honda radiator(14" * 10" * 1.5") the engine overheated(the temperatures shoot up when we put the restrictor on) and also the chasis could not give us the required width for placing the radiator(too big) and placing it at an angle seriously decreased the radiator efficiency so i came up with the idea of using two raditors (paralell arrangement will be better as the air flow through both the radiators will never remain constant due to turning of the car).now I have to decide about the fan size and the piping.
Which fans have u guys used?

[BeaverGuy]

lets not use the term efficiency when talking about radiators but effectiveness instead

effectivness=q/qmax
where q is the amount of heat rejected and qmax is the maximum amount of heat that could be rejected in a perfect world

qmax can be defined as Cmin*(T,h,i-T,c,i)

effectiveness can also be written as:
effectivness=Ch*(T,h,i-T,h,o)/(Cmin*(T,h,i-T,c,i)

effectiveness is also related to NTU and R, IT increases with an increase in NTU and decreases with an increase in R

Where R=Cmin/Cmax

and NTU=U*A/Cmin

regardless of the type of setup a dual raditor system using identical radiators will have the same values for A and Cmin

because the only value on which U is dependent that changes between setups of radiators is the flow rate of water, It is higher for the series setup and lower for the parallel setup, U will be largest for the radiators in series. Because Cmin is the same for all the radiators as is A this shows that NTU is largest for the radiators in series. Also of note is that R is largest for the radiators in parallel because Cmax has decreased by almost half from the value for the series radiators.

With a larger value for NTU and smaller value for R the effectiveness for the radiators in series will be higher than those in parrallel.

If we look at how effectiveness is defined,
effectiveness=q/qmax and qmax is the same for both arrangements, we can say that the amount of heat rejected by a system of radiators in series is greater than those same radiators plumbed in parallel provided that the flow rate of the system in series does not drop too close to that of the individual radiators in parallel.

Note: I was writing a much clearer and well defined post but about 2/3rds of the way through I bumped the ESC key and it cleared everything I had written. So if you have questions feel free to ask as I imagine it is quite confusing.

[Chris Allbee]

Yes, I have a question. If you have the exchangers plumbed in series would that not increase the pressure (back-pressure, resistance, head-loss, pick your term here) more so than in parallel? Assume there is not a whole lot of loss difference between the piping systems for the time being.

Now lets assume for a moment that the water pump is at max efficiency when the pressure in the system is close to the stock set-up. If running the exchangers in series raises the pressure in the system then we could actually have less volumetric flow through the system, right? So, in this instance it could be beneficial to have the lower back-pressure of parallel radiators in order to increase the efficiency of the pump and therefore volumetric flow through the system, improving that U value and the effectiveness of the system.

In addition to all that the other variables that come into play are the external aero. What quality of air flow do you have through the radiators? Are you actually exposing all of that surface area to positive air flow? I can (and have) seen both arrangements work and each team swear that it is the best way. I think that in the end its the package deal that matters most. If the packaging on your car favors single radiator set-up then you probably won't see a gain by switching to duals, and vice versa. And...my thoughts have trailed off. Need some sleep. This is an interesting discussion though, hope it goes on.

[Gasperini]

Chris, two parallel radiators will get you more flow. It can also get you into trouble with balance like it did for us in 04.

There's a lot of little details to address here, and these are what set different systems apart.

I would love to show you our cooling system next time you're in Pomona. (Or next time we're in Detroit!)

And to answer an earlier question about flow speed during turns, I would recommend you assume a reasonable uniform speed for airflow rather than get too caught up in an unsteady flow design scenario.

[Chris Allbee]

Yes, as you say its the little details that can make or break each individual system. And I would love to take a gander at your cooling system. We're going to be at the Detroit and Cali competitions this year, so there should be an opportunity...

[BeaverGuy]

Two radiators in parallel will have more overall flow than two in series. However, the U is dependent on the flow through each individual radiator. Which will be higher for the radiators in series. Now if the pump is operating in a region where a large change in pressure causes a large change in flow, for example doubling the system pressure drop would cut the nearly in half then you would be better off with radiators in parallel rather than series.

[Wesley]

I have a question. If you had a given radiator with a specific flow rate, and you could change the volume of the radiator while maintaining flow rate, what would this do for temperature difference (or more appropriately, change in internal energy?)

Obviously coolant would be circulating more slowly (at least, time for total circulation of the system would be greater if you were to track the path of a single molecule) but would this added time in the radiator actually accomplish anything, or would it be a waste of volume?

I apologize for ignorance as I haven't had any heat transfer courses and got a B in Thermo. Maybe I'm just not thinking it through.

[BeaverGuy]

are you saying increased volume with out increased surface area if so under a steady state situation there would be no difference.

In a transient situation you would have a delay in system temperature increase and it might not be as large and you would also have a delay in the temperature decrease of the system. However, it is difficult to calculate as most data and heat transfer correlations are for steady state and not transient situations as adding time complicates the equations quite a bit.

[Yehoram]

Did you think about just using a larger single rad and angling so it fits in your duct? The rad doesn't always need to sit exactly square to the car, if you design your sidepod duct right, you can get the airflow to hit the rad square

[Chris Allbee]

The larger radiator may not fit even if you angle it. And at extreme angles you either have to have to pretty radical duct geometry or accept the reduced effectiveness of the radiator.

If a single radiator was THE ONE way to do it, then I'm sure EVERYONE would do it that way. However, it may not be the best option for every application. Weigh all the variables, see what compromises you can afford to make, and make a decision on which system to use. Implement, test, refine. Go to comp and kick ass. Those are the steps to a good cooling system.

[Bill Kunst]

All these equations without a good real market application. How about the Rc51 and the Vtr100f superhawk. 1 liter bikes running two wimpy arse radiators that barely come close to the size, when combined, of the rads used in fsae. And do they cool? I can't, even under very extreme weather, get my temp to the red, or 230 degrees F. How about a 7,500rpm burnout for a few minutes, no overheating.

As for fans, they only have one dinky little guy pushing through one radiator.

[Marshall Grice]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">I can't, even under very extreme weather, get my temp to the red, or 230 degrees F </div></BLOCKQUOTE>

perhaps this is due to the fact that you can't put any load on your motor without very quickly approaching 100mph(or doing a 360 degree wheelie)? I don't think any FSAE teams are having trouble with their cars overheating at idle either. Only sustained running at extreme load and low speed. In which case your dinky RC51 radiators would not be sufficient.

Might as well be comparing to an F1 cooling system...

[mtg]

We used RC51 radiators a few years back and still had overheating problems because of insufficient airflow across them (which was caused by a large air shovel on the front of the car).

On the same lines as Marshall, motorcycles get WAY more airflow for a given heat output than an FSAE car. Even on an RC51 with the radiators mounted on the side of the engine, you're going to get much more airflow than an FSAE car.

On my CBR, I've stuck my hand down in the fairing outlet on each side while riding- they are ducted quite well.

[Mike Flitcraft]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Yehoram:
Did you think about just using a larger single rad and angling so it fits in your duct? The rad doesn't always need to sit exactly square to the car, if you design your sidepod duct right, you can get the airflow to hit the rad square </div></BLOCKQUOTE>

If you run dual radiators, you have a better split on your CG. Not by an insane amount, but every little bit that gets the CG on the axis splitting the wheels is a good thing.