Hymee

iTrader: (1)

After reading some of Richard Pauls material, I decided I would take some of my actual measurements, and plot them.

I have used my own ODBII logging s/w and a CAN interface I aquired. I logged the MAF values (grams per second) against the measured RPM. (All measurements taken in real time from the PCM via OBDII).

I also plotted the Theoretical mass air flow, and calculate a % efficiency. At a couple of spots this is slightly greater than 100%. My reaction to this is the SDIAS in operation.



Notes:[list=1][*]MAF & RPM are actual as reported by OBDII.[*]Measurements taken at ~20 degrees C and 102 kPa BARO, virtually at sea level.[*]Theoretic displacement = 1.3 litres per revolution.[*]Theoretic mass flow calculated at 1.3 grams / litre (mass of air at sea level)[/list=1]

Thought it was interesting, and thought y'all might like to take a peek. I guess the actual displacement is acctually 1.308 litres, so I am out a tinsy bit, but you get the drift.

Hope you like it.

Cheers,
Hymee.

PS - Let me know if I suffed up, so I can make amends!

IKnowNot'ing

Good info, thanks.

Did you correct it for your ambient conditions?

Re VolEff > 100%, Yamagushi says VolEff max = 105 %.

Can you tell us with OBD tool you used?

brillo

I'm not an expert in this field, what can we learn from this? How efficient is the MAF? can anything here be improved?

IKnowNot'ing

It doesn't tell you anything about the MAFS, but about the efficiency of the rotary engine to pump air into the combustion chamber.
Usually, 4-stroke reciprocating engine have a VolEff between 80 and 90.

babylou

Is this an average VE throughout the rev range? I ask because I have seen in excess of 100% VE at a specific engine speed (usually the torque peak).

IKnowNot'ing

Don't know, it's a quotation from Professor Heywood. But I guess you could get 100% at a given RPM on some higly tuned and efficient piston engines. But I wouldn't know, as I used tp work on basic Ford engines ...

shelleys_man_06

BTW, how is VE measured for rotary engines? I can't find any good resources on this.

Hymee

iTrader: (1)

I don't see how it would differ from a piston engine. It is a black box positive displacement air pump. The VE is a measure of the theoretical displacement v's the practical displacement.

Also note, my graph is on Mass Air Flow, not volume. The theoretic mass flow is calculated from the displacement volume based on the mass of air at sea level is 1.3 grams/litre.

Cheers,
Hymee.

Hymee

iTrader: (1)

I didn't make any corrections, but I noted what the temp was, and the BARO according to OBDII.

The CAN interface is manufactured specifically for us.

The software is written by me.

Cheers,
Hymee.

bgreene

Was this run moving or sitting?

From another thread, it was mentioned that intake temp jumped 120F when the car stopped, and matched ambient when moving with clear flow in front of the vehicle. That big of a delta-T could presumably have some impact on intake mass flow (or does the MAF sensor reading somehow compensate for that?)

Hymee

iTrader: (1)

It was on the road, full throttle, under load (I used the brakes to load it up like a dyno does).

As far as I know the MAF reading is the actual grams/second. Temperature does not affect the mass air flow. It affect the density and pressure, but a gram of air is always the same number of molecules of air. That is what is important.

Cheers,
Hymee.

shelleys_man_06

Well, it depends on what formula you are using for mass flow rate.

For example, remember the always abused Ideal Gas Law? If not, here it is -

pV=mR[G]T

'p' is the pressure.
'V' is the volumetric flow rate.
'm' is the mass flow rate.
'R[G]' is the ideal gas constant divided by the molar mass of the element.
'T' is the temperature.

Rearraging slightly, you get

m=pV/(R[G]T)

As you can see, the mass flow rate is a function of pressure and temperature.

The other form of the equation for mass flow rate,

m=pAV

where p and V are the same as the above equation, and A is the area, is only dependent on pressure and volume (flame me if I'm wrong).

Here's the real question: does anyone care what I just wrote?

shelleys_man_06

Actually, my real question

was pretty vague. Sorry about that. To find the volumetric efficiency, you must divide the actual displacement by the theoretical displacement. This is one of my favorite efficiencies because it can exceed 100% :D. Same thing with intercooler efficiency. Thermal efficiency is so depressing . Carnot efficiency, I can sort of live with . Anyways, I have the formula for the theoretical displacement somewhere on my computer, so that's one side of the equation. I suppose the only way to calculate the actual displacement is through experiment, or maybe some sophisticated simulation. Man, I'm a nerd .

Hymee

iTrader: (1)

I think you abused it all right.

I might need some help here, but the formulae is PV = nRT, is it not?

The mass of x molecules of gas is invariant. Keeping all other variables constant, If you increase the pressure, the volume will decrease. If you increase the volume, the pressure will decrease. If you increase the temperature the pressure will increase etc.

500 pounds of a gas will [b]always[b] contain the same number of molecules. 5 grams of air is always 5 grams of air.

That is why I was talking about the Mass Flow Rate, not the volumetric flow rate. That is why the car uses a Mass air flow sensor, not a volume sensor.

Where is RAP when I need him?

Cheers,
Hymee.

Hymee

iTrader: (1)

You will find the "actual displacement" in my very first post.

I used the sophisticated sensors and electronics on the car to tell me the Mass Air flow at each RPM point, and it is plotted on the graph.

The theortical displacement does not really require a computer, as we know each rotor displaces 654cc. That makes the theoretical displacement for the engine 1308 cc / revolution, or 1.3 litres. At 1.3 grams / litre for air (at sea level etc.), the theoretical mass flow can be calculated. That is what I did.

Then I plotted theoretical mass flow against actual mass flow and determined the % efficiency (by diving the theoretical mass flow by the actual mass flow).

So you could calculate the actual displacement from the g/s readings the engine was consuming, based on 1.3 g/litre.

So I gave you all the info you wanted...

Cheers,
Hymee.

bgreene

That's true, depending on how the MAF sensor actually functions (anyone out there know, or know of a post/thread describing it?).

As shellys_man said, the air density varies with temperature. So if you're using a volumetric method, your theoretical mass flow prediction would need to be adjusted (and, depending on how the sensor works, your "real" value may need some adjustment, although that may be done within the ECU since there's also a data channel for intake air temp).

Of course, that's all academic since you were doing these tests moving, because your intake air will be closer to or at ambient temperature if there's good flow in through the grill.

Elevated intake temps would be more due to sitting with the hood closed since the intake ducts are in locations that are conducive to drawing air from inside the bumper, which would get heated by the engine without good flow-through.

Hymee

iTrader: (1)

All I know is a correctly calibrated MAF sensor is a very sensitive and accurate device. I believe it works on the cooling effect of the air flow on a hot wire.

With the ODBII requirements being enshrined in law, I can't see why I should not have a fairly high degree of confidence in the
reported readings.

Cheers,
Hymee.

shelleys_man_06

Hymee,

Yes you are right. The Ideal Gas Law follows the form

pV=nRT

The form above is as general as it gets. The 'n' here stands for the number of moles of bananas, goats, whatever. There is also a mass flow rate form of the Ideal Gas Law, which was my equation. Remeber that mass flow rate is a derivative with respect to time, or in layman's terms, dm/dt, or an m with a 'dot' over it, which I normally call m[dot]. If I could employ MathType on this forum I would . Also, volumetric flow rate is dV/dt, or V[dot]. There isn't just one form of the Ideal Gas Law. High school physics is wrong. Thermodynamics is right . Please forgive me Hymee. I don't want to make you feel as if I am insulting your intelligence. Also, you did note that you found the actual displacement from experimentation. Sorry that I didn't see that .

Hymee

iTrader: (1)

SM6,

No worries, mate.

Cheers,
Hymee.

shelleys_man_06

In a nutshell, it doesn't really matter how you calculated the volumetric efficiency. The point is, you have a result, which you can use to manipulate other variables such as power, torque, heat transfer, thermal efficiency etc. Happy tuning .

Hymee

iTrader: (1)

Thanks - that is exactly the point! Happy tuning!

I actually did it as I was thinking about a blower of some sort, and wondered what the actual breathing requirements of the engine are in naturally aspirated form, before considering mass flow rates of various puffers.

So I went and got me some real data, and thought I would share it with you all.

Cheers,
Hymee.

IKnowNot'ing

You said too much or not enough! Come on spit it out!
What does the soft do?

Hymee

iTrader: (1)

IKN,

So far the software just makes OBDII PID requests for the operating parameters I am interested in. It decodes the data, and displays the info on a simple GUI on my laptop. It also logs the info to a file. From the file I load the data into Excel, and produce the graph as shown.

The GUI is far from "production quality". But it does serve it purpose - i.e. functional for what I want.

Next step will be to find the correct commands to query/change/reflash the PCM Calibration. No mean feat.

Cheers,
Hymee.

Turbine_pwr

Hymee,

Great information here. This data is exactly the kind of information people need to understand this beast.

By the way, I'm sporting a 6mt velocity red 8 too.

To another comment you made. I also agree. Most MAF systems use a hot film anemometer (sp?). What these devices do is they heat the sensor wire to maintain a constant temperature. This results in a change in voltage level to the sensor (note: some are constant current devices and some are constant voltage). As the airflow moves past the wire film. The energy convected away reduces the heat. More airflow causes higher heat loss which results in a need for more input power. Less airflow obviously reduces the power to maintain the sensor temperature. By measuring the power input and knowing the behavior of the hot film sensor. The manufacturer has developed a correlation between the power input and a corresponding air mass flow rate. That's it in a nutshell. However, one of the interesting features of these devices is that they are/can be sensistive to airflow direction. If the flow is prependicular to the hot film sensor, they give good readings. If there is significant distortion or bulk swirl of the flow as it passes over the MAF... them the mass flow reading will be compromised. This is why some "not so well designed" aftermarket inlet systems do not perform well. They have not paid enough attention to the affect they have on the MAF system and air-fuel mixtures are compromised as a result.

Hymee

iTrader: (1)

Turbine_pwr,

That is an incredibly important point you make about the laminar flow required over the MAF to obtain the correct signal.

I saw so many debates in LS1 land about removing the MAF altogether - but this requires a totally reprogrammed PCM, and requires a MAP sensor. Myself - I am a bit of a believer in having the MAF for it's accuracy, so long as it is not causing a restriction to the flow. If it is, get a bigger MAF - and make sure it is calibrated for the PCM (or vice versa).

I'm glad you like my input of the air flow facts into the FI discuission.

Cheers,
Hymee.