Lets Discuss Intercooler Pipe Sizes!

[S130Z]

Choosing the Correct Intercooler/Charge Pipe Size

 

I have been thinking of this topic for quite some time now and after running a few searches, I feel like this would be good to discuss. There must be a precise and correct pipe size for your specific application, correct? Depending on the number of cylinders, the size of your intake ports, and size turbo you are running, etc... you should have a specific sized pipe for maximum flow.

 

Having too large of piping can yield boost lag. Your turbo is trying to fill the pipe instead of feeding the cylinders. The longer it takes to fill the piping, the longer it will take to achieve peek boost in the cylinder. On the other hand, have too small of piping can have a negative effect too. Although you decrease boost lag by being able to pressurize your entire system faster, you loose high flow capabilities.

 

There must be a mathematical equation that can give you a specific sized piping. Thus giving you maximum flow. I would like to hear some ideas from others on this as I have a few thoughts in my head about how to calculate something like this.

 

Last question, from where are you getting your PSI readings from? Are you measuring from the intake plenum, or from an individual runner. I ask this because it makes sense to me that the closer to the cylinder you are, the more accurate of a reading you are going to get.

[SHO-Z]

Try this calculator it will give you an idea of the flow looses in various pipe sizes. I know it is for hose but it will work the same. Every elbow needs to be added in ft of pipe. A 90 is about 3 ft if I remember right. Use your max RPM to calculate the CFM at your boost pressure.

 

http://www.gates.com/industrial/pressure/airflow.cfm?location_id=3043

[hughdogz]

...There must be a mathematical equation that can give you a specific sized piping. Thus giving you maximum flow. I would like to hear some ideas from others on this as I have a few thoughts in my head about how to calculate something like this...

 

It has been awhile since I took a fluid dynamics course in College (~1996) but this is what I can remember...

 

All else being equal, the mass flow rate through a pipe is m_dot = rho * V * A (rho = density, V = velocity, A = cross-sectional area). Due to the conservation of mass, it doesn't matter what diameter pipe you have, they will both flow the same CFM (mass in = mass out). But as you mentioned, the dynamic pressure drop will probably be greater for a smaller pipe, depending on the Reynolds number (whether it is turbulent flow or laminar, friction causing the turbulence).

 

To illustrate this point, say that you have a reducer going from 3" to 2". The mass flow rate will be the same, but since the cross-sectional area decreased, the velocity has to increase. Since the velocity has increased, the dynamic pressure has increased, while static pressure decreased (total pressure being the same = Bernoulli's principle).

 

When in boost, I'm pretty sure that the flow will be turbulent, to make sure, you can calculate the Reynolds number using V*D/nu (V = velocity, D = pipe diameter, nu = kinematic viscosity). You can check if the flow is in the turbulent region or not, using the moody diagram. On average, transition to turbulence will occur at a Re > ~2300. It will depend on the diameter of pipe, velocity, pipe roughness, etc.

 

• Moody diagram explained by Wiki: http://en.wikipedia.org/wiki/Moody_chart
  

• More Moody info: http://www.engineeringtoolbox.com/moody-diagram-d_618.html
  

• Moody friction factor calculator: http://www.lmnoeng.com/moody.htm
  

Pressure (head) loss depends on fittings / bends, you can add the "K-factors" for the bends / elbows. This will determine the equivalent length of the head (pressure) loss for fittings and elbows, as Sho-z already mentioned. They go into a detailed example here: http://www.cheresources.com/eqlength.shtml

 

To calculate the head loss due to friction through a straight section of pipe, you can use the Darcy equation using the friction factor from the Moody diagram:

 

 

 

You can also take calculate (seperatly) the static and dynamic pressure changes as well, using Bernoulli's equation.

 

Like you said, it is a tradeoff, and will ultimately depend on your HP (flow) goals to choose the proper diameter tubing.

 

Hope this helped...

 

Last question, from where are you getting your PSI readings from? Are you measuring from the intake plenum, or from an individual runner. I ask this because it makes sense to me that the closer to the cylinder you are, the more accurate of a reading you are going to get.

 

For recording the boost pressure (MAP) I just take it at the intake manifold plenum. For the boost controller, I think it is best to take it nearest the compressor outlet, if not at the compressor outlet itself.

[GrayZee]

You can make calculations all day but in the end I don't think the differences are going to matter that much at all. I usually go by the simplest route, transition the pipes to match the outlet of the turbo and the size of the throttlebody.

If your turbo has a 2" inch outlet and your IC has 2.5" in and outlets and you have a 2.75" TB... then just start with 2" and in logical places (I like to buy silicon reducer hose) and work your way up to 2.5 at the IC and up to 2.75 after the IC or near the TB.

[WizardBlack]

You can make calculations all day but in the end I don't think the differences are going to matter that much at all. I usually go by the simplest route, transition the pipes to match the outlet of the turbo and the size of the throttlebody.

If your turbo has a 2" inch outlet and your IC has 2.5" in and outlets and you have a 2.75" TB... then just start with 2" and in logical places (I like to buy silicon reducer hose) and work your way up to 2.5 at the IC and up to 2.75 after the IC or near the TB.

 

Bingo. In that situation (a very common one), you can either go the smaller size (2" pipe from turbo to IC with the step-up coupler on the IC) or the larger size (2.5" pipe from turbo to IC with the step-up coupler on the turbo outlet). That's about it. Your parts dictate your pipe size to a great degree. The power rating on your turbo and intercooler will dictate the typical acceptable pipe size they need.

[ShaggyZ]

Bingo. In that situation (a very common one), you can either go the smaller size (2" pipe from turbo to IC with the step-up coupler on the IC) or the larger size (2.5" pipe from turbo to IC with the step-up coupler on the turbo outlet). That's about it. Your parts dictate your pipe size to a great degree. The power rating on your turbo and intercooler will dictate the typical acceptable pipe size they need.

If you have NO money to spend, then that all makes sense.

 

In the case of a standard L28ET install, there is no sense in moving to 3" intercooler piping when one has to transition from the turbo outlet size of about 2" to a 3" pipe and back down to a 240SX throttle body. Plus, the bends on a 3" pipe to fit are generally going to be tighter. I think calculations down to the last tiny bit cost too much in time, money, and worry, especially when it's easy to overlook something and end up unknowingly making a sacrifice that puts you right back where a thoughtless setup would have - this is something along the lines of cost/benefit analysis.

 

Use the ideas gathered from the calculations and do what you can to stick close to ideal - fewest transitions (make them long and smooth when possible), smoothest bends, best fits (ie fit pipe to pipe at silicone connections rather than having a huge gap "bandaided" by the silicone hose).

[S130Z]

If you have NO money to spend, then that all makes sense.

 

In the case of a standard L28ET install, there is no sense in moving to 3" intercooler piping when one has to transition from the turbo outlet size of about 2" to a 3" pipe and back down to a 240SX throttle body. Plus, the bends on a 3" pipe to fit are generally going to be tighter. I think calculations down to the last tiny bit cost too much in time, money, and worry, especially when it's easy to overlook something and end up unknowingly making a sacrifice that puts you right back where a thoughtless setup would have - this is something along the lines of cost/benefit analysis.

 

Use the ideas gathered from the calculations and do what you can to stick close to ideal - fewest transitions (make them long and smooth when possible), smoothest bends, best fits (ie fit pipe to pipe at silicone connections rather than having a huge gap "bandaided" by the silicone hose).

 

I had been thinking about all of these aspects before I had posted this thread. The main killer in this situation is running an intercooler. The intercooler completely ruins the pipe diameter relevance.

 

I have been seeing people running a fairly small diameter pipe coming from the turbo to the intercooler, then having a larger pipe on the other side to match up with their TB. What is the reason for this?

[S130Z]

By the way, I just got Maximum Boost by Corky Bell in the mail. I am sure this book should clear some of my thoughts up.

[GrayZee]

My piping setup is actually a very confusing setup as far as pipe sizes go, however it was done in the most practial way IMO. The following points effected my decisions.

 

1. I have a Holset with a 2.75 OD outlet (but the ID is more like 2.5)

 

2. I had a choice between two vertical flow intercoolers, one was too big for the car with

3" in/outlets, the other was a little on the small side with 2.5 in/outlets. I decided to

get the bigger one and then just cut it to the perfect size and weld on plates to cap it

off.

 

3. My SX throttle is 2.75

 

4. I wanted to do minimal cutting to the sheet metal of the car, so the smaller the pipe the

better.

 

 

The Holset starts off with a 2.75 to 2.5 hose, 2.5" piping until it goes through the rad cradle, then two 90 degree silicon reducing bends. 2.5 to 2.75 and 2.75 to 3.0 (thus making a very gradual increase to 3.0 Coming out of the top side of the intercooler I use 3.0" piping and eventually get to a 45 degree reducing bend at the throttle, taking it back down to 2.75 a very minor transision that might even provide a venturi effect when I add methanol this year. I did not go with 2.75" piping anywhere except 3" worth to joint two hoses together, 2.75 is generally not as common and the 2.5" stuff meant very little cutting of the sheet metal.

[ShaggyZ]

Grayzee, I'm no fluids expert, but I can tell you that sounds terrible - it sounds just like what I warned against as the basic type of stuff to avoid.

 

Having two silicone transitions (they're just steps, really) from 2.5 to 2.75 and 2.75 to 3.0 does not make a gradual transition. Silicone parts are not the best place to do turns or reductions because they are usually very abrupt. When I referred to "long and smooth transitions," I meant doing something like cutting a 12" long, 3" diameter pipe lengthwise (removing a very narrow triangle) and welding it back so that one side remains 3" and the other is now 2.5", for example.

 

Doing it this way costs very little (less than silicone 90° transitions), looks groovy, reduces complexity, and causes far less disturbance than standard step transitions. A lot of guys think they're saving themselves money by avoiding a welder and, instead, having 10 silicone connections with 20 worm gear clamps - yuck!

 

 

I'm betting there's a thread around here somewhere with the likes of John Coffey and similar guys with clout and experience explaining this.

[GrayZee]

I have welded parts on my IC piping, I only used silicon where I NEEDED to use one anyway.. If that area happened to require a bend, then I did it in silicone. There is only one spot that is kinda overkill in the silicon area and that is the two reducing bends right together, and there was no way that I had room for any other option. To make two 90 degree bends AND go from 2.5 to 3.0? I didn't exactly have 24" of space to play with. Do a search for 280zforce's IC setup and you will see what I mean. Mine is very similar. The idea of cutting a pipe lengthwise 12" and putting a gradual taper on it seems like a big waste of time to me. Although in the most technical/mathematic way it might be correct, the question is... At the end of the day what does this "superior flow" characteristic really matter? Will you notice a difference? I doubt it.

Another thing? have you actually seen a silicon reducing bend? The are not at all like the straight ones, the transision is VERY smooth, infact with a .25" difference you have to look carefully to even notice.

[ShaggyZ]

I didn't intend to rip into you.

 

I have seen the silicone transitions you're referring to - those are far better than the standard reducers people use. Regardless, I think you'd be served well to have fewer transitions. Just fixing that likely won't make any noticeable performance difference (even looking at a dynograph), but it all adds up, which was the point of my "main" post about transitions and bends and pipe size.

 

For what it's worth, splitting a pipe as I described and welding it back together shouldn't take much more than an hour, if that, for a regular joe with a Dremel and a welder.

[WizardBlack]

I had been thinking about all of these aspects before I had posted this thread. The main killer in this situation is running an intercooler. The intercooler completely ruins the pipe diameter relevance.

 

I have been seeing people running a fairly small diameter pipe coming from the turbo to the intercooler, then having a larger pipe on the other side to match up with their TB. What is the reason for this?

 

The reason is to simply let the hard parts dictate the pipe sizes and keep it with as few of changes in size as you can. Bend minimization is important but this is 20 psi air (with no fuel in it), not (comparatively) ultra dense fluids like water. Bends in a head are critical, but you are talking about all that air in a much smaller cross section with fuel vapor and you have about a 1.5" turn radius.

[GrayZee]

The reason is to simply let the hard parts dictate the pipe sizes and keep it with as few of changes in size as you can. Bend minimization is important but this is 20 psi air (with no fuel in it), not (comparatively) ultra dense fluids like water. Bends in a head are critical, but you are talking about all that air in a much smaller cross section with fuel vapor and you have about a 1.5" turn radius.

 

 

I will agree with that, there is a huge difference between charge pipe and the air/fuel mixture flowing in a cylinder head. I really think the IC piping differences are too minor to worry about that much. I went from a stock l28 turbo with a J pipe (no intercooler at all) to a Holset and 3" IC piping. Did I notice a difference in lag? Barely... and was that extra lag from the larger air volume of the IC and piping? I very much doubt it, it would have alot more to do with the size of the larger turbo.

[SHO-Z]

Ok lets look at it this way if you have around 6 elbows on each side of the intercooler and 10 ft of run piping gives an equivalant length of about 60 ft of pipe. At 15 psi of boost at 6000 RPM gives the following pressure drop between the turbo and the intake. This does not include the intercooler loss.

 

2" Pipe 1.78 psi

2.5" Pipe 0.54 psi

3" Pipe 0.21 psi

 

It looks like 2.5" should be the optinal size for low turbo lag and pressure loss in most cases.

[jeffp]

I set my car up with 2 1/4" pipe in and out. I haven't had any problems making boost or power. I run a GT35R on the car and @ 26psi of boost the engine was making right around 700hp. the dyno @ 22psi indicated my power to be right at 658Hp. I could only run 22psi on the dyno, but after I was finished I ran it to 26psi on the road and not surprizingly, the engine was going lean.

So in short, 2 1/4" pipe will get you to 600hp no problem.

Is it the best efficiency, maybe not, was it easy to install, no it was not, but plumbing in a 2 1/2" inlet and outlet would have forced me to plumb the car with alot more mods and longer piping. I say 2 1/4" is the best all around sixe to run.

Now if you have a race car and are not to concerned with routing, then by all means go with 2 1/2" pipe.

[ShaggyZ]

I set my car up with 2 1/4" pipe in and out. I haven't had any problems making boost or power. I run a GT35R on the car and @ 26psi of boost the engine was making right around 700hp. the dyno @ 22psi indicated my power to be right at 658Hp. I could only run 22psi on the dyno, but after I was finished I ran it to 26psi on the road and not surprizingly, the engine was going lean.

So in short, 2 1/4" pipe will get you to 600hp no problem.

Is it the best efficiency, maybe not, was it easy to install, no it was not, but plumbing in a 2 1/2" inlet and outlet would have forced me to plumb the car with alot more mods and longer piping. I say 2 1/4" is the best all around sixe to run.

Now if you have a race car and are not to concerned with routing, then by all means go with 2 1/2" pipe.

 

You did this on a L28ET? I'm guessing this was another engine just by the horsepower number because it is exceedingly rare to hear of a L-series motor (even up to 3.1L) making that power.

[Evan Purple240zt]

You did this on a L28ET? I'm guessing this was another engine just by the horsepower number because it is exceedingly rare to hear of a L-series motor (even up to 3.1L) making that power.

 

yes, he did that on an L28ET... if this is the Jeff Priddy (sp) that I am thinking of.

[slown280z]

yes, he did that on an L28ET... if this is the Jeff Priddy (sp) that I am thinking of.

The one and only.

It's bored and stroked by the way...

http://www.angelfire.com/extreme/280zxt/

[ShaggyZ]

The one and only.

It's bored and stroked by the way...

http://www.angelfire.com/extreme/280zxt/

 

Oh ya! I remember that car. I remember reading his whole freaking restoresupermodificationchangeeverything writeup on that. Incredible!

 

Thanks!

 

Time to update that site, though. Old data and... ugh... Angel Fire!? Seriously? What year is this!?