87 13b. Does anyone know if the air flow meter needs to be mounted horizontally. when i did my transplant i ran out of room to mount the stock air box up front. i can get a aftermarket k&n cone filter in there at an angle though. i know that the opening and closing of that gate controls the fuel pump but, i thought i read somewhere in here that the meter had a sensor in it that would shut off fuel if it was flipped over. (in case of an accident i think). any info is greatly appreciated. thanks. mike

 

Always mount flapper typ AFM's horizontally, otherwise the flap is affected by gravity, screwing up mixtures. Don't bother with the stock arbox, it's very restrictive. Go with pod filter, heat shield and cold air duct .

 

You may also mount the AFM rotated 90 degrees, with the plug pointing straight up. This is the position that is used by the Bonez system sold by Rotary Performance! This has worked perfectly in my 87 TII and my 88 SE. I compared performance in my TII with the AFM mounted horizontally and rotated, the performance was actually better rotated. It damped out AFM fluctuations (Thank you gravity). Just my 2 cents worth. Have fun all!

Irv, Keith's dad

 

Thanks guys, that what i was loooking for!!

mike e

 

I really do not recommend doing this.&nbsp It might be that this "trick" is hiding some other problem you are having.

Damping is a function of the air filter itself.

The ECU is calibrated to run with no gravity effect on the flapper door.&nbsp You're probably running too rich and significantly hurting mileage due to this trick.&nbsp The rotary engine runs fine on the rich side - try running around with an unplugged O2 sensor; I'd bet it runs a little better than with it plugged in!



-Ted

 

Hey RETed,

Your theory sounds good. Now, let me acquaint you with the facts. At anything below 5psi boost, I run perfect stoich. Above that, I run rich, of course - but you and I both know that the AFM is not in the loop at this point. As far as my running way rich, I am now running NGK BUE spark plugs in leading. This plug is a full surface gap plug - it has no heat range (I think it is equivalent to a 12 or 13 NGK heat range number), and I do not miss or foul at idle, cruising, or WOT. If I was running so rich, that would not be possible. Also, the car performs better than any stock TII I have ever driven. If this is how a "problem" car runs, then everybody should have this "problem" imho. Oh yeah, I average about 12-15 MPG in town. Maybe this is low, but considering I can't seem to keep her out of boost, and Keith is even worse than I am in this department, I believe that this is quite reasonable mileage under the circumstances (the engine has over 120K miles on it). Oh yeah, and while we're on the subject, my car consistently exceeds emission specifications every year, as certified by the State of Georgia via dyno test. Yes, I agree with you - gravity does affect the flapper door when the AFM is not level - but not significantly imho

Irv, Keith's dad

 

Humm... I never ran an flap AFM sideways... You both do know you can adjust the flap and how easy it is to open and close. Just pull the black plastic off the top and move the white wheel... If you run it sideways i would think you should turn the wheel, to make it harder/softer depending which way the flap is, to compinsate. I did have to make the AFM softer on my SE when I added the turbo because it would not idle correctly.

 

How did you measure stoic?&nbsp Wide-band?



-Ted

 

Narrowband O2 sensor installed in the car. About 6 months old. You aren't going to suggest that I need a wide-band sensor to measure Stoich, are you???

Irv, Keith's dad

 

The only time you can be sure you're running stiochiometric mixtures is when the ECU is in closed loop and you can see the LEDs swinging back and forth aroung the middle of the gauge. The ECU will be correcting for the AFM being mounted sideways. Under load when the ECU is refering to maps for mixtures (open loop) you shouldn't be running stiochiometric, so a narrow-band meter is pretty useless. This is when a sideways mounted AFM will cause errors because the relationship between airflow and flapper movement (hence calculated fuel quantity) has changed.

 

Do you even have to ask?


-Ted

 

Ok, now I am confused. If you look at the Autometer site, tech info section, it clearly states that the problem with narrowband guages is that there are errors when the sensor tries to indicate lean or rich. Yet, at all temperatures Stoich is shown as defined with a voltage swing of .2 to .7 volts for 14.6-14.8, regardless of temperature. Silly me, that strikes me as incredibly accurate (that is, if the voltage is within that very wide range, you have stoich exactly!!!). Oh well, I gotta hurry up and finish the DOY wideband sensor I am working on so I can measure Stoich even more accurately!
OK, people here seem to be intent on pointing out the holes in what I am saying. Funny, I seem to be the only one that has actually tried both positions! The only difference between the 2 positions is that there was slightly more surging in the horizontal position when running closed loop (amazing that I even know and understand what closed loop is, isn't it? ). Yep, musta been because of the different position of the air filter . Also, I've tried this on 2 different cars (one NA, one TII) with no noticeable differences. So, dear readers, it is up to you - you can take the advice of the theoreticians who will tell you why this won't work, or the advice of someone who has actually tried it both ways and is currently running just fine in the vertical position, thank you. I am done with this thread. Have fun, all!

Irv, Keith's dad

 

I hate being misquoted. I never said it wouldn't work, I said it would affect your mixtures. Gravity does have an effect on the output of the AFM if it's turned om it's side. This will have an effect on the mixtures calculated by the ECU. This ain't theory, this is fact. You will no longer be running the mixtures tuned by the factory. I'll let you decide if this is a good thing or not.
As for narrow-band O2 sensors, if you've ever seen the ouput vs. mixture curve, you'd know why these are only used for closed-loop tuning for low-medium loads. They're only suitable for seeing if you're running above or below stoichiometric, nothing else.

 

I've run it backwards and almost every configuration you can think of.&nbsp It always ran the "best" when the flapper door swung horizontally for me.&nbsp I didn't claim it was not going to work either; I just said it wasn't a good idea, because you're changing conditions that Mazda had set through R&D.&nbsp Now before this comes down to a big re-engineering debate, I'd like to repeat that all I said was that I didn't think it was a good idea.



-Ted