Mazda MX-6 Forum banner

vaf gear + set screw = perfection

29K views 78 replies 19 participants last post by  BMW530i  
#1 · (Edited)
EDIT: entire write up changed for clarity


why, what, and how




WHY:

since the turbo and injector swap, its been running well, but RICH, very rich.

bad for mpg mostly, and fouls plugs quicker..not great for making optimum power either.


WHAT:

inside the VAF, under the plastic cover, is a gear with a click stop, the gear and spring are told how far to move, by how much the flapper door opens



Image


Image



turn the gear clockwise, puts more tension on the spring, makes flapper door more resistant to the pull of air opening it.

less flapper door pull, less fuel.

Image


to account for and adjust idle richness or lean, there is a small set screw, hidden by a plug that must be drilled out.

Image

(OLD PIC, PLEASE DONT HURT ME, no more zip ties)


counter clockwise leans, (more air let through), clockwise more rich (less air)


HOW:

hooking a volt meter up to your o2 sensor will give you a better ballpark of a/f than a narrow band, but still not perfect like a good wideband.

but its a good start, you dont want to do this blind.

volt meter = positive to o2 sensor
negative to any body ground

Image



this chart (basic google search for o2 sensor + voltage.)

Image


so now you have monitoring.

idle will bounce between rich and lean, so its really just getting it lean enough, or not too rich.

low rpm driving will give you a fairly solid voltage to go off of.

WOT will read a steady voltage, makes it easy.

i turned the gear clockwise, one tooth at a time between runs.

ended up going 6 clicks, and turned idle bypass screw a turn and a half or so clockwise.

END RESULT:

flawless idle, and near flawless rpm range.

touch rich coming from idle to throttle, WOT is near perfect.

keep in mind this is a temporary fix for insane richness, to aid a DD tune. if and when i get a wideband and/or fmu, life will be more exact and simple.
 
  • Like
Reactions: gavin
#2 ·
hmm..

do you still offer these sick?

I also cut out some extra's, $8.00 each shipped. Seems like it's easy to do, but each piece actually takes awhile to get fitment correct:

Image
 
#4 ·
Sick hasn't logged in since he flaked on the box.

You'd have better luck finding him on ClubProtege where he's trying to advertise his services without paying the advertiser's fee.
 
#33 · (Edited)
Allow me to bump this thread. As of a couple hours ago, I installed my MTX-L wideband.

After a year or so of 'ball parking it', and over a week of finally driving the car legally (registered on the 19th). I've been tuning it by ear, spark plugs, coolant temp and vacuum readings.

For the most part, it's just a stock GT with some goodies, but my MPG was lacking, and I knew it ran rich.

So for the past week or so I've been leaning it down, adjusting timing and changed spark plugs. And just today, I finally have a wideband.

My air/fuel at idle and cruise?


Image


Out of boost, I don't see under 13.5, or over 15.

I understand this is on the lean side, and translates to higher engine temperatures, coolant temperatures, and risky boost situations.

I can assure you I have an above adequate cooling system, also take into account the hood louvers, overheating is not a possibility.

While in boost, up to 15psi, I dont see over 12.5. No pinging or detonation, just smooth power.


I have a driving video (with too much wind noise) if anyone is interested, I'm going to upload it to youtube tonight.
 
#5 ·
Meh for 8 bucks, I'm not ganna start a search party.

Funny though lol.
 
#10 ·
There's threads about this dating back to 2003 and 2005.

sure this has been out, but no one supports it, no one says, yeah go for it.

For you GT people I can understand, there's no need to lean out, maaaybe just richen.

But for n/a people going forced induction, this is very useful info.

But again, no one to say, yeah go for it, just nay sayers.
 
#11 ·
Sure there's a reason for "us GT people" to lean out. (What do you mean, "you people?") This is a quick and dirty way for us to run larger injectors while maintaining a resemblance of driveability around town without resorting to a full standalone. Many have done it, and maybe people will in the future again. :p

It's not the best way to do things, but if it works, it works. I've seen videos of 330whp+ Escort GTs using nothing but the VAF to "tune," and they hold together for at least a few months at a time.


I'd be lying if i said i hadn't considered doing this on my car at some point or another. I've seen a success story of this type of tuning on this cars in real life, hell, even ridden in one, and it was massively fast, had been together for a long time, and drove like stock.
 
#15 ·
im sorry, but just to be a touch off topic for a quick sec...

as of midnight last night, or 12:01 am at the grocery store at time of purchase to be specific..

turned the ol' 21 years old.

heheheheheh fun fun fun.
 
#18 ·
Meh, i'm all for tampering with the factory tune by any means possible. I can't speak for the LX cars, but the factory maps on the GT are nothing to be proud of. Of course, i would expect no less from any 20+ year old Japanese turbo car.
 
#19 ·
Eh...I guess it depends on what your expectations are and which side of the fence you want to look from.

Its a zippy commuter car with an ECU designed in 86(?)...With very limited memory to write code (when you have an 8 or 16bit processor, you cant afford to be sloppy with code) we have a cars that are still chugging along admirably....despite age, despite questionable maintainance, despite ham-fisted hot-rodding attempts. Turbo cars in the 80s gad gained a bad rep as being unreliable. While that poor rep was mostly crappy maintainance(user error), the point stands that a lot of these cars survived that era of crappy maintainance as well. Thats a testament to good engineering and a robust tune thats proved its worth.

I think thats pretty impressive.....but, it doesnt give the car 500hp instantaneously so, you know...it sux :shrug:

Gavin
 
#20 ·
Heh, i meant power-wise, and by modern standards. But then again, back then, turbochargers STILL weren't understood as well, nor did they have the means to create the ideal maps. But yes, they did very well in terms of longevity by staying on the safe (very) side. :)
 
#22 ·
now how exactly do you get a gain out of bigger injectors if you lean it out with the vaf? max duty cycle? temp fix to run it like normal?

Since we are inside of it... any way to stop the fuel cut from the vaf?
 
#23 ·
The gain comes from the change in injectors, is you need 20% more fuel get 20% larger injectors.

But at idle and cruise you don't need 20% more fuel so you use the VAF idle screw mixture screw so that the VAF reads lower load at idle and cruise and you adjsut the spring tension so that the VAF reads whatever load it requires to get the right amount of fuel out of the larger injectors.

The problem with this whole method is that the ECU also uses the VAF load signal to determine ingition advance.
It is therefore a long way from ideal and you cannot get optimum results.
 
#24 ·
The problem with this whole method is that the ECU also uses the VAF load signal to determine ingition advance.
It is therefore a long way from ideal and you cannot get optimum results.
Interesting...

I was not aware of that.
 
#25 ·
I have done this using 440cc injectors and let me say it works brilliantly. I was a bit worried that the ignition timing would be too advanced, but I am yet to notice any pinging.

This car was never intended to be a daily driver, so I didn't care about fuel economy, but I have found that the a/f ratio of around 11.5 at full boost means I am running around 16 or leaner when cruising and idle around 18 on my wideband. Which is giving me fuel economy that I cant complain about.

The only issue is I think I have maxed this setup already. My setup is a VF10 hybrid, big fmic and a 2.5" cat back running around 18psi. I have found that on cold nights (16*c) I run out of adjustment on the vaf. The reason I say this is because when it approaches 5000rpm it starts getting too lean (up to 12.5) and if I try to make it richer it will boost cut, meaning I have reached the maximum duty cycle that the ecu will run the injectors at, while I still need more fuel.

It is still early days and I still have some work to do checking that there aren't any other factors changing my results. I need to change the clutch and do a lot of little things like test for boost leaks and change the fuel filter before I start chasing more fuel.

Can anyone tell me what is the maximum duty cycle that the ecu will run the injectors at?
 
#26 ·
one more question with the vaf too then, is if the computer will still read the same duty cycle from a bigger injector as a stock one. I imagine it would.

See I have a set of 440s sitting in my toolbox and a t3/t4 ready to go in but with a stock ecu. I suspect a bad injector so this has helped nicely so far. thanks for the info.
 
#27 ·
The ECU doesn't read the duty cycle from the injector, it supplies the duty cycle to the injector.

The duty cycle it supplies to the injector depends mainly on load (VAF angle) and RPM.
Generally the higher the load and RPM the higher the duty cycle.
If you throw in larger injectors and change nothing else then yes the ECU will supply the same duty cycle to them hence creating an overly rich A/F ratio.
Except for at closed loop (idle and cruise) when the ECU will reduce the duty cycle to try and get the O2 sensor output to stoich.
If the A/F ratio is too far off though, the ECU will not be able to trim it back enough.

By altering the VAF spring tension you are aiming to change the load signal from the VAF so that the ECU is fooled into suppplying a lower duty cycle to the injectors when not in closed loop.
Likewise by adjusting the VAF bypass screw (idle mixture screw) you are making the idle and cruise A/F ratio closer to stoich so that the ECU has an easier time adjusting the duty cycle.
 
#29 ·
From what i remember, at a certain load, it just kindof dumps all the fuel in there.
 
#34 ·
12.5 at full boost is a bit lean, in my opinion. You aren't gaining anything by running it that lean. I usually shoot for 11.8-12.0 on these motors.

There's nothing lean about your cruising/out of boost AFRs. Mine don't dip below 14.2, normal cruise is 14.6-15.2.
 
#35 ·
You aren't gaining anything by running it that lean.
My in-boost MPG has skyrocketed, I used to blow through a tank of gas after a couple days or a week of 18psi drives, now I barely see the needle drop.

But yes, your advice is noted, as always. Thanks.
 
#37 ·
Depends. If you tighten the spring so much that under full throttle/full boost, it doesn't trigger the cut, then yes.

But at that point, you're probably way too lean and the spring is very tight. There are simpler ways of bypassing cut.
 
#40 ·
Don't make me sound like an idiot. Obviously MPG is not my only concern while at WOT, I'm not naive.

The compression ratio on the GT is tractor status, it's rock bottom low. It's damn near fool proof.

Running 12.5 A/F @ 16PSI, on a motor with anything higher than 9:1 compression ratio, then yes I would be a little nervous. I'd have an EGT gauge and maybe a standalone to fully control timing maps.

12.5 A/F @ 16PSI, on the GT 7.8:1 pistons does not pose any risks. As long as the ignition isn't advanced all the way lol.


(My friend bought a blown GT of a clueless boost junkie. PO blew a hole in the piston at 12PSI, detonated and pulled over. Thought he just needed to change a spark plug. When I finally got to take a look at the motor, the first thing I looked at was the distributor, fully advanced. #1 cyl had a golf ball hole in the piston)

Clearly, you have to be careful with these motors. Without getting too narcissistic, I'm not dumb, and I am very careful. Check, double check, triple check, test, tune, check, repeat.
 
#41 ·
That wasn't my intention... untwist. ;)

The thing is that with your method of tuning, you're also making your timing maps more aggressive. So i'm just saying that because 1) you don't know what your timing is and 2) it's definitely advanced from stock throughout the range, there's absolutely zero point to running it at 12.5:1.

Do what you want, i'm just pointing out that it's not really the greatest idea. 12.5:1 is the target of perfectly tuned motors running a standalone ECU with efficient combustion/quench areas.

Neither your car or mine satisfy ANY of those requirements.
 
#42 · (Edited)
Well duh it's advanced, I have a timing light that tells me that, I've tuned for that. It's at OEM 9dbtc, knowing that the probinator ECU advances, I made sure to be careful about that and treat the motor accordingly.

There IS a reason to run 12.5, I literally reap the benefits on a daily basis. I have multiple gauges and sensors telling me there's a difference. I'll hesitantly mention the butt dyno, and I'll leave it at that.

If this was the N/A block, with even that little bump up in compression, I would richen the tune a tad. I speak from experience in that department.

Having run 12PSI on my N/A-T for 2 years hand tuned with no wideband, problem free, I feel confident about the tune on this GT. If you'd like to see piston tops, spark plugs and cylinder walls from that motor, I could provide the pictures. I've fixed the leaks from bad gaskets and it's currently waiting to replace this slow ass motor.

If this was a honda motor (or even the FE3) with 9:1/10:1 compression, it's common sense that you wouldn't even be near stoich in boost unless complete control over the maps was in place..even then, it would probably be pig rich just to be safe.

OEM ignition timing, OEM 7.8:1, normal and stable coolant temps, 16PSI @ 12.5 A/F, will not detonate, will not explode. I'm a living example of this. Perfectly tuned? I can't say that, but it's really damn close. If it wasn't, I wouldn't risk it. I wouldn't even be defending this point if I wasn't fully confident of what I'm doing and how the engine is doing.


Abnormal combustion temps cause pre-ignition explosions that blow holes in pistons, advancing the ignition is a precursor to imminent damage while boosting. I am not advanced enough for this to be an issue, nor am I running enough boost at extended intervals for this to be an issue. 20-25PSI? Yes, I would richen the tune to compensate for the combustion temps and chipped ECU advance. I would even run the MSD BTM with the neato-keen retard function at that point.

Yes, a standalone is of course the final answer. I'm not denying that.

I know I'm easily the least liked or respected person on this forum, but I beg you to give me just a little more credit.
 
#43 ·
Well duh it's advanced, I have a timing light that tells me that, I've tuned for that. It's at OEM 9dbtc, knowing that the probinator ECU advances, I made sure to be careful about that and treat the motor accordingly.
That's not what i'm talking about, though.

There IS a reason to run 12.5, I literally reap the benefits on a daily basis. I have multiple gauges and sensors telling me there's a difference. I'll hesitantly mention the butt dyno, and I'll leave it at that.
I'm afraid you're not understanding, or maybe i'm not explaining myself clearly, but it's your car, do what you want.

If this was the N/A block, with even that little bump up in compression, I would richen the tune a tad. I speak from experience in that department.
You mean if you go from... an incredibly low compression motor with a horrible combustion chamber, to another incredibly low compression motor with a horrible combustion chamber? :p

Having run 12PSI on my N/A-T for 2 years hand tuned with no wideband, problem free, I feel confident about the tune on this GT. If you'd like to see piston tops, spark plugs and cylinder walls from that motor, I could provide the pictures. I've fixed the leaks from bad gaskets and it's currently waiting to replace this slow ass motor.
I don't need pictures. It's your car, i was simply saying that there's no point to running the ragged edge when it comes to AFRs unless you truly have FULL control over everything the motor is doing. Which, you don't. Neither do i.

If this was a honda motor (or even the FE3) with 9:1/10:1 compression, it's common sense that you wouldn't even be near stoich in boost unless complete control over the maps was in place..even then, it would probably be pig rich just to be safe.
That's not really common sense. That's actually... not correct.

Source: How i tune my 9.5:1 turbo motor on a Haltech Platinum Sport 1000.

OEM ignition timing, OEM 7.8:1, normal and stable coolant temps, 16PSI @ 12.5 A/F, will not detonate, will not explode. I'm a living example of this. Perfectly tuned? I can't say that, but it's really damn close. If it wasn't, I wouldn't risk it. I wouldn't even be defending this point if I wasn't fully confident of what I'm doing and how the engine is doing.
I'm saying point blank that you do NOT have OEM ignition timing. Because you messed with your VAF. Said it before, and it's the truth, cannot be argued, which is why i'm not entirely sure why you're getting angry over this.


Abnormal combustion temps cause pre-ignition explosions that blow holes in pistons, advancing the ignition is a precursor to imminent damage while boosting. I am not advanced enough for this to be an issue, nor am I running enough boost at extended intervals for this to be an issue. 20-25PSI? Yes, I would richen the tune to compensate for the combustion temps and chipped ECU advance. I would even run the MSD BTM with the neato-keen retard function at that point.

Yes, a standalone is of course the final answer. I'm not denying that.

I know I'm easily the least liked or respected person on this forum, but I beg you to give me just a little more credit.
It has nothing to do with respect or liked or any of that. I don't care.

You're advancing the ignition. Also: MSD BTM is a great way to cripple an F2T. Seen it installed first hand. Ran like crap. Removed it. Ran not like crap.



Think about it: Modifying the VAF setting as you're doing is "tricking" the VAF, and therefor, the ECU into seeing less load and air. Therefor, if the timing map is anything like every other motor on the face of the planet, at less load, the ECU will increase ignition timing. You've created MORE ignition timing at your airflow levels than stock, on top of your advance from the chip. And then you're leaning it out to 12.5:1 on top of that. Just to put this in perspective, you're running MORE timing than i am, and i saw no benefit in terms of power and was unwilling to sacrifice safety moving from 11.8-12.0 range to 12.2-12.5. (Yes, i've done testing at both.)


Here's the timing map to my 9.5:1 turbo motor, to illustrate the relation between load and timing.

Image