heres a top to bottom flow 24"L x 12"H x 4.5"W intercooler that PhoenixTurboRick just finished up...
going on a $200K long travel sandrail, built big block, twin T70's... the setup can do 1300whp, but with the paddle choice, they're only gonna push it to ~900whp....

acosta motorsports in NJ used this same core on their 3rotor RX-8 and made 1142whp...

 

looks sweet got any pics of the sandrail ???

 

i wish i had that kinda money to waste on a sandrail, imagine his house/car/boat/motorcycles/rv's/vacation house/jet ski's/snowmobiles/atv's and all the other ****

 

im hoping there are a lot of sponsors....

post some pics of the sandrail

 

haha, you don't normally get sponsors for a toy like that. Sandrails are just rich mans toys.

 

damn.... 200000 dollar toy.... i wish i was rich

 

So Phoenix Turbo... 'bout them there end tanks... What gives? Maybe you could enlighten me, but those things look like ****. Not the welds or anything; I can't weld aluminum, so I'm not going to go there, but the design. I mean, a sharp right angle? No internal baffling to direct the air towards the holes? Would you put a 90° angle in the charge pipe like this?



I mean, that's basically what those end tanks look like to me, and without any baffles, I see tons of turbulence as well. Maybe some turbulence is desired to aid in heat transfer, but that's what the internal fins in the rows do, right? End tanks are meant to evenly disperse the air to all the rows, and then collect them at the other end of the core in hopes of maybe even "suggesting" to the air that it go through the hole or holes, right? Or am I totally off my rocker here? If I am, please let me know and point me in the right direction to correct my misinformation.

 

It is identical to a standard backdoor style IC. Which has proven faster spool-up time and time again.

Just like this...

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by integra_gsr98 &raquo;</TD></TR><TR><TD CLASS="quote">It is identical to a standard backdoor style IC. Which has proven faster spool-up time and time again.</TD></TR></TABLE>

Why has it proven faster spool up time and time again? Because of a lower charge tract volume? I don't buy into that "bigger ICs and longer charge pipe makes more lag" camp. See, I went from a top mount IC on my AllTrac that was about 11" X 19" with about 4" of pipe between the turbo and IC and about 2" between the IC and the TB to a 30X10X3 Spearco FMIC with 11' (that's eleven FEET) of charge pipe, excluding the intercooler. Add the intercooler to the equation, and that's damn near 14 FEET of charge tract between the turbo and the throttle body. Did I notice any more lag? Nope. I had fuel lag, though, since I was using the crap *** stock vane MAF setup before I went to Megasquirt management.

But if it's acceptable to have right angles in the intercooler end tanks, why bother with mandrel bent charge pipe? Why not just angle cut and weld everything? It'd be a shitload easier and cheaper just to buy some 14 ga. mild steel straight pipe and go to town, eh?

Yeah, I know people are making one million HP on back door setups. They're also making 400 HP on log manifolds. Doesn't mean either is an efficient design...

 

Geoff from Full-Race has the facts to back up the backdoor design, however I do not know if he has posted them yet.

Remember that two single hard 90 degree bends like that tend to eliminate most bends in the charge piping providing a straight short shot to the IM from the IC.

The DSM world has been using this design for YEARS with great success. I will see if I can track down the last of the information you are looking for.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by integra_gsr98 &raquo;</TD></TR><TR><TD CLASS="quote">Geoff from Full-Race has the facts to back up the backdoor design, however I do not know if he has posted them yet.

Remember that two single hard 90 degree bends like that tend to eliminate most bends in the charge piping providing a straight short shot to the IM from the IC.

The DSM world has been using this design for YEARS with great success. I will see if I can track down the last of the information you are looking for. </TD></TR></TABLE>

I'd be very interested in reading the raw data, as they say. Thanks!

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Loco Honkey &raquo;</TD></TR><TR><TD CLASS="quote">I'd be very interested in reading the raw data, as they say. Thanks! </TD></TR></TABLE>

Faster spool my *** the air is smacking into a wall then going to do it again as it exits.

 

think pressure, not flow.
of course a straight line is always ideal, but the losses are minimal from what we and many others have seen.
this intercooler core is rated by Garrett to support 1000hp. we built a v-mount custom with the same core for Acosta Motorsports 3-rotor rx-8 that did 1142whp! same inlet and outlet style, but single 3" in and out.

pictures of the sandcar very soon!

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by PhoenixTurbo.com &raquo;</TD></TR><TR><TD CLASS="quote">think pressure, not flow.</TD></TR></TABLE>

I follow what you're saying, kind of. In the interest of helping me further my knowledge, humor me here... if we're thinking in terms of pressure, and the aerodynamics of the charge tract don't matter a whole lot (that's what I'm getting, at least), then why bother with smooth, round intake manifolds? Why bother with bell mouth inlets to the intake runners? Why worry about porting and polishing (upstream of the injectors, of course)?

I hear what you mean about thinking pressure, not flow, but pressure doesn't mean much if the flow isn't there, right? IE- small turbo making big boost and low CFMs vs. larger turbo making medium boost and big CFMs. And in the whole scheme of things, isn't the CFM rate what matters more, regarding how much fuel you can burn?

I suppose you're right, that it doesn't matter what shape the end tanks are if all that's being done is pressurizing the volume, but since that volume of air is actually moving...?

Don't get me wrong... they make big HP with end tanks like that, and I'm not saying they don't. I'm just curious about it. I accept answers from good sources, but I like to know why the answers are what they are.

Thanks for the info!

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by PhoenixTurbo.com &raquo;</TD></TR><TR><TD CLASS="quote">think pressure, not flow.

</TD></TR></TABLE>

i'm thinking along these lines also, why do we pressure test ICs and not throw them on flow benches. I'm sure with radius'd turns it would make a difference, but that could be lost by the extra piping you have to add to achive this. The lesser, properly sized piping, the faster the spool.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by b16hybridsol &raquo;</TD></TR><TR><TD CLASS="quote">i'm thinking along these lines also, why do we pressure test ICs and not throw them on flow benches.</TD></TR></TABLE>

To make sure they don't leak or burst a weld under boost? I'm a big fan of manometer testing, myself.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">I'm sure with radius'd turns it would make a difference, but that could be lost by the extra piping you have to add to achive this. The lesser, properly sized piping, the faster the spool.</TD></TR></TABLE>

See, now, I don't nessicarily agree with that. Yes, the larger the volume, the longer it will take to pressurize. That's a given, and there's no way around that. However, what's the difference between a back door setup and a "conventional" setup with bends and smoother, larger end tanks? Maybe a cubic foot or two? And how many cubic feet per minute does that turbo move? Depends on the turbo size and HP level, but let's say 400 CFM for the sake of pulling a number out of the air. At that flow rate, that extra cubic foot or two is the equivalent of an extra 0.2 to 0.3 seconds to spool. In a drag application, that's insignificant, and in a road race car or daily driver where response matters, that's hardly felt.

Again, here's what I posted a few posts back about my personal setup, which is probably the most extreme case of increased charge tract volume on Honda Tech:

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Loco Honkey &raquo;</TD></TR><TR><TD CLASS="quote">I went from a top mount IC on my AllTrac that was about 11" X 19" with about 4" of pipe between the turbo and IC and about 2" between the IC and the TB to a 30X10X3 Spearco FMIC with 11' (that's eleven FEET) of charge pipe, excluding the intercooler. Add the intercooler to the equation, and that's damn near 14 FEET of charge tract between the turbo and the throttle body. Did I notice any more lag? Nope. I had fuel lag, though, since I was using the crap *** stock vane MAF setup before I went to Megasquirt management.</TD></TR></TABLE>

What I failed to mention was that I'm using 2.5" charge pipe, too. I'd also like to point out that since I went with a speed/ density system with the Megasquirt, the fuel lag issue has been eliminated, and the turbo response is akin to a T25 on a D16.

I don't have all the answers, though. I'm just recounting my personal first hand experience. I also believe that there's lots of room for improvement on existing designs, and would like to learn more so that I can make some sense of my ideas.

 

I have always wondered why tank shape hasn't come up as an issue on H-T. Tank shape definately has an effect on flow and spool up time, think of all that wasted space in a square cooler tank as a good place for turbulence to get happening(especially when you havent made the transition up onto boost), my thinking is kinda like the effect that really huge ports have on low spead driveabillity as opposed to a well designed port that uses all its space efficiently in an NA application. A back door cooler definately reduces the plumbing length this is a definate . I'd be intersted to here other peoples comments on this topic, it is one that everyone can benifit from.

Here was one of my early attempts at a backdoor style cooler, its actually a side mount cooler for an Toyota MR2. We've come a bit further with our tanks now but it worked very well for one of our early efforts.


As was mentioned earlier you can also use diffusers/deflectors to help direct the air where you want it go, they definately work and help flow.

I'm not knocking you guys sandrail cooler. The workmanship and welding look very good and you are making great power with it, good luck with the project and finding someone crazy enough to drive it!!

 

like i said, a straight line is always best, but the losses are so minimal, if even noticable.
there are alot of things that people dwell on, when their time would be better spent looking into other areas of their setups...

the spool gains are so minimal, 200-300rpms, that its not even noticable, unless on a datalog. if you have some datalogs showing a controlled test environment on the same vehicle, with on the changed variable being the endtank design, i'd be interested in seeing those. (no sarcasm ment)

ken

 

Well, thanks for the input, Ken. Perhaps you could point me towards some books and learning materials that you've found to be reliable sources of learning?

 

nice shirt

 

Wait, you're in Phoenix? PM sent.

 

Any pics of the sand rail?

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Trooper &raquo;</TD></TR><TR><TD CLASS="quote">Any pics of the sand rail? </TD></TR></TABLE>

word, i'm with this guy, enough arguing about ****, lets see the what 200k buys us

 

backfire.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by O3DigitalBath &raquo;</TD></TR><TR><TD CLASS="quote">word, i'm with this guy, enough arguing about ****, lets see the what 200k buys us</TD></TR></TABLE>

I'm not arguing about ****. I'm trying to learn.