MAF
Turbo Mini Forums

Schrödinger's cat - so which one am I ???

Dazed and Confused....

Yo-Han,

From a very quick and dirty calculation from L/sec to kg/hr, that is not a lot, just over 1/4 of what this MAF measures......

I'll have to re-check the maths but I guess an internal combustion engine on boost does draw in an awful lot of air.

I think we are looking to calibrate up to ~140l/s at ambient temperature - but I do need to check the maths.

I'm talking to Robert at the moment about his flow bench but I'll have to drop this shortly as I'm off out of the country for a while very soon.

Schrödinger's cat - so which one am I ???

Rod,
I thought as much.
Just thought I'd illustrate how much a (very good ) vacuum cleaner actually can pull.
I'd say Robert's magic black box is probably your best bet.

Escaping the fast approaching winter weather are we..?

Han

Dazed and Confused....

On 23rd Oct, 2012 Yo-Han said:
Rod,
I thought as much.
Just thought I'd illustrate how much a (very good ) vacuum cleaner actually can pull.
I'd say Robert's magic black box is probably your best bet.

On 23rd Oct, 2012 Yo-Han said:
Escaping the fast approaching winter weather are we..?

Schrödinger's cat - so which one am I ???

Is there anyone having calibration curve of HFM 6 or 7?

I can't answer the question above because I'm not quite sure what is meant by HFM 6 or 7 but, totally co-incidentally, and two years after I started this thread, myself and Graham T spent a very interesting afternoon with Robert and his flow bench yesterday.

As mentioned earlier, non-genuine Bosch copies both different configurations but both based on the 640Kg/hr models (characteristic curve 3 from the Bosch datasheet).

There are about 80 varients of this model, we assume the same sensor in the housing (whether genuine or not) but the housings are different to suit different OEMs who use them.

The genuine "test" model is 0 280 217 531, the rest are all variants on the same basic sensor but the published voltage vs mass air flow is for the model above.

Bowling and Grippo (the original Megasquirt authors) have published a document stating that the housings etc. have a significant effect on the voltage/flow curve. They have also published a programme that can hopefully turn an actually measured calibration set of points into the 1024 point table required to give an accurate ADC to a CPU such as Jean's IOx which, along with all other key measurements (temperatures and pressures around the whole intake/exhaust route) will allow us to data log the whole of the engine cycle and actually calculate the real BSFC rather than just a simulation.

Graham has pretty much all of the additional sensors installed, mine is still work in progress but we both have all the electronic hardware (24 additional ADCs on Jean's full size IOx) to log it all.

So we set off with the intention of measuring two completely different MAFs (but both based on the 480Kg/hr versions) and in two different installations, Graham's having a fairly straight run of piping and mine having a sharp 90 degree bend immediately after it, in order to compare to the standard Bosch curve.

Mine remained “work in progress” as we had to cobble one of the transition pieces out of plasticine and duct tape as I ran out of time the evening before to finish machining my adapter (the MAFs are larger diameter than our turbo inlet) but I believe the results are still valid.

Graham's, photo taken before being at Roberts


Mine on Robert's flowbench


Graham and Robert discussing the results


And the preliminary data which Graham has put into a spreadsheet (Jpeg is pretty poor, I'll try to insert the excel file below)


There are two curves for mine because we rotated the sensor by 90 degrees relative to the sharp bend to see if the installation does have an effect, especially at low flow rates as has been suggested by Bowling and Grippo.
That effect was quite small but the 90 degree bend certainly seems to have a major effect.

The three of us are still thinking about all this but it seems (a) the non-genuines aren't that far off the Bosch data, and (b) installation has the most effect in skewing the factory curve.

More later.

EDIT - flow figure corrected (I read the wrong column in the datasheet)

Edited by Rod S on 16th Nov, 2015.

Schrödinger's cat - so which one am I ???

Attached, hopefully

well maybe not..............

OK, managed to zip it, I think.

Attachments:

33.67kb

Edited by Rod S on 15th Nov, 2015.

Schrödinger's cat - so which one am I ???