MSefi.com read only forum (Jan.15,2018)

I've got a turbocharged BMW K-75 motorcycle running using hybrid Alpha-N. I'm thinking about a way to convert the bike to use a mass air flow sensor instead.

My intake consists of two 90 degree bends between the air filter and the turbo, with no straight runs where a MAF could be easily mounted. This bike is very limited for space, so a normal automobile sensor housing won't fit. I want to make my own housing for the sensor.

I'm trying to decide between two different sensors.

The first is a Toyota sensor. It's commonly found on Camrys and Priuses (Prii?) It looks like this.



The second choice is the Nissan/Ford sensor, which looks like this.



The Toyota model would be an easy install, assuming that it fits in the 1.75" pipe that leads from my air filter to the turbo. And this also assumes that the airflow is smooth enough for the sensor to operate correctly. The bends in my intake may cause a problem.

The Nissan sensor is smaller. It's got a design difference. It is normally installed in a side passage, with the upstream opening of the passage in dead air, and the downstream opening located so that there is a venturi effect. This draws a small amount of air past the sensor.

I'm not sure if this sensor can be mounted directly into the intake. It may not operate correctly. I suspect that it will work fine if I fabricate a side passage system similar to what the stock sensor housing has. That would be a lot of work, though.

So, on to questions.

1. What are the dimensions on the Toyota sensor? (If nobody knows, I can go by NAPA and measure one.)

2. Will the Nissan sensor work if placed directly into the intake airstream, or does it need the side passage?

Watch this space, I'll try and post progress reports.

Fred

Mega Squirt has a built in MaP sensor.

MAF sensor. Not MAP sensor.

I'm pretty sure that if you were able to get your intake system (with a fabbed MAF housing) onto a flow bench while measuring MAF output you would be able to characterize your system fairly accurately. Any nonlinear behavior should be consistent and allow you to tune for said behavior.

Mike, thanks for your reply.

I suspect that you're right about this. My only concern is what will happen if the airflow changes in such a way that it stalls or partially stalls at the sensor. That could mess me up. But I won't know until I try.

The Nissan sensor is smaller, but I don't think it's going to be happy unless it's set up with the proper bypass system.

I'm going to give the Toyota sensor a go. It looks like an easier install since it can go directly into the airstream. I'll probably put it in and do some datalogging with it before I actually give it control of the fuel map.

Cheers,
Fred

You can't win of you don't play! Let us know your progress if you don't mind.

-Mike

Fred,
I did a bit on characterizing the MAF sensors. On the inlet tube, it might a good idea to look at the cross section and determine the velocity of the air flow to make sure that sensor will be operating in its range. Mike's idea with the flow bench should work really well. If you don't have a flow bench, it is possible to use the engine and a narrow band sensor and let MS data log MAF voltage with everything else and then back calculate for the MAF curve. I did this and had excellent results - see viewtopic.php?f=64&t=33823&start=60 page 4 or so. I don't think I would want to run a turbo with a narrow band sensor; a wideband could work but will have some more offsets in it. The math for back calculating the air flow would be a little different.

Good luck,
- Charles.

Charles,

You and I appear to be thinking the same way. I just finished calculating the intake airspeed. With my 750 CC engine at 9000 RPM and pushing 6 pounds of boost, I'll see an intake air speed of 33mph in the 1.75" 16ga intake pipe.

If I up the boost, I could potentially see a 50+ mph wind in the intake.

What I think I'm going to do is mount the Toyota maf and do some test rides to see what my range looks like. The Toyota MAF looks like it could be scaled for different air flow rates just by scaling the size of the pipe it's stuck into.

My next step will be to get into my neighbor's Prius and measure the intake pipe where the MAF is located, and see what airspeed range Toyota sets it up for. I think that if I can match their airspeed at redline, then the MAF should behave itself.

I'm going to go through the thread you posted and try out your excel file. It looks interesting!

Frederic