JBL930

Wouldn't it make sense for there to be vanes splitting the air up, surely without it the air will just go through the most direct route causing hot and cold spots in the cores, i'm only trying to apply logic (which is probably flawed)

Jussi

Do you mean vanes like that?

sfl993t4

I think that he means vanes in the end tanks (the inlet side). Some intercoolers have flow distribution issues (often because packaging has compromised the end tank design and the inlet pipe entry), and so have vanes to split the incoming airflow between different parts of the core.
The vanes themselves can cause other compromises.

Steve

MOD500

Let me know what dimensions you want Kevin, and I will do the measurement this weekend.

Thanks alot for the offer on the turbos, after 13k miles they shouldn't be so bad though?!

Felix

Martyn, your Secan IC was v dry insde unlike mine.

JBL930

Yes i meant vanes in the end tanks, i thought it would make sense to split the end tanks into three compartments internally, so the air coming in from the pipe was forced to go through all of the intercooler, i can just imagine the air finding the easiest route which would leave some sections of the intercooler colder/hotter than others.

Jussi, i've wondered what that "V" section in the FVD intercooler is all about, that would make you think that the back rows of that cooler are more restrictive than the others, is it there to allow clearance for the engine lid catch?
Here's a picture of one i tried on my car, it wouldn't fit with the standard tail so i didn't use it, i think it was Toby's at one point many moons ago




I guess if the Secan cooler doesn't have internal vanes/fins/compartments dividing the air then it isn't a good idea, it just seams logical to a layman like myself



Edit to change the name on my response

935racer

Toby, I thought I recognised the coolers in the picture! Not sure I can add much to this discussion except to remind people there are at least two types of manufacture for intercoolers. The first is "tube and fin" and is by far the most common method. Tubes are extruded to certain dimensions and the "cooling fins" are then added afterwards. Its a relatively cheap form of manufacture and can produce coolers with good performance. The main issues mitigating against this for optimum performance are; The extruded tube tends to have thicker wall than required just to withstand the internal pressure thus inducing a "resistance" to heat flow, the extruded tube is only available in pre set dimensions unless you go to the trouble of making a specific extrusion die and thirdly, the materials used in the extrusion process may not be the ones you would choose for a heat exchanger. The second method, and the one chosen by Secan for their high performance coolers, is called "bar and plate". In this constuction the tubes are made by brazing thin plates (thus reducing the thermal resistance) into tubes of the correct size for the given application. The cooling fins - to add area - are then added later. The important point here is that the core is specifically designed for the application. So, you need to know the revs, boost, temperatue, capacity and range of the engine (all easy) but you also need to know the volume and temperature of the cooling air going across the cooler. The latter is much more difficult to calculate. Once all these are known it is down to some thermodynamics to calculate the intercooler. Nusselt, Prandtl and all sorts of other interesting things come into play here!!!!
I used to use some old, heavily modified, lorry cores for my race car. They were of tube and fin construction and worked quite well. However, they got old and damaged and I thought I would change them. Secan were the obvious choice but are stupidly expensive. Secan are a French aerospace company and make coolers for Eurocopter I believe. I looked around and found other manufacturers making "identical" things for other military applications and had my cores made by them. They looked beautiful but really that does not matter. I made up the tanks and fitted them. The proof is how they work and so far the cooled air is around 15 centigrade lower than with my old units measured under conditions that are the same as possible as before. (i.e same ambient, same boost and same compressed temperature). The engine is slightly different this year but I am happy they work well. I can't tell you if they are as good as, or better than, the equivilant Secan core but on the other hand I cannot see why they should be worse - afterall the technology they use is the same and the workmanship looks excellent so the end result is probably similar. Material choice is sensible for the application (I doubt Secan would use anything different) and the really good news is that they are considerably cheaper. Deliveries are a problem though as most of their production goes on Eurofighter - ah let the cat out of the bag there!
If you want to know more email me.
Keep cool in this lovely UK weather,
Richard.

Jussi

Thanks for professional comments.
Can you tell that core supplier?
I like to do also better IC from those.
PM or email or reply here

Jean

Ahemm...Marston? Heard that name before

935racer

Yes, they are Marston. However, I reiterate that the biggest benifit of bar and plate construction is that they can be optimised for the specific application. Thus a core that is suitable for my car may not have the optimal internal dimensions for another car even if the outside dimensions are the same! That is why "off the shelf" intercoolers must be a compromise no matter what their construction unless the engines / installations are all identical. That said, I suspect a bar and plate cooler will outperform a tune and fin one almost every time.
Nothing is simple when you are trying to extract the last few percent of performance!!!
Richard.

Jean

Richard, and I think that is the morale of the story! Which intercooler is better is too broad a question. Some cores are better than others that's for sure, but the meat is the testing and build sheet.

Core alloys used, density, size, flow, pattern etc.. they all work together, this is why I find it hilarious when I see all this theory behind how "restrictive" this or that end tank looks like, or how nice and smooth this intercooler looks, great welding, etc.. even bench flow testing does not make any sense in itself....Who is testing pressure drops in a racing environment amongst these $2k intercooler suppliers... Great BS marketing.

The success of the likes of Secan in Motorsports and more specifically with Porsche, is the experience and data that some people have brought behind it throughout the few years of racing the aircooled GT2s by independent houses and getting some backing from the factory who supported the efforts of these racing houses.

When there is someone like yourself willing to put so much effort and time (= money) researching and testing to find the best setup both on the engine dyno and in real life racing, everyone else benefits.. Unfortunately in this case, you run two intercoolers ala 996TT, so not much people with a top mounted intercooler can benefit from... We need to convince you to design one for us heat challenged individuals

Thanks for sharing.

935racer

Jean, I think you are right about all the BS that abounds. Trouble is the Porsche world seems to attract this in vast quantities!!!
As for a top mounted intercooler a la 993 I don't even have any initial designs. However, some people may know that I have built an engine for a chap who does the V max thing at Bruntingthorpe. He has a 964 and he wants to break 200 mph on a "standard" aero car!! So far we have only achieved 193 so there is a way to go. The reason I mention this is because I am designing a single cooler to mount over the engine for his car. Its a single turbo so it won't be the same as a 993 but the results should be extremely useful if I ever do a "real" 993 version.
Anyway, back to coolers. If you want a really good cooler my advice would be to talk to someone who has done it and can prove the results. Don't be taken in by off the shelf products as they cannot and will not work optimally and if you are contemplating this sort of money surely you will want the best possible. If this boils down to Secan then so be it - just save up the pennies first.
Good luck,
Richard.

TB993tt

Richard, thanks for sharing your expertise and results

Whilst I understand the "off the shelf" vs "carefully matched using measurements and calculation" argument is it not likely that for example the cores which Richard has developed to run on his 935 racer would be effective in Jussi's 650hp engine also (if one could get the packaging to work) since both are air cooled tts, or is it more complicated than that, does one have to take into account the air volume curve per rpm and stuff like that ?

935racer

Toby, I think the answer to your question is yes, they will show an improvement. My coolers are a bad example as to be really good they are too thick (an ideal cooler would be very thin but still able to pass enough compressed air to support the engine) but my point here is that when you are spending a lot of money (be it on Secan, Marston or others) it is best to get the most performance possible and therefore each cooler should be designed specifically. I had to have "thick coolers" as I have a special packaging constraint so it goes to show that real world issues have to take priority over theoritcal improvements! In real life I doubt if we would notice much difference as the odd degree here and there is well within normal experimental accuracy!!!!
Obviously a cooler can only be designed to be optimal at a particular point. On my car I chose this to be flat out at maximum boost but if you have a road car you may decide to compromise some of this to obtain better performance (i.e. cooler air) at lower throttle. Afterall, a road engine spends much of its time at less than 40% throttle. Of course, if you are upgrading the cooler you may want the optimum performance at 100% throttle - its up to you!
What I have proven to myself is that there is a definate and repeatable improvement in performance by using coolers of bar and plate construction. This should not be a suprise as theory points us in this direction but its always good when theory and practice coincide!
I saw another debate on water to air coolers. Fundamentally, these will be better than air to air as water is a better thermal conductor but then there are all the downsides of weight and complexity - especially on an air cooled car! If I had water in mine I would use them, as I don't the penalty of adding all the systems to support them is not worth the extra efficiency thay would give. (All in my opinion I hasten to add!!!).
Cheers,
Richard.

sfl993t4

I agree that air/water coolers can be better from a packaging point of view, but I don't think that there can be any gain in fundamental efficiency if the heat has to be transferred back to the ambient air through a second water/air cooler somewhere else.
I guess that the exception would be a situation in which you didn't need to transfer the heat back to the air, for example in a boat installation, or in a short run/drag run with a good supply of water or an ice chest or similar.
Steve