Air filter pressure drop results

[pboglio]

Well,

I did a little pressure drop experiment. I run a very large new K&N cone filter (9.5" long & 6.6" dia) on a 2g turbo eclipse. This air filter has a flat area of about 168 sq inches, enough flow for 970 cfm or 650 h.p. I basically run a big K&N airfilter, HKS VPC, NO MAF, and a big 4" aluminum pipe tapering into a RRE 2.25" turbo inlet pipe. I used a Magnehelic Pressure Differential Gauge (0-100 inches of water) to measure the air filter pressure drop, ultra accurate. Measured at the little nipple right on the rubber MAF inlet hose. I figure my pressure drop should be next to nothing. My car was run on 20 psi and maybe kicking out 360 crank h.p., according to DSMLINK.

I was expecting maybe 4-6" of water pressure drop thru the huge K&N airfilter which is 1 week old. Boy was I in for a surprise. Light cruise there is no pressure drop, under WOT there is minimal until 4000 rpm then it hits 10" pressure drop. Max pressure drop is at 5000-7000 rpm at 24" of water on 20 psi boost pressure on a 45*F night. For you guys that don't know, 29" of water equals 1 psi.

This goes against every thing I've read, that the air filter has little to no pressure drop. I say bullshit. I'm thinking of running a 14" long cone filter instead. Basically I have .82 psi of pressure drop just thru the airfilter. After the pressure ratio rise in the turbo compressor this would be almost 2 psi lost after the turbo. You cannot go too big on an airfilter.

Gene

 

[pneumo]

If this is a differential gage I assume it has two sensors. Where was the second sensor placed?

I'd also be interested to know if there is any pressure drop between the outside of the car and the area just before the filter.

Thanks

 

[pboglio]

The second port was left open with the gauge sitting in the passenger seat inside the car. I'll try and run the high pressure port to the engine compartment just outside the airfilter to rule out any variables. I'm having a hard time believing myself an airfilter can drop that much pressure, means there might just be horsepower to be gained from a huge airfilter. I'm running a 4" cold air hose to the engine compartment too. Fluid dynamics is not always intuitive. I'll run a couple more variations on this test to verify the measurements.

Gene

 

[thegreatms]

I'll try and run the high pressure port to the engine compartment just outside the airfilter

sounds like a good idea.

definately keep us updated, this is really good info.

 

[pneumo]

Thanks for the info. I remember Dave Johnson from Dejontool did some testing with a differential pressure gage many years ago. The results were posted in the Talon Digest. Archives can be found at TeamNABR.com. He uncovered some good info, too.

It's possible the passenger compartment can see slightly higher pressure if the ventilation controls are set for fresh air. The area just in front of the windshield is a high pressure area, which is where the air enters the ventilation system. I wonder if you would see a difference in the results if you changed from 'fresh' to 'recirc'?

You've got me dreaming about all the fun experiments I could do with that gage. IC pressure drop, exhaust backpressure, IC piping pressure drop, etc. Where do you get a gage like that?

 

[pboglio]

You can buy one from dywer instruments $60 in any range from 0-1" all the way to 0-150" of water. Its got a blow off plug rated at 25 psi so no high boost action. Vacuum is also rated at 20" HG, so no hooking up to the intake manifold. Exhaust backpressure is out of the question since its rated for maybe 150 *F.

I'm doing the I.C. pressure drop test tomorrow so I'll post those results. Intercooler efficiency tests I'll post when I get my temperature data logger next week with 2-3 temp sensors. Yeah, this thing has got all sorts of uses. I think I might have to buy a 0-10" gauge too for testing pressures inside Intercooler & Oil cooler ducts and such. Cheers.

 

[pneumo]

$60? I thought something like that would be $600! You gave me another idea- pressure in ducts. Is there enough of a pressure difference between the sidemount IC duct and the passengers wheelwell to pull enough air through for good cooling of the sidemount IC? How about for a drivers side oil cooler? I guess I'll have to find out for myself if my custom Rubbermaid and milkjug ducting works! Thanks

 

[pboglio]

Um, the duct work pressure differentials has got to be a pretty untested subject as far as amateurs goes. If you get the low pressure (0-15") gauge you can test it and tell me Probably be able to prove whether ram air ducts work or not too. I've got a whole slew of tests coming up in the next month, mostly temperature related.

Read this website SDS EM-4: Aircraft , got some good info on intercooler duct design and testing using the gauge I have. This guy is running a subaru turbocharged engine in a single engine prop plane. This website has more test data than the entire DSMTUNERS website. He's got pages on cylinder head flow testing, prop design, cowling design, you name it. Just look around. Cheers.

 

[ICUB4UCME]

For you guys that don't know, 29" of water equals 1 psi.

Gene

i always thought that 29" of water was roughly equal to one atmosphere, not one psi. and one atmosphere is aproximstely 14.7psi.

please, correct me if i'm wrong and by all means if thats the case explain to me why i;m wrong and how it all works

 

[mosier]

You can find dwyer magnehelics on ebay for around $30 shipped. For testing before the turbo, you'd want to be able to measure pressure differences less than 60 inches of water ( ~2 psi), since using a 0-150" gauge doesn't offer much resolution.

dejontool (http://www.dejontool.com/dsm-cip-faq.htm) tested the stock 1G maf and found a 3 psi drop at 6000rpm and 18 psi. I'd like to see someone confirm these results and also take readings directly before the turbo.

For the exhaust back pressure, you need the bigger gauge, and a big coil of thin copper piping from a hardware store--it'll keep the temps down enough to read with the gauge.

I plan on getting one to see, like said above, if you can build any pressure in front of the stock sidemount by sealing the ducting around the IC. If a decent amount of pressure can be made, then a cutout above that area and under the air filter could make a nice pressurized intake, using a sealed airbox.

1 psi = 27.6 inches of water
1 in Hg = 13.6 inches of water

 

[pboglio]

A sealed air box on a turbo is restrictive. I ran one and tore it out and have not looked back. Only part I kept was the 4" cold air duct blowing on the air filter.

I don't know what the pressure drop is on a 2g MAF but I know pulling it out and replacing it with a 4" pipe gave me 28 whp.

 

[habitatguy187]

Sorry to bring this thread back from the dead but basically what this means is that a smaller intake/air filter makes the turbo work harder to achieve the same airflow as a large intake/air filter correct? So switching over to a larger intake/air filter would decrease intake temps allowing for more boost on pump gas right?

Pboglio, when you say you gained 28whp (according to link I'm guessing) by removing the 2g mas did your boost raise when you removed it? (without touching the boost controller)

 

[pboglio]

Yep, you want a big airfilter, lower restriction means less compressor work required.
My boost always increases 1-2 psi switching from MAF to speed density. Problem is the
timing and fuel also change so drastically that a good part of the gains comes from the leaner A/F's and high ignition timing. Technically, not all of the power gain was from the lack of intake restriction, but from inherent tuning changes that go along with the MAF to speed density conversion.

 

[turboglenn]

I've always thought about checking this out with my mercury manometer (for syncing ITB and Bikes) but have always had something that was just a tad more pressing when it came to having time to play with the car. That's very surprising info to say the least... No wonder the dragh cars just run a screen :

I've always tried to run the largest K&N that I could find to fit the piping i run (mandrel bent 3'' AL from OEM SMIC location) but have been re-thinking the whole system. At least now i'm SURE that i'm finding the most massive element i can find.

 

[MIdsm]

A water gauge would be more accurate than a mercury gauge - more movement for the same amount of pressure . When pressure testing filters the pressure probes ought to be on both sides of the air filter, not inside the pipe. The pressure in the pipe drops as the air velocity increases.

Browser Warning

 

[turboglenn]

^^^ Since mercury has become harder to obtain due to legislation in california and other places, I was recently forced to use the newer water type manometer and was just not pleased with the level of accuracy it gave when compared to my mercury filled unit (tended to jump around too much, probably because the water/dye solution is so uch lighter that merc. So after a few calls to industrial supply places i was able to set up an acount to order some mercury and sold the POS blue water version... call me picky, call me oldschool, but i'll take mercry over water any day.

 

[RipperXX]

Did anyone ever do any further research? I knew this was true, but I didn't realize that as large of a filter as the OP was using would still cause such a large pressure drop in the intake to the turbo as he encountered. Thats simply nuts to me!

 

[We're on Boost]

Did anyone ever do any further research? I knew this was true, but I didn't realize that as large of a filter as the OP was using would still cause such a large pressure drop in the intake to the turbo as he encountered. Thats simply nuts to me!

It might not be so far off. Garrett, in their turbo tech 103, figures 1 psi pressure drop in the intake before the turbo compressor. This could include a MAF though, they don't really say. Their numbers are pretty general.

I called K&N one day to ask them what the pressure drop is that they use for rating the flow of their air filters, since it makes no sense to give a flow rating without an accompanying pressure drop figure. You know, xxxCFM @ xx psi or whatever. But the guy gave me a number that was ridiculous in the opposite direction. Something like .01 psi. I didn't even write it down whatever it was, don't think it was right.

 

[pneumo]

Yes, the maf has some pressure drop. The 2g and EVO maf have a baro sensor. Recently I logged the baro sensor during WOT pulls on my Laser. With a 2G maf pulling around 2200HZ it showed between .49-.56 psi pressure drop at the maf. I swapped an EVO maf on the same setup and logged .32psi pressure drop with boost pressure rising 1/2 to 1psi over the 2G maf setup. I'm sure the rest of the turbo intake pipe has a little pressure drop too.

 

[Budget90GST]

If you don't mind running without a filter for a run or two so you can eliminate the pressure drop from the whole intake. Basically, run one without filter - record, then another one with filter - then record. The difference is the filter pressure drop. Just a thought.

 

[We're on Boost]

If you don't mind running without a filter for a run or two so you can eliminate the pressure drop from the whole intake. Basically, run one without filter - record, then another one with filter - then record. The difference is the filter pressure drop. Just a thought.

I was thinking about that too, but then I thought, would I do that?
Nah.

The other thing I'm not clear on is where is the baro located in the 2g and evo maf. I don't know. Is it upstream or downstream from the restrictive part of the maf?

 

[pneumo]

On a 2G maf the reference hole for the baro sensor is located just behind the honeycomb in the main metering chamber.

 

[gixrman]

Intercooler efficiency tests I'll post when I get my temperature data logger next week with 2-3 temp sensors. Yeah, this thing has got all sorts of uses. Cheers.

Did you ever do this? I know it was 2yrs ago but I was looking into this for before and after IC temps as well as W/I and could never really find a reasonable temp. datalogger.

 

[We're on Boost]

Did you ever do this? I know it was 2yrs ago but I was looking into this for before and after IC temps as well as W/I and could never really find a reasonable temp. datalogger.

Yeah, hmm, the Innovate DL-32 looks interesting.
DL-32 Multi-sensor Universal Data Logger.

http://www.innovatemotorsports.com/support/manual/DL32_Manual.pdf


"Each of the 5 channels on the DL-32 can be configured to accept input from an external 0 - 5V sensor. Raw sensor data can be converted into meaningful units and values using the input configuration features of LogWorks on a PC."