Cooling Efficiency

[Rick Iverson]

Dave;

Firstly, did you have to strip the Ni before the Kephos process? I had Kephos at one time, but was mildly disappointed in that rust spots routinely appeared under the coating.

I have a friend in Mn that recommended a salt nitride bath coating that is quite durable, but for the life of me I cannot remember the name of the process (cripes sakes, I can’t remember what I had for breakfast today). I have it somewhere in my notes, post tomorrow.

Last, ever try Dupont DP-90 in a flat or semi gloss black.

[Rick Iverson]

I can think of 4 possible advantages of mounting the vents on the side vs the top:
o having the air change direction less than if they were on the top may increase their efficiency, and
o mounting them in an inward-sloped area (low pressure) may be easier on the sides v finding a down-sloped area on the top
o being on the sides may better keep the hot air off the rear wing(s)
o less chance of water getting into things if it's raining

I LIKE it.

[Neil_Roberts]

Richard, and others ... it's tight here for turning vanes. Still a worthy project?

You want *all* of the air that goes in the inlet to pass through the radiator core. Because the core fins and tubes create a large pressure change, if it is allowed to, most of the air will take the easy path around the radiator. It takes some effort to block off all of the airflow paths around the radiator with foam and tape, but it is worth the effort to improve both cooling effectiveness and efficiency.

Turning vanes in front of an angled radiator almost never help. The air speed through the core is so low that even a major direction change does not cause much energy loss.

[DaveW]

Dave;

Firstly, did you have to strip the Ni before the Kephos process? I had Kephos at one time, but was mildly disappointed in that rust spots routinely appeared under the coating.

I have a friend in Mn that recommended a salt nitride bath coating that is quite durable, but for the life of me I cannot remember the name of the process (cripes sakes, I can’t remember what I had for breakfast today). I have it somewhere in my notes, post tomorrow.

Last, ever try Dupont DP-90 in a flat or semi gloss black.

IIRC, we never used Kephos on a previously electroless NI-plated part. Steve Lathrop had them done near Indianapolis - he would know more about it. I've never used DP-90.

[DaveW]

...Turning vanes in front of an angled radiator almost never help. The air speed through the core is so low that even a major direction change does not cause much energy loss.

Interesting - I didn't know that.

[S Lathrop]

My thinking on turning vanes on radiators that are at such a steep angle compared to the incoming air do help cooling, a bit.

The radiator will pass just so much air through the cores, given the difference in the pressure on the inlet side of the core and the exit side. The idea of the turning vanes is to distribute the high pressure inlet air more evenly over the surface area of the core.

But the bottom line is that there is no substitute for surface area of the core to deal with the cooling requirements of and engine. With both the Zink Z10 and Zink Z16 we had several radiator designs until we got it right for those cars. And in the end, it was large surface areas of the cores that solved the problems.

[Purple Frog]

Then there was Remy's vent in 2010...

[BeerBudgetRacing]

Then there was Remy's vent in 2010...

I assume that was that deemed illegal - an extra aero device?

[R. Pare]

Y

Turning vanes in front of an angled radiator almost never help. The air speed through the core is so low that even a major direction change does not cause much energy loss.

Key word being "almost" - it sometimes does, as when I installed them on Larue's car (his old 1600) way back when - it was an immediate drop in water temps on hot days.

[R. Pare]

Gents;

Since I do not have the room to mount the radiators horizontally, they will be mounted vertically in the stock fashion, with the fore end 30 degrees or so outboard of the chassis centerline. So, what if I considered mounting the vents on the outside of the left and right pods as opposed to the tops?

Thoughts?

Most likely the tops will be the more effective place for them. On the sides is possible, but that will depend greatly on what the airflow is there ( same requirement as for the top surfaces - turbulent air flow will screw things up), and how far back you place them. One issue, if placed too far back, will be the high pressure wake off of the front of the rear wheels.

[Purple Frog]

Maybe folks should look at the "inner liner" of a '98 Tatuus. Wizzy ducting.

Remy's attempt was never ruled on. A few reasons: 1. no one protested it. I think because everyone was interested to see if it worked. The zetec exhaust filled up so much of a VD sidepod exit. 2. He was running in the pro series

Interesting that the following year when he won the championship, he never used them.

[Hardingfv32]

Turning vanes in front of an angled radiator almost never help. The air speed through the core is so low that even a major direction change does not cause much energy loss.

Google: 'Performance of Compact Heat Exchanger inNon-Perpendicular Cooling Airflows' for a paper on the subject.'

A radiator can be laid over to a 30 deg angle before there is a loss in efficiency. Also discusses the inefficiency of turn vanes.

For most radiator designs there is no gain in efficiency after air flow exceeds 40 mph.

Brian

[mark elder]

I have a Ray FF and it does not have the greatest cooling. What seems a little odd is that engine temp rises with car speed. Yes, the engine is working harder at speed but it is still working hard off slower corners etc.

It was suggested that at speed the large pods cannot get rid of the air out the back (it does seem smaller than the inlet area). If so, then outlets such as Ricks vents on the top of the pods would help?

Thoughts?


As an aside, I had thought the biggest virtue of Keyphos and parkerising is that the surface is porous and therefore holds oil very well. The implication is that all these surfaces need a coat of oil from time to time.

[DaveW]

...Thoughts?...

2 things come to mind:
o The radiators may be too small/thick to transfer enough heat to the air to cool the car at high constant load/speed, or
o As you mentioned, there may need to be more air exit area to allow sufficient airflow

or a combination of both.

[R. Pare]

It could also be that the shaping of either the inlet or the exit creates a lot of turbulence, restricting the flow through the radiator, and at higher speeds, the restriction is too much for the flow needs.

[mark elder]

Thanks, the radiators are huge. Biggest I have ever seen on a FF, maybe twice the size of a 1990 VD. And they are at a reasonably gentle angle in the pods (the pods are wide) plus things are well sealed around the radiators.

[Rick Iverson]

Thanks, the radiators are huge. Biggest I have ever seen on a FF, maybe twice the size of a 1990 VD. And they are at a reasonably gentle angle in the pods (the pods are wide) plus things are well sealed around the radiators.

Mark;

Photos, please.

[mark elder]

The side pod inlet is large; 12 inches wide, the outlets are half that size (excluding air losses around the rear cover)

[Rick Iverson]

The side pod inlet is large; 12 inches wide, the outlets are half that size (excluding air losses around the rear cover)

Now that car is super clean

[Stan Clayton]

The side pod inlet is large; 12 inches wide, the outlets are half that size (excluding air losses around the rear cover)

Mark, when I managed Arnie Loyning's Swift 008 FA-based CSR in 2005 to the SCCA Runoffs Championship we did some experimentation to determine how much cooling we needed to provide adequate cooling on max temperature days, which turned out to be about 4hp per square inch for the inlet. For our 240hp that was 60 sq/in, or an inlet measuring 6"x10". After consulting old RAF Mosquito and USAAF P-51 documents we set the exit area at 60% of the inlet. It worked perfectly.

So what does that mean for your car? Well, guessing from the photos and your comments above it looks like you have about 36 sq/in of inlet and about 18 sq/in of outlet, MINUS some area for the exhaust tube and for the oil lines. In other words, you might have way too much inlet for the effective outlet area. The easiest way to test this would be to block off the outer 2" of the inlet (with duct tape) to see if that sets things right.

I realize this sounds counterintuitive, but give it a try on a test session and let us know the result.

Cheers, Stan

[Rick Iverson]

……we did some experimentation to determine how much cooling we needed to provide adequate cooling on max temperature days, which turned out to be about 4hp per square inch for the inlet. For our 240hp that was 60 sq/in, or an inlet measuring 6"x10". After consulting old RAF Mosquito and USAAF P-51 documents we set the exit area at 60% of the inlet.

Cheers, Stan

Stan;

Can this data be used for any open-wheel or SR class car? Seriously good stuff here.

BTW, I thought you guys were on the left coast.

[steve zemke]

Stan;

Can this data be used for any open-wheel or SR class car? Seriously good stuff here.

BTW, I thought you guys were on the left coast.

They are moving/ alomost finished moving to North Carolina. Brand new facilities

[S Lathrop]

Over the decades that I have been working on small bore formula cars, like FF and F2000, the most important measurement in the cooling system for me is the surface area of the radiators.

With the Zink Z10, we started with cores that were 4 inches thick, mounted perpendicular to the air flow. But they only had 60 square inched of surface area at best. The total cooling area was 120 square inches, even though the radiators had 4 rows of tubes. These did not work. The next step which was the 2" thick radiators that were maybe 10 inches square and 2 rows of tubes work very well. This gave 200 sq. in of core surface. In this situation we had almost the same core density as the original design but we had double the surface area for the radiator core. The cores were 90 degrees to the airflow and the ducting was not much. Not even close to efficient. But they cooled well.

The Swift DB1 had something like 160 square inches of surface area but in a single core and they , to my knowledge, had no cooling issues. Again good surface area.

No matter how poor or good the airflow is through the core, a thinner core of the same volume, cubic inches of core, will cool better.

My first FF's were front radiator cars. I don't remember having cooling issues with those cars like the issues I had when I started racing the Zink Z10. But I will bet that the adequate cooling came at a much higher cost in aerodynamic drag that the side mounted radiators.

[David Clubine]

Maybe folks should look at the "inner liner" of a '98 Tatuus. Wizzy ducting.

Remy's attempt was never ruled on. A few reasons: 1. no one protested it. I think because everyone was interested to see if it worked. The zetec exhaust filled up so much of a VD sidepod exit. 2. He was running in the pro series

Interesting that the following year when he won the championship, he never used them.

We did a few things on that car that made no measurable difference either way, but made people wonder lol......

[Jerry Kehoe]

Went through all kinds of solutions when I installed the BMW V12 in my F3000 March. Originally thought the the radiators for the DFV would work just fine but quickly found out that even with only about 2/3 of the HP of the DFV that the cooling provided was not enough for the larger motor. Ended up with Saldana making larger 4 row radiators with about 25% more surface area along with much larger oil coolers solving the problem. The DFV was far more efficient and was physically smaller than the V12 which is my guess as to why the original radiators did not work. The other thing I found out was that bargain radiators are made of much softer core material which will inflate the cores blocking air flow across them. Buyer beware! On the V12 oil cooling was a major factor in bringing temps in line.

[Stan Clayton]

Stan;

Can this data be used for any open-wheel or SR class car? Seriously good stuff here.

BTW, I thought you guys were on the left coast.

As to your first question...probably...within reason. It worked for the Swift's physically small 1600cc I-4 engine, but may not work as well on Jerry's V12. The development was completely empirical, with no time taken to rigorously analyze the project.

The Swift's radiator core is about 10" x 16" and is just 2" thick, so Steve Lathrop's comments need to be taken into account since this 'rule of thumb' may not work for the small-but-thick radiators commonly used in early F & SR cars.

Finally, as Steve Zemke notes we moved earlier this year to near Mooresville NC, where we are within a few weeks of moving into a new shop.

[Rick Iverson]

No matter how poor or good the airflow is through the core, a thinner core of the same volume, cubic inches of core, will cool better.

Gents;

My radiator base lengths are 20.875”, heights of 8.125”, and a widths of 2.0”. So may I presume I am in the safe zone?

[S Lathrop]

Gents;

My radiator base lengths are 20.875”, heights of 8.125”, and a widths of 2.0”. So may I presume I am in the safe zone?

If I read your post correctly, you have 339 squire inched of core surface area. that is double a Swift BD1. My bet is you are golden. One caveat is that 2 radiators in series are less efficient a single radiator of the same surface ares. But if they are parallel, you are is the super cooling range for a FF and a FC engine. And I think you are in great shape for the same engines.

[DaveW]

If I read your post correctly, you have 339 squire inches of core surface area. that is double a Swift DB1. My bet is you are golden. One caveat is that 2 radiators in series are less efficient a single radiator of the same surface area. But if they are parallel, you are is the super cooling range for a FF and a FC engine. And I think you are in great shape for the same engines.

There have been several threads on parallel v series plumbing - the series setup is generally more efficient.

Link to lengthy discussion on this:
http://www.apexspeed.com/forums/show...ntial+radiator

[Rick Iverson]

Gents;

I checked my drawings and photos, and the feed from the water pump enters a tube that feeds both LH and RH radiators. So, I have that going for me.



Here is a photo of the Carter car showing the main pump line feeding both LH and RH radiators.