MAF testing

Ask for and share advice on using the PicoScope kit to fix vehicles here.
Post Reply
User avatar
Fat Freddy
TwoWaves
TwoWaves
Posts: 335
Joined: Fri May 04, 2012 4:52 am

MAF testing

Post by Fat Freddy »

Hi all.

Just a question about analogue MAF testing. I can't think of a reliable way of checking them. TPS against MAF only shows outputs change not how good the value is. Looking at a capture of a snap throttle may show real bad ones. Comparing along side O2 is reliable either.
Say for example as you did a snap throttle, the MAF did the initial good gulp but the egr opened. The engine then starts to misfire, the O2 shows lean and the air flow doesn't increase so the MAF rate levels (this is a hypothetical scenario) and the engine runs rough.

How can you work out the chicken from the egg. Does the MAF under report in a certain area causing the miss. Or did another (egr) component cause the miss and the MAF levels because the airflow increase has slowed.

What am I missing? :?

Cheers
FF

tode
OneWave
OneWave
Posts: 48
Joined: Fri Jun 20, 2014 6:59 pm

Re: MAF testing

Post by tode »

Hello FF, I to see if the MAF is calibrated, using a very simple method:
- Ignition on and engine off, I read the voltage of the MAF if Bosch must be 1.0V and remain stable for a few minutes, sometimes it happens that the sensors start at 1.2V or 0.8V, if not calibrate or waiting a few minutes the voltage rises slowly . The sensors Siemes start from 0.6V.
For this test, the engine must be cold or unplug it from the suction hoses, to avoid noise in the measurement.
-Disconnect the MAF from the pipes and the blow with air-gun in it, the voltage must rise at least above 4.8V usually exceeds 5V.
Con this test I know that part of the MAF reading calibrated and unable to read the maximum air flow.
This test also control the digital MAF, watching them in frequency, I made a small database for Bosch, Siemes, Hitachi ....
regards

User avatar
Fat Freddy
TwoWaves
TwoWaves
Posts: 335
Joined: Fri May 04, 2012 4:52 am

Re: MAF testing

Post by Fat Freddy »

Thanks Tode.

So are you saying a Bosch sensor should sit at 1.0V and a Siemens at 0.6V
If either are above or below their specified voltage or start to drift after a period of time, they are no good?
Also what margin would you allow, =/- 0.1V ?

Something I will definitely have a look in to.

Cheers
FF

Valhalla
TwoWaves
TwoWaves
Posts: 66
Joined: Sat May 18, 2013 10:22 pm

Re: MAF testing

Post by Valhalla »

Hi FF,
I think that it would be almost impossible to predict what airflow a MAF is reading (digital or analogue) without (as Tode says) having an extensive library of signal conversion files, and therefore it would be difficult to say "chicken or egg" from the operation of the engine over a speed and load range.

However, there are principles that can be 100% applied to testing of MAF meters;
a) A MAF meter will not suddenly peak or drop-out during normal operation, so any such reading is a sure sign of meter or wiring problems. Most engine softwares filter this sort of error out with a LowPass, but they can get away with that because they have absolute limits from engine calibration data during development. The maximum theoretical rate that the signal can change in a fully-functional engine system depends on its airpath layout (trunking dimensions, etc.), but a common-sense approach is usually good enough to spot errors.
b) The only reliable area to test a MAF meter on an engine is at hot idle; this implies the minimum engine speed and load, held steady by the engine controller, below a speed+load that is likely to be boosted (turbocharged). As you allude, EGR is a major problem, and that needs to be eliminated either mechanically (blocking-off the valve, removing the vacuum signal from the valve control if it has such a system, etc.) or electronically (switching EGR function over diagnostic tool, waiting a set time to "time-out" the EGR, etc.). If you have a Landrover, then sticking the vehicle in LowRange will do the job of eliminating EGR - assuming the Transfer Box switch is working!. Many vehicles will detect the presence of a trailer over the electrical socket, and remove EGR at the same time. Once you have EGR removed, and you are sure that the breather system is not belching gas back into the engine, then it is a simple job to calculate the air displacement from the engine swept volume and speed, and, in the case of a throttled Gasoline, manifold vacuum. However, as Tode says, you will need to know what the signal conversion ratio is for that particular vehicle, meaning that a known "good" car is invaluable alongside the "unknown". No two vehicle models are the same, unless they share the same airbox and duct layout around the MAF meter, as each new model will have a unique look-up table calibrated during engine development.

tode
OneWave
OneWave
Posts: 48
Joined: Fri Jun 20, 2014 6:59 pm

Re: MAF testing

Post by tode »

Fat Freddy wrote: Also what margin would you allow, =/- 0.1V ?
FF
Yes the margin is +-0,1V.
Valhalla wrote:Hi FF,
I think that it would be almost impossible to predict what airflow a MAF is reading (digital or analogue) without (as Tode says) having an extensive library of signal conversion files, and therefore it would be difficult to say "chicken or egg" from the operation of the engine over a speed and load range.
For diesel engines that can calculate the theoretical air mass with simple programs, where enter rpm, manifold pressure and engine capacity, the result is compared with the reading with diagnostic, clearly the value is with egr closed.
If diagnosed I observe that egr is OFF and air mass too low (compared to the theoretical value), I perform the first test MAF then control circuit air / EGR.
For gasoline is more complicated to calculate the theoretical air mass because there is throttle valve, but the test is the same for the MAF and if the sensor is OK and Short / Long term fuel are off the scale, the problem is elsewhere.
Only database, in my opinion, is only for the various models of MAF voltage/frequency departure and voltage/frequency maximum, in the car park there are few models.
regards :D

Valhalla
TwoWaves
TwoWaves
Posts: 66
Joined: Sat May 18, 2013 10:22 pm

Re: MAF testing

Post by Valhalla »

I think that the point of note here is the airflow at "idle + EGR off" measurement is only worth taking with the scantool reading over OBD or the manufacturer's diagnostics; the Pico in this case is almost worthless as a measure!

Why? Because trying to extract any meaningful data from a PWM or voltage over the Picoscope trace is useless without a look-up table for that particular application. If you want to see a deviation of airflow from normal, an error of +/-0.1v (or PWM equivalent deviation) is not going to make things any clearer.

To clarify, one could the same airflow meter and place it behind two different airboxes (air cleaner assemblies) and have two different voltage outputs for the same mass airflow. That is why every engine development programme has a period of refining the look-up tables after each phase of development, to account for prototype and production mouldings, air-cleaner elements, etc. etc. This sort of work is done with a positive displacement measurement meter at steady-state (similar to a Rootes supercharger), and takes a long time to complete, but that is the only way of confirming the signal-to-measurement table. I have personally been through several engine programmes where the calibration of the table has had to be refined 3 or 4 times to get the accuracy correct for conformity of emissions and engine homologation.

User avatar
hexibot43
TwoWaves
TwoWaves
Posts: 123
Joined: Tue May 10, 2011 5:27 pm
Location: Redwood City, Californa, USA

Re: MAF testing

Post by hexibot43 »

The way I look at MAF sensors at our shop is a little different. But first thing I would agree is that a scantool is the better tool for diagnosis of the MAF. There is quite a bit of info on this subject out there. What you do is Calculate a theoretical Volumetric efficiency, and then compare that to what you can get based on readings from the car. It just takes a quite run down the street at WOT. Recording the MAF and RPM readings at peak, and then plugging them into a calculator for VE. I found this one here...

http://www.lindertech.com/downloads.htm

Run that and plug in RPM, Engine size, and Ambient Temperature. and run it for VE. It will give you values that you can compare to your taken readings. More than 15% off and I'd be seriously looking at the MAF.

This test doesn't work for Forced Induction vehicles such as ones with Turbo Chargers.

Just something to check out. These tests have worked very well for me.

arjen
TwoWaves
TwoWaves
Posts: 53
Joined: Wed Mar 06, 2013 7:18 pm
Location: Drachten, Holland

Re: MAF testing

Post by arjen »

You can also download the mafcalc app from diagnose centrum to calculate the value.

I believe the Bosch sensors have to be 1 volt exactly when you close both sides of the sensors housing.

When I have a maf reading I don't trust, I take a scope capture, then clean it with a maf cleaning spray, then take another capture. The maf is not always 100% afterwards, but if the second capture shows improvement then you know you're on the right way.

Post Reply