AIr flow meter and mass airflow 1

[yoshimitsuspeed]

It seems to be pretty common terminology to call an AFM a MAF especially in the automotive world. It's my understanding that on it's own there is no way for an AFM to calculate mass airflow.
An AFM could be used with other components like a temp sensor and MAP sensor to calculate mass airflow but on it's own it cannot. It seems to me akin to calling a hotwire a MAF. A hotwire MAF cannot calculate mass airflow without the other included components to calculate mass airflow and that make it a mass airflow sensor. A hotwire is nothing more without those things.
It is very hard to find information talking about technical details and technical function in regards to the airflow's effect on the vane on an AFM. I have thought in the past that a vane AFM measured volumetric airflow but in thinking about it I realize that's not entirely correct either. X CFM passing through the vane at the fringes of space would open the sensor much less than the same CFM flowing through it at sea level so the the AFM should respond to a reduction in mass airflow at the same volumetric airflow but not in a way that can calculate volumetric airflow without other sensors right?

Many AFM devices also have a temp sensor. This will help but it still seems to me you wold need a pressure sensor to calculate mass airflow.

So is the whole entire internet wrong? Or am I missing something?

 

[gruntguru]

Hotwire airflow meters do in fact measure mass airflow. The filament loses heat in proportion to the number of air molecules interacting with it. This heat loss is proportional to average air velocity, and air density ie massflow. Air temperature is not so much "measured" as "used as a reference to maintain the filament at a fixed temperature above the airstream". This can be simply achieved by incorporating the filament and the reference (air temp) resistor as two of the four legs of a wheatstone bridge, using bridge current to heat the filament and maintaing the bridge in balance with a feedback cct.

je suis charlie

 

[Lou Scannon]

gruntguru is quite right, though some may quibble with his first sentence. The hot wire element, together with the reference temperature element, allows a signal that is a function of the velocity of the stream passing the element to be generated. The velocity can be used for a surrogate of mass flow, assuming that the transfer function from velocity to mass flow is known for the unit in which the hot wire element is employed. This is going to be the case for a released production unit. In fact, the transfer function may well be developed directly from output signal voltage (or current or frequency, as may be the case) to mass flow, ignoring the intermediate step of velocity.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[yoshimitsuspeed]

Okay well I guess I blew that part lol. I thought that hot wire sensors required more sensory input to calculate actual mass airflow.
This does still set the baseline though because a hotwire MAF does calculate mass airflow.

So the real point of the thread and what I was trying to get at is if a vane type air flow meter should be categorized as a mass airflow meter. I say no it shouldn't but almost everything you can find on them online puts them in the category of mass airflow sensors.

 

[gruntguru]

Hemi.(I did mention velocity in my post.) Apart from that:

Name 5 sensors that directly measure the physical property stated on the box - without measuring some intermediate physical property. I will start you off. Tape measure.

For every one you can name, I will name ten that measure some intermediate physical property and infer the measurement of the stated physical property.

je suis charlie

 

[yoshimitsuspeed]

Does a vane AFM infer the measurement of the stated property though? That is the whole question.
You can look at the information coming from a hotwire MAF and estimate within a fairly accurate margin the mass airflow right?
Looking at a vane type air flow meter without data on air pressure or temp can you actually know what the mass airflow was?

 

[gruntguru]

No. You need to know air density as well.

je suis charlie

 

[yoshimitsuspeed]

That was exactly my thinking. It could be used in conjunction with other sensors to be a mass airflow sensor but on it's own could not be considered to be one.

Then the question is what is an easy accurate title to describe what an AFM actually measures? It's not mass airflow, and it's not volumetric airflow.

 

[gruntguru]

I doubt there is an easy title. I think it is safe to say that its output is volumetric and not sensitive to atmospheric pressure within a normal range.

je suis charlie

 

[Tmoose]

Hi yoshimutuspeed,

By "vane AFM" do you mean the single disk, like in Bosch K Jetronic?

http://www.ffp.fi/contentimages/tekstit_kejet_2.jpg

Most charts show this system measures engine temp to determine if enrichening is required for starting and the first several minutes of running, but NO air temp sensor in the air stream to correct airflow > mass flow.

Or, the curious right angle flapper in L Jetronic?

http://www.alfabb.com/bb/forums/att...roblem-help-bosch_fuel_injection_1_sample.jpg

This one does have an air temp sensor just before the flapper.

I was thinking some systems had a barometric pressure sensor, but on first glance these to vintage Bosch system had none.
Maybe they began relying on the Lambda sensor to make it all right.

 

[BrianPetersen]

Those mechanical air-flow meters dated back to a different era in emission standards. I don't know of any current applications of these (that doesn't mean they don't exist - it's just been a long time since I've seen one.) They could probably get away with a lot back then that would not fly today.

Carburetors (before emission controls) have no temperature/barometric pressure correction, either, and the engines still (mostly) ran.

Also bear in mind that quite a number of modern, emissions-compliant engines have no MAF sensor at all. They infer the amount of airflow from barometric pressure, manifold pressure, intake temperature, throttle position, etc. The view of these manufacturers is likely that pressure and temperature and position sensors are more robust. From my own experience, I can't say that I disagree.

If you can get the system to work with no MAF, surely you can get it to work with an imperfect one!

 

[yoshimitsuspeed]

Tmoose I am interested in the differences and defining characteristics of any unit that falls under the title of AFM.
I do not believe either of those could fall properly under the title of mass airflow meter. I am much more familiar with the L Jetronic style and I know that most have intake temp sensors. Some L Jjetronic cars do also have rudimentary MAP sensors. This is where I was saying an AFM could be part of a system that can estimate mass airflow but on it's own shouldn't be considered a mass airflow sensor.
You are correct about most systems using the O2 sensor as the final piece to do the fine tuning compensation. In this case the EFI system does know enough to calculate for mass airflow but the AFM is only one of the tools is uses to do so.
Interestingly enough most first generation MR2s had L Jetronic with O2 sensors but in at least some areas of Europe and possibly other design markets they had no O2 sensors. These required manual baseline setting of the tune. I think these may have used a rudimentary MAP sensor but I don't remember for sure.

If you search for airflow meter the most common thing you will come up with is something like a portable weather station. Now this unit should be able to calculate mass airflow because it is usually also going to be collecting temp and pressure readings. I would still argue that an airflow meter can be used to calculate mass airflow but that it should not be automatically labeled as a mass airflow sensor because many AFMs don't provide enough information to calculate mass airflow.
As another example a wind speed meter like a sniper might use will be a similar air flow meter but the sniper won't care about temp or pressure or the then achievable calculation of mass airflow. He only cares about the winds effect on the bullet and that will have the same effect on the meter as it will the bullet so those other variables are really irrelevant.

 

[Tmoose]

Hi Yoshimitsuspeed,

What did you mean when on 15 Mar 15 01:07 you said "vane AFM" ??
regards,

Dan T

 

[yoshimitsuspeed]

PrianPeterson
I completely agree with everything you said. My question is in regards to nomenclature and proper understanding of the principle and function. I wanted to make sure I wasn't missing something and that I understood within an acceptable level of accuracy what an AFM did. And apparently also had a little to learn on the basic operation of a hot wire MAF as well hehe.
I hate it when wikipedia is the best source I can find to cite but in this case I haven't really found anything that better describes the operation and differences between an AFM and a MAF and Wikipedia definitely has messed it up pretty bad. I really feel like there were sources I could easily find a couple years ago that made it clear that an AFM was not a MAF but I can't find them anymore.
Wikipedia lists vane type airflow meters under MAF and in a recent edit someone has suggested they be completely merged under mass airflow sensor.
I believe this is completely incorrect but can't really find anything that backs that in a substantial manner.

http://en.wikipedia.org/wiki/Mass_flow_sensor

http://en.wikipedia.org/wiki/Air_flow_meter

An air flow meter, is a device that measures air flow, i.e. how much air is flowing through a tube. It does not measure the volume of the air passing through the tube, it measures the mass of air flowing through the device per unit time. Thus air flow meters are simply an application of mass flow meters for a special medium. Typically, mass air flow measurements are expressed in the units of kilograms per second (kg/s).

Now as far as motor operation these systems did work pretty dang well especially compared to the capability of the carbs and mechanical fuel injection that preceded them.
I can drive my MR2 from 4000 feet to 12000 feet and under cruise maintain stoichiometric AFRs under cruise the whole time. When I increased my injector size by 20% I really learned how much even basic 80s technology was able to adapt and what it's limits were. I was surprised to see that it was able to use closed loop AFRs to trim open loop operation and was able to largely adapt to the larger injectors. It never got all the way there though. The size increase was just outside of the ECUs range of adaptability.
I do beleive one of the biggest things that has allowed newer cars to get rid of the AFMs and MAFs is more advanced 02 sensors. Of course paired with the other technological advancement as well but I think the medium band and wide band sensors have done a lot in allowing cars to really tighten up their efficiency and emissions with less monitoring on the intake side.

 

[BrianPetersen]

The O2 sensor and monitoring strategies are necessary for achieving the last bit of accuracy that is necessary for compliance with today's emission standards ... but the base tune that is used for open loop operation (cold start, full load, and in the event that the O2 sensor fails) is usually pretty darn close these days! It has to be close, because generally the O2 sensor is used to determine short-term and long-term trim maps, and these are only allowed to have a certain percentage before the diagnostics chime in with fault codes indicating that something is wrong. Your 1980s car probably has a wider allowance in its trim maps than newer models do.

Between one car (which is a DaimlerChrysler product), one van (which is a FiatChrysler product), and two fuel-injected motorcycles (one Kawasaki, one Honda), I own no MAF sensors. All but one bike is closed-loop EFI with 3-way catalyst. The one bike (which is from 2004) is open-loop EFI with secondary air injection and oxidizing catalyst.

I don't know how the cars do it, but the bike systems use a combination of both N-alpha (i.e. RPM and throttle position) at high engine load, and speed-density (i.e. RPM and intake manifold pressure) at light load, and some sort of blend in between. If a sensor conks out, it switches over to the other algorithm (and turns on the fault-warning lamp).

 

[140Airpower]

In fact, thermal conductivity sensors cannot ever measure mass. Of course an ECU can infer mass flow from the operating point of a circuit that includes a conductivity sensor. If the flowing gas is air, the calculation is reliable. If anything other than air is flowing, the calculation will be wrong. For example, atmospheric humidity throws it off. Fortunately for fuel metering, the error in fuel metered due to humidity is in the same direction as the needed correction to the calculated air mass, so no correction need be made (at least at the current level of required accuracy).

I once debated a colleague about this and he cited a Mercedes Benz PR statement that made the claim that hot wire MAFs measured mass while the old vane style MAF did not. The curious thing is that the vane style sensor is an aerodynamic device where mass is expressly part of the force equation. Oh well...

 

[gruntguru]

The point is that hot-wire MAF sensors generate a mass-flow-dependent output regardless of temperature and pressure variations. Vane style AFM's do not, their output being sensitive to both temperature and pressure.

A hot wire MAF is actually quite insensitive to Humidity - the output still delivering a fairly accurate MAF. The problem as humidity increases is that dry air flow (and therefore oxygen flow) is reduced - having been displaced by water vapor.

je suis charlie

 

[SomptingGuy]

Hot-wire vs vane strikes me as either a brilliant bit of lateral thinking or the fruits of random experimentation and investigation of unexpected correlation. Either way, who have thought it could work so well?

Steve

 

[140Airpower]

Gruntguru, are you saying that the output of an aerodynamic flow sensor is dependent on temperature and pressure in a different way than mass flow is dependent on temperature and pressure?

A 1% error in conductivity due to humidity would be a problem, IMHO, if it were not for the error largely overlapping the reduced air due to displacement.

The hot wire anemometer does not measure mass, but as said, since you know the gas is air, you can infer mass and then the device itself is more precise and reliable than the vane.

 

[gruntguru]

The output of a hot wire AMM has a fixed relationship to air mass flow. This relationship holds even if air pressure or temperature vary.

The output of a Vane AFM has a relationship to air mass flow. Any variation of air pressure or temperature will affect that relationship. Compensation will be required.

je suis charlie