Ok, to finish off the last stuff, so rudely interrupted by a crying child (cute, though she may be....).

Really two numbers charecterize IC merit, pressure loss and thermal efficiency. This is because molecular density is a function of both pressure and temp.

So, what does the graph at the beginning of this thread mean? It's a graph of pressure loss vs RPM with a car sitting still and the motor revving. But since this is only half the picture, it in itself isn't enought to know what is the "best" IC! You also need the thermal efficiencies......

Now, some comments on thermal efficiency measurements. What sucks is that there isn't a single number that is the thermal efficeicy of a given IC! Now, why is that?

1) the amount of charge air flowing through the core is a function of engine RPM, post SC boost and post SC temps.

2) The amount of cooling air flowing through the core is a function of car speed, scoop geometry, diverter geometry, barometric pressure, ambient tempurature and relative humidity! (who knows, the phase of the moon may even be in the mix!)

So, what does this mean for IC thermal efficiency numbers? Well, they aren't absoulute. They are relative. If you have roughly the same test conditions, you can compare them to see how one IC stacks up vs another IC. So the fact that Randy is seeing numbers that are higher than me isn't too surprising, we just have different test conditions! But it does mean that you can't take a number from Randy's testing and compare it directly to numbers from my testing.

So, how could one make numbers that compare? One would have to set up a test rig. It would have to be able to flow a controlled amount of air through both the combustion charge path, and the cooling path. The temps of these two paths would have to be controllable, and then you could measure things like "at a post SC pressure of X and temp of Y flowing Z CFM of air through the charge path, a flow of A CFM of cooling air at a pressure of B and a temp of C resulted in a thermal efficiency of D". This is a royal pain in the butt!

The on car testing tries to recreate this control, but is really only a crude approximation of it. But to try to get things to compare... I use the speed and RPM of redline in second gear. Since I mostly use my car, (other than the GTT stuff Bart ran), the SC is the same as is the motor. The ambinet temps varied a bit (but not much) and the humidity varied a bit (how much I don't really know). But one thing that sucks for sure is that the differential pressure measurment I did doesn't like the dynamic envirement of HP runs to red-line, so I can't really measure density efficiency (what we really care about) while doing the HP runs. To get a look at this, I look at how much fuel the ECU is trying to squirt, because it tries to squirt roughtly proportional to molecular flow at a given RPM and load. (Even though it tries to squirt a certain amount, there's no garantee that it acutally does, the injectors may be the limit here).

Hope this helps....

Matt

 

Actually, it does... a lot......

a sincere thanks,

bb

 

Matt,

While I realize that any of the tests are not perfect because the enviroments were not 100% controlled, I still cannot fathom the Webb pre-IC numbers of 262 - 270 degrees F. The difference between these numbers and my testing done at appr. the same ambient temperature are HUGH!

Upon revieing all of my runs, the one run with pre IC temps which equalled 200 degrees F (my highest pre IC temp) was also my most thermal efficient!!! Had I not attempted to normalize the temps between runs my thermal efficiency would have continually improved.

 

I was getting up to 240 F in my testing.... But some runs were a good ~40 degrees cooler.... Some of this scales with boost, and higher post SC pressures will have higher temps, all else being equal..

Matt

 

Matt, which pulley do you run?

Bob

 

Dr O - any news on the M7 IC yet??

 

Bump

 

with an ambient temp of 80 Degree F, my
PRE-IC temp range was 187-200 (avg 187).
Post IC temp range was 138-145 (avg 142).

As I said earlier my highest pre-ic temp run produced the best thermal efficiency number, had I not done cool downs between runs, the GTT would have produced better thermal efficiency numbers.

 

Numbers asside, how do you feel the intercooler does on your car?

 

I'D say the GTT does quite well on the car.. never did any numbers but the test (if you want to call it that) that i did.. was after a ride me and several other MINI owner done. Back roads and such. when we reached where we were going. mine was the only one you could put your hand on. other that were there was Stock with stock pully, Alta with 15%, GRS with 15% and mine GTT with 19%. Thats all i needed to know that i made the best choice for me.

 

From a GTT hand test perspective, I can always hold my hand on the post IC horn when I can't touch the pre IC horn. This was not the case with the stock IC so this is why I was shocked when the thermal efficiency numbers came out lower than stock. We all want to understand what a MOD will do for us and that is why I think Matt's IC work (and others like Andy (R.I.P.) and Webb) is so important. Clearly, my testing was flawed with the cool down between runs. There is no doubt in my mind that if i re-tested and did not perform any cool down between runs that the thermal efficiency numbers jump significantly. The reason I did the GTT testing as opposed to sending it to Matt was because the GTT requires custom install (i.e. cutting) to get it in place. I may just buy thermometers or ask Matt to send them back out because its clear to me that the thermal efficiencyt numbers i oproduced are not reflective of the GTT IC.

 

How much cutting and cutting of what?

 

Given your conclusion, wouldn't it be fair to alter the opening post of this thread? Saying the GTT has the lowest thermal efficiency of the i/c's tested doesn't seem fair at this point.

 

And that's not necissarily bad. The GRS has lower thermal efficiency than the stock, and lower pressure drop, and results in higher charge density.

I don't know if the cooling down is a culprit. When I did my testing, I drove about 5 miles at 70 MPH to really cool down the intake path, and get the IC to a stable repeatable baseline. So all the other data was done with cool-downs as well.

When I crunch the numbers, there sure are some surprises in there that I didn't expect, but such is life! But in general, if you have less back pressure, you'll have less thermal efficiency, all other contraints being equal. That's just the nature of the beast.

Matt

 

There's no data to show that it isn't. Just because it has been theorized that the car was cooled between runs [and this changed the outcome], doesn't mean that this is the case. Rather, it would be highly suspect for the IC to miraculously wake up and cool any better if it saw hotter air. For it to pass the hand test after a run doesn't say much, in light of this data. Perhaps it doesn't absorb heat all that well. This is speculation, just the same.

If it meant enough and we had the raw data, we could use some inferential stats, but we'd need to know the number of readings and to have the individual data, ideally. Point is, the data says that this IC has the lowest efficiency, the testing rig (including temp sensors and hardware) was the same, presumably, and all that changed was the testing rig.

 

Two, unrelated to each other, points:
1. I have noticed that the IC horns get *hotter* over time after parking the car in the shade. I believe this is due to heat transfer, from the air and engine block and other hot masses under the bonnet. Those horns are pretty thick castings, and so can hold quite a bit of heat, and need some time to transfer and dissipate all that under bonnet energy. I also note that Matt refers to a cooling down period of driving at a constant speed to pass air over the IC, whereas Bart seems to refer to cooling down by parking the car between runs--these would produce very different results ceteris paribus.

2. From what I've seen about data here on NAM (and Webb's site) you'll never get enough observations to do any inferential statistical analysis. While I've seen referreed publications with as few as 18 observations, I've never seen that many for HP, 1/4 mile times or IC temperatures here. And to produce a 95% confidence interval that wasn't bigger than the point estimate you'd need substantially more observations (30 or 60 or even 100.) Not knowing much about the process generating the data, I'd guess that the reason we don't see any standard errors published along with the point estimates is because they would be just too big for anyone to swallow.

cheers, dr. phil

 

This makes sense and I've asked about this before.
As many have said air acts like a fluid. If fluid passes too quickly through your radiator you reduce time for conduction. Anybody that has ever driven without a thermostat can tell you how that works Don't ask...it was the 70's
Based on this there has to be an optimal time period for the air to stay in the IC. This would equate to the pressure drop across the IC.

I think we're seeing the results of this compromise.

 

Conversely, there's no (apparently reliable, repeatable, statistically valid) data to show that it is. Isn't that the point?

 

deleted

 

Bump

 

Red-Line Efficiency
38.7%
43.9%
41.9%
40.1%
35.8%
42.3%
48.5%
42.4%
34.7%
46.7%

41.5% Average
4.4% Standard Deviation

Cut and paste from Excel is a pain! These are the IC thermal efficiency calcs from pre-post IC temps at 2nd gear redline.

Matt

 

We'd need comparable data for the other ICs to find a difference between the means. That standard deviation is not very high, though- from that data, it's a 97.5% chance that the real efficiency is below 50.3% [x<2 standard deviations from mean], as you know, I'm sure. That's not very high...

 

which IC is this?

 

that Bart took. But really, one should expect a lower thermal efficiency with the lower backpressure unless there are tons more tubes, and the ends terminate as nicely as the stock unit. The stocker is very well built. Look in the ends, and the tubes are crimped in such a way that there's very little "blocked area". Each crimp forms a bit of a funnel into each tube. Look in the end of some of the others. There are square faced ends between each tube. This isn't good from a flow perspective.

Matt

 

I think you and I are waiting for the DFIC....... Peter tells me it should be next week......