Buying an intercooler
#1
Buying an intercooler
I was planning on using AI to chemically cool thr FC, but for some reason I now want to keep things simple. A off the shelve intercooler kit will not work on my application, as my setup is out of the ordinary.
When looking for an intercooler, is bigger better? or is there a formula for determining the "proper size"?
When looking for an intercooler, is bigger better? or is there a formula for determining the "proper size"?
#5
Trending Topics
#8
I was planning on using AI to chemically cool thr FC, but for some reason I now want to keep things simple. A off the shelve intercooler kit will not work on my application, as my setup is out of the ordinary.
When looking for an intercooler, is bigger better? or is there a formula for determining the "proper size"?
When looking for an intercooler, is bigger better? or is there a formula for determining the "proper size"?
Bell Intercoolers is a really good resource when thinking about a custom IC. I spoke with them when I was doing my custom V-mount.
#10
Then again, ALKY looks so much easier
BTW - I've been looking at the bellintercoolers site, cool place
A core of 24 might be at the limit, depending on the size of the endtanks, and 4.5 is too thick for the location, I believe 3 inch is the max. You build intercoolers, correct?
#12
There is alot to sizing a charge air cooler for turbocharged vehicles. I am an application engineer for a radiator/charge air manufactuer for all diesel and gas generators in the power generation field. Sizing a charge air cooler has two vital requirments.
1. Turbocharger compressor map
2. IMT(Intake manifold temp)
Those are the two criteria we size for, as well as ambient conditions. When given the compressor map we can then see it's optimium peak and mid range of flow as well as pressure the turbo will be outputting. At that rate after taking those variables, we then are asked what tempertaure will the turbo be operating at, at its peak temp and boost.
The intake manifold temp is very important for several reasons, the EPA requires a certain temp to meet its certification at 86F day. There for as an example some of the certified Caterpillar enignes require a 122F @ 86F day to return back into the enigne. In our case we don't have to worry about that.
What we need is to bring the temperature down as far as we can. Not really knowing this temp, we are left with a void. But the only thing we could compare and contrast is to see the IMT on the stock charge air setup with the twin turbos. If someone can help on getting this info I can help people size cores.
Another item for sizing cores is the effiecncy of the fin design. You can have a "louvered" design which is small slits in the fin to slow the air down, or you can have a "dimpled" design whihc only allows dimples or rivets in the fin and will not slow air as fast there for not allowing a complete fast cool. The main ingridient when looking at a core is knowing the fin to tube bond ratio or percentage. Most of us wouldnt have a clue as we are not taking cores apart. But the fins bonded to the tubes is what the true heat sinks as the removal.
One last note another major contributor is pressure drop, on a FMIC there will be a higher pressure drop then the V-mount, since it is further away, the pressure drops and there for you lose that psi or so.
I hope this clarifies alot when people talk about well buy the biggest baddest core...that is not nesscarly the truth and can yield more problems then advantages.
Amir
1. Turbocharger compressor map
2. IMT(Intake manifold temp)
Those are the two criteria we size for, as well as ambient conditions. When given the compressor map we can then see it's optimium peak and mid range of flow as well as pressure the turbo will be outputting. At that rate after taking those variables, we then are asked what tempertaure will the turbo be operating at, at its peak temp and boost.
The intake manifold temp is very important for several reasons, the EPA requires a certain temp to meet its certification at 86F day. There for as an example some of the certified Caterpillar enignes require a 122F @ 86F day to return back into the enigne. In our case we don't have to worry about that.
What we need is to bring the temperature down as far as we can. Not really knowing this temp, we are left with a void. But the only thing we could compare and contrast is to see the IMT on the stock charge air setup with the twin turbos. If someone can help on getting this info I can help people size cores.
Another item for sizing cores is the effiecncy of the fin design. You can have a "louvered" design which is small slits in the fin to slow the air down, or you can have a "dimpled" design whihc only allows dimples or rivets in the fin and will not slow air as fast there for not allowing a complete fast cool. The main ingridient when looking at a core is knowing the fin to tube bond ratio or percentage. Most of us wouldnt have a clue as we are not taking cores apart. But the fins bonded to the tubes is what the true heat sinks as the removal.
One last note another major contributor is pressure drop, on a FMIC there will be a higher pressure drop then the V-mount, since it is further away, the pressure drops and there for you lose that psi or so.
I hope this clarifies alot when people talk about well buy the biggest baddest core...that is not nesscarly the truth and can yield more problems then advantages.
Amir
#13
I would look at staying 18" or less in length in your application I try and stay under 20" if I can for flow reasons. If you can fit 13.5" height that would go well. 3" 3.5" in width.
Most cooling is done in the first half of the core adding length has some merit in getting the coolest charge air but at a cost of flow. If it is assumed the core is a "heatsink" then thicker would work better than longer for pulling out more heat and not effecting flow as much.
Here's a example,
3x12x18 flow is rated at 550cfm
go longer
3x12x24 flow is rated at 490cfm
go thicker
3.5x12x18 flow is 640cfm
thicker still
4.5x12x18 flow is 820cfm
Stepping this core up to a 24" length and you lose roughly 100cfm.
Keep in mind I am not saying going to a deeper core is better than a longer core, just showing the tradeoff. Turbo systems are nothing BUT give and take.
I think there is a bigger issue. There is no "standard" for measurement. ie Spearco will rate their cores higher than say bell, or garrett. I don't like this it can be misleading.
Here is the comparison, for a customer a while back.
Spearco 3"W x 7.125"H x 18.5"L 670CFM
Bell 3.5"W x 12.1"H x 18"L 635CFM
Funny how the much smaller spearco core flows more. Truth is it doesn't really.
There is not a difference in flow-capacity if sizes were equal (spearco's FPI ratio isn't that far off of bells to make that profound a difference), but a difference in the basis of comparison. One uses lower boost for its data and takes a more conservative approach, the other uses higher boost thus higher flow numbers as well as a higher acceptable pressure loss for the given size. I would say a acceptable pressure loss is roughly 10% of max boost. ie 10psi drops 1psi, 20psi drops 2psi, 30psi drops 3psi.
Taking the more conservative approach you have less of a chance of under sizing your core.
I would also say it is not a bad idea to bell mouth your charge air side of the core with some small half round. it can make a huge difference in flow and help you pick some of it back up from stepping up to a longer core. Also it isn't a bad idea to step IC outlet diameter down a notch from inlet diameter, (around 15% less cross section is good) because of the lower volume of flow at the outlet versus the IC inlet.
As example 3" inlet pipe and a 2.75 " outlet pipe is about a 16% reduction. This will help more than hurt flow in most applications.
Hope that info is helpful to someone
Most cooling is done in the first half of the core adding length has some merit in getting the coolest charge air but at a cost of flow. If it is assumed the core is a "heatsink" then thicker would work better than longer for pulling out more heat and not effecting flow as much.
Here's a example,
3x12x18 flow is rated at 550cfm
go longer
3x12x24 flow is rated at 490cfm
go thicker
3.5x12x18 flow is 640cfm
thicker still
4.5x12x18 flow is 820cfm
Stepping this core up to a 24" length and you lose roughly 100cfm.
Keep in mind I am not saying going to a deeper core is better than a longer core, just showing the tradeoff. Turbo systems are nothing BUT give and take.
I think there is a bigger issue. There is no "standard" for measurement. ie Spearco will rate their cores higher than say bell, or garrett. I don't like this it can be misleading.
Here is the comparison, for a customer a while back.
Spearco 3"W x 7.125"H x 18.5"L 670CFM
Bell 3.5"W x 12.1"H x 18"L 635CFM
Funny how the much smaller spearco core flows more. Truth is it doesn't really.
There is not a difference in flow-capacity if sizes were equal (spearco's FPI ratio isn't that far off of bells to make that profound a difference), but a difference in the basis of comparison. One uses lower boost for its data and takes a more conservative approach, the other uses higher boost thus higher flow numbers as well as a higher acceptable pressure loss for the given size. I would say a acceptable pressure loss is roughly 10% of max boost. ie 10psi drops 1psi, 20psi drops 2psi, 30psi drops 3psi.
Taking the more conservative approach you have less of a chance of under sizing your core.
I would also say it is not a bad idea to bell mouth your charge air side of the core with some small half round. it can make a huge difference in flow and help you pick some of it back up from stepping up to a longer core. Also it isn't a bad idea to step IC outlet diameter down a notch from inlet diameter, (around 15% less cross section is good) because of the lower volume of flow at the outlet versus the IC inlet.
As example 3" inlet pipe and a 2.75 " outlet pipe is about a 16% reduction. This will help more than hurt flow in most applications.
Hope that info is helpful to someone
#14
we can talk about how a core flows, but it still doesn't yield the truth in sizing up the appropriate core for the appropriate turbocharger. Again its not just about flow. Those flows given, what was the pressure drop? Was that the flow through each core? there are plenty of tube designs which will not nesscarly yield you the flow rates you have provided per the sizes. We don't know the tube designs hence if we did , we would know the pressure drops and the accurate flow rates. I can bet 10 to 1 that these flows are at optimal with less to none on pressure drop. I think your flow rates vs size is a needle in a haystack still. no one here would know exactly the true math behind the accurate sizing. The only way you would know is if you knew the "exact" tube design and at that rate the air flow can be stated.
But no joe schmo is going to buy charge air cooler X from vender Y and get the same flow rates as you had stated becasue of the certain tube designs..
My .02 cents
By the way the manufactuer I work for is in the process of looking to providing charge air coolers for the aftermarket world, as we are the leading provider for Caterpillar, Cummins, MTU, Jenbacker, John Deere, Kohler and other diesel/gas engine manufactuers.
But no joe schmo is going to buy charge air cooler X from vender Y and get the same flow rates as you had stated becasue of the certain tube designs..
My .02 cents
By the way the manufactuer I work for is in the process of looking to providing charge air coolers for the aftermarket world, as we are the leading provider for Caterpillar, Cummins, MTU, Jenbacker, John Deere, Kohler and other diesel/gas engine manufactuers.
#15
Please step back take a breath and relax, your E-***** is secure no worries. Those flow rates are at 10psi with a PL of roughly 1psi tested by the manufacturer. So 640CFM is 640CFM at 10psi with a measured pressure loss of roughly 1psi through the core. Joe schmo now knows what he needs to know as far as core flow goes (ROUGHLY) He now needs to start piecing together the rest of the puzzle. As I clearly stated I was showing the trade off of going one way with size versus the other. Not whether X size was more appropriate for Y application.
~S~
~S~
Last edited by Zero R; 02-05-09 at 12:49 PM.
#17
There is alot to sizing a charge air cooler for turbocharged vehicles. I am an application engineer for a radiator/charge air manufactuer for all diesel and gas generators in the power generation field. Sizing a charge air cooler has two vital requirments.
1. Turbocharger compressor map
2. IMT(Intake manifold temp)
Those are the two criteria we size for, as well as ambient conditions. When given the compressor map we can then see it's optimium peak and mid range of flow as well as pressure the turbo will be outputting. At that rate after taking those variables, we then are asked what tempertaure will the turbo be operating at, at its peak temp and boost.
The intake manifold temp is very important for several reasons, the EPA requires a certain temp to meet its certification at 86F day. There for as an example some of the certified Caterpillar enignes require a 122F @ 86F day to return back into the enigne. In our case we don't have to worry about that.
What we need is to bring the temperature down as far as we can. Not really knowing this temp, we are left with a void. But the only thing we could compare and contrast is to see the IMT on the stock charge air setup with the twin turbos. If someone can help on getting this info I can help people size cores.
Another item for sizing cores is the effiecncy of the fin design. You can have a "louvered" design which is small slits in the fin to slow the air down, or you can have a "dimpled" design whihc only allows dimples or rivets in the fin and will not slow air as fast there for not allowing a complete fast cool. The main ingridient when looking at a core is knowing the fin to tube bond ratio or percentage. Most of us wouldnt have a clue as we are not taking cores apart. But the fins bonded to the tubes is what the true heat sinks as the removal.
One last note another major contributor is pressure drop, on a FMIC there will be a higher pressure drop then the V-mount, since it is further away, the pressure drops and there for you lose that psi or so.
I hope this clarifies alot when people talk about well buy the biggest baddest core...that is not nesscarly the truth and can yield more problems then advantages.
Amir
1. Turbocharger compressor map
2. IMT(Intake manifold temp)
Those are the two criteria we size for, as well as ambient conditions. When given the compressor map we can then see it's optimium peak and mid range of flow as well as pressure the turbo will be outputting. At that rate after taking those variables, we then are asked what tempertaure will the turbo be operating at, at its peak temp and boost.
The intake manifold temp is very important for several reasons, the EPA requires a certain temp to meet its certification at 86F day. There for as an example some of the certified Caterpillar enignes require a 122F @ 86F day to return back into the enigne. In our case we don't have to worry about that.
What we need is to bring the temperature down as far as we can. Not really knowing this temp, we are left with a void. But the only thing we could compare and contrast is to see the IMT on the stock charge air setup with the twin turbos. If someone can help on getting this info I can help people size cores.
Another item for sizing cores is the effiecncy of the fin design. You can have a "louvered" design which is small slits in the fin to slow the air down, or you can have a "dimpled" design whihc only allows dimples or rivets in the fin and will not slow air as fast there for not allowing a complete fast cool. The main ingridient when looking at a core is knowing the fin to tube bond ratio or percentage. Most of us wouldnt have a clue as we are not taking cores apart. But the fins bonded to the tubes is what the true heat sinks as the removal.
One last note another major contributor is pressure drop, on a FMIC there will be a higher pressure drop then the V-mount, since it is further away, the pressure drops and there for you lose that psi or so.
I hope this clarifies alot when people talk about well buy the biggest baddest core...that is not nesscarly the truth and can yield more problems then advantages.
Amir
As far as IAT, at what point should they be measured? under boost? under vac?
I would look at staying 18" or less in length in your application I try and stay under 20" if I can for flow reasons. If you can fit 13.5" height that would go well. 3" 3.5" in width.
Most cooling is done in the first half of the core adding length has some merit in getting the coolest charge air but at a cost of flow. If it is assumed the core is a "heatsink" then thicker would work better than longer for pulling out more heat and not effecting flow as much.
Here's a example,
3x12x18 flow is rated at 550cfm
go longer
3x12x24 flow is rated at 490cfm
go thicker
3.5x12x18 flow is 640cfm
thicker still
4.5x12x18 flow is 820cfm
Stepping this core up to a 24" length and you lose roughly 100cfm.
Keep in mind I am not saying going to a deeper core is better than a longer core, just showing the tradeoff. Turbo systems are nothing BUT give and take.
I think there is a bigger issue. There is no "standard" for measurement. ie Spearco will rate their cores higher than say bell, or garrett. I don't like this it can be misleading.
Here is the comparison, for a customer a while back.
Spearco 3"W x 7.125"H x 18.5"L 670CFM
Bell 3.5"W x 12.1"H x 18"L 635CFM
Funny how the much smaller spearco core flows more. Truth is it doesn't really.
There is not a difference in flow-capacity if sizes were equal (spearco's FPI ratio isn't that far off of bells to make that profound a difference), but a difference in the basis of comparison. One uses lower boost for its data and takes a more conservative approach, the other uses higher boost thus higher flow numbers as well as a higher acceptable pressure loss for the given size. I would say a acceptable pressure loss is roughly 10% of max boost. ie 10psi drops 1psi, 20psi drops 2psi, 30psi drops 3psi.
Taking the more conservative approach you have less of a chance of under sizing your core.
I would also say it is not a bad idea to bell mouth your charge air side of the core with some small half round. it can make a huge difference in flow and help you pick some of it back up from stepping up to a longer core. Also it isn't a bad idea to step IC outlet diameter down a notch from inlet diameter, (around 15% less cross section is good) because of the lower volume of flow at the outlet versus the IC inlet.
As example 3" inlet pipe and a 2.75 " outlet pipe is about a 16% reduction. This will help more than hurt flow in most applications.
Hope that info is helpful to someone
Most cooling is done in the first half of the core adding length has some merit in getting the coolest charge air but at a cost of flow. If it is assumed the core is a "heatsink" then thicker would work better than longer for pulling out more heat and not effecting flow as much.
Here's a example,
3x12x18 flow is rated at 550cfm
go longer
3x12x24 flow is rated at 490cfm
go thicker
3.5x12x18 flow is 640cfm
thicker still
4.5x12x18 flow is 820cfm
Stepping this core up to a 24" length and you lose roughly 100cfm.
Keep in mind I am not saying going to a deeper core is better than a longer core, just showing the tradeoff. Turbo systems are nothing BUT give and take.
I think there is a bigger issue. There is no "standard" for measurement. ie Spearco will rate their cores higher than say bell, or garrett. I don't like this it can be misleading.
Here is the comparison, for a customer a while back.
Spearco 3"W x 7.125"H x 18.5"L 670CFM
Bell 3.5"W x 12.1"H x 18"L 635CFM
Funny how the much smaller spearco core flows more. Truth is it doesn't really.
There is not a difference in flow-capacity if sizes were equal (spearco's FPI ratio isn't that far off of bells to make that profound a difference), but a difference in the basis of comparison. One uses lower boost for its data and takes a more conservative approach, the other uses higher boost thus higher flow numbers as well as a higher acceptable pressure loss for the given size. I would say a acceptable pressure loss is roughly 10% of max boost. ie 10psi drops 1psi, 20psi drops 2psi, 30psi drops 3psi.
Taking the more conservative approach you have less of a chance of under sizing your core.
I would also say it is not a bad idea to bell mouth your charge air side of the core with some small half round. it can make a huge difference in flow and help you pick some of it back up from stepping up to a longer core. Also it isn't a bad idea to step IC outlet diameter down a notch from inlet diameter, (around 15% less cross section is good) because of the lower volume of flow at the outlet versus the IC inlet.
As example 3" inlet pipe and a 2.75 " outlet pipe is about a 16% reduction. This will help more than hurt flow in most applications.
Hope that info is helpful to someone
I will measure the depth, hopefully I can go with a 3.5 (I seriously doubt that a 4inch will fit). What do you mean by "bell mouth your charge air side of the core with some small half round?"
BTW - Thanks for the help!
#18
EDIT: Too much work. Form erased white space in front of each line of "art" making it crappy. . .had to screenshot and photobucket.
Last edited by calculon; 02-05-09 at 02:36 PM.
#19
Originally Posted by calculon
He means that you can really help flow through the intercooler by welding half rod (half of a cylinder of aluminum) with a diameter equal to the IC's tube height so that incoming air is not just slammed into what looks like a picket fence and forced to sort itself out and flow through there.
Sean, do you have any back to back flow numbers with standard cores vs an identical core with this done?
#20
Please step back take a breath and relax, your E-***** is secure no worries. Those flow rates are at 10psi with a PL of roughly 1psi tested by the manufacturer. So 640CFM is 640CFM at 10psi with a measured pressure loss of roughly 1psi through the core. Joe schmo now knows what he needs to know as far as core flow goes (ROUGHLY) He now needs to start piecing together the rest of the puzzle. As I clearly stated I was showing the trade off of going one way with size versus the other. Not whether X size was more appropriate for Y application.
~S~
~S~
No need for name calling, especially from a "reputable" rotary shop in chicago. I wouldn't want to see a worker from a shop acting in such manner.
Besides the point, you can rate the air flow through the core and estimate its pressure drop, but if I were to purchase the same size core as one you have listed from ebay, and you have purchased one from spearco there not nesscarly going to be the same becasue of the physical size. Each manufactuerer will have their own specific tube design, there for my point that is being overlooked is that you can't base airflow through a core based on core size.
#21
[QUOTE=KNONFS;8941465]I'll be happy with 375hp, anything over 400hp
Thanks for the info, here is the compressor map for my turbo:
As far as IAT, at what point should they be measured? under boost? under vac?
IMT should be measured under boost conditions and at idle. But to get an effective cooling temperature you would want it at full load. As we all know the cooler the air the much more dense it is. The more dense the air the more air molecules you have the more HP you will achieve. Ever notice your car is much quicker on a 60F night then a 90F day...its becasue of that. Especially with turbocharged vehicles, will notice it more becasue the compressor works much colder at that temp.
Thanks for the info, here is the compressor map for my turbo:
As far as IAT, at what point should they be measured? under boost? under vac?
IMT should be measured under boost conditions and at idle. But to get an effective cooling temperature you would want it at full load. As we all know the cooler the air the much more dense it is. The more dense the air the more air molecules you have the more HP you will achieve. Ever notice your car is much quicker on a 60F night then a 90F day...its becasue of that. Especially with turbocharged vehicles, will notice it more becasue the compressor works much colder at that temp.
#22
If you want an intercooler with actual thought into their design.... just buy an ARC. It is what I have an will probably continue to use for any performance car now on.
http://www.arcinter.co.jp/english/ar...gie/index.html
http://www.arcinter.co.jp/english/ar...gie/index.html
#24
No name was called, I was referring to your ego you chose to come at me. I wasn't even commenting on your posts when I put that up. I was merely stating no need to worry
As for the ebay core example, your preaching to the choir... Ebay cores don't have the fin density that a well made core does, I never mentioned ebay cores in my example for that reason, they.... are JUNK on a properly tuned setup.
"Tube design" is non existent on a Bar and Plate style core (then again could be semantics), which again is what I was referring to. Fin density and fin design matter which I will say you covered well.
All this is a moot point for the most part, for the guy on the street building his car going into this type of core analysis is pointless. If he really wants actual flow loss data for the need to compete for best lap time. He can give me
max. Charge-Air Flow (or max. BHP)
Charge-Air Pressure at that Condition
Ambient Temperature
Vehicle Speed
then we can start from there.
Enzo hit it on the head. For this turbo and this guy, his choice is simple.
KNONFS We ran a 3.5x12x18 on the red 20B 42R(Tial)car in the 20B section. Car made 500whp@10psi 405ft/lb's easy work, slight clean up on the ports is all that was done. That core would be more than fine for what your after. Even a 3" would work for you. How many FD's were in the 350-430whp range on M2/ASP cores? that's your answer.
As for the ebay core example, your preaching to the choir... Ebay cores don't have the fin density that a well made core does, I never mentioned ebay cores in my example for that reason, they.... are JUNK on a properly tuned setup.
"Tube design" is non existent on a Bar and Plate style core (then again could be semantics), which again is what I was referring to. Fin density and fin design matter which I will say you covered well.
All this is a moot point for the most part, for the guy on the street building his car going into this type of core analysis is pointless. If he really wants actual flow loss data for the need to compete for best lap time. He can give me
max. Charge-Air Flow (or max. BHP)
Charge-Air Pressure at that Condition
Ambient Temperature
Vehicle Speed
then we can start from there.
Enzo hit it on the head. For this turbo and this guy, his choice is simple.
KNONFS We ran a 3.5x12x18 on the red 20B 42R(Tial)car in the 20B section. Car made 500whp@10psi 405ft/lb's easy work, slight clean up on the ports is all that was done. That core would be more than fine for what your after. Even a 3" would work for you. How many FD's were in the 350-430whp range on M2/ASP cores? that's your answer.
No need for name calling, especially from a "reputable" rotary shop in chicago. I wouldn't want to see a worker from a shop acting in such manner.
Besides the point, you can rate the air flow through the core and estimate its pressure drop, but if I were to purchase the same size core as one you have listed from ebay, and you have purchased one from spearco there not nesscarly going to be the same becasue of the physical size. Each manufactuerer will have their own specific tube design, there for my point that is being overlooked is that you can't base airflow through a core based on core size.
Besides the point, you can rate the air flow through the core and estimate its pressure drop, but if I were to purchase the same size core as one you have listed from ebay, and you have purchased one from spearco there not nesscarly going to be the same becasue of the physical size. Each manufactuerer will have their own specific tube design, there for my point that is being overlooked is that you can't base airflow through a core based on core size.
Last edited by Zero R; 02-06-09 at 10:30 AM.
#25
It can be for sure, I spent time going over it with two people who I know and trust well one has built and designed stuff for F1 and Cart, the other runs over 200psi. Both said they measured 2-5% increase depending on the application with as much as 7% was seen. Doesn't seem like much but plug in that 640CFM number and add 5% and you've got 672CFM. I did it to my core on my car ask ned how noticeable it was.
Better flow equals better response