markjmd

iTrader: (11)

Question aimed mostly at mappers, tuners and other fueling experts, but which I hope could eventually be of benefit to a lot of members:

Assuming some way could be found to significantly increase the laminarity of the air-flow passing through a MAF (hot-wire type), where would you expect to see the most obvious potential gains - reduced fuel consumption, higher power, less DET, all or none of the above?

The project I've got in mind is an evolution or improvement over the wire-mesh grille that normally sits on the filter-side of a standard Subaru MAF.

PMs welcome if anyone wants to know more.

sunny1989

Sounds very interesting, how do you plan on improving the properties of the flow? are you looking at using flow straighteners? any computational modelling etc..

i'm just interested as i'm a grad aerospace engineer, researched and developed flow straightening methods previously so this seems like a nice project..

dunx

iTrader: (3)

But the system is calibrated "as is" ? Improving the mass flow rate would instantly involve either re-scaling the MAF calibration, or a re-map. In extreme use a blow-through set-up may provide a far superior flow regime... (?)

dunx

P.S. How about pre-swirl into the turbo entry, to alter the momentum of the air entering ?

markjmd

iTrader: (11)

Yep, that's pretty much it, an improved form of flow-straightener. Looking at the standard piece of mesh on a stock MAF again, that apparently isn't there to prevent foreign objects from getting into the air tract (the air filter should be doing that), but to smooth and straighten the flow of air. In my view though it's a pretty crude piece of kit, and could potentially be greatly improved.

Based on some designs I looked at that are used in water jet applications, the kind of things most often used are small-bore tubes packed together and various grades of mesh, so I took that as my starting point while I hunted around for ideas. I'm now at the point where I think I've found a convenient and more or less off the shelf product that could fit the bill, that'll neither be prohibitively expensive nor restrict air-flow any more than the stock mesh, so I want to get a clearer picture of what's to be gained from going ahead and fitting the thing.

markjmd

iTrader: (11)

The point isn't to actually change the flow rate measured relative to the real volume of air entering, but to avoid unwanted up and down spikes in measured flow rate that would be caused by turbulence in the air as opposed to an actual change in air flow volume. Basically, trying to make the MAF more accurate, I suppose.

sunny1989

sounds good, we've used off the shelf honeycomb matrix for a few developments (small scale wind tunnels, aircraft anti icing bleed air simulation rigs), different matrix forms are implemented in star ccm to simulate their capabilities as a flow straighteners based on the application. It may be worthwhile you implementing your straightener in a cad package then cfd package to see its benefits in real time (if you've got access to them that is). Its a very worthwhile step as the straighteners sometimes have adverse effects that can't be predicted or these effects just take ages to identify through hand calculations, restrict the flow too much etc. The mesh screens that can be used sometimes cause a considerable pressure drop, only if several screens are used though..

When I replaced my MAF I was wondering what the purpose of the mesh was, I was under the impression it was a redundancy filtration feature. But if its a flow straightener it could definitely be improved upon!

I'm sure by creating a more laminar airflow the accuracy of the hot wire anemometer would be improved, thus improving the running of the engine as the reading would be more accurate. Not sure on these side of things though as its not my field..

great idea anyway buddy, hope all goes well

joz8968

iTrader: (13)

Cram the "filter side" of the MAF tube with loads of drinking straws.

saiklon

You want to improve the MAF accuracy but I don't know of anyone complaining about MAF accuracy. (I understand that the fueling stability when using MAP based methods is due to the MAP sensor measuring conditions in the manifold as opposed to the MAF sensor that measures conditions pre-turbo i.e. further away from the cylinder).

Have you quantified the inaccuracy of the MAF sensor under different conditions so as to know if you've improved/worsened accuracy? How will you measure the difference between your suggestion and the stock wire mesh?

I'm not trying to discourage you here but I think that if you're going to attempt to improve something as complex as airflow to the MAF, you need to carry out proper scientific experiments if you're going to convince any one else.

dunx

iTrader: (3)

And offer a free re-map ? Who has ever complained about MAF inaccuracy ?

dunx

Alan Jeffery

You can easily improve the air flow through a MAF by removing all the internal obstructions. Then you fit a MAP sensor.

markjmd

iTrader: (11)

I have to admit some of the comments did make me chuckle, and the drinking straws idea isn't as crazy as it sounds.

In all seriousness though, are MAFs really not in the slightest bit prone to under or over reading slightly, if the air-flow isn't as smooth as it could be for one reason or another? I appreciate we might be talking about quite marginal amounts, but if there's a practical way to improve things, why not do it.

dunx

iTrader: (3)

Alright, if Subaru installed a deliberate "spoiler", to avoid laminar flow, WHY ?

dunx

markjmd

iTrader: (11)

Not sure what you mean here. The wire-mesh grille which comes in a stock MAF will make the air-flow more laminar, but only to a certain extent. Were you talking about something else?

Alan Jeffery

I'd say the point is that the MAF works just fine as Subaru intended it. What it's doing is feeding information regarding intake air volume, required by the ECU to determine load correction. If you change the air filter system or MAF intake in any way, without getting at the mapping, the output will differ and the ECU won't know which way is up!

Having a MAF that flows more actual air isn't the point of it. What you would do is use a larger MAF, then rescale accordingly, but you'd only do that if the incoming air is of a much greater volume, and overcomes the MAF range.

Alan Jeffery

The spoiler is intended to avoid vorticies in the tube disrupting the sensor's ability to measure air flow across a range of air speeds.

stedee

iTrader: (4)

i have seen that straw thing work albeit with water but afaik air acts the same as liquid dynamically wise but wouldnt it be beneficial as in more air can be flowed to the turbo.

i did have a similar idea but with a fan kind of set-up, the point being to spin the air as in whirlpool effect to again flow more air, that was more for a n.a engine, couldnt be arsed in the end

dunx

iTrader: (3)

Do some maths....

8000 rpm. 2 litres, 2 bar of boost = 24 m^3 per minute through a tube less than 100mm in diameter...

dunx