- Joined
- Feb 18, 2016
- Points
- 231
- Model of Z
- Z3M Imola and Z4 (e89)
Maybe, just maybe on a stock engine BMW actually designed and fitted this component with the best compromise?
MAF screens removal
- Thread starter windym
- Start date
- Joined
- Dec 8, 2011
- Points
- 145
It was certainly a compromise. Ideally, there would be a MAF for each throttle body.
Only the McLaren F1 and the Bugatti Veyron were not built to a budget.
- Joined
- Dec 8, 2011
- Points
- 145
Well, I can make a standard screen...
Now I need to work out how to test it against a modified one.
I'm thinking of logging airmass data via TestO to measure airflow through the MAF, an anemometer to measure airspeed and the manometer to measure pressure drop.
I'm still not sure about how to create enough smoke to visualise straightness of flow.
[edit] I've just ordered a smoke machine + smoke fluid (£10.60 to pay, using £10 eBay voucher + 20 Nectar Points + 10p cashback from Quidco - every little helps
)
Now I need to work out how to test it against a modified one.
I'm thinking of logging airmass data via TestO to measure airflow through the MAF, an anemometer to measure airspeed and the manometer to measure pressure drop.
I'm still not sure about how to create enough smoke to visualise straightness of flow.
[edit] I've just ordered a smoke machine + smoke fluid (£10.60 to pay, using £10 eBay voucher + 20 Nectar Points + 10p cashback from Quidco - every little helps
)
Last edited:
If you are able to "straighten" the airflow, what difference do you anticipate it will make to the MAF sensor reading?
- Joined
- Dec 8, 2011
- Points
- 145
I have to assume that the airflow is already straight enough for the MAF to give accurate results.If you are able to "straighten" the airflow, what difference do you anticipate it will make to the MAF sensor reading?
The trick will be to use the best shape that produces the same straightness of airflow, but has the lowest restriction to airflow.
Probably but this is fun to see!
By the way don’t you have a heavily improved by Alpina Z4 ..
- Joined
- Feb 18, 2016
- Points
- 231
- Model of Z
- Z3M Imola and Z4 (e89)
Still has a stock MAF.Probably but this is fun to see!
By the way don’t you have a heavily improved by Alpina Z4 ..
- Joined
- Dec 8, 2011
- Points
- 145
I've done some tests with a smoke machine. My conclusion is that you want the holes to be as small as possible with the thinnest side walls.
I've just started making a screen with a 4mm hex hole and has 0.5mm side walls...
Hopefully, I will be able to test it tomorrow.
I've just started making a screen with a 4mm hex hole and has 0.5mm side walls...
Hopefully, I will be able to test it tomorrow.
- Joined
- Oct 10, 2015
- Points
- 123
Thinking things through, why not place the honeycomb screen in the air filter exit, in my case I have a simota type.
This would allow a larger screen area to be used, looking at the video of the airflow/smoke test through the screen,
the smoke appeared to be straighter or less turbulent slightly further passed the screen into the MAF.
By placing the screen further away from the MAF it looks to be possible to avoid the initial settlement of turbulence in the now straightened air flow
Andy
This would allow a larger screen area to be used, looking at the video of the airflow/smoke test through the screen,
the smoke appeared to be straighter or less turbulent slightly further passed the screen into the MAF.
By placing the screen further away from the MAF it looks to be possible to avoid the initial settlement of turbulence in the now straightened air flow
Andy
- Joined
- Dec 8, 2011
- Points
- 145
Some tests...
I think that the purpose of the screen is to straighten the flow. I am still to test the resistance to flow of the different screens using a manometer, but it looks like Hex_06 straightens the flow more than is necessary, so I could make its holes bigger.
An approximate estimate to flow resistance is the cloud on the windward side of the screens. It looks like both of my screens are worse than standard, but only the manometer will tell the truth.
[Video updated on 25/07/18 @ 21:26. It now includes Hex_07. This is a 4.5mm hex with 1mm side wall. The video also shows slow-motion details at the end of the video.]
I think that the purpose of the screen is to straighten the flow. I am still to test the resistance to flow of the different screens using a manometer, but it looks like Hex_06 straightens the flow more than is necessary, so I could make its holes bigger.
An approximate estimate to flow resistance is the cloud on the windward side of the screens. It looks like both of my screens are worse than standard, but only the manometer will tell the truth.
[Video updated on 25/07/18 @ 21:26. It now includes Hex_07. This is a 4.5mm hex with 1mm side wall. The video also shows slow-motion details at the end of the video.]
Last edited:
- Joined
- Dec 8, 2011
- Points
- 145
I've done a lot of testing and I can't make any improvement over standard.
The OEM screen is 3.5mm square holes with 0.65mm side walls.
By observation of the smoke tests, a 4.5mm honeycomb straightens the air as efficiently as the 3.5mm square screen.
The thinnest side walls that I can make are 1.0mm. I tried to make a screen with 4.5mm hex honeycomb with a 1.0mm side wall, but my rendering software couldn't draw the image without leaving gaps. The best screen that I made was Hex_07 (4.5mm hex with 1.0mm side wall), but the honeycomb was joined at the tips rather than the edges.
I've used my homemade manometer and the measured differences are so small as to be insignificant.
I've just bought a digital manometer which will be delivered in about a week. I plan to retest everything using it.
The OEM screen is 3.5mm square holes with 0.65mm side walls.
By observation of the smoke tests, a 4.5mm honeycomb straightens the air as efficiently as the 3.5mm square screen.
The thinnest side walls that I can make are 1.0mm. I tried to make a screen with 4.5mm hex honeycomb with a 1.0mm side wall, but my rendering software couldn't draw the image without leaving gaps. The best screen that I made was Hex_07 (4.5mm hex with 1.0mm side wall), but the honeycomb was joined at the tips rather than the edges.
I've used my homemade manometer and the measured differences are so small as to be insignificant.
I've just bought a digital manometer which will be delivered in about a week. I plan to retest everything using it.
- Joined
- Dec 8, 2011
- Points
- 145
For the record - using my homemade water manometer, the measurements (cm H2O) were...
01. Open Leaf Blower = 4.0
02. MAF (no front screen, no rear screen, no sensor) = 4.1
03. MAF (no front screen, rear screen, no sensor) = 4.0
04. MAF (OEM front screen, rear screen, no sensor) = N/A
05. MAF (Hex_02 front screen, rear screen, no sensor) = 4.2
06. MAF (Hex_07 front screen, rear screen, no sensor) = 5.0
07. MAF (OEM front screen, OEM rear screen, with sensor) = 4.0
08. MAF + Airbox (OEM front screen, OEM rear screen, with sensor, OEM air filter) = 6.0
09. MAF + Airbox (OEM front screen, OEM rear screen, no sensor, OEM air filter) = 6.0
10. MAF + Airbox (OEM front screen, no rear screen, with sensor, OEM air filter) = 6.2
11. MAF (Hex_02 front screen, no rear screen, no sensor) = 5.2
12. MAF (Hex_07 front screen, no rear screen, no sensor) = 4.8
13. MAF (OEM front screen, rear screen, no sensor) = 4.0
Conclusion: It is very difficult to measure the very small distances using my homemade apparatus, but the OEM MAF is very efficient as it produces no measurable drag (see Test 07 compared to Test 01)
01. Open Leaf Blower = 4.0
02. MAF (no front screen, no rear screen, no sensor) = 4.1
03. MAF (no front screen, rear screen, no sensor) = 4.0
04. MAF (OEM front screen, rear screen, no sensor) = N/A
05. MAF (Hex_02 front screen, rear screen, no sensor) = 4.2
06. MAF (Hex_07 front screen, rear screen, no sensor) = 5.0
07. MAF (OEM front screen, OEM rear screen, with sensor) = 4.0
08. MAF + Airbox (OEM front screen, OEM rear screen, with sensor, OEM air filter) = 6.0
09. MAF + Airbox (OEM front screen, OEM rear screen, no sensor, OEM air filter) = 6.0
10. MAF + Airbox (OEM front screen, no rear screen, with sensor, OEM air filter) = 6.2
11. MAF (Hex_02 front screen, no rear screen, no sensor) = 5.2
12. MAF (Hex_07 front screen, no rear screen, no sensor) = 4.8
13. MAF (OEM front screen, rear screen, no sensor) = 4.0
Conclusion: It is very difficult to measure the very small distances using my homemade apparatus, but the OEM MAF is very efficient as it produces no measurable drag (see Test 07 compared to Test 01)
In conclusion
BMW (or bosch) knew what they were doing
BMW (or bosch) knew what they were doing
Good that logic prevails. Interesting finding.Logic says there must be a reason for it (its been designed and manufactured at a cost). The reason can't be a filter as its too course for that.
I'd leave well alone but will be interested in your findings.
- Joined
- Oct 10, 2015
- Points
- 123
Hi, certainly very interesting shame you could not test with thinner side walls, did you remove the wire mesh from behind the front screen?
Andy
I think most of us would find that the results are of no surprise
- Joined
- Oct 10, 2015
- Points
- 123
Dear Sean,I think most of us would find that the results are of no surprise
What makes people carry out research? This is a question of fundamental importance.
Curiosity is an intrinsic characteristic of the human brain and a powerful drive in the exploration of man’s surrounding environment. To undertake research is to discover and understand the facts. By researching, conclusions may find for the original, this may be due to current limitations but does not detract from the legitimacy or the results of the research. Accordingly, curiosity for the known vs the unknown is a fundamental element in research.
Science sits on the implicit and unshakable belief that all events have cause. It can be argued nothing can occur or exist in an uncaused state. Research by design makes use of multiple techniques to examine various problems. Studying effectiveness of a technique, to apply to a cause or effect either known or unknown, is primarily responsible for results in any research.
As a result, research and the researcher are keenly interested in analysing existing techniques looking to improve and advance in them. Thus the refinement of techniques is another reason for conducting research.
Andy