How to build an 8 sec Air to Air intercooler
01-12-2011, 01:48 PM
#1
FormerVendor
Thread Starter
iTrader: (1)
Join Date: Oct 2010
Location: Dallas, Tx
Posts: 138
Likes: 0
Received 0 Likes
on
0 Posts
How to build an 8 sec Air to Air intercooler When you pick an intercooler, you should figure out a few important things first.
I realize this isnt an LS setup, but the math, sizing, and fab would be the same regardless of the platform.
This particular post is referring to a customers car we have in the shop. We had the opportunity to think outside the box, take data from Ryans GSS supra and improve on it.
The core I could fit was thicker than the biggest core on the market with a little tom foolery. Why this wasnt used or been made yet is beyond me.
Airflow to the actual core, from the turbo, is VERY important. Using 1/3 the core to cool 70% of the total flow is counter productive. This was one thing we wanted to improve on. End tank design is very important in efficiency. As you can see by this CFD model we did, the core is not used very efficiently. We aim to change that.
The shorter the core, the better. The first few inches are the most important, as that is where the most heat transfer occurs. This is why vertical flow intercoolers (top to bottom) are more efficient, but can be near impossible to fit due to car/chassis/platform constraints and often horizontal flow is used.
Shorter means less pressure drop. Less pressure drop=more power and over all a more efficient forced induction setup.
We ended up shortening the core 3.5", and adding an inch to core thickness. This allowed us a proper end tank to feed the entire face of the core, to help heat transfer while keeping pressure drop across the core in check.
here is the actual core, its a bar and plate piece sourced from Bell Cores. Coke can for reference.
You can see the cardboard cover to keep the fins from being damaged as I fabricate the tanks and mounts for the core.
I used a Jack to position it, it was the easiest thing to use and it allowed me to move it where I wanted it. This particular core is 37 pounds, so its no lightweight.
I mocked up the first end tank using a thinner aluminum that I could cut with hand shears and form very easily. The purpose was to save myself from killing my hands forming it and to have a proper template for later use. Using cardboard is great, and its what I use most of the time, this is just another alternative.
I then took that aluminum template, and mocked it up on my table. The inlet was 3.5", the core face was 6". This allowed me to draw the most progressive and gradual line to connect the two, and make another template.
Here is the result of that sketch. This is a great start.
Once I made the inner and outer arcs, I was able to make the lower shelf of the tank.
And here is that cardboard transferred to the final material. In theory, this is the right way to do it, leading into that compressor outlet tube. Im being optimistic...read on to find out more
LOL!
Here is our template, and our final lower shelf together. I tacked the bottom shelf to the core so it would not move.
And the inner and outer walls of the tank complete.
I went ahead and cut the "tail" off the shelf/floor. It provided to be too difficult to form the material all the way around the curve, for all 4 sides. I cut it where I did, as that was the centerline of a 70* bend, and a Vband would line up perfectly as I neared completion.
I went ahead and removed the core from the car and started the roof of the tank. This is a tricky piece, much like a puzzle. There was a lot of post primary cut trimming. this was the hardest piece.
Before I put that roof on, I intended to weld as much as I could on the inside of the tank. You can see here, as I preheat the aluminum ( Makes is MUCH easier to weld) the moisture forming. If you dont do this, it will make for a terrible weld. Preheat it to atleast 175* and your life will be much easier.
This is the roof, it bends and curves in 3 different directions. this took me over an hour to make this one part.
With the Roof welded on, you can see now that I can mock up the bend leading into the tank, and the necessary Vband. I will explain the pie shaped cuts shortly. I will tack this Vband in place, and weld it later off the car.
On to the drivers side tank, or the "cold side"
I didnt take pics of the template in progress, but this is the result. You can see my plan here, making the transition to the tube as straight as possible. Less bends = better flow= less restriction.
You can see what Im after here- one nice curve to keep the airflow happy flowing into the tube that feeds the TB.
Since I knew that my outlet was 3.5", and I had a small window to work around, I went ahead and pulled the core back off, and tacked on the inside most wall of the cold tank.
Once that was all finished, I started the roof for the cold tank.
Back to the hot tank- You can see here, the pie shapes are cut, and the Vband is tacked in place
Remember the Pie shapes? This allows a nice transition to the tubes that feed and draw from the core.
Sometimes what you want the aluminum to do and what it does are two different things....
Here she is all finished!
Because I welded on this, the original 6061T6 is now annealed somewhat, and the weld is definately NOT T6.
What do we do? We see boot pressures in the mid 40s, sometimes even 50 psi of boost. The assembly must be strong and not fail. Heat treating it is a great way to add substantial strength with out much work.
The assembly was heated for one hour at a high temp in order to bring it back up to T5. Its OK to treat the entire core, as per bell.
Now to finish the rest of the car!
Louis
I realize this isnt an LS setup, but the math, sizing, and fab would be the same regardless of the platform.
- Power
- boost level
- available frontal area
- footprint for core and end tanks
- pressure drop
- intended use
This particular post is referring to a customers car we have in the shop. We had the opportunity to think outside the box, take data from Ryans GSS supra and improve on it.
- Power level will be 1250-1300 flywheel HP
- boost level will be a compressor limited 76mm inducer billet wheel
- available frontal area is good, lots of room.
- footprint for core and end tanks Is tight, there is a compromise on tank sizing vs core sizing.
- pressure drop is important, as every pound of boost matters here.
- intended use is Standing mile and drag
The core I could fit was thicker than the biggest core on the market with a little tom foolery. Why this wasnt used or been made yet is beyond me.
Airflow to the actual core, from the turbo, is VERY important. Using 1/3 the core to cool 70% of the total flow is counter productive. This was one thing we wanted to improve on. End tank design is very important in efficiency. As you can see by this CFD model we did, the core is not used very efficiently. We aim to change that.
The shorter the core, the better. The first few inches are the most important, as that is where the most heat transfer occurs. This is why vertical flow intercoolers (top to bottom) are more efficient, but can be near impossible to fit due to car/chassis/platform constraints and often horizontal flow is used.
Shorter means less pressure drop. Less pressure drop=more power and over all a more efficient forced induction setup.
We ended up shortening the core 3.5", and adding an inch to core thickness. This allowed us a proper end tank to feed the entire face of the core, to help heat transfer while keeping pressure drop across the core in check.
here is the actual core, its a bar and plate piece sourced from Bell Cores. Coke can for reference.
You can see the cardboard cover to keep the fins from being damaged as I fabricate the tanks and mounts for the core.
I used a Jack to position it, it was the easiest thing to use and it allowed me to move it where I wanted it. This particular core is 37 pounds, so its no lightweight.
I mocked up the first end tank using a thinner aluminum that I could cut with hand shears and form very easily. The purpose was to save myself from killing my hands forming it and to have a proper template for later use. Using cardboard is great, and its what I use most of the time, this is just another alternative.
I then took that aluminum template, and mocked it up on my table. The inlet was 3.5", the core face was 6". This allowed me to draw the most progressive and gradual line to connect the two, and make another template.
Here is the result of that sketch. This is a great start.
Once I made the inner and outer arcs, I was able to make the lower shelf of the tank.
And here is that cardboard transferred to the final material. In theory, this is the right way to do it, leading into that compressor outlet tube. Im being optimistic...read on to find out more
LOL!Here is our template, and our final lower shelf together. I tacked the bottom shelf to the core so it would not move.
And the inner and outer walls of the tank complete.
I went ahead and cut the "tail" off the shelf/floor. It provided to be too difficult to form the material all the way around the curve, for all 4 sides. I cut it where I did, as that was the centerline of a 70* bend, and a Vband would line up perfectly as I neared completion.
I went ahead and removed the core from the car and started the roof of the tank. This is a tricky piece, much like a puzzle. There was a lot of post primary cut trimming. this was the hardest piece.
Before I put that roof on, I intended to weld as much as I could on the inside of the tank. You can see here, as I preheat the aluminum ( Makes is MUCH easier to weld) the moisture forming. If you dont do this, it will make for a terrible weld. Preheat it to atleast 175* and your life will be much easier.
This is the roof, it bends and curves in 3 different directions. this took me over an hour to make this one part.
With the Roof welded on, you can see now that I can mock up the bend leading into the tank, and the necessary Vband. I will explain the pie shaped cuts shortly. I will tack this Vband in place, and weld it later off the car.
On to the drivers side tank, or the "cold side"
I didnt take pics of the template in progress, but this is the result. You can see my plan here, making the transition to the tube as straight as possible. Less bends = better flow= less restriction.
You can see what Im after here- one nice curve to keep the airflow happy flowing into the tube that feeds the TB.
Since I knew that my outlet was 3.5", and I had a small window to work around, I went ahead and pulled the core back off, and tacked on the inside most wall of the cold tank.
Once that was all finished, I started the roof for the cold tank.
Back to the hot tank- You can see here, the pie shapes are cut, and the Vband is tacked in place
Remember the Pie shapes? This allows a nice transition to the tubes that feed and draw from the core.
Sometimes what you want the aluminum to do and what it does are two different things....
Here she is all finished!
Because I welded on this, the original 6061T6 is now annealed somewhat, and the weld is definately NOT T6.
What do we do? We see boot pressures in the mid 40s, sometimes even 50 psi of boost. The assembly must be strong and not fail. Heat treating it is a great way to add substantial strength with out much work.
The assembly was heated for one hour at a high temp in order to bring it back up to T5. Its OK to treat the entire core, as per bell.
Now to finish the rest of the car!
Louis
01-12-2011, 03:34 PM
#6
LS1TECH Sponsor / 7 Second Club
iTrader: (2)
Join Date: Dec 2001
Location: Houston
Posts: 971
Likes: 0
Received 0 Likes
on
0 Posts
Very nice piece, but why not put a air flow divider in the inlet side end tank. that would help even the air flow across the core. I would also liek to know what the pressure drop across the core is at high boost levels. I guess it being so thick it shouldn't be that bad.
Trending Topics
01-12-2011, 07:17 PM
#13
SSU Moderator
Join Date: Nov 2001
Location: Austin, TX
Posts: 4,857
Likes: 0
Received 0 Likes
on
0 Posts
I'd like to see the flow model on the new one too. Amazing work...and i know you're working within the confines of the car structure, but i wonder how much flow is still 'stuck' on the bottom half of the core.
01-12-2011, 07:56 PM
#14
Very nice Louis. This gives people an idea of what it takes to make power. Often people can't understand the cost of custom fabrication, because they don't know what goes into making one-off pieces such as this. It is a great example to share with new customers to explain why things cost so darn much!
01-12-2011, 09:38 PM
#16
Launching!
iTrader: (5)
Join Date: Mar 2004
Location: Chippewa Falls, WI
Posts: 292
Likes: 0
Received 0 Likes
on
0 Posts
looks good. I like it. Lets say I could fit a vertical flow intercooler. Would a 25x9x3.5 thick core work good? with opposite facing end tanks for flow? I would like to build my own end tanks, but the tools and time currently are against me.
01-12-2011, 10:08 PM
#18
FormerVendor
Thread Starter
iTrader: (1)
Join Date: Oct 2010
Location: Dallas, Tx
Posts: 138
Likes: 0
Received 0 Likes
on
0 Posts
Very nice piece, but why not put a air flow divider in the inlet side end tank. that would help even the air flow across the core. I would also liek to know what the pressure drop across the core is at high boost levels. I guess it being so thick it shouldn't be that bad.
Very nice Louis. This gives people an idea of what it takes to make power. Often people can't understand the cost of custom fabrication, because they don't know what goes into making one-off pieces such as this. It is a great example to share with new customers to explain why things cost so darn much!
01-12-2011, 11:17 PM
#19
TECH Addict
iTrader: (16)
Join Date: Jul 2004
Location: Nevada, TX
Posts: 2,218
Likes: 0
Received 0 Likes
on
0 Posts
Looks very clean. I would love to hear what an apples to apples difference would be with the same core and typical "square" end tanks. Thats alot of damn work.
