Intercooler shape/size questions...
Discussion
I'm trying to work out what is best for an intercooler.
If you go big, what are the downsides vs small?
I have read things like don't expand the incoming air body up too much more than maybe 3x otherwise you generate excessive losses through flow seperation or reverse flow, but then you can end up having the intercooler stack facings inside quite square (a bit like the PWR chargecoolers) and then you generate turbulence there.
Is it optimal to have the internal core cross-section as close to the intercooler intake cross-section as possible, and then simply make it as long as it needs to be?
What is the tradeoff with respect to length and stack area? Is it linear both ways? Ie, double length and halve stack area and it's the same (assuming laminar flow)?
I've seen some cars have short core's but huge cross-sections from a small inlet and they work really well, yet I've seen long thin intercoolers on other engines that work well too.
Kinda having a discussion with someone who suggests that bigger is simply worse, and having a large stack cross-section (ie, RS500 intercooler) is bad, and using a small stock Cossie intercooler say is better for pressure drop.
Pressure drop matters of course, but if it's 2psi on a small cooler and 4psi on one twice as big, is it really relevant vs the much greater cooling efficiency if the intercooler is flowing 500cfm and at 1.5 bar that drop change is going to be offset by much better cooling as a trade off?!
Any rules of thumb or best practice trends for things to look out for would be cool (literally)
Thanks
Dave
If you go big, what are the downsides vs small?
I have read things like don't expand the incoming air body up too much more than maybe 3x otherwise you generate excessive losses through flow seperation or reverse flow, but then you can end up having the intercooler stack facings inside quite square (a bit like the PWR chargecoolers) and then you generate turbulence there.
Is it optimal to have the internal core cross-section as close to the intercooler intake cross-section as possible, and then simply make it as long as it needs to be?
What is the tradeoff with respect to length and stack area? Is it linear both ways? Ie, double length and halve stack area and it's the same (assuming laminar flow)?
I've seen some cars have short core's but huge cross-sections from a small inlet and they work really well, yet I've seen long thin intercoolers on other engines that work well too.
Kinda having a discussion with someone who suggests that bigger is simply worse, and having a large stack cross-section (ie, RS500 intercooler) is bad, and using a small stock Cossie intercooler say is better for pressure drop.
Pressure drop matters of course, but if it's 2psi on a small cooler and 4psi on one twice as big, is it really relevant vs the much greater cooling efficiency if the intercooler is flowing 500cfm and at 1.5 bar that drop change is going to be offset by much better cooling as a trade off?!
Any rules of thumb or best practice trends for things to look out for would be cool (literally)
Thanks
Dave
Mr Whippy said:
I have read things like don't expand the incoming air body up too much more than maybe 3x otherwise you generate excessive losses through flow seperation or reverse flow, but then you can end up having the intercooler stack facings inside quite square (a bit like the PWR chargecoolers) and then you generate turbulence there.
Can you explain that in English ?
Mr Whippy said:
I'm trying to work out what is best for an intercooler.
If you go big, what are the downsides vs small?
If you go big, what are the downsides vs small?
Virtually none, aside from packaging issues. Some claim there may be more lag, but it would rarely be noticeable.
Mr Whippy said:
Kinda having a discussion with someone who suggests that bigger is simply worse, and having a large stack cross-section (ie, RS500 intercooler) is bad, and using a small stock Cossie intercooler say is better for pressure drop.
Pressure drop matters of course, but if it's 2psi on a small cooler and 4psi on one twice as big, is it really relevant vs the much greater cooling efficiency if the intercooler is flowing 500cfm and at 1.5 bar that drop change is going to be offset by much better cooling as a trade off?!
Any rules of thumb or best practice trends for things to look out for would be cool (literally)
Thanks
Dave
So he thinks a small restrictive intercooler is better ??? I'm guessing his Cossie, is extremely slow then ? if thats his thoughts on tuning.
Pressure drop is very important. The more you have, the slower your car will be. With a restrictive intercooler, you are throwing away basically free power.
Not sure how you figure your logic, in a larger intercooler being more restrictive than a smaller one.
Can you flow more water ( air or anything ) through a straw, or a 4" sewer pipe ? Which is more restrictive. Its quite simple really.
You appear to be reading some odd literature, or misinterpreting it in some way, as its pretty simple really.
A bigger intercooler will cool better.
A bigger intercooler will flow better.
Point out the downsides to a bigger intercooler ?
Size doesn't really mater, its' the flow rate you need to keep an Eye on, you could go huge but end up with either some of it not being used or the air, I can't think how to out this, could get lost!
On the other side if it's too small it won't cool the air enough, I'd be inclined to keep it pretty much the same size, get a free flowing one and fit water jets/Nirous oxide.
On the other side if it's too small it won't cool the air enough, I'd be inclined to keep it pretty much the same size, get a free flowing one and fit water jets/Nirous oxide.
I think one of the issues hinted at for large intercoolers is that the larger it is, the longer distance you need to expand the air cross section over. If you just dump a small pipe into a large box, you get a higher pressure drop than if run it through at a small diameter all the way. I'd agree though, slightly too large is better than too small.
stevieturbo said:
Mr Whippy said:
I have read things like don't expand the incoming air body up too much more than maybe 3x otherwise you generate excessive losses through flow seperation or reverse flow, but then you can end up having the intercooler stack facings inside quite square (a bit like the PWR chargecoolers) and then you generate turbulence there.
Can you explain that in English ?
Mr Whippy said:
I'm trying to work out what is best for an intercooler.
If you go big, what are the downsides vs small?
If you go big, what are the downsides vs small?
Virtually none, aside from packaging issues. Some claim there may be more lag, but it would rarely be noticeable.
Mr Whippy said:
Kinda having a discussion with someone who suggests that bigger is simply worse, and having a large stack cross-section (ie, RS500 intercooler) is bad, and using a small stock Cossie intercooler say is better for pressure drop.
Pressure drop matters of course, but if it's 2psi on a small cooler and 4psi on one twice as big, is it really relevant vs the much greater cooling efficiency if the intercooler is flowing 500cfm and at 1.5 bar that drop change is going to be offset by much better cooling as a trade off?!
Any rules of thumb or best practice trends for things to look out for would be cool (literally)
Thanks
Dave
So he thinks a small restrictive intercooler is better ??? I'm guessing his Cossie, is extremely slow then ? if thats his thoughts on tuning.
Pressure drop is very important. The more you have, the slower your car will be. With a restrictive intercooler, you are throwing away basically free power.
Not sure how you figure your logic, in a larger intercooler being more restrictive than a smaller one.
Can you flow more water ( air or anything ) through a straw, or a 4" sewer pipe ? Which is more restrictive. Its quite simple really.
You appear to be reading some odd literature, or misinterpreting it in some way, as its pretty simple really.
A bigger intercooler will cool better.
A bigger intercooler will flow better.
Point out the downsides to a bigger intercooler ?
I'm asking the questions, giving some basic statements for people to suggest reasons why they may be wrong.
You've kinda answered what I assumed was correct, just someone else made the points which I put across in my initial post, and I wanted clarification as to if they were correct or not.
The English bit, well if we have a tube entering a large open area with a sharp transition is it generating excessive turbulence and drag? Ie, smooth flared transition or a sudden one?
Stack faces, well a recent discussion about the PWR chargecoolers (tubular) went along the lines of that the stack face ends were very flat and large vs the actual air passages, which might cause excessive pressure loss?
Is it important that the cross-sectional area of the feed pipe is at least matched by the cross-section of the internal passages for air within the intercooler? And then I assume making the air move from it's relatively laminar flow in the tubular feed pipe to travelling through the intercooler cores in the most smooth way possible is also optimum?
Assuming the above are all considered in a small or large intercooler, is it then simply that as well as it's well designed a big intercooler is just better full stop?
Are there any quick rules? Like if the intlet pipe is 10cm.sq and the overall cross-section of the intercooler cores is 10000cm.sq, is that still reasonable for the intercooler being better?
Ie, does pressure drop get effected by the ratio of inlet area vs core area? Can you have a too big core cross-section?
Same as above for core length.
Just got in a discussion here www.pug306.net/forum/index.php?showtopic=22680&st=0 and now I'm left baffled with what appears to be a student studying fluid dynamics and the thousands, nay millions of hours of knowledge of the writers at Autospeed who seem to know what they are talking about( http://autospeed.drive.com.au/cms/A_1 ) might need to subscribe there, and also what I've read here in bits and pieces over the years.
Dave
As suggested above, larger (within reason) is always better. Also make sure the bends are smooth, mandrel bends, not tight 90 degree bends where a 60 bend would do. Don't believe that "it's just a pressurised system - what goes in one end comes out the other"
You need to think of it as runny porridge flowing through a series of pipes, each bend slows it down, sharp intercooler core pipes will cause turbulance etc.
You need to think of it as runny porridge flowing through a series of pipes, each bend slows it down, sharp intercooler core pipes will cause turbulance etc.
The best way I've seen to do this is to feed the air in across the top of the matrix, with a tapered end tank spreading the air across the open face of the matrix, with a similar collector at the far side. This means the matrix works like lots of tiny turning vanes which greatly reduces the problem of the air flow separating when the cross section increases. I understand a similar trick is often used with cats.
I agree that in principle bigger than necessary is better than smaller than necessary. The reason that bigger isn't always better is that increasing the distance that the air has to travel through the matrix increases the drag. You can reduce the drag by increasing the cross section area, but there's a limit to how far it is sensible to take this. The benefits from cooling the charge down get into diminishing returns so you can get to a point where the extra cooling from a bigger matrix isn't enough to justify the extra drag. Similarly, in practical terms you get to a point where the extra cooling benefits aren't enough to justify the extra costs and packaging difficulties.
I agree that in principle bigger than necessary is better than smaller than necessary. The reason that bigger isn't always better is that increasing the distance that the air has to travel through the matrix increases the drag. You can reduce the drag by increasing the cross section area, but there's a limit to how far it is sensible to take this. The benefits from cooling the charge down get into diminishing returns so you can get to a point where the extra cooling from a bigger matrix isn't enough to justify the extra drag. Similarly, in practical terms you get to a point where the extra cooling benefits aren't enough to justify the extra costs and packaging difficulties.
Thanks for the replies.
As I expected really, and what I've already read and assumed from intercoolers from high-end performance cars and how they are designed!
I guess assesing the pressure drop and temp drop efficiency for a variety of coolers for your chosen app is best, and then go with the one which performs best... trial and error basically.
Just when you can pickup R34 Skyline GTR intercoolers for £100 or a Mondeo TDCi one for £50, you have to wonder if you'd be better with the Skyline one for obvious reasons, but it seems there is a stigma that bigger = lag and pressure drop while ignoring the fact the Mondeo one may well be built to the same cut price level as the OEM top mount on the given car (this case 1.9Tdi Peugeot 306) vs a relatively high spec intercooler for a high performance car!
Dave
As I expected really, and what I've already read and assumed from intercoolers from high-end performance cars and how they are designed!
I guess assesing the pressure drop and temp drop efficiency for a variety of coolers for your chosen app is best, and then go with the one which performs best... trial and error basically.
Just when you can pickup R34 Skyline GTR intercoolers for £100 or a Mondeo TDCi one for £50, you have to wonder if you'd be better with the Skyline one for obvious reasons, but it seems there is a stigma that bigger = lag and pressure drop while ignoring the fact the Mondeo one may well be built to the same cut price level as the OEM top mount on the given car (this case 1.9Tdi Peugeot 306) vs a relatively high spec intercooler for a high performance car!
Dave
Mr Whippy said:
Thanks for the replies.
stigma that bigger = lag and pressure drop
stigma that bigger = lag and pressure drop
Where do you get this from ??? ( or am I reading you wrong ?? )
A bigger intercooler, would flow better, so have much less pressure drop, which is good.
I dont know much about fluid dynamics, but opinion seems to be, that for heat transfer, turbulence is good.
Long tubes will be more restrictive, but if there's enough of them, it isnt an issue, but the longer the air is in the tubes, the more it will be cooled..
Short tubes will be less restrictive, which is good, but they wont cool as well. So you need LOTS more of them, to slow the air enough through them, to allow the air to get cooled.
As for ensuring the core has enough area, to at least match the IC pipework...well if it didnt, it would be a damn small intercooler !!
As for quick rules etc etc....
Pipework to a degree isnt the concern. You need an IC core, suitably sized firstly, to allow xxxx cfm of air to pass, with minimal restriction. That in itself will dictate actual core size requirements.
It may come in various designs, but engine airflow is the main parameter. The rest are packaging matters.
If you pump air into a dustbin it will expand and pressure will drop. That said, you'd need a really big intercooler to get a huge pressure drop. So a big intercooler may cause a pressure drop of a couple of psi in which case you could wind up the boost or make use of the cooler air you already have.
Boosted.
Boosted.
Boosted LS1 said:
If you pump air into a dustbin it will expand and pressure will drop. That said, you'd need a really big intercooler to get a huge pressure drop. So a big intercooler may cause a pressure drop of a couple of psi in which case you could wind up the boost or make use of the cooler air you already have.
Boosted.
Boosted.
I dont follow your analogy.
Yes the pressure will drop, but only for the duration of time it takes to full the "dustbin"
People would term that the same as lag I think ???
You want the air to slow down inside the IC, the longer its in there, the more heat will be dissipated.
I dont see how a big IC will cause a pressure drop. IMO a large free flowing, non restrictive IC would free up some boost, causing a guage increase as measured at the intake manifold.
Now if the IC was small and flowed crap. You may see a guage pressure drop at the intake, due the air not being able to get through the IC fast enough. That would be a very bad scenario.
stevieturbo said:
Mr Whippy said:
Thanks for the replies.
stigma that bigger = lag and pressure drop
stigma that bigger = lag and pressure drop
Where do you get this from ??? ( or am I reading you wrong ?? )
Yes, for the 2nd time your reading me wrong
I don't get that bigger=more lag and pressure drop. THAT is what someone else has suggested to me, someone who is making an arguement on the basis of their fluid/thermo dynamics education in their 1st year of university.
I do have a genuinely good interest in that field, with aerodynamics and aeronautical engineering being one of my chosen subjects before I went to Uni, and several friends working in the F1 industry and using finite analysis in CFD. So I'm quite sure I have a good grasp and that he is wrong, but I came here to seek clarification on the idea that "bigger = more lag" which is being put across to me on the linked thread...
There doesn't appear to be any real rule of thumb here, but I doubt having a 1m cubed intercooler core would be very effective for lag, but it's cooling capacity would be pretty high, vs a 10cm cubed intercooler which would have almost zero lag but very poor cooling ability.
I'm assuming that you can go quite big with an intercooler. Just because a base model has a low volume one it's not because it needs to be small to be non-laggy, but because it's done to cost and to requirement (TDCi Mondeo will be wafty motorway work and some overtaking). Chances are it'd be just as good if not better with respect to sustained hard use with one four times bigger.
Again, there is a stigma it seems amongst many that using an intercooler of a similar spec mainstream car is optimum because it has the same size capacity and power output range. Rather than use a 300bhp 2.6 Twin Turbo intercooler from an R34 Skyline GTR...
Dave
stevieturbo said:
Boosted LS1 said:
If you pump air into a dustbin it will expand and pressure will drop. That said, you'd need a really big intercooler to get a huge pressure drop. So a big intercooler may cause a pressure drop of a couple of psi in which case you could wind up the boost or make use of the cooler air you already have.
Boosted.
Boosted.
I dont follow your analogy.
Yes the pressure will drop, but only for the duration of time it takes to full the "dustbin"
People would term that the same as lag I think ???.
That's what I mean. I think it will drop because the dusbin has a leak in it's side , the engine. There will also be a temperature drop assuming the coolers working well. You could measure it but in real life it might not be noticeable.
Boosted
The engines leak, is totally seperate from the IC's "leak", and remains the same, regardless of IC fitted ( again unless the IC is ridiculously small )
With regards to lag.
Yes, a large IC etc may have more volume. But seriously, consider the quantities of air the turbo is pumping. Even with a huge IC, its relatively small.
Is the Skyline IC better ??? depends how well its specced in the first place.
Would the Mondeo one do ?? Depends how much air/boost you intend using/flowing on your specific engine.
If the engine is a 1.0 turbo, then good chance the mondeo IC would be more than adequate.
Its all back to choosing an IC with adequate core size to allow unrestricted flow. , and of course that will offer good cooling.
Just buy the biggest baddest intercooler ya can, and be done with it lol.
With regards to lag.
Yes, a large IC etc may have more volume. But seriously, consider the quantities of air the turbo is pumping. Even with a huge IC, its relatively small.
Is the Skyline IC better ??? depends how well its specced in the first place.
Would the Mondeo one do ?? Depends how much air/boost you intend using/flowing on your specific engine.
If the engine is a 1.0 turbo, then good chance the mondeo IC would be more than adequate.
Its all back to choosing an IC with adequate core size to allow unrestricted flow. , and of course that will offer good cooling.
Just buy the biggest baddest intercooler ya can, and be done with it lol.
Boosted LS1 said:
stevieturbo said:
Just buy the biggest baddest intercooler ya can, and be done with it lol.
Given the lack of reliable data to assist decision making that's probably the best advice out there.
Boosted.
Yep, after lots of reading in the past and present that is what I suggested, get the biggest one that will fit and you can afford and that will be the best bet!
Just people started getting all technical and worrying about lag and I just wondered if I'd been reading the wrong stuff.
I think they need to do some testing and stop being pansies with their crappy little intercoolers (because there is room for one four times as big)!
Dave
Give Forge Motorsport in Gloucester a ring, They were going to make me one, a direct replacement (what it came out the garage with, from a rad specialst) was £150 they would make me a high flow one for £500, but at the moment it won't quite fit!! (They intercooler itself was the same size, it's just they couldn't do oval entry/exit points so it wouldn't pass infront of the rad!)
combemarshal said:
Give Forge Motorsport in Gloucester a ring, They were going to make me one, a direct replacement (what it came out the garage with, from a rad specialst) was £150 they would make me a high flow one for £500, but at the moment it won't quite fit!! (They intercooler itself was the same size, it's just they couldn't do oval entry/exit points so it wouldn't pass infront of the rad!)
I have just had an intercooler made by Pro Alloy in Haverhill (round the corner from Pace !) with oval pipes, custom core size and tanks, all pipework, modification to plenum chamber, silicon joiners etc, fantastic quality, fitted to the car for £600 all in. Their welding is amazing, everything is pressure tested and fit and finish is perfect, couldn't have asked for more. Highly recommended.
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff