Twin MAF readings to the ECU

[allmotor8v]

Say I wanted to design/fab a custom intake manifold that utilizes two throttle bodies. I'm just curious on how you would be able to wire two MAFs together to read the load.

What I was thinking is having two throttle bodies. The first throttle, which would be an OBDII (with the integrated TPS) would handle all the closed loop, low rpm acceleration, and most of the daily driving duty. Then, somehow, rig up the second throttle (this time an OBDI throttle so I don't have to worry about TPS input) to open as a second stage for more air to feed the motor.
The thing is, I want to keep the primary OBDII throttle intact with the velocity bumps so it runs perfectly smooth. Im experimenting with a completely honed out throttle right now, and while the WOT power is really nice, idle quality sucks. I was thinking if I could keep the primary for 80% driving duty, then cable up the secondary to open at a certain throttle plate position and just open up.
But my main problem is how the engine sees the load.....which lead me to my original question:

How to wire up two MAFs.


How would the ECU read two MAF wiring harnesses spliced together? What does the voltage on each wire to the ECU do? Does it add the readings together giving the ECU one voltage reading and then go off that? Or is the voltage averaged out? Do I need something to combine/amplify the signals in to one usable signal?

Not quite sure how this works to be honest and would appreciate some insight.

*edit* This is for an OBDII VW 2.0l engine utilizing Motronic 5.9.

Thanks -Trav

 

[70btdc]

Why does the second throttle have to be so big? As long as the area of the throttle (minus the projected area of the spindle, screws and plate when they are at at WOT position) is larger than the combined area of all inlet valves (circumference*lift) then it cannot be a restriction.

Maybe use a factor of safety of 1.2 to account for other losses around the throttle location, but any larger than that and you may be just sacrificing response and stability at anything less than WOT.

Also, how are the throttles arranged? Are they at opposite ends of the manifold? That would probably give interesting flow dynamics when the second throttle starts to open...

As a solution to the issue though, why not feed them both back to a large single maf and then v-pipe off from that? TPS is a different matter, throttle pedal position instead of throttle plate position may suffice.

I know old carbs had the twin-choke principle, but you dont see many twin throttle setups these days. Maybe try camming the actuation of a suitably large throttle to give you some contorl at smaller openings.

 

[patprimmer]

You are in way out of your depth.

Blending the signal will not add them, but power will flow from high to low voltage sensor..

Using two throttle bodies will not work better for idle than one large throttle body. To fix idle you need a very precise bore and good finish in the throttle body and a very good fit of the throttle plate. Matching throttle plates to bore is not a trivial matter unless you have the machines and knowledge of exactly how to do it.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[allmotor8v]

For manifold design, I was considering two ram pipes with throttle body flanges on either side of the plenum ends. Run them to either side of the bay into their own MAFs/filter housings. I was considering the 2 throttles into 1 MAF idea, but it's already been done and I wanted to try something outside the box.

As for the twin throttle idea, let me clarify a little...

With OBDII VW throttle bodies, the inside bore is designed with these velocity humps/ramps along the trailing edges of the throttle plate till about 40* of throttle angle. Also, on the edge of the lip of the front ramp, there is a dip, which acts as the first opening at lower throttle positions. It's all for air velocity for a snappier throttle response. All this helps with the drive-ability of the vehicle at low-mid throttle pressure. But for me, running a large profile cam, a professionally ported head, and the custom short runner intake manifold I designed, the throttle bore design is a hinderence. I got my hands on a spare throttle to play with and I completely de-rapmed the bore and modified the butterfly shaft to be very thin. Now while my WOT power has dramatically increased, the idle quality has suffered. The TPS is constantly hunting and produces an erratic idle, even when warmed up. So for the hell of it, I thought, what better way to increase the breathing characteristics than to run two throttles. Keeping the primary throttle untouched, drivability is retained. Having a secondary throttle open up at a specific primary throttle position to allow more air flow would more or less have the same effect as a bored out single throttle, but without any issues. Again, leading to my question of how to handle two MAF voltage signals to the ECU. That's the question I needs answered, not throttle design, or if it will work or not....thats the fun part of this project.

You are in way out of your depth.

I'm actually not as dumb as you think I am, so please don't paint me with the 'stupid' brush. I just don't know a whole lot about the voltage side of things.

 

[BrianPetersen]

How much flexibility do you have as far as the electronics are concerned; how much flexibility with the pipe routing? Are we talking drive-by-wire or cable operated throttle bodies? At idle, does it regulate the idle speed by cracking open the main throttle (not normal on modern cars) or does it do it using a separate idle-air control motor (conventional modern design)?

Why does having two throttle bodies imply having two MAF sensors? The MAF sensor is (intentionally) remote from the throttle body in many/most installations. Have all the air flow go through the MAF then split in a "Y" and go through your throttle bodies.

 

[allmotor8v]

Idle is operated by a TPS sensor built into the OBDII throttle (which my car uses) The secondary throttle would use an OBDI throttle which, in normal circumstances, is controlled by an ISV (idle stabilization valve. Both throttles are DBC.

The whole reason for two MAFs is to have two throttles on opposing sides of a custom intake plenum. Like I stated above, having two throttles Y-pipe into one MAF housing has been done. I want to try something different. Having two throttles on either side of a plenum and going to their own MAFs and air boxes. This is just for a fun side project to see if it can be done.

I just need some info on how, or if it's even possible, to utilize two MAFs to feed the ECU with voltage....

How would the ECU read two MAF wiring harnesses spliced together? What does the voltage on each wire to the ECU do?
Does it add the readings together giving the ECU one voltage reading and then go off that? Or is the voltage averaged out? Do I need something to combine/amplify the signals in to one usable signal?

 

[mattsooty]

allmotor8v

Listen to Pat - he is right.

What sort of EMS system are you going to utilise to run this engine? I would think that you will struggle to find an aftermarket unit that will cope with your application. If you are considering using the standard ECU then forget that straight away.

The OEM application that I work with is a two MAF, single throttle set up. In this unit the MAF signals are read in on two separate pins on the ECU and there is clever stuff that happens in the software to determine the measured airmass. Splicing the wires together wont do this.

Whilst I am in no was doubting your intellect I really dont think you have the necessary understanding to even begin to see this through to completion.

MS

 

[bigpaul]

hi fi you use tps only forget maf it is too slugish in response use a good ecu with good idle speed control you can set your idle speed to suit your engine, for super smooth O E M style control use idle bypass valve.

 

[DocDyno]

Several thoughts. First, the TPS usually only indicates a closed throttle (TPS less than 2%) so that the ECU can know when to manipulate the IAC motor and thereby control idle speed. So it is possible that the TPS may not be totally responsible for your erratic idle condition.

Idle fueling is traditionally controlled by a mass air or speed density calculation. Which causes me to wonder- if your ECU will support speed density fuel mapping (MAP vs. RPM) why not use a MAP sensor and forget about the MAF altogether. With a high performance-likely high overlap cam, your MAF signals may be confusing to your ECU anyway.

My suggestion would be to consider a MAP sensor in leiu of MAF.

 

[patprimmer]

I once had erratic idle from water in the TPS after the manifold was cleaned during an engine rebuild. It dried out after a week and is now operating as designed. Is your problem permanent.

This is not my are of expertise, but my understanding is idle speed is normally controlled via an adjustable air bypass valve on most EFI engines. The TPS controls fuel cut at closed throttle overrun idle then reinstates fuel supply for idle at a preset RPM on closed throttle to avoid a stall. If the TPS is not working exactly as designed, the idle fuel supply will be upset and maybe the idle air will be upset and maybe an anti stall throttle kicker might activate if RPM gets to low. All this will depend on make and model.

Matty is the man on EMS. Take anything he says on this subject as gosphel.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[bigpaul]

if you have a good ecu with idle strategies which have
target idle speed and also has high idle capture,tempreture offset and idle spark scatter.Also for the bypass valve
coolant duty table,rpm duty table and idle feedback loop.
The idle fuel table on tps load based engines do not measure the extra airflow due to the opening of the idle motor so does not compensate with extra fuel in the main fuel map, to maintain the target air fuel ratio an o2
sensor cannot make adjustments to fueling fast enough to cope with these changes the table allows fuel trim based on idle duty to mantain constant idle condition you would also need the abilty to change load site resolution around the
idle area. hope thats not to foggy an explanation.
One of you main problems is to compensate for throttle plate movement and small angles the problem is exaggerated with two t/b a small inbalace as in prentage terms the movement is far greater over these small openings hence the idle valve.

 

[allmotor8v]

Thanks for the info guys.

The poorish idle situation is definitely a product of taking all the material away from the immediate area surrounding the throttle plate. Im guessing bigpaul hit the nail on the head as to why its so irratic now.

With the idea I have, the secondary throttle isnt going to open until the primary is open at least 75%, so the secondary wont be opening with the primary, thus causing a whole slew of issues.

The OBDII throttle that is used on my motor has the TPS and ISV integrated into the throttle itself. Where-as the OBDI throttle, has an external TPS and a idle stabilization valve setup that comes off the stock intake plenum. The reason for using the OBDI throttle as the secondary is because its smaller (no electronics bulking it up) and it doesnt have the ramps on either side of the throttle plate. I realize that even with both throttles hooked up, they will cause a higher idle speed, but I have software that allows me to adjust the idles speed up or down as I see fit.

Honestly, the throttle situation isnt what is concerning me. I can figure it all out with trial and error (the fun part) I mainly wanted to know about the MAF signal(s) and how it works. Is this the wrong forum to be asking this in? Is there another forum on this site better suited to answering my question?

 

[70btdc]

You have 2 sensors that have the same output range. This means that the sensors need to be re-scaled to operate within the same range, but over the new higher airflow. The the second sensor needs to match the first sensor max voltage at zero angle (whatever WOT voltage is rescaled on small sensor), and increase to its max through the remaining voltage range left from the first sensor.

Second problem is you need to add the two voltages together when the second throttle is active, and ignore it when it is not. Naturally the higher voltage signal will just earth through the other sensor when it becomes larger.

Some simple electronics is all that is required to bodge a solution, made from cheap components. But you will spend the rest of your life fiddling to get it set up. Much easier is a single MAF for twin throttles.

Or,, as I said, get proper quadrant actuation on your large throttle body which will offer more adjustability off idle and improve the part throttle jerkiness you probably have now with increased flow/decreased flow control. Idle quality as suggested is not taken care of by the throttle valve, but a separate valve. You may have created a very leaky seal around your throttle plate by removing the 'step' and 'polishing'. You cant fix this now though. What method was used to port the throttle?



For info, please have a look here before diving in:

http://www.jenvey.co.uk/

 

[mcgyvr]

Your idle issue could simply be because you modified/knocked the position of the IAC motors pintle (sealing end) during your honing of the throttle body and now the ECU thinks it's position is not where it really is.
Quite a few modern ECU's need a recalibration procedure performed on the IAC motor pintle position when it's position is modified. (Chrysler/Dodge/Jeep dealer DRB3 tool has this functionality)
You could also simply move it in or out manually depending on if your idle is too high or too low and see if the idle is fixed.

Is it idling too low or too high?

or you have messed with the sealing of the throttle plate and the IAC motor cannot adjust enough to compensate for your leak. Boring a throttle body should not effect idle unless done improperly.

 

[allmotor8v]

I never touched any material around where the plate rests at idle. Didnt remove or even polish the area. So that isnt the issue. I just think the fueling map is set up for, and around, the ramps designed into the throttle bore. My idle on a stock throttle is about 850rpms (give or take due to the lope from the cam). With the modified throttle, it settles to around 880-900rpms. Sometimes it will jump up to 1100, then settle back down. All I removed was the ramps themselves, fore and aft of the plate. I should take a picture to show what Im talking about.

Thanks for the voltage info. I'll have to hunt around on the internet for some more info on how to equalize a MAF voltage output.

 

[mattsooty]

allmotor8v

I guess it is the way that these sort of threads progress & evolve but, to be honest, I'm not 100% sure if I can follow what it is, exactly, that you are struggling with here.

Yes some people have written all sorts of 'automotive buzzwords' - but most have not even the faintest understanding of what they are talking about. (If you look through the forum you will notice what I mean).

Anyway, I think that you will find some genuine help but to even begin to answer your questions I for one feel that more understand of the limits of your application, resources & more importantly ems is required.

After all is said and done, just about anything is possible with the right amount of sterling behind you.

MS

 

[mcgyvr]

To answer the twin MAF readings.
You will simply need circuitry that intercepts the stock primary MAF signal to the ECU and when the secondary TB is closed simply passes the signal to the ECU uncorrected. Then when the secondary throttle plate is opened (signaled by an increase in the secondary MAF signal) adjusts the primary MAF signal to the ECU proportionally to the secondary MAF signal. How much needs to be determined by extensive testing. So your ECU still only gets 1 signal from the MAF sensor but it will be electronically altered to show a higher reading depending on the secondary MAF signal to this "electronic box". Very similar to how a Apexi SAFC works.