Intercooler: horizontal or vertical core

[gui67]

Did someone already made a comparison between intercoolers with horizontal core ( smaller sectional area and longer exchanger) vs vertical core (short exchanger but with bigger sectional area)
here are some picture to help you understand what I mean:
vertical:
46898
horizontal:
46899

for the same frontal area, the vertical would be better because it causes less pressure losses (lower air speed in the channels and less length.

But in the real life, on my 67 camaro, I only have limited space and need to put one pipe on the side to go to the upper tank for a vertical core, which means less frontal area.
I made some calculations using the space I have on my car:

the horizontal core would have 3 times more pressure losses then the vertical one, but also 48% more frontal area.

here are some details about the data (for 6000RPM, 6.3l, 85%volumetric efficiency):
horizontal Vertical
air speed: 15.9m/s 12.5m/s
core length: 470mm 250mm
frontal area: 0.21m2 0.14m2

Is there a rule for maximum air speed in the core in order to dimension it?
Did someone already made some comparison of 2 solutions in the same car and saw a difference?
The pressure drop from horizontal core could be overcome with more pressure from turbocharger, but that again creates more heat...

What is a common pressure loss in intercoolers, maybe it is so small that the 3 factor in this case would change nothing:hammer:

[DeltaT]

Look at the cfm ratings delta between the same (almost exact) size core horizontal or turned vertical:

Using the core ratings at: http://bellintercoolers.com/pages/AAMain.html

Horizontal orientation (charge air moving across the IC, through maybe 10-12 channels): 3.00" Deep, 6.20 Height, 18.00 Width, 279cfm
Vertical orientation (charge air moving up or downwards through the IC, maybe 30-32 channels): 3.00" Deep, 17.90" Width, 6.00" Height, 1033 cfm!

Both have an almost identical frontal area, but the internal charge area is much less on the horizontally oriented IC.

An acceptable pressure loss is generally <1 psi. More is manageable but not desirable.

That being said, there are people successfully making big power through a horizontal IC. They would just be making bigger power (or the same power at a lower boost level) with a more efficient IC. You can help a bit by making the IC thicker but it is really the front half of the IC that does most (~75%) of the work, and the aire gets warmer and warmer as it travels through picking up heat.

Jim

[gui67]

thank you for the link, Jim

I compared the cfm given by bell for the 2 solutions I have:
765CFM for the horizontal core and 1184CFM for the vertical one, even with 33% less frontal area.

Even if the CFM rating is lower, I still think the horizontal one will cool better (because of the larger area) but cause much more pressure losses. the CFM rating must be done for a given pressure drop (certainly 1PSI), and not according to thermal performances.

How are CFM calculated for a turbo engine: I assume it is air at atmospheric pressure on the suction side of compressor that is measured.
For my 383 engine, 6000RPM, 15PSI boost, assuming 85% volumetric efficiency, I get 1091CFM, does this sound right?

[DeltaT]

IC's work best when the temp differential is at its greatest, which is the point where the charge air enters the IC. With good air throughput, and excellent charge air distribution to all parts of the IC core (in your 2nd picture, the IC really would need an air splitter inside the inlet side to avoid having the bottom section of the IC do most of the work. The air will not naturally want to make that turn and go through the top section.), the vertical one will still shed more total heat. The horizontal one may shed more heat for the air it passes, but since it passes a lot less air, the total heat rejection should be lower. To say it a different way, the degrees shed per cfm will be lower.

Your cfm calc seems about right. I have the equation if you need me to check.

Sounds like you have some height to put the IC in, how come you can't go with vertical (or lots of short) channels? The turbo Buick guys use a vertical style IC turned sideways and put the inlet and outlet coming up under the radiator. Check out Cotton's setup, I think he has a picture on his site.

Jim

[67 ls1 vert]

just read in my instruction book from ATI today that horizontal better.

[DeltaT]

That's interesting because their site shows that they mostly sell IC's with vertically-oriented tubes:

https://static1.pt-content.com/images/noimg.gif

https://static1.pt-content.com/images/noimg.gif

Jim

[gui67]

I will defitively going to use a vertical core.
I think I will manage to avoid a vertical tube going to the the top tank: it should be possible to pass between the radiator and the passenger side light door when opened, if I remove the battery (it will go to the trunk).
I should then be able to use a core with the width of the radiator.

[gui67]

I was thinking about using this one:
https://static1.pt-content.com/images/noimg.gif

- Core Dimension
3.50'' thick x 18.48'' wide x 9.0'' tall
- Hose Connection – 2.50'' OD
760HP

It should be gig enough for my application. I don't think I will ever need more then 750 HP, and I already have water injection for more cooling.
The only problem is that I was looking for at least 3" inlet and outlet, 2.5" seems a bit small

[DeltaT]

Yes, I agree, try to find one w/3" inlet/outlet.

Jim

[gui67]

Yes, I already did: treadstone performance has the TRV25 and TRV259 with 3" inlet and outlet at a better price.
And the core has good comments on forums about quality and heat transfert.
TRV25 (http://www.treadstoneperformance.com/product.phtml?p=34&cat_key=63&prodname=TRV25+Series+Intercooler)
TRV259 (http://www.treadstoneperformance.com/product.phtml?p=485&cat_key=63&prodname=TRV259+Series+Intercooler)

I will check that I have enough place for the 9" long and order it if it fits

[gui67]

Next question:
these intercoolers are available with the inlet and outlet either on the same side or opposite side.
https://static1.pt-content.com/images/pt/2011/06/img4850large-1.jpg

https://static1.pt-content.com/images/pt/2011/06/img4852large-1.jpg

Intuitively, I would say that the IC with the opposite side inlet and outlet would have a better distribution of air inside the core and a better efficiency.
Did someone already have experience of comparing both or made a numeric simulation? how different are both solutions in efficiency? Is it negligeable or does it have a big impact?

on my car, the easiest solution would be to have inlet and outlet on the same side, but I could also make the piping for the other solution, it would just be a little more complicated

[gui67]

No-one has a comment on tank orientation? or is there no effect on internal flow?

I made some fake IC in cardboard to check position ans interference in the camaro: the9" heidh could be OK, but very tight, I will go for the 6", it is rated for 1000HP: enough for my application

[gui67]

I received my intercooler today.
I don't think there will be a problem to make it fit in the car, and it should be big enough for my power goal.
Some pictures (the scale is in cm not inch)
https://static1.pt-content.com/images/pt/2011/08/largeurJPG-1.jpg

https://static1.pt-content.com/images/pt/2011/08/hauteurJPG-1.jpg

https://static1.pt-content.com/images/pt/2011/08/entree_airJPG-1.jpg
internal fins:
https://static1.pt-content.com/images/pt/2011/08/faisceau_intJPG-1.jpg
external fins:
https://static1.pt-content.com/images/pt/2011/08/faisceau_extJPG-1.jpg

[DeltaT]

Looks good - I haven't been around for a while - but you can make a long IC like that more efficient by putting an air splitter in the inlet side that forces ~1/2 the flow farther down the inlet tank to the second half of the length of the IC. Not needed on the output side. It wouldn't be a hard mod even with end tank on.

Jim

[gui67]

That would make the flow inside more even.
You mean by putting a piece of sheet-metal inside that divides the inlet in 2, right? how would you be sure it does not move after ordoes not get pushed by airflow?

[WS6]

The psi inside the intercooler will be basically the same through out the intercooler. So there will be no forces pushing on one side of the splitter that aren't present on the other side. Weld the splitter into the inlet and let it extend into the intercooler a little and call it good. Maybe even put a slight down curvature to the splitter. 1/16 to 3/32 thick aluminum should work just fine.