Flowbench design

[Richard Lewis]

I'm in the midst of designing a flow bench to use with our intake manifold and restrictor design this year. I was wondering if anyone had any experience with building this type of thing, or is using a home built one. I have access to a horde of sensors, and so on.

Specific questions:

1) What are you using for an air supply? Would shop air be adequate?
2) What sensors/information (other than pressures) are important?
3) What configuration of the apparatus is the most useful? (vertical pipe?)

Much appreciated...

-------------------------
UVIC Formula SAE Team

http://www.engr.uvic.ca/~fsae

[David Money]

We did exactly what you are talking about. Unfortunately, it is very hard to get it to do transients without using a real engine. We used a small windtunnel that pulls air (much like an engine) instead of pushing air. We used pressure sensors in front of and after the restrictor to determine the pressure differential. We only did analysis on different types of restrictors, but it could easily be done to integrate throttle controls and even manifolds with some work.

[K Thomas]

Richard, it looks like me and you have a nearly identical project.I will PM you a couple of links that I found that you may find useful, but I'm not that far into the design phase yet. If anyone has built one, I (and Richard) would greatly appreciate the input. I'm currently working on making a rudimentary flow bench to test our intake designs.

University of Louisville FSAE - Engine Team
<A HREF="http://louisville.edu/speed/rso/sae/fsae/">

[Marc Jaxa-Rozen]

Guys,

I would be very interested in this area as well. My needs are fairly limited- two pressure sensors and shop air are probably all I need- but I'd be interested in any further specific info on this.

[Richard Lewis]

I looked into some stuff today, and it looks like a decent (2hp or better) shopvac should produce right around 125cfm, which is theoretically what we can flow through the 20mm restrictor... Other than that, I am just going to use a thermocouple, pressure transducer, and a mass flow meter to get my data... At least that is the plan as of today. [img]/infopop/emoticons/icon_smile.gif[/img]

-------------------------
UVIC Formula SAE Team

http://www.engr.uvic.ca/~fsae

[Marc Jaxa-Rozen]

Seems like our shop air will only provide less than 100 CFM, but my goal at first will be to compare the pressure loss for various restrictors, so it should do OK.

Hopefully I'll find some time to work on the project, it's not a given these days.

[Frank]

I built a flow bench using Pitot probes (in an attempt to get the most from the vacuum pump available).

I test air cleaner, throttle, and restrictor.

I attach these devices to a 300 mm diameter 300 mm long piece of pipe

From here I used about 1 meter long piece of 28 mm ID pipe before the Pitot probe (in my case a dynamic only probe and wall tap). The pipe was reasonably rough on the inside (to ensure fully developed flow)

28 mm internal diameter seemed about right to be able to ignore compressibility effects yet still ensure a reasonably accurate measurement

I then had bypass valves and a workshop vacuum

I can pull 11 inches in the 300 mm pipe

I test at 10 inches (I recon this is probably a reasonable pressure. Perhaps 10-15 inches would be best. Think about [even ATTEMPT to measure] the pressure at the base of the restrictor when maximum power occurs)

I get about 75 CFM at 10 inches test pressure

The static pressure at the probes (wall tap) is about 18 inches

The Pitot pressure (dynamic minus static) is about 40 inches

I tested on a "superflow SF 600" the other day to get a comparison and got 77 CFM at 10 inches, so it's ok accuracy for a dodgy student built piece of junk

The main benifit in having a flow bench as powerful as a sf-600 is that, once the restrictor is near choked (say 40 inches, 134 CFM), you can distinguish the effects IN THE VINCINITY OF AND UPSTREAM OF THE 20mm DIAMETER ORIFICE... consider ISO 9600 data, and indeed the theory of mass flow rate devices.

Most importantly, a simple student built flow bench SHOULD be accurate for repetitive testing, expect about 1 % accuracy, the greatest source of error is due to incorrectly read test pressure

Don't forget to use a barometer and thermometer in ambient air for correction

And I recommend very long clear plastic tubes (because you sometimes get fluctuating pressures when a restrictor is flowing a bit unsteady)

I simply used water that was coloured red using Condy's crystals

In my opinion a new team should concentrate on getting the restrictor, throttle, and air cleaner ok before worrying about other areas of the engine

From there I'd cut some duration off the cams, then raise compression ratio a bit and clean up the ports at the same time.

Beware large plenums, 2 Litres max is my opinion (I mean from end of runners to throttle plate). I prefer 1 litre plenum volume, but was getting better "primary pulse tuning effects"� using more volume.

And I'd recommend using a throttle under 35 mm internal diameter (preferably with a progressive rate linkage)

[This message was edited by Frank on February 23, 2003 at 10:05 AM.]

[This message was edited by Frank on February 23, 2003 at 10:17 AM.]

[This message was edited by Frank on February 23, 2003 at 10:22 AM.]

[This message was edited by Frank on February 23, 2003 at 10:23 AM.]

[This message was edited by Frank on February 23, 2003 at 10:23 AM.]