MAF killed :(

[Jay]

[QUOTE=SR71]Paul,

It doesn't actually matter where you get the MASS flow from because it doesn't change through the engine. Thats the advantage of plotting the maps against MASS flow!

But you are quite right....the engine is just an elaborate air pump. For maximum power you want to force as much air as possible into it. Your formula for volume is correct and all a turbocharger does is essentially tweak the VE.

Then you use the ideal gas law to convert to MASS flow. It doesn't matter where you do the conversion as long as you use the associated correct values of pressure and temperature.

Here is a good link on the process: Turbo Calculations

S2driveruk,

Indeed it is a 737. Like you, always wanted to be a pilot. I have a PhD in aerodynamics/mathematical modelling but its certainly not a pre-requisite for the job! In fact, it was a positive hinderance to getting one as they always thought I should be in an office with no windows and a UNIX station!

You do have to enjoy numbers though but I get by with a few simple rules of thumb!

Jay!,

Funnily enough thats my name too!

I have seen a titan grey S2 in Cardiff...once. In fact, I thought someone had stolen my car until I saw the plate. I was "WTF!" Apart from that I only know of one green S2 Avant in the neighbourhood.

But your avatar suggests Shropshire?

I love my 12R. Not ridden the 10R yet although I'm lookin'. That said, I wanna get a sub-10 1/4 with a 145mph trap before I get rid of it. I've seen 220 on the clock @ RAF Woodbridge which equated to 197 through the lights. I'm afraid this is hardly impressive when my mate runs ~8 secs for the 1/4 @ ~180.

Now that is quick!

Drag racing....fastest sport on the planet!

Yep i'm from shropshire, I have just brought the new beast from there and just wondered if youed seen!

You need to try te 10R its a lot lighter and can produce very good power with out much expense- if your into your drag racing add two links in the chain to lengthen the swing arm to try and stop it wheeling (power wheels in 3rd- 1st tops out at 106mph, 2nd 132 mph 3- who cares!!)

Me, i prefere to go round in circles, on supermotos (i instuct) or road racing.

P.S your thread was good but will lose most on here unless like Lee, myself and yourself we fried our minds at coll/uni.

Jay

[S2central.net]

I think what the bods are trying to say is after the compressor the mass is greater which at a lower speed could be worse for the MAF wire Is that correct?

YES - We all agree the MAF is in the right place...

But there's more !


Paul

[S2central.net]

Paul,

It doesn't actually matter where you get the MASS flow from because it doesn't change through the engine. Thats the advantage of plotting the maps against MASS flow!

But you are quite right....the engine is just an elaborate air pump. For maximum power you want to force as much air as possible into it. Your formula for volume is correct and all a turbocharger does is essentially tweak the VE.

Then you use the ideal gas law to convert to MASS flow. It doesn't matter where you do the conversion as long as you use the associated correct values of pressure and temperature.

Here is a good link on the process: Turbo Calculations

Right to take this one more time point by point -

FIRSTLY -

It DOES matter where you get MASS flow information from. It MUST...
We all agree that the S2 engine would suck 234 cfm of air at 7000rpm without any boost at VE=85% (limitations of completely filling cylinders). Add one bar of boost (under the ambient conditions below) gives a Pressure Ratio (PR) of 2.27 for a boosted volume flow rate of 531cfm ingested by the engine.

Pa = 1020mbar (14.79 psi)
Pb = 1bar (14.5 psi)
Pi = 2 psi
Pf = 1 psi

PR = (Pa+Pb+Pi)/(Pa-Pf)

So now we have 531cfm. Depending on what point in the inlet tract you apply temperature and pressure adjustment you get very different values of MASS flow in lbs/min.

From the ideal gas law, we agree m = PV/RT

And to get MASS flow in lbs/min I have been using

m = [Pressure(psi) x Volume(cfm) x 29] / [10.73 x Temp(rankin)]

So where do we take as the reference point for temperature and pressure -
1. Ambient conditions as presented to the air filter ?
2. At the intercooler output / throttle body input ?
3. At the compressor exit / intercooler input ?

Method 1 gets me 40.19 lbs/min if ambient temp is 20 Celsius (528 rankin)

Method 2 gets me 32.65 lbs/min with 1bar boost and IAT=42 celsius

I won't even bother calculate the ludicrous value by method 3.

So by simplying applying the ideal gas law at different locations I cannot see how your statement that MASS flow never changes is correct.

Results get further skewed when you do the temp correction for the scaling on the compressor map, but its my understanding the compressor maps get corrected for ambient conditions as per method 1... Which means we have to do the calculations twice - once to get compressor efficiency values off the map, and then a second time to get the numbers needed for Method 2 (from assumed intercooler efficiency or measured IAT and we already know boost).

Furthemore, other calculations that estimate crank horsepower from MASS flow give *MUCH* more believable figures with Method 2 for an S2 at 7000rpm with 1bar boost using an intercooler good enough for 42C IAT. It comes out about 310-320bhp which is on the money for this setup. This makes logical sense to me as Method 2 MASS flow information is what the engine 'sees' nearest the combustion chambers.

It is my belief that VOLUME flow is the static parameter here, as the cylinders can do nothing but pump that exact same volume of air as defined by their physical size (ably assisted by the turbo of course).

In my mind, the MASS flow is very different depending on where you measure it because of pressure and temperature changes through the system as proved above with simple application of the ideal gas law.

Your statement about MASS flow being the same only holds true at the same volume and pressure and temperature.


SECONDLY -

The turbocharger does not 'essentially tweak the VE'. I'm sure you've said that by accident. The turbo increases the VOLUME of air drawn in by nature of a dirty big air pump that is driven by exhaust gasses. VE is the efficiency factor for how much of an engines swept capacity can be filled with air every cycle.


I'm probably the only one here losing sleep over this you know, but I think I'm right enough !


Paul

[Jay]

Right to take this one more time point by point -

FIRSTLY -

It DOES matter where you get MASS flow information from. It MUST...
We all agree that the S2 engine would suck 234 cfm of air at 7000rpm without any boost at VE=85% (limitations of completely filling cylinders). Add one bar of boost (under the ambient conditions below) gives a Pressure Ratio (PR) of 2.27 for a boosted volume flow rate of 531cfm ingested by the engine.

Pa = 1020mbar (14.79 psi)
Pb = 1bar (14.5 psi)
Pi = 2 psi
Pf = 1 psi

PR = (Pa+Pb+Pi)/(Pa-Pf)

So now we have 531cfm. Depending on what point in the inlet tract you apply temperature and pressure adjustment you get very different values of MASS flow in lbs/min.

From the ideal gas law, we agree m = PV/RT

And to get MASS flow in lbs/min I have been using

m = [Pressure(psi) x Volume(cfm) x 29] / [10.73 x Temp(rankin)]

So where do we take as the reference point for temperature and pressure -
1. Ambient conditions as presented to the air filter ?
2. At the intercooler output / throttle body input ?
3. At the compressor exit / intercooler input ?

Method 1 gets me 40.19 lbs/min if ambient temp is 20 Celsius (528 rankin)

Method 2 gets me 32.65 lbs/min with 1bar boost and IAT=42 celsius

I won't even bother calculate the ludicrous value by method 3.

So by simplying applying the ideal gas law at different locations I cannot see how your statement that MASS flow never changes is correct.

Results get further skewed when you do the temp correction for the scaling on the compressor map, but its my understanding the compressor maps get corrected for ambient conditions as per method 1... Which means we have to do the calculations twice - once to get compressor efficiency values off the map, and then a second time to get the numbers needed for Method 2 (from assumed intercooler efficiency or measured IAT and we already know boost).

Furthemore, other calculations that estimate crank horsepower from MASS flow give *MUCH* more believable figures with Method 2 for an S2 at 7000rpm with 1bar boost using an intercooler good enough for 42C IAT. It comes out about 310-320bhp which is on the money for this setup. This makes logical sense to me as Method 2 MASS flow information is what the engine 'sees' nearest the combustion chambers.

It is my belief that VOLUME flow is the static parameter here, as the cylinders can do nothing but pump that exact same volume of air as defined by their physical size (ably assisted by the turbo of course).

In my mind, the MASS flow is very different depending on where you measure it because of pressure and temperature changes through the system as proved above with simple application of the ideal gas law.

Your statement about MASS flow being the same only holds true at the same volume and pressure and temperature.


SECONDLY -

The turbocharger does not 'essentially tweak the VE'. I'm sure you've said that by accident. The turbo increases the VOLUME of air drawn in by nature of a dirty big air pump that is driven by exhaust gasses. VE is the efficiency factor for how much of an engines swept capacity can be filled with air every cycle.


I'm probably the only one here losing sleep over this you know, but I think I'm right enough !


Paul

Substitute mass for density- mass stays the same but density changes(the turbo packs more air in but increases temp so if you cool it the air is denser!! or is that me? ), as soon as the air has entered the engine its mass does not change (but adding fuel and then burn it does change the mass!!)
Hope this helps

Jay

[SR71]

Paul,

Confusion ensues!

Returning to your original formula:

I can think of the engine as being a big elaborate air pump - that sucks up a volume of air as defined by the tried and trusted equation :

Volume = (RPM/2) * Engine Size * VE

Remember the idea of an engine as an air pump is a crude approximation.

The correct point at which to utilize the figures is location 2 as you suggest. It is this air that the cylinders are effectively seeing.

What I mean by the MASS flow being constant is that, if at any point in the engine now, you measure density, cross sectional area and velocity you will obtain a constant MASS flow via the calculation: density multiplied by cross sectional area of the tract multiplied by the velocity.

This MUST be the case unless you have a leak. Strictly as Jay says, during combustion you are introducing additional mass into the system, but you can imagine that mass (fuel mass) is small.

The discrepancy you have discovered in your calculations is because its almost as if you've considered two different design conditions: one where the air pump/engine is seeing ambient pressure air at 20C, the other 1bar (gauge) air at 42C.

(Remember you cannot only use the equation of state to evaluate the flow properties of the air through the engine. You have to use the isentropic relationships as well.)

Like you said, to compare one compressor map against another we need to compare them for a given particular set of ambient conditions and therefore have to correct the flow to standard. This corrected mass flow is merely the mass flow through the engine that it would theoretically see if the ambient conditions were the standard atmosphere conditions.

In conclusion, the displacement/volume of air being consumed at any particular rpm/design point is constant. Agreed.

Depending on ambient conditions and turbocharger boost, this then fixes the mass flow rate through the engine at that rpm/design point.

Depending on thermal, volumetric and mechanical efficiency, your ultimate power is then fixed at that rpm/design point.

As for turbochargers and VE, technically you are, of course, right Paul. The comment obtains its justification from the fact that if VE is an expression denoting how easily a design allows air into the cylinder, the effect adding a turbocharger to the design has, is an effect analogous to an increased VE.

Any help?

[S2central.net]

Paul,

Confusion ensues!

Returning to your original formula:



Remember the idea of an engine as an air pump is a crude approximation.

The correct point at which to utilize the figures is location 2 as you suggest. It is this air that the cylinders are effectively seeing.

OK - we are on the same page on this point !

What I mean by the MASS flow being constant is that, if at any point in the engine now, you measure density, cross sectional area and velocity you will obtain a constant MASS flow via the calculation: density multiplied by cross sectional area of the tract multiplied by the velocity.

Perhaps density is the missing link in the understanding here. I have not taken that into consideration... nor any concerns about velocity changes because of different x-sectional area along the way. The only references I have seen to air speed:density is that such airflow sensors are not as accurate as MAF devices.

So the mismatch I think we have is that 'your' method of measuring mass, by knowing density, velocity and port area will produce the same answer regardless of where you measure it in the system... whereas 'my' method uses the ideal gas law under for the prevailing conditions.

This MUST be the case unless you have a leak. Strictly as Jay says, during combustion you are introducing additional mass into the system, but you can imagine that mass (fuel mass) is small.

The discrepancy you have discovered in your calculations is because its almost as if you've considered two different design conditions: one where the air pump/engine is seeing ambient pressure air at 20C, the other 1bar (gauge) air at 42C.

(Remember you cannot only use the equation of state to evaluate the flow properties of the air through the engine. You have to use the isentropic relationships as well.)

You've lost me with the isentropic bit there... I'm an electronic systems guy remember - this aero stuff is a hobby for me remember

Does that factor in the speed & density information ?

Like you said, to compare one compressor map against another we need to compare them for a given particular set of ambient conditions and therefore have to correct the flow to standard. This corrected mass flow is merely the mass flow through the engine that it would theoretically see if the ambient conditions were the standard atmosphere conditions.

Agreed - but I use the MASS calculated at ambient before correcting to compressor map standard, as opposed to the MASS calculated after the intercooler. So the MASS info being used on the compressor map is two steps removed from the MASS flow actually ingested by the engine. But this OK as long as we remember NOT to base fuel injector estimates and power claims on the compressor map values.

So in essence the MASS reading from the MAF does not directly correlate to the MASS value seen by the engine - the ECU must make adjustments for engine speed, ambient pressure, IAT and boost in order to calculate the MASS flow entering the engine by this method. That's easy enough - I can do it on Excel, so a trusty old 8081 micro will crunch that no problem as the ECU provides ALL of those inputs.

In conclusion, the displacement/volume of air being consumed at any particular rpm/design point is constant. Agreed.

Happy daze - its what I understood from day one !

Depending on ambient conditions and turbocharger boost, this then fixes the mass flow rate through the engine at that rpm/design point.

Depending on thermal, volumetric and mechanical efficiency, your ultimate power is then fixed at that rpm/design point.

As for turbochargers and VE, technically you are, of course, right Paul. The comment obtains its justification from the fact that if VE is an expression denoting how easily a design allows air into the cylinder, the effect adding a turbocharger to the design has, is an effect analogous to an increased VE.

Any help?

OK - what a tortured week of a thread this has been, but I think we are on the same page now. Sincere thanks for your inputs to this as it helped me get more things straight in my head (and convinced me to carry on ignoring speed:density stuff for the purposes of better evaluating what performance can be had from specific turbos) and forced me to create a more accurate calculator tool on Excel - which you can have to play with and test if you like.

Paul

[SR71]

Perhaps density is the missing link in the understanding here. I have not taken that into consideration... nor any concerns about velocity changes because of different x-sectional area along the way. The only references I have seen to air speed:density is that such airflow sensors are not as accurate as MAF devices.

So the mismatch I think we have is that 'your' method of measuring mass, by knowing density, velocity and port area will produce the same answer regardless of where you measure it in the system... whereas 'my' method uses the ideal gas law under for the prevailing conditions.

We should , of course, always arrive at the same answer! Your comment about how it does matter what input conditions you enter into your equation of state belies the inadequacy of the air pump approximation.

Suffice to say in the practical world, typically there are two places we might access information with which to calculate the mass flow through the engine: the MAF and at the TB. Extracting the info to do so is tricky as you know and we may need an extra sensor or two.

Agreed - but I use the MASS calculated at ambient before correcting to compressor map standard, as opposed to the MASS calculated after the intercooler. So the MASS info being used on the compressor map is two steps removed from the MASS flow actually ingested by the engine. But this OK as long as we remember NOT to base fuel injector estimates and power claims on the compressor map values.

So in essence the MASS reading from the MAF does not directly correlate to the MASS value seen by the engine - the ECU must make adjustments for engine speed, ambient pressure, IAT and boost in order to calculate the MASS flow entering the engine by this method. That's easy enough - I can do it on Excel, so a trusty old 8081 micro will crunch that no problem as the ECU provides ALL of those inputs.

I'm not sure you need to do the normalization twice... Evaluate the mass flow at location 2, then correct back to ambient and that "corrected" mass flow should be the one used when reading the maps.

OK - what a tortured week of a thread this has been, but I think we are on the same page now. Sincere thanks for your inputs to this as it helped me get more things straight in my head (and convinced me to carry on ignoring speed:density stuff for the purposes of better evaluating what performance can be had from specific turbos) and forced me to create a more accurate calculator tool on Excel - which you can have to play with and test if you like.

No problem.

A quick observation....the ideal gas law essentially relates 3 variables: p, T and V. Keep one fixed and the relationship between the remaining two is fixed. Its most clearly demonstrated if we consider a fixed volume of fluid whereupon V is constant. The relationship between p and T is now fixed whereupon if the temperature T of our fixed volume of fluid increases, p increases by the corresponding factor (n*R/V)*T. Its a static relationship.

If we now consider the dynamic situation where fluid is flowing through a converging diverging duct (essentially what our intake is) for instance, and we want to know what the pressure and velocity of the fluid is at any particular point in the duct, the ideal gas law is no longer helpful because it doesn't relate p or T with velocity.

These relationships are the isentropic relationships which in incompressible flow reduce to Bernoulli's equation.

By all means fire off the Excel spreadsheet.

I'm waiting to sell my house before my RS2 upgrade but why the hell shouldn't I see what bolting a T78 to the EM would do for me.

[S2central.net]

I'm not sure you need to do the normalization twice... Evaluate the mass flow at location 2, then correct back to ambient and that "corrected" mass flow should be the one used when reading the maps.


By all means fire off the Excel spreadsheet.

I don't do the normalisation twice, but I do calculate MASS flow at a few places for the prevailing conditions.

The attached ZIP explodes to an Excel file. It is password protected to prevent accidental data input in the wrong places. I'll PM the password to anyone that wants to see the formula.

Its heavily commented along the way so you can follow the logic.

Cheers,

Paul

[SR71]

Paul,

I'll have a look at the spreadsheet in the next few days. Looks good.

Cheers,

J

[audiman]

For ppl like me, is my MAF in the wrong place then?? (after the turbo). I dont understand those equations, well i could, but i dont have the time to study them!!

[S2central.net]

The MAF is in the best place. The equations aren't *that* bad - and using the spreadsheet is intended to make it dead easy to crunch numbers - it does all the hard work - all you need know is how to read compressor maps and get efficiency info.

Sorry for the thread hijack BTW

Paul

[audiman]

Not a problem Paul, its nice to get some good debates/tech info on the forum. There are some ppl out there with a lot more mathematical skills than myself, sometimes its good to hear a few ppl thoughts on a particular idea. Especially if they can prove what they are saying is correct. Good work everyone, and thanks again.

[newsh]

[A80Avant]

after developing sever blindness by reading all this,i have a major question here...(and yes sorry to bring up all the talk about MAF and mas/volume of air etc.

Are the ambient conditions capable of producing noticable results on boost and boost pressure and acceleration when everything is working properly on your car?
thanks