Contradicting Intercooler Designs
#1
Elite Member
Thread Starter
iTrader: (16)
Join Date: Aug 2007
Location: Houston, TX
Posts: 9,339
Total Cats: 519
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Guys I realize there are many threads about intercoolers. I've read and searched and just have a couple questions. I've got a 99 NB that I'm putting a 1 Liter twin screw supercharger on and it's gonna be intercooled. I'm shooting for 300 WHP with the blower spinning between 16-18K. I've read Corky Bell's book 'Supercharged' and it includes a chapter on intercooling. He suggest channel lengths of 6, 8, 10, and 12 inches. He says anything longer is just increasing drag and not helping to cool the charge as much, which makes since. I understand his logic and explanations. He suggest increasing the number of channels as opposed to increasing the lengh of them. This will increase internal flow area while. He also points out long tubes inherently have more drag than short tubes.
Anyway, I've Read Braineak's Faq's on intercooling and he's suggesting 18-24 inch channel lengths. This contradicts what Bell suggest. Braineack seems to agree with the idea that overly long runners are'nt the answer, but his numbers of ideal are twice that of Bells. It seems most people on this forum follow Braineaks numbers and there are IC's on Ebay like this cheap.
I'm wondering if I'm missing something. I suppose backpressure on a turbo isn't costing you as much HP as it does on a SC, so perhaps that's part of it. I understand the Ebay IC's are cheap too, so that's also a factor. Are these the reasons most use 24"+ channel lengths?
I'm wanting to design and build my own intercooler, sort of. The plan is to purchase the core, and build the endtanks and weld it all together. My old machine shop teacher can weld aluminum, so I've got a good welder. That shouldn't be a problem.
Based on my HP goals and using Corky's numbers from his book, I decided on IC core dimensions of 8" channel lengths running vertically, 22 inches wide with a 3 inch thick core. Bell's formulas suggest this will support 565 CFM with less than 1 PSI of pressure drop through the IC. This will fit in my car, but the endtanks will be mounted on top and bottom, increasing it's total height from 8" to something between 13-15 inches, depending on end cap design.
I was almost done making a scaled drawing when I realized the total height of my IC may be a problem. I think I can make the IC 13 inches tall and still keep the 8" channels. Fromt he measurements I took without removing the bumper or anything else, I have 13-14 inches of height to work with. Does this sound about right?
Thanks.
Edit: Here's what I came up with. Mounted off center it would measure 12" tall.
Anyway, I've Read Braineak's Faq's on intercooling and he's suggesting 18-24 inch channel lengths. This contradicts what Bell suggest. Braineack seems to agree with the idea that overly long runners are'nt the answer, but his numbers of ideal are twice that of Bells. It seems most people on this forum follow Braineaks numbers and there are IC's on Ebay like this cheap.
I'm wondering if I'm missing something. I suppose backpressure on a turbo isn't costing you as much HP as it does on a SC, so perhaps that's part of it. I understand the Ebay IC's are cheap too, so that's also a factor. Are these the reasons most use 24"+ channel lengths?
I'm wanting to design and build my own intercooler, sort of. The plan is to purchase the core, and build the endtanks and weld it all together. My old machine shop teacher can weld aluminum, so I've got a good welder. That shouldn't be a problem.
Based on my HP goals and using Corky's numbers from his book, I decided on IC core dimensions of 8" channel lengths running vertically, 22 inches wide with a 3 inch thick core. Bell's formulas suggest this will support 565 CFM with less than 1 PSI of pressure drop through the IC. This will fit in my car, but the endtanks will be mounted on top and bottom, increasing it's total height from 8" to something between 13-15 inches, depending on end cap design.
I was almost done making a scaled drawing when I realized the total height of my IC may be a problem. I think I can make the IC 13 inches tall and still keep the 8" channels. Fromt he measurements I took without removing the bumper or anything else, I have 13-14 inches of height to work with. Does this sound about right?
Thanks.
Edit: Here's what I came up with. Mounted off center it would measure 12" tall.
Last edited by olderguy; 11-09-2007 at 09:11 AM.
#2
Boost Czar
![](/mt2011/images_td/ranks/supporter1.gif)
![](/mt2011/images_td/ranks/admin.png)
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,688
Total Cats: 4,113
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
corky's new revision will have those numbers. read any of corky's new posts over at m.net. they are straight from his word of mouth.
he suggests, if as long as there is ample flow area, a core with 12" runners will be around 80% efficient, 18-20" 90%.
he also doesn't do vertical runners for a reason now too.
he suggests, if as long as there is ample flow area, a core with 12" runners will be around 80% efficient, 18-20" 90%.
he also doesn't do vertical runners for a reason now too.
#4
Elite Member
Thread Starter
iTrader: (16)
Join Date: Aug 2007
Location: Houston, TX
Posts: 9,339
Total Cats: 519
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
corky's new revision will have those numbers. read any of corky's new posts over at m.net. they are straight from his word of mouth.
he suggests, if as long as there is ample flow area, a core with 12" runners will be around 80% efficient, 18-20" 90%.
he also doesn't do vertical runners for a reason now too.
he suggests, if as long as there is ample flow area, a core with 12" runners will be around 80% efficient, 18-20" 90%.
he also doesn't do vertical runners for a reason now too.
![](https://www.miataturbo.net/images/smilies/confused.gif)
Braineak, your saying the numbers in your faq are from Bell, correct? Also, what's the reason for not having vertical runners? I buy a book to learn and he goes and changes this
![ugh](https://www.miataturbo.net/images/smilies/ugh2.gif)
#5
Boost Czar
![](/mt2011/images_td/ranks/supporter1.gif)
![](/mt2011/images_td/ranks/admin.png)
iTrader: (62)
Join Date: May 2005
Location: Chantilly, VA
Posts: 79,688
Total Cats: 4,113
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
you forget when he published Maximum Boost dontcha....1997.....
yes. search Corky Bell as a username and intercooler as a search term, he has a lot to say about them.
he switched to verticals because he realized that the runner length is more important than keeping the pressure loss to a minimum, because the longer the runner (until reaching diminishing returns) the better it cools. A bunch of short runners will never fully cool the charge as well as less long runners....something like that.
yes. search Corky Bell as a username and intercooler as a search term, he has a lot to say about them.
he switched to verticals because he realized that the runner length is more important than keeping the pressure loss to a minimum, because the longer the runner (until reaching diminishing returns) the better it cools. A bunch of short runners will never fully cool the charge as well as less long runners....something like that.
![wink](https://www.miataturbo.net/images/smilies/wink.gif)
Last edited by Braineack; 11-09-2007 at 08:51 AM.
#7
Elite Member
Thread Starter
iTrader: (16)
Join Date: Aug 2007
Location: Houston, TX
Posts: 9,339
Total Cats: 519
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
Ok, I searched as Braineak suggested and indeed, Bell now suggest 18-24" charge tubes on various post on m.net over the last year. I wish I had known the IC section of his book was under development and due for a revision before I purchased it. Oh well. Overall I learned a lot so I can't sweat it too much.
I'll keep doing research. Thanks for setting me straight guys.
Savington, I said 300 whp, not one thousand
I'll keep doing research. Thanks for setting me straight guys.
Savington, I said 300 whp, not one thousand
![](https://www.miataturbo.net/images/smilies/smile.gif)
#10
Tour de Franzia
![](/mt2011/images_td/ranks/supporter1.gif)
iTrader: (6)
Join Date: Jun 2006
Location: Republic of Dallas
Posts: 29,085
Total Cats: 375
![Default](https://www.miataturbo.net/images/icons/icon1.gif)
yes, but don't forget rpm based ignition retard. Its also important to remember that you can't use pc-pro with megasquirt (which doesn't use all the stock sensors) because it cools so efficiently that it will flop over to the cold start map from frosting the head and lower water temps.
Thread
Thread Starter
Forum
Replies
Last Post
stoves
Suspension, Brakes, Drivetrain
5
04-21-2016 03:00 PM