calibrating a maf sensor

[dieselgeek]

Hm, just read that it should be at 90° ATDC.

Make sure that's on the right portion of the cycle. Physically 90* ATDC, but the ECU might see the intake stroke as something different.

[Belgian1979]

Hm, just read that it should be at 90° ATDC.

Make sure that's on the right portion of the cycle. Physically 90* ATDC, but the ECU might see the intake stroke as something different.

Not sure how I have to determine that as I have no way of telling where in its cycle it is.

[mk e]

Not sure how I have to determine that as I have no way of telling where in its cycle it is.

Is the ECU doing spark? If so you have the crank position set when you set the timing I'd think.

With a timing window you will get a stronger signal at 90ATDC in theory, but MAP sensors don't react very fast to start with and for you with a large plenum I'd be very surprise if it makes much difference.

[dieselgeek]

Hm, just read that it should be at 90° ATDC.

I do not have a crank wheel, so I cannot measure crank degrees.

I thought you had a cam sensor in place? that is all you need, and a trigger in the distributor for the crank position works fine. Cam signal needs to happen after 8th cyl in the firing order, before the 1st cylinder fires.

[Belgian1979]

No, conventional small cap distributor of a GM TPI (MSD 8366) It does run ignition.

I think that the 40° is after firing of the plug. In this case I run 23° initial timing, so the vacuum pulse is measured from 17° ATDC to 27° ATC which in combination with a short stroke and more dwelling of the piston would create less of a vacuum pulse. I think I need to change that to 90 + 23 = 113 °

Anyhow, to come back to vacuum readings. I think some people make the mistake between absolute vacuum measurements and relative vacuum measurements. A vacuum gauge measures relative. The relative vacuum of 69 kPa if I'm not mistaken is 9-10" inch vacuum. In absolute this is 4.5 psi.

Mike Jones, when designing the cam said it would only produce around 10" of vacuum, so this seems pretty normal.

Doesn't mean I like the way the map runs the car. I really need to try ITB load out first and then maybe maf. Will see what this brings.

[mk e]

Don't forget you have a V8...so 1 cylinder is at peak vacuum every 90 degrees. So if cylinder x is at TDC with 0 vacuum yet, cylinder y is at peak so you'r good unless the ecu is collecting a separate MAP for each cylinder and I'm pretty sure it's not.

Really there is nothing magic about 90ATDC either.....vacuum is pretty good from about 30-120ATDC (120 degrees) and very good 45-135 (90 degrees) and this happens every 90 degrees, so there is absolutely no point in the cycle that you shouldn't have a solid vacuum signal. Then you have a large volume plenum that will damp any variation that might have existed. There is a reason no ecu I know of other than MS has the MAP timing feature....very few setups benefit from it and just switching to TPS based load sense fixes them making the feature more trouble than it's worth..

I'm pretty confident there is nothing wrong with you MAP signal. If you have a scope or can borrow one you could confirm this very quickly by looking at the signal coming off the MAP sensor, but I don't expect you'll see anything.

Set the MAP signal and fuel pressure questions aside and test the fixes you've already made. Make certain all the throttle or accel or whatever enrichments are turned OFF! and see what you get.

If you still can't get a decent tune something else is still wrong and we'll need to figure out what it is. A log with MAP voltage, MAP reading, baro voltage, baro reading, VE and then the fuel pulse would be the next step I think so we can have a look at noise in the readings and how that is becoming noise in the fuel pulse...and at this point you're going to NEED a scope because you'll be hunting signal quality. The most expensive ECUs fix all this with better circuits and more importantly high sample rates and post processing the signal before handing if off to be used in calculations...you do get stuff when you spend more.

[Belgian1979]

OK, back to my original question :

I have a expander board coming that has 4 additional map sensors. These were intended to measure pressures in the airboxes and tubes and such.

Could I devise a way in which I could mount one sensor after the maf, and another 2 after an orifice of fixed size in the inlet tube and calculate airflow based on that ? The added advantage would be that I can use other sensors outputs like air temp and baro. It would make creating a suitable maf transfer curve easier. I would only have to move sensor pickup points around, make an orifice plate and connect the whole deal on my efi ecu.

[user-23911]

Get rid of your ITBs, build a proper manifold with 1 or 2 throttle bodies, mount a MAP sensor and you'll get " normal" readings of 18 to 20 inches Hg when it's idling.

[mk e]

OK, back to my original question :

I have a expander board coming that has 4 additional map sensors. These were intended to measure pressures in the airboxes and tubes and such.

Could I devise a way in which I could mount one sensor after the maf, and another 2 after an orifice of fixed size in the inlet tube and calculate airflow based on that ? The added advantage would be that I can use other sensors outputs like air temp and baro. It would make creating a suitable maf transfer curve easier. I would only have to move sensor pickup points around, make an orifice plate and connect the whole deal on my efi ecu.

yes. there is an on-line calculator here:

http://www.efunda.com/formulae/fluids/c ... wmeter.cfm

[Belgian1979]

Get rid of your ITBs, build a proper manifold with 1 or 2 throttle bodies, mount a MAP sensor and you'll get " normal" readings of 18 to 20 inches Hg when it's idling.

Joe,

You're completly wrong on this. I had a carb and manifold on her before converting to FI and it had no more than 9 inch Hg vacuum as well.

[dieselgeek]

the cam he insists on running in this, is lucky to pull 6" at a normal idle. If I was replacing anything, I'd give up the extra 4hp on top for something that was a LOT easier to tune on. Just my $.02

[dieselgeek]

OK, back to my original question :

I have a expander board coming that has 4 additional map sensors. These were intended to measure pressures in the airboxes and tubes and such.

Could I devise a way in which I could mount one sensor after the maf, and another 2 after an orifice of fixed size in the inlet tube and calculate airflow based on that ? The added advantage would be that I can use other sensors outputs like air temp and baro. It would make creating a suitable maf transfer curve easier. I would only have to move sensor pickup points around, make an orifice plate and connect the whole deal on my efi ecu.

This is a lot less work than adding a cam sensor pickup.

Also, do not expect a MAF sensor to properly provide airflow or load information to an ECU with a cam profile like yours. It doesn't work well in OEM cars that receive big cam swaps either. I.e. I think you're traveling down the wrong path. Either try ITB load mode, or better yet - MAP window sampling after you upgrade to a cam sensor.

[Belgian1979]

http://www.bimmerforums.com/forum/showt ... Conversion

Interesting read on itb's and maf's from someone with a 4 pot bmw and aggressive cam.

PS : engine vacuum is in essence a measurement of the restriction in the inlet tract. It's needed but not as much for FI where atomization is largely done by the injector.

[mk e]

PS : engine vacuum is in essence a measurement of the restriction in the inlet tract. It's needed but not as much for FI where atomization is largely done by the injector.

That's true...but not a great way to think about it I don't think. Restriction implies you are reducing flow, and you are reducing mass flow, but you aren't changing volumemetric flow much and that is the key point SD or MAP based tuning uses. If the volume is constant then the mass in the cylinder has to be proportional to the change in pressure, aka MAP. Now it's not truly constant volume due to dynamic effect which is why you need the VE table, but load axis is generally pretty linear

[Belgian1979]

PS : engine vacuum is in essence a measurement of the restriction in the inlet tract. It's needed but not as much for FI where atomization is largely done by the injector.

That's true...but not a great way to think about it I don't think. Restriction implies you are reducing flow, and you are reducing mass flow, but you aren't changing volumemetric flow much and that is the key point SD or MAP based tuning uses. If the volume is constant then the mass in the cylinder has to be proportional to the change in pressure, aka MAP. Now it's not truly constant volume due to dynamic effect which is why you need the VE table, but load axis is generally pretty linear

Well, a factor to consider regarding vacuum is that it's also influenced by valve overlap to some extent, but in an ITB engine not so much since the volume in the intake tract is rather limited.

MAF's are a dime a dozen, so there is not much harm in trying things out. I'm going to get me a 3.5" unit and try it on one box with a dummy on the other. See how it works out.