This is another little project of mine I've been meaning to write up for a while but just haven't had the free time. Part I discusses the test setup I used for the measurements (in detail in case any of you guys want to build something like it) and Part II is the results. Enjoy!
Ok, background: Routinely you hear people talk about "how big XXX MAF is" or "how much HP XXX MAF can measure", but these statements are just best guesses and only vaguely have something to do with reality. Realistically you shouldn't even be talking about sizing MAFs in terms of HP; HP is not a unit of mass flow. What should you be talking about when you compare MAFs? Units of mass flow, like gm/sec or lb/min (just like fuel injectors!).
So to clear the air of the smell of bullsh*t I started measuring the different common Nissan MAFs. To date I've measured the following: Stock DE, Stock GTi-R, Z32, and Q45. I'd like to get my hands on bored versions of the DE and GTi-R MAFs, and also the VE MAF. Its just a matter of time. Anywho, I had an ulterior motive in mind for this project too. A lot of you guys know the k-value has something to do with the ratio of the maf size to the injector size. Well, it turns out the k-value is exactly the ratio of MAF size (defined as the maximum measurable airflow) and injector size, both in units of mass flow, x 1000. Thus if you know the exact MAF size (and assuming you know the injector size) calculating the correct k value becomes significantly less guess work than it currently is.
So getting down to business, how do you measure the how much a MAF can read? Well, in a nutshell, hook the MAF up to some calibrated controllable airflow, monitor the output voltage of the MAF, and then step the airflow through a given range, resulting in nice table of voltage vs. mass air flow. If only it were as easy as it sounds.
There are two big problems: first, you've got to have a calibrated source of air flow. In my case, I had this covered by already owning a precision anemometer of suitable range. More on this dude later. Quick side note, if your looking to build a similar test setup you've got two options: hunt eBay for suitable calibrated mass flow meter (new is very expensive) or use a stock MAF sensor based on the data I provide and live with the margin of error.
Second problem: your air source needs to be able to max out the MAF. This is tricky, your engine can flow a crap load of air, so the MAFs have a huge range (especially the big ones). Not having a way around this problem I cheated.
We've actually already got a mass flow vs. voltage table: the VQ maps from the rom! However, they're not scaled (well, they are by the k-value, but without already knowing the answer we can't use that), so what we've got is the shape of the curve but no values. So what I did was measure the lower half of the curve (using a dust collector as my air source) and do a little curve fit to extrapolate the larger values. This adds some error, but this is ballpark stuff anyways.
Here's my test setup. From left to right: the dust collector (air source), a valve to adjust the flow, my flow meter, and finally the test MAF with power supply and multimeter (and I finally found a use for my old HKS mushroom
).
Here's a shot of the valve I used to change the air flow. It’s just a simple slide valve. I placed it downstream of both air flow meters so any leaks don't add error.
Here's my Dwyer Anemometer. This dude is hot wire sensor just like our MAFs use but with no outer pipe. It also has a temperature probe in the tip, so it reads in standard velocity (m/s). If you know the diameter (and thus surface area) of your pipe you can convert this measurement to mass air flow. Note also the long pipe length on both sides of this sensor, the longer the better.
...and here's the end of the setup. Power supply, multimeter, and MAF sensor with mushroom filter. What’s not shown is the vacuum grease I used to seal each fitting. Any leaks contribute error, so everything needs to be nice and tight. Vacuum grease is wonderful for sealing connections but its pretty nasty stuff.
Operation of the setup is pretty straightforward: adjust the valve incrementally and record the MAF voltage and measured airflow.
I'll post Part II tomorrow, stay tuned!
