80mm MAF Sensor for SOHC V6

[2000StreetRod]

Still too rich

I've run several more 90mm MAF sensor tests and determined that the MAFSAmp output voltage was too high at idle instead of too low. For the latest test I reset the PCM by disconnecting and reconnecting the battery. The MAFSAmp gain had been adjusted for 0.55 volts out with richness pot at midrange, ignition ON but engine not running. The engine started fine but the idle was unstable. Adjusting the richness to minimum (lean) smoothed the idle throughout warmup. The MAFSAmp output at pot midrange was 1.05v and caused the idle to become unstable compared to pot lean (0.95v). With the pot at lean the engine seemed stronger and more responsive than with the stock (55mm) MAF sensor.
Below are the 55mm vs 90mm with MAFSAmp voltages to the PCM with the richness pot at minimum (lean).

rpm 55mm 90mm
550 0.8v 0.98v
1000 1.0 1.2
1500 1.25 1.4
2000 1.4 1.55
2500 1.55 1.7
3000 1.65 1.9

You can see from the table that the MAFSAmp voltage is still too high even for the pot lean setting. I'll try to readjust the gain so the output is 0.8v at idle with the pot set to mid range.

 

[2000StreetRod]

Proof of MAFSAmp concept

Below is a table showing the latest results of the voltage to the PCM for my 90mm/MAFSAmp compared to the stock 55mm MAF sensor for a warmed up engine with the richness potentiometer at midrange.

RPM 55m 90m
0550 0.8v 0.42v
1000 1.0v 0.85v
1500 1.25 1.12v
2000 1.40 1.30v
2500 1.55 1.45v
3000 1.65 1.65v
3500 1.90 1.85v
4000 2.10 2.10v

The engine started fine from cold and idled fairly well but could benefit from slightly higher voltages at the lower speeds. I adjusted the throttle for 2000 rpm and then did not move my foot as I slowly changed the richness pot. The engine speed changed according to the pot (full rich = + 300 rpm, full lean = - 300 rpm). The engine seemed very strong up thru 5000 rpm (the fastest I tried).

I will tweak the gain some more but am satisfied that the concept is valid. Next I will proceed with the permanent installation of the wiring harness from the MAF sensor to the yet to be installed pillar pod.

 

[CDW6212R]

You have it very close, try to set the idle at about 700rpm. Mine hits a low of about 620rpm most days when hot, but just under 700 when cool. I also still have the stock PCM.

 

[2000StreetRod]

Idle speed

You have it very close, try to set the idle at about 700rpm. Mine hits a low of about 620rpm most days when hot, but just under 700 when cool. I also still have the stock PCM.

No matter what configuration I'm running (55 or 90mm MAF, rich or lean, different air filters) once the engine is warmed and the idle is stable it idles at 550rpm with A/C off. I assumed this was controlled by the PCM/IAC valve for closed loop. During warmup if conditions (too rich or too lean) exceed some normal limits then the idle speed is erratic.

I've seen a procedure posted on this forum for adjusting the idle for modifications such as racing. It involves disabling the IAC valve, adjusting the closed throttle plate stop, adjusting the TPS to zero, resetting the PCM and then enabling the IAC valve. The only other way to change the idle speed that I can think of would be with a custom tune. Am I overlooking an alternative?

 

[CDW6212R]

Yes I believe that is in the tune. I think they are a little low stock, but that 550 is very low. You could benefit from getting that up some, but you might have to wait until you get a flasher.

 

[2000StreetRod]

Pillar pod assembly

The photo below shows the completed internal wiring and assembly of my multifuntion pillar pod assembly.The left half is associated with oil and the right half is associated with richness. The connectors will allow for the installation/removal of the entire assembly.

The potentiometer on the left controls the MAFSAmp gain (richness). The test jack on the right allows connection to a voltmeter for monitoring the MAFSAmp output voltage to the PCM.
The photo below is a side view showing the richness control knob and the test jack.

Also visible near the bottom are two of the four mounting holes. I decided to use small black screws instead of the supplied friction fasteners since it was likely I would be removing/installing the pod multiple times.
The photo below shows the wideband Air/Fuel ratio meter mounted below the multifunction oil pressure/temperature indicator.

The tasks remaining for the 90mm lightning MAF sensor are:
1. Cutting holes in the pillar trim for the pillar pod mating connectors and wiring.
2. Wiring the power for the pillar pod from the instrument cluster wiring near the connector in the passenger kick-panel area.
3. Installation of the pillar pod.
4. Assembly of a custom 4 inch diameter to 3 inch diameter air intake system to include ports for the IAT sensor, IAC valve and PCV air inlet hose.

Before I drive with the 90mm MAF sensor I want to become more familiar with the characteristics of the stock 55mm MAF sensor. To do that I will get an additional bung welded downstream of the first O2 sensor on the driver side exhaust downpipe and install my wideband O2 sensor so I can monitor A/F ratios under various operating conditions. I will also learn how to use my laptop hosted Dyno-Scan to monitor and record associated OBD-II parameters. I will then collect similar data with the 90mm MAF sensor installed and compare the two datasets.

 

[2000StreetRod]

Initial installation & results

Yesterday I became inspired to resume work on the 4 inch diameter intake system. The photo below shows the results as of this morning.

The red arrow indicates the MAF sensor amplifier (MAFSAmp) that makes the 90mm Lightning MAF sensor electrically similar to the PCM to the stock 55mm MAF sensor. The green arrow identifies the loom that routes the MAFSAmp output to the pillar pod mounted richness control. The blue arrow identifies the new hose that connects the valve cover to the intake tube for PCV inlet air to the crankcase.

The red arrow in the photo below identifies the added separate IAT sensor from an Escort I purchased used with connector and pigtails for $0.99. The stock 55mm MAFS has an integrated IAT sensor.

The green arrow indicates the IAC valve hose connection to the intake tube.

The vehicle started immediately this morning from a cold start with an ambient temperature of about 55 degrees. It stumbled after a few seconds and then the IAC valve stabilized the idle which became very smooth. After a little warm up the PCM switched to O2 closed loop and A/F ratios were normal. I went for a short test drive. There is significantly more power at below 3,000 rpm! Driveability is excellent except for mashing the accelerator. If the accelerator is floored very quickly the mixture goes lean and the engine hesitates. Reasonable throttle change is smooth and satisfying. The A/F ratio is only affected by the richness control for changes in the throttle. Once the engine rpm stabilizes the O2 sensors bring the A/F back to normal. Overall, I'm very pleased but will need a custom tune for mashing the throttle to utilize the full response cabability of the intake system.

 

[CDW6212R]

Very good, that's the best I could have imagined with an "altered" MAF signal.

BTW, where did you source the intake hoses? I will need to make an "L" shape like my OEM V8 hose, for 90mm MAF to a 75mm TB. I hope to find it in one piece of about 3.75" and reduce it at the TB. If what you used comes in black I could use something like it.

 

[2000StreetRod]

Intake source

I purchased the intake 45 degree 4 inch to 3 inch diameter reducer from ebay seller Votion, Inc. It was designed for a turbocharged intake system and is thick and strong but flexible. There is a much smaller selection available for 45 degree reducers and even less for 4 to 3 inch. I recall that red and blue were available so I went with blue to match the Spectre cone air filter. I cut off several inches of the 3 inch diameter end to get it to fit between the throttle body and the cooling fan shroud.

I performed my first Dyno-Scan recording this morning of the engine idling from a cold start with the new intake system. The intake temperature initially was just under 54 degrees. During the 1000 seconds of recording the IAT increased steadily to 72 degrees even though the ambient temperature only increased a degree. The data supports the justification of a cold air intake to increase efficiency during idle. A box to enclose my cone filter is on my low priority to do list.

I started the test with the richness control set at the midrange detent. The engine started fine but then immediately almost died but the IAC valve "saved" it. That is an indication that the mixture is too lean during post start. During the first 20 seconds the idle speed rapidly increased to 1500 rpm and then dropped to 1100 rpm. During the next 180 seconds the engine speed gradually decreased to 920 rpm. At 200 seconds there was a sharp drop to 780 rpm and then a gradual decrease to 700 rpm at 330 seconds when there was another sharp drop to 640 rpm and the idle became unstable. The A/F ratio increased to 16:1 so I increased the richness control to keep the engine from dying and the idle climbed back up to 700 rpm and remained stable. Except for the brief period when the engine almost died, the A/F ratio varied reasonably around 14.7:1.

At 880 seconds I decided to increase engine speed to get the long term fuel trims (LTFTs) to change from their max of 25%. As soon as I opened the throttle from idle the LTFTs dropped to 0% in less than 20 seconds. STFTs all looked reasonable fluctuating between +5% and -5%.

I ran the engine at 2000 rpm for 100 seconds and the A/F ratio stayed good and no abnormalities were recorded. No CEL was illuminated.

 

[2000StreetRod]

Awesome response!

This morning I briefly tried my latest MAFSAmp adjustment. The ambient temperature was 54 degrees and the engine started immediately and idle was steady. The A/F ratio was a little rich at slight throttle so I adjusted the richness control toward the lean position to bring the A/F ratio within normal range in closed loop. I restarted the engine twice to clear the stored LTFT and then went for a short test drive. The engine response was awesome when I went to full throttle from light throttle at about 20 mph! No hesitation before or after automatic downshift. I was watching the A/F meter and before I realized it I was going 65 mph in a 45 mph zone so backed off immediately. The A/F ratio is about 12:1 at full throttle which is near ideal for the SOHC according to info I've received from those that know. I want to lower the MAFSAmp voltage slightly to be able to drive with the richness control in the midrange detent position.

Next is to see how it does at various ambient temperatures for starting and idling. I also want to record scan data on a short test drive to look at long and short term trims and O2 readings. I also want to make one of my 100 mile fuel economy runs. Overall at this point I'm quite pleased with the progress.

 

[2000StreetRod]

Temperature compensation

It became apparent that the MAFSAmp gain decreases with temperature increase. Subsequent testing shows that at the lower limit of the operating voltage output range (idle) the gain decreases almost 15% from initial power up over a twenty minute period with about 8% occurring in the first three minutes. This variation is undesireable since it impacts the idle speed and stability. I have ordered a package of thermistors and when it arrives will experiment with adding temperature compensation to the MAFSAmp circuit. In the interim, I will have to adjust the richness control slightly during the first few minutes of engine operation.

 

[2000StreetRod]

Thermistors, torque & mpg

My package of thermistors arrived yesterday so I spent some time adding temperature compensation to my MAFSAmp. I ended up using two in parallel to achieve (on the test bench) a less than 2% change in gain from turn on thru a 10 minute period. I don't know yet how well the compensation will work for the temperature variation experienced in the engine compartment.

I have no dyno data yet but it seems like my low engine speed torque has definitely increased. The transmission does not shift down as often as it used to going up hills. Also, it seems like the vehicle accelerates faster under light throttle from a full stop.

I filled my fuel tank yesterday and computed my gas mileage. I was surprised that I got 15 mpg considering I had floored the accelerator multiple times on almost every drive, I ran the engine in the driveway numerous times at high speed, and all of my driving was city (no highway). I am determined to drive conservatively on this tank to obtain a representative city mpg. My A/F meter serves as a constant reminder!

 

[Jakee]

This looks like a fun experiment but, I have to tell you, you'll have more fun once you purchase the pro racer package. After you get into the tuning you'll realize very fast that you can do everything you want with a keyboard. Trust me, you'll love all the data you get from the datalogs the SCT flasher gives.

You can even change the stoich to 15.1 if you'd like.

And yes, there's a seperate table for air/fuel during warm-up.

There are MANY tables that go off the load calculation.

Load = VE = % cylinder filling =....

VE = Normalized Fill Volume / Theoretical Fill Volume

(Intake Charge Mass/min) / ((CID)*(ρSTP)*(RPM)*(0.5))

Where denisty of dry air (ρSTP) = 0.00004413 lbm/in3

For Example, a 346ci engine idling at 1.0lb/min airflow at 700rpm:

VE = (1.0) / ((346)*(0.00004413)*(700)*(0.5))
0.187265918
= 0.187 or 18.7% load

The air is measured with the MAF so when you "trick" the MAF by using some sort of resistor, you distort the load calculation, which can result in bad things. For example, if the PCM thinks the load is 20% when the load is actually something else, the timing applied to the engine will be wrong....etc.

 

[2000StreetRod]

changing stoich

. . . Trust me, you'll love all the data you get from the datalogs the SCT flasher gives.

You can even change the stoich to 15.1 if you'd like.

And yes, there's a seperate table for air/fuel during warm-up. . . .

The air is measured with the MAF so when you "trick" the MAF by using some sort of resistor, you distort the load calculation, which can result in bad things. For example, if the PCM thinks the load is 20% when the load is actually something else, the timing applied to the engine will be wrong....etc.

Thanks for all of the information Jakee! I didn't realize that the Pro Racer Package had a scanner capability. I thought it only had the capability to program PCM stored data parameters. It sounds much more useful than what's described on the SCT website. If I had known that it had monitor/recording capability I might not have purchased the Auterra Dyno-Scan software.

Being able to change the desired A/F ratio is a huge capability! I've just started investigating the possibility of building an O2 voltage modifier to provide that capability. SCT should advertise that capability to attract interest from HHO users.

While watching my A/F ratio when my MAFSAmp gain was off it became obvious that the PCM was recomputing load and adjusting the STFT to compensate. I've just started looking at ignition advance to try and determine how it is changed by the PCM.

One of the reasons I started developing my MAFSAmp is I wanted the ability to change performance characteristics in real time. I guess I could have a similar capability with SCT's Eliminator Multiprogram (5) Switch Chip and Dial Selector. It looks like I would have to buy a switch chip programmer to interface to the Pro Racer Package. If I went that route then I might not need a Flasher. I guess I should check with James Henson before buying anything because the SCT published data is rather vague.

 

[Jakee]

The Pro Racer Package allows you to download the stock tuning "strategy" and gives the ability to modify. There are many tips on what does what and gives you a general idea of how things work.

The SCT xcal II or SCT livewire is where you data log. I think of these as just a flash drive or transporter to hold the new tune created from the Pro Racer Package that allows you to plug into your vehicle and upload the new tune strategy. You'll also have the ability to read and clear codes, and you can change a few tuning parameters (just the basic and it depends on if the turner that wrote the tune allows this during creation)

The SCT website will describe this alot better than I can but I wanted to clarify that the Pro Racer Package does not do the datalogs, sorry.

From what I see, you'll do good in the tuning field.

 

[2000StreetRod]

load vs throttle position?

. . . There are MANY tables that go off the load calculation.

Load = VE = % cylinder filling =....

VE = Normalized Fill Volume / Theoretical Fill Volume

(Intake Charge Mass/min) / ((CID)*(ρSTP)*(RPM)*(0.5))

Where denisty of dry air (ρSTP) = 0.00004413 lbm/in3

For Example, a 346ci engine idling at 1.0lb/min airflow at 700rpm:

VE = (1.0) / ((346)*(0.00004413)*(700)*(0.5))
0.187265918
= 0.187 or 18.7% load

The air is measured with the MAF so when you "trick" the MAF by using some sort of resistor, you distort the load calculation, which can result in bad things. For example, if the PCM thinks the load is 20% when the load is actually something else, the timing applied to the engine will be wrong....etc.

That's the same point that James Henson made to me - If the air flow information is incorrect then timing and A/F ratio will be incorrect. I'm careful to watch my A/F ratio to avoid engine damage.

I think I understand how the PCM uses air flow and engine speed to index tables to control pulse width and ignition timing. I don't necessarily agree that volumetric efficiency (VE) is a valid method to determine engine load. However, it can be used to index tables based on dynomometer testing that reflect engine load vs VE. I'm glad that Ford chose to use a MAFS based system instead of just throttle position and engine speed like some manufacturers. The MAFS allows the system to "adapt" more to changes.

I still don't understand the throttle position index (ranging from 100 to 700) in the 2001 stock air to fuel table you provided me.

 

[2000StreetRod]

Extreme temperature range

The engine compartment temperature ranges from about zero degrees (in Greenville, SC on a cold morning) to 150 degrees (after a drive on a summer day). I have been experimenting with thermistors to try to stabilize the gain of my MAFSAmp over this wide range. I have been only marginally successful. The latest attempt results in a slightly low gain (low voltage) on cool mornings and a slightly high gain (high voltage) for a warm engine compartment.

For nominal operating conditions the MAFSAmp output voltage to the PCM is slightly higher than the stock 55mm MAF for engine speeds greater than 2000 rpm. I've read the output as high as 4.4 volts at 4000 rpm accelerating at WOT. The MAFSAmp output results in a nearly ideal A/F ratio (12:1) at or near WOT. For moderate to light acceleration the A/F ratio stays within control of the closed loop (between 14.4:1 and 15:1). With the new intake system (4 inch dia intake, 90mm MAFS, MAFSAmp & 75mm throttle body) my Sport has actually become "sporty".

However, I suspect that I'll never be able to achieve the temperature stability required to start and drive the vehicle without having to adjust the richness control from very cold mornings to very warm afternoons. One of my modification criteria for success is that my wife can drive my Sport without any special instructions. So I'll soon be looking at alternatives to the MAFSAmp.

 

[2000StreetRod]

I concede!

The engine compartment temperature ranges from about zero degrees (in Greenville, SC on a cold morning) to 150 degrees (after a drive on a summer day). . . However, I suspect that I'll never be able to achieve the temperature stability required to start and drive the vehicle without having to adjust the richness control from very cold mornings to very warm afternoons. One of my modification criteria for success is that my wife can drive my Sport without any special instructions. So I'll soon be looking at alternatives to the MAFSAmp.

This is to notify techieman33 and Carguy3J that you were correct when you said that trying to use an amplifier to make a 90mm MAF sensor look like a 55mm MAF sensor would not work. I've been able to add temperature compensation to keep the gain constant over the large temperature ranges that occur in the engine compartment. However, I have not been able to compensate for the gain change that occurs during the first few minutes of operation after the MAFSAmp is turned on. I still have to adjust my richness control a few minutes after start up.

Last weekend I ordered an SCT X3 from Henderson Performance and I have discussed with James a custom tune for my configuration without the MAFSAmp and potentiometer control. I've also ordered the 00M12 intake kit and will install it before testing begins on the custom tune so that will not be a complication factor.

 

[angstrom08]

... The problem was obtaining an adequate settable voltage gain range (0.5 to 3.0) with a minimum voltage output of 0.8 volts. After numerous redesigns and reworks I now have a three stage transistorized prototype to evaluate...

Just idly curious, why did you make a discrete transistor amplifier instead of using an op-amp? If it was it just the parts you had on hand, what would you have used ideally?

Later, you said:

... This is to notify techieman33 and Carguy3J that you were correct when you said that trying to use an amplifier to make a 90mm MAF sensor look like a 55mm MAF sensor would not work. I've been able to add temperature compensation to keep the gain constant over the large temperature ranges that occur in the engine compartment. However, I have not been able to compensate for the gain change that occurs during the first few minutes of operation after the MAFSAmp is turned on. I still have to adjust my richness control a few minutes after start up.

Last weekend I ordered an SCT X3 from Henderson Performance and I have discussed with James a custom tune for my configuration without the MAFSAmp and potentiometer control. ...

So, was that the end of MAFSAmp?

Did you already have plans for a permanent installation of a final MAFSAmp unit? What I mean is, do you have any thoughts on how mitigating danger to a handmade circuit mounted in the engine compartment from the cold, heat extremes, vibration and humidity/leaks? What about supply power spikes, dropouts, and noise? PCM input protection?

Thank you very much for all of your contributions to the forum!

 

[2000StreetRod]

why not op-amp

The op-amps I remember from many years ago required negative and positive power supplies and I think the output voltage had a bias relative to the input voltage. No doubt there are numerous single supply op-amps available now but they may still have the offset.

Tuning software is designed to accommodate different MAF sensors. The PCM has a MAF transfer function table that can be altered for the characteristics of various MAF sensors. I eventually purchased by own software package so I could perform my own tuning. Below is the table for my stock 55 mm MAF sensor.

Below is the bench flow test results for my 90 mm Lightning MAF sensor.

It is so much easier to just change the values in the table than to "tweak" an amplifier circuit.

I avoid placing electronic circuits in the engine compartment due to the temperature and moisture exposure. It is difficult to build a moisture proof box that allows the heat generated by electronic components to escape. I installed an electronic fuel pump controller behind the lower dash for that reason. It has reliably operated for several months using automotive power.