Maf scaling

[teamcolors30]

Is there any way to edit an existing maf scaling in the vts? I have a v7-9 Subaru maf in a larger housing and a cal table has poor resolution along with increments that make a curve sharp.

[Dave Kriedeman]

Hi,

are you using the default V7-9 MAF Calibration, or have you chosen a custom CAL TABLE and created your own Cal Table to suit.

Are you intending on using a MAP sensor or just MAF,BAP and TPS.

What are you using for the Y axis on your Main Fuel Table.

Regards

Dave.

[teamcolors30]

Hi,

are you using the default V7-9 MAF Calibration, or have you chosen a custom CAL TABLE and created your own Cal Table to suit.

Are you intending on using a MAP sensor or just MAF,BAP and TPS.

Regards

Dave.

I have a 80mm intake vs the oem 65mm intake. I cannot use the oem calibration in the software because my intake tube is ~1.51 times the size as stock. A custom cal table offers very poor resolution. I've tried to enter a custom calibration for a sensor but it forces equal spacing which I need specifically more resolution where the change in slope is greatest in order to gain accuracy.

I believe at this point there is no solution. But I thought I would ask the pros.

[Dave Kriedeman]

Hi,

I personally never use MAF, however if you can attach your PCL file so I can have a look and do some testing for you.

Are you setting up the Custom cal input as a voltage or ohms.

Is the output scaled as g/s.

How much resolution are you after and what type of scaling.

EG: 0 Volts = 0 gm/s

5 Volts = 1000 gm/s etc.

Regards

Dave.

[teamcolors30]

Hi,

I personally never use MAF, however if you can attach your PCL file so I can have a look and do some testing for you.

Are you setting up the Custom cal input as a voltage or ohms.

Is the output scaled as g/s.

How much resolution are you after and what type of scaling.

EG: 0 Volts = 0 gm/s

5 Volts = 1000 gm/s etc.

Regards

Dave.

Thank you Dave for the offer. At this point I do not have a PCL set up to run a MAF. This is my starting rough maf scale for the 80mm intake tube. My turbo is a 51lb/min unit and I am looking to use the MAF primarily as an airflow validation, and down the road in my fresh air anti-lag system. I have been tinkering with that for a couple years. I would like potentially to use g/s for fueling during this systems operation. Im tinkering quite frequently and nothing in stone has been set just yet. I hoped to have airflow be measured correctly, I may be able to just install a smaller intake tube and be done with it. However id obviously like the 80mm one

[Dave Kriedeman]

Hi,

OK,

I am working on this now, I believe I have a solution.

I will confirm with the ViPEC engineers before posting.

Regards

Dave.

[Dave Kriedeman]

Hi,

I have submitted my solution and I am just waiting confirmation from the engineers on my solution.

Regards

Dave.

[Dave Kriedeman]

Hi,

I have not received confirmation from the engineers as yet as they are very busy.

I have run this setup on my ENGINESIM and it works.

I used your values you supplied on your spread sheet.

I converted your LBS/MIN to GMS/SEC to create the Calibrations.

1 LB/MIN = 7.55987283 GMS/SEC.

I set up an AN VOLT I/P as a G/P INPUT, I then used CAL TABLE 4 as this gives you a better set up, simple volt = grams/sec.

I then rescaled your list of LBS/MIN to GMS / SEC.

I setup the FUEL MAIN TABLE with AN VOLT 2 as the Y AXIS and set as GRAMS/SEC.

I then chose some progressive grams/sec values and created a basic fuel table.

With 20 Y axis GRAMS/SEC load sites available it is up to you what you wish to use.

Remembering the ECU will interpolate between these sites.

Personally I would delete the MAF and use a MAP sensor and tune using MGP or MAP,

However this is your call.

I have attached an example of how I have set it up.

Unfortunately the axis is only able to be setup as a single decimal place, so the software rounds off values to the closest single decimal place.

I hope this helps you out.

Regards

Dave.

[Dave Kriedeman]

Hi,

here is a screen shot of the MAF setup running on my ENGINESIM.

Everything seems to run fine.

The low end resolution scaling numbers I have used on the FUEL TABLE are very sensitive.

Take note I am using an i88 for the test ECU here.

Regards

Dave.

[teamcolors30]

I think I understand, but does that make a linear correlation between mafv and g/s? For instance my wbo2 is linear and I have it set up with a cal slot using 2 input points and 2 output points. I'm still wrapping my brain around what you did though and I could definitely be off. I will go through the steps later(after work) myself to emulate what you did.

[Dave Kriedeman]

Hi,

in regards to if this is linear, well it is linear to the values you have supplied me with.

If you look at my cal 4 I have used your lowest voltage value for input A and used your highest value for input B.

I do not know the true full voltage scaling etc of your MAF, so I have set this example up with the information you have supplied.

The output values in grams/sec are calculated values derived from your supplied pounds/min values converted to grams/sec.

These values are rounded to 1 decimal place only, by the software.

So the software knows that voltage input A = 0.94 v = 1.3 grams/sec and voltage input B = 4.61 v = 425.8 grams/sec.

The end result output of grams /sec will be derived from any given input voltage between input value A and B.

Exactly the same as your wideband setup, only your input A value would normally be 0v and your input B value would normally be 5.00 v

WIDEBAND EXAMPLE:

0v = 10.0:1

5v = 30.0:1

I have no 0 volt or 5 volt reference supplied by you, hence for the input A and B values being what they are.

If you look at the Y AXIS of the Main Fuel Table you will notice it is set as AV1 which is grams/sec and these are values derived from my CAL 4.

Regards

Dave.

[teamcolors30]

Ok I will digest that a few more times and post up how I did it by trial. I know it came out as a curve when I made that trial entry. Maybe you can identify my error.

I fully appreciate the help.

[teamcolors30]

This is the original scale for a Subaru v7-9 maf as it appears in the factory software.

This is my attempt to scale it 1.51 times roughly for the increased pipe size, and with many fewer data points.

I wanted to know if I could get more data points for this kind of situation or somehow edit what VTS is using for V7-9 maf scale. I assumed it was more data points than a standard cal table.

[Dave Kriedeman]

Hi,

I understand that you are trying to replicate closely the linear curve of the factory sensor with the correct scale offset for the larger diameter MAF.

Unfortunately you are restricted to 16 sites in CAL 1,2 and 3.

Amongst that using CAL1,2 or 3 you must nominate a start value and a table increment amount.

This is somewhat of a limitation to try and achieve your goal.

This is why I have setup CAL 4 in my supplied example.

You have a minimum and maximum volt scale and then in respect to the voltage scale we are able to nominate given (known) output for value A and B.

Output scale is grams/sec .

Input value A = 0.94 volts = 1.3 grams/sec

Input value B = 4.61 volts = 425.8 grams/sec.

Therefore the ECU generates a linear curve itself derived from the known data., how accurately this replicates the factory curve I have no idea.

"""WHEN I REFER TO A LINEAR CURVE IT WILL REALLY BE A LINEAR LINE""""

Using CAL 1,2 or 3 limits you to 16 sites to make up a replication of your desired curve.

As you notice using CAL 1,2 or 3 limits you to given data entry points, the ECU must then try to interpolate between these given values.

Using CAL 4 as I have enables the ability for the software to create many more points, considering it would change the output grams/sec in relation to every 0.00001volt or whatever the resolution of these CALS allow.

This allows for more ECU generated interpolation between the given points, compared to CAL 1, 2 or 3.

I am no engineer so my example of 0.00001 volts is purely that, an example.

Having a true known value for 0 volts and 5 volts for the MAF would give an even more accurate curve.

Then when setting up your main fuel table as I have using CAL 4 grams/sec on the Y AXIS you have 20 load points to enter your desired grams/sec you wish to use.

However as I stated earlier the use of MAF sensors is not that common in practice by myself and many others.

Speed density (MAP) sensed inputs have greater potential and are far easier to setup and tune with.

Some may argue this point and that is fair as MAF sensors do have a place in this area however in my experience and many others the use of MAF and standalone ECU's no mater how good, normally lead to less than desirable end results.

Considering if you removed the MAF, you wouldn't have that restriction to start with.

Will the 80 mm MAF definitely allow you enough resolution for further mods or higher boost levels without bottoming out(running out of range).

Creating flow vs voltage curves accurately is also a limitation of using MAF.

I am just letting you know, you may go through the trials and tribulations of this and end up going to a MAP SENSOR at the end anyway.

I hope this has explained it all a little better for you.

Regards

Dave.

[teamcolors30]

Thank you Dave. we agree to see the same light, I am restricted as the function I seek is more exponential than linear. This is something I will deal with for now, using my limited resolution. I already have my car running well by MAP. Was looking for a different way to skin a cat. ( my lord I hope that crosses cultural barriers - for I don't really kill cats).

I am in this process to learn. Sometimes I will have questions that don't make sense as much as they should

Dana

[Dave Kriedeman]

Hi Dana,

No worries, we are all the same.

You never stop learning when it comes to this type of thing.

As for asking questions, that is a very important step in learning.

Sometimes deciphering what someone else has written can be tricky to understand, the joys of the internet and time zone differences from country to country etc , etc

Your cat comment I am sure is known world wide and always taken as a joke.

All the best.

Regards

Dave.

[teamcolors30]

A follow up if anyone could indulge:

I will be supplying maf sensor power and ground as well as an auxiliary map sensor from the expansion harness. The oem stuff is shielded. What if any steps are taken internally for signal filtering? The maf signal and map signal will be right next to each other and if I was going to create a shield for these two wires where do I ground it and what ground to use?

[Dave Kriedeman]

Hi,

your shield grounds should go to sensor ground at the ECU end.

Do not ground the shields at the sensor end.

In relation to your setup, you are intending to run an extra MAP sensor back onto your expansion cable on the PNP board.

May I ask why you are going to use the expansion cables for the MAP which is internal and the MAF which is AV 2.

Regards

Dave.

[teamcolors30]

I'm using the internal map already for engine load , this is to monitor pre tb pressure.

But you bring a good point up the factory maf has iat baro and hz signal so there should be enough inputs ( will double check). I was thinking there was not enough room.

The primary purpose of this is researching turbocharger performance when torque request is non existent. I am trying to produce what WEC WRC WTCC etc cars use that maintains a less destructive method than people may be used to, in order to minimize loss of turbo speed. I'm careful to say anti lag as much as people think it turns 4 bangers into v8 torque monsters, I'm aiming at a more linear engine vs turbo response pattern, and not trying to lower the boost threshold like many conceive is the goal.

Thank you for the input Dave!

[Dave Kriedeman]

Hi DANA,

I agree with you, DATA is KING.

The more DATA you can analyse the more powerful the abilities of the ECU become.

Considering the spare inputs available on the main connector if certain items are deleted and configured for other uses, you have an extra 2 digital inputs on the expansion connector as well as 4 AN V inputs, Which can be setup as temp inputs with the use of a resistor if required.

I sometimes setup multiple air temp sensors for checking intercooler efficiency, 1 pre cooler,1 post cooler and 1 at he inlet plenum to progressively monitor heat (thermal) rises etc.

Spare DI 's enable you to also fit off an extra wheel speed sensors etc.

As for normal anti lag operation, extra air bypass at throttle for raised rpm and ignition retard etc, etc.

Have you ever considered adding bypass air into the exhaust manifold itself, the extra air keeps the exhaust temp up and the turbine spinning fast due to the added heat created by the added air.

I have no experience with rally cars and next to no experience with WRX's and EVO's.

I normally work on custom high performance setups, V8's Turbocharged/Supercharged and highly strung turbo 6 cylinders.

I did however work on an EVO fitted with an aftermarket ECU, bigger turbo and a custom inlet plenum, many , many years ago.

I remember seeing 4 pipes going into the exhaust manifold, (My initial thought was it was some type of EGR setup) when I asked about what they where and did a little research it was an early form of anti lag used on that model.

Just thought I would mention it, as your response re anti lag triggered my memory.

Regards

Dave.

[teamcolors30]

I have the oem set up, and I've made my own a couple times over. Unfortunately then I was not capturing the data to monitor the systems or even improve upon it. "Here I am";)

[Dave Kriedeman]

Ahh,

you are all over it already.

Your a smart man.

Regards

Dave.