The Subaru SVX World Network

The Subaru SVX World Network (https://www.subaru-svx.net/forum/index.php)
-   Proven Engine Enhancements (https://www.subaru-svx.net/forum/forumdisplay.php?f=62)
-   -   tune your afr and stop your MAF from maxing out for $100 (https://www.subaru-svx.net/forum/showthread.php?t=29386)

michael 08-12-2007 09:41 PM

Similar to the idea of routing the bypass from the airbox to right behind the MAF. Why not play with different sized breathers where it's routed now. This would cut down on varying velocity speeds.

Trevor 08-13-2007 05:36 AM

Another Option.
 
Consider how the MAF transducer works. Those aware of the what is involved please skip this data. (I have been nastily accused of copying and pasting text from other sources and in particular the internet. What follows I have posted previously and it is copied and pasted from my files, but I am the absolute originator of same.) :(


------ MAF measurement is achieved by heating with an electric current, a wire/gauze/grid whatever, that is suspended in the engine’s air stream. The wire's electrical resistance increases with the temperature, which limits electrical current flowing through the circuit. When air flows past the wire, the wire cools, decreasing its resistance, which in turn allows more current to flow through the circuit. As more current flows, the wire’s temperature increases until the resistance again reaches an equilibrium.

N.B. The amount of current required to maintain the wire’s electrical resistance, is directly proportional to the mass of air flowing past the wire.

If air density increases due to an atmospheric pressure or temperature change, but the moving air volume remains constant, the denser air will conduct more heat from the wire, so that as a result, higher mass airflow will be indicated, thus offsetting what could otherwise constitute an error. -------


Having gone through this thread I have withdrawn the most pertinent points:-

“I've been trying to figure out how I can get some control over the air-fuel ratio and stop the MAF from maxing out quite as soon.”

“The problem is that the MAF has a limit on how much flow it can detect. Once the flow exceeds that value, then the MAF output stops increasing. The bypass fixes the problem by allowing some of the air to get to the motor without going through the MAF. The only problem I found was that it does mess with the idle in very cold weather.”

The suggestion has been to bypass the MAF, which must result in upsetting calibration, leading to methods ot trying to overcome this obvious defect in the proposal.

It occurs to me that if a deflector was fitted at a distance in front of the transducer, the mass of air flowing through the transducer, as a proportion of the main flow, could be reduced. It would seem that the ratio of this reduction would increase with velocity, and this should fall into line with that which is required. At idle and restricted throttle openings there could be very little affect. The distance the deflector is ahead of the transducer would have a marked affect in this regard.

Therefore there would be two contingencies providing fine tuning and requiring experiment. Size and aerodynamic shape of the deflector and distance ahead of the transducer.

It will be argued that a deflector will inhibit airflow at full throttle, but as it should not be required to exceed the area of the existing transducer and mounting assembly, this should not prove of consequence.

Trial and error would again be the order of the day, but surely would involve a great deal less work than the proposed bypass set up and it appears to me to be a more logical approach. Would it work? :confused:

Trevor 08-16-2007 04:36 AM

This rather interesting thread appears to have been bypassed, but meantime the the subject has been irritating my grey matter. :rolleyes:

The Final output signal in the form of a varying voltage of, I gather from approximately 1 - 4.5 volts, is the product of internal electronics in front of the actual transducer. I would suspect that each MAF requires fine tuning during production, and there could well be a resistor which is adjustable, or is interchanged, so as to cater for tolerances and component variations.

As I recall, a member has posted to the affect that what could be termed the amplifier, is easily exposed by unsoldering a cover plate. If someone could examine same, there may be evidence relating to what I suggest.

If this hunch is off line, it could still be on to make a change, or add a component, in order to modify the output parameters. What is required is the internals drawn as a schematic, so that analysis is possible. Any takers? ;) However often much is hidden within an IC, making critical information impossible to obtain. :(

An alternative and simpler but cruder line of attack, is to modify the signal at the output terminal, but this could unduly adversely upset the parameters. However the problems in this respect are unlikely to exceed those involved with an air bypass arrangement. What is more a throttle operated switch, could be easily incorporated, so as to override any added circuitry, until full or near full throttle.

There are very cheap voltage regulator IC’s available, which could be used to limit the output signal to a fixed 5 volts, as well as types which could be adjusted to requirements, by means of an associated potentiometer. This is rather extreme lateral thinking, but not all that out of line. :rolleyes:

Does anyone have figures indicating the maximum voltage of the output signal at full airflow and an idea of the current involved? The latter not so important as it can be expected to be mimimal. Both could be easily measured with an engine in operation. A low impedance voltmeter would be prudent requirement in this regard.

Exactly applicable comments will be awaited with interest.

Phast SVX 08-16-2007 11:42 AM

why not simpler approach that is more effective then bleeding air(not a good idea) and far easier then adjusting the voltage side of things.

One small but important aspect you are leaving out of this equation of your definition Trevor is the intepretation of the voltage can be overcome by increasing or decreasing the diameter(ultimately volume) that the meter resides in.

The ECU is using this constant to properly calculate airlfow into the engine based on the voltage the meter supplies. While this trick works better for hot-foil meters, its nothing a little R&D couldnt take care of. Note: Like the rest of the proposed adjustments, this will no doubt affect the timing. How much? small adjustments not too much, increased diameter 33% and installing 33% larger injectors(370cc nissians lets say it is a straightline relationship for ease, you may want to increase the radius by 30% and injectors by 33% to obtain your desired air fuel ratio in closed loop), and I really would be afraid to watch you take off on your maiden beat run.

food for thought

Phast SVX 08-16-2007 11:46 AM

Quote:

Originally Posted by Trevor
Does anyone have figures indicating the maximum voltage of the output signal at full airflow and an idea of the current involved? The latter not so important as it can be expected to be mimimal. Both could be easily measured with an engine in operation. A low impedance voltmeter would be prudent requirement in this regard.

Exactly applicable comments will be awaited with interest.

IIRC, the stock MAF meter increases too and becomes irratic over 4.7v, as measured by Sam(SVXtasy) via his greddy emanage ultimates log. This has been varified, by mychailo in another thread, and sparked him to creat his bleader valve.

Trevor 08-16-2007 04:42 PM

Quote:

Originally Posted by Phast SVX
why not simpler approach that is more effective then bleeding air(not a good idea) and far easier then adjusting the voltage side of things.

One small but important aspect you are leaving out of this equation of your definition Trevor is the intepretation of the voltage can be overcome by increasing or decreasing the diameter(ultimately volume) that the meter resides in.

The ECU is using this constant to properly calculate airlfow into the engine based on the voltage the meter supplies. While this trick works better for hot-foil meters, its nothing a little R&D couldnt take care of. Note: Like the rest of the proposed adjustments, this will no doubt affect the timing. How much? small adjustments not too much, increased diameter 33% and installing 33% larger injectors(370cc nissians lets say it is a straightline relationship for ease, you may want to increase the radius by 30% and injectors by 33% to obtain your desired air fuel ratio in closed loop), and I really would be afraid to watch you take off on your maiden beat run.

food for thought

Phil,

You say ---- " One small but important aspect you are leaving out of this equation of your definition Trevor is the intepretation of the voltage can be overcome by increasing or decreasing the diameter(ultimately volume) that the meter resides in.

In my first post you will see that I agree that the best approach is likely to be in relation to altering the ratio of air flow to output signal, and this is the exact feature of my suggestion involving a deflector to shield the transducer. This idea was in fact put forward as a still simpler alternative to altering the diameter of the body, such that a proportion of air will flow outside of the transducer.

I also agree that messing with the MAF signal is a far from ideal and adjusting parameters can never be exact and caution will be the name of the game.

We are on the same wave length. Trevor. ;)

Trevor 08-16-2007 05:08 PM

Quote:

Originally Posted by Phast SVX
IIRC, the stock MAF meter increases too and becomes irratic over 4.7v, as measured by Sam(SVXtasy) via his greddy emanage ultimates log. This has been varified, by mychailo in another thread, and sparked him to creat his bleader valve.

Phil,

I would expect that the maximum operative cut off point for the signal would be 4.5 volts, so that in excess of this would have no affect. Whatever, stumbling in the dark will never produce answers relative to any modification involving the electrical signal, so would you please point me to the thread you have mentioned.

Thanks, Trevor.

XT6Wagon 08-18-2007 01:57 AM

MAF diamiter vs Flow is highly non-linear. even worse sensor height in the body is also higly non-linear.

For example, lowering the sensor hieght in a stock diamiter maf-body just a little bit produces a large change in the reading. AKA why you can have "fuel tuning" with aftermarket MAF bodies for the WRX with stock diamiters. Even worse Subaru has gone to highly asemetrical bodies for the LGT and whatnot.

I'm not sure how it does it, but the unichip does its tuning by screwing around with the MAF voltages the ECU sees. So you can space the sensor on a normal body, get more headroom, and then pull down if you need to.

Even better if someone can find a cheap "translator" that has plenty of RPM/load points you can skip the stock MAF setups altogether and just run a Ford aftermarket MAF body and sensor. AKA what many people with those old worthless flapperdoor MAFs do. Course the kit for the BMW's I saw was nearly the same cost as a cheap stand alone so....


=============================
On the FPR, you can just stuff in a nice rising rate FPR (with 1:1 ratio please) into the return fuel line. The stock FPR just makes for a "floor" you can't set the fuel pressure below, but once you go over that, the added regulator does all the control. Done this on a good many cars. Even got a AVO500 to 22psi with good fueling by doing a FPR this way behind my stock FPR on my STi.

For really high HP setups you may want to replace the stock FPR with just an outlet pipe. I've seen a FPRless outlet pipe somewhere, but forgot what it was on. Likely a nissan of somekind. I know my XT6 engine swap the stock FPR was having hell with the fuel pump as it was sized for far more than it was designed to handle. Poor thing was screaming.

Phast SVX 08-18-2007 07:57 AM

Quote:

Originally Posted by XT6Wagon
MAF diamiter vs Flow is highly non-linear. even worse sensor height in the body is also higly non-linear.


Even better if someone can find a cheap "translator" that has plenty of RPM/load points you can skip the stock MAF setups altogether and just run a Ford aftermarket MAF body and sensor. AKA what many people with those old worthless flapperdoor MAFs do. Course the kit for the BMW's I saw was nearly the same cost as a cheap stand alone so....


=============================
On the FPR, you can just stuff in a nice rising rate FPR (with 1:1 ratio please) into the return fuel line(top left of the picture, mounted under the CC). The stock FPR just makes for a "floor" you can't set the fuel pressure below, but once you go over that, the added regulator does all the control. Done this on a good many cars. Even got a AVO500 to 22psi with good fueling by doing a FPR this way behind my stock FPR on my STi.

For really high HP setups you may want to replace the stock FPR with just an outlet pipe. I've seen a FPRless outlet pipe somewhere, but forgot what it was on. Likely a nissan of somekind. I know my XT6 engine swap the stock FPR was having hell with the fuel pump as it was sized for far more than it was designed to handle. Poor thing was screaming.

As i said, this works much better on a hot-foil sensor, and a little R&D would do the trick. The straightline relationship was for ease of discussion...... Going the MAF translator route is pointless, these injectors are not worth the trouble, far too small.

Regarding the RRFPR....

with stock SVX injectors its not a good option. The ~272cc injectors do not perform well over 75% DC, as with any side feed setup.

I used a Vortech RR when running 3.5lbs of boost for a while. I assume you meant an Adjustable rate fuel pressure regulator, since you are saying 1:1, meaning no pressure rise?
http://www.subaru-svx.net/photos/fil..._SVX/35676.jpg
i used anywhere from 7-12:1 discs while tuning. from a roll the RRFPR is horrid, dropping under 10:1 (my wide-band was reading in the 10's but as we all know a Bosch is not good for reading much under 10.5:1) accurately in the lower RPM range, to keep fuel safe up top. It functions decently for drag racing, but just dosn't cut it with the amount of pressure required on our rig.

Phast SVX 08-18-2007 08:15 AM

Quote:

Originally Posted by Trevor
Phil,

I would expect that the maximum operative cut off point for the signal would be 4.5 volts, so that in excess of this would have no affect. Whatever, stumbling in the dark will never produce answers relative to any modification involving the electrical signal, so would you please point me to the thread you have mentioned.

Thanks, Trevor.

Trevor,
I even posted earlier in this thread :rolleyes: should have really gone back and recapped before i posted.

I think all the info you would want on this is in mychailo's locker.

(http://www.subaru-svx.net/photos/user.php?mbtoloczko|27876)

Looks as though i was wrong, and you seem to be incorrect. There is something serious going on with the maf above a certain level, you can really see it oscillating above a certain load

Myxalplyx 08-18-2007 08:47 AM

Quote:

Originally Posted by Phast SVX
Looks as though i was wrong, and you seem to be incorrect. There is something serious going on with the maf above a certain level, you can really see it oscillating above a certain load

This is why I mentioned what I did previously. Adding a resistor in line to cap the voltage right before that 4.5 voltage point would solve have of the problem. The other half is to measure how lean the car is running and compensate with adding fuel (A FPR, RRFPR, SAFC, bigger injectors, etc).

No need to get extra tubing to let more air around the maf sensor. Let all the air you want go through it but compensate for the leaner air/fuel ratios.


Bah, it's just talk. I'd love to test this out.

Phast SVX 08-18-2007 09:15 AM

Quote:

Originally Posted by Myxalplyx
This is why I mentioned what I did previously. Adding a resistor in line to cap the voltage right before that 4.5 voltage point would solve have of the problem. The other half is to measure how lean the car is running and compensate with adding fuel (A FPR, RRFPR, SAFC, bigger injectors, etc).

No need to get extra tubing to let more air around the maf sensor. Let all the air you want go through it but compensate for the leaner air/fuel ratios.


Bah, it's just talk. I'd love to test this out.

you would need an adjustable rate fuel pressure regulator. You are running a stronger vacuum than stock. A rising rate regulator addx x psi of fuel pressure for every 1 lbs of boost and only works on a pressurized system. Te adjustable rate you can increase overall pressure in the system. Capping the system may not help, would just have to see how it reacts to reaching the threshold of 4.5v, it may still oscillate.. Think of it this way; The sensor is going through the motions, you are merely capping what it sends out to the ecu. This is why increasing the larger ID tubing is a better and often used solution.

oh and whoops, Trevor, I meant u were correct. I should really start paying more attention

XT6Wagon 08-18-2007 02:53 PM

Quote:

Originally Posted by Phast SVX
As i said, this works much better on a hot-foil sensor, and a little R&D would do the trick. The straightline relationship was for ease of discussion...... Going the MAF translator route is pointless, these injectors are not worth the trouble, far too small.

Regarding the RRFPR....

with stock SVX injectors its not a good option. The ~272cc injectors do not perform well over 75% DC, as with any side feed setup.

I used a Vortech RR when running 3.5lbs of boost for a while. I assume you meant an Adjustable rate fuel pressure regulator, since you are saying 1:1, meaning no pressure rise?
http://www.subaru-svx.net/photos/fil..._SVX/35676.jpg
i used anywhere from 7-12:1 discs while tuning. from a roll the RRFPR is horrid, dropping under 10:1 (my wide-band was reading in the 10's but as we all know a Bosch is not good for reading much under 10.5:1) accurately in the lower RPM range, to keep fuel safe up top. It functions decently for drag racing, but just dosn't cut it with the amount of pressure required on our rig.

1:1 rising rate means that if you gain or lose 1psi of pressure on the manifold pressure signal then the fuel pressure will rise or lower by 1psi. And yes the ones I use are adjustible, but I've never seen one that wasn't that you would want to buy. Talking about an aeromotive or other one that is designed to be connected in the return fuel line, not attached to the back of the OEM fuel rail.

I've also never encountered a fuel injector that had issues past 75% duty cycle. I'm wondering if its not an issue with the injector drivers as thats honestly the item that gets hit the hardest with high duty cycles. At 100% duty cycle the injector basicly has a free ride as it goes full open and is continualy cooled by fuel. The driver though has it rough as it handling alot of power with no interuption.

Also on the subject, possibly its time to inspect/test/replace the wires for the injectors. I know of newer cars that have corrosion issues with thier injector wiring. Given that atleast my SVX is hardly the least corrosion and wiring problem free car I've seen, running new wires for performance applications might be a good performance mod.

On your setup, The reason I don't like running more than 1:1 is the fuel system gets bruatly non-linear, coupled with what is typicly a brutaly non-linear MAF curve, and you now have fueling issues as the two are basicly out of your control. While I don't advise it, with a good highpressure Walbro fuel pump, 100psi at the rail is possible and makes for a hell of alot of headroom over the stock injectors at stock fuel pressure. Problem gets to be pulling down the maf signal at idle and low rpm since thats going to be about 70psi static at idle for a 20psi peak boost setup (-10psi vac idle => 20psi on boost). So yes, I'd advise larger injectors for forced induction SVX's. Problem gets to be with cars like the STi where idiots put in 850cc/min injectors to replace the 470/480 cc/min stock injectors thinking that the stock ECU running them 100% on boost is bad. So they end up with huge drivablity issues as the 850's are complete crap and a tuning nightmare. Where as a FPR is good for tossing on an easy to tune 10% more fuel if you are paranoid, or have a one in a hundred car that actualy needs more injector on the stock turbo. The STi also is apparently designed for this as I've seen alot of cars rack up alot of miles running 100% injectors with safe A/F ratios.

So moderation in your injector choice is a good idea too. Good news is that the selection of nissan sidefeed injectors is good and those injectors are not the complete pile that is the 850cc/min injectors the STi people love.

Trevor 08-19-2007 06:28 AM

The problem relative to cold idle surely involves the auxiliary air control valve. This normally opens when the engine is cold and the throttles are closed, so as to override control via the ECU. At start up from cold the mixture is therefore fixed by aperture values, prior to a heating period arranged as correct for the original set up.

When open the AACV valve bypasses the shut throttles, and in so doing notably draws the require air via the MAF. The fitting of a bypass means that at cold idle, the AACV valve will not draw all air supplied to the engine via the MAF, as would normally be the situation at idle with the throttles shut. As a result any signal from the MAF at cold idle, will be distorted accordingly

The MAF voltage spikes and swings in AFR at cold idle which have been noted with a bypass tube in place, could well represent a resultant feedback oscillating cycle. When the running temperature becomes normal, the AACV shuts, the ECU takes complete control of the AFR at idle via the bypass solenoid valve, and all is well as has been recorded.

In the event that a MAF bypass arrangement is used, this should be closed when the throttles are shut, so that the original idle system remains preserved. It is interesting that Greg touched on this issue way back in this rather confused thread, but his idea was discounted.

He also mentioned the idea of shutting off the bypass in line with closed throttles. This could be arranged mechanically, using a throttle assembly from a small engine ex lawnmower, or what have you. A suitable cam and lever arrangement would provide a quick opening at the initial movement of the main throttles. Alternatively this arrangement could be extended to provide control throughout the range of throttle opening, as a means of overall compensation.

It has been agreed that as is obvious, the addition of a bypass introduces extra non metered air, post MAF and prior to the AACV input, which must upset the original parameters. The difficulty, again obvious, is that this aberration must be corrected. I can not see how adjustments in respect of the routing, shape, or dimensions of a bypass tube, can in any way assist in this regard.

I still contend that shielding the MAF element is a well worth trying, as this would be much less complicated than a bypass system in several critical areas. What is latter advice within the thread, confirming the affect of element position within the MAF, tends to give credence to my suggestion.

The maximum output voltage delivered by the MAF can be cheaply controlled by means of voltage regulator IC as I have previously suggested. However if the ratio of air flow to sensor signal can be modified, this should result in a clean signal from the MAF and negate this requirement.

Inserted edit P.S. It has been suggested that a resistor could be inserted in the signal circuit, but this would reduce the voltage throughout the range of operation. Critically the affect would be most severe at the low voltage end and in reverse of requirements.

A resistor in parallel, i.e. inserted across the MAF output as a bleed, would modify the signal such that, as voltage increased there would be an increased reduction in voltage. Just how the resulting output curve would stack up is anybody's guess, but experimenting would not be difficult.

As there are no figures available covering signal current, as a precaution, it would be wise to commence with an overly high resistance of say 250 ohms, no responsibility accepted. A prior current measurement would be wise to enable calculations. A potentiometer/rheostat would present a hazard as it could be inadvertently set too low, but once a safe minimum value was established, a fixed resistor could be included as a safety stop during experiments.

In any event, I hope the above may serve as a summary, so as to tidy up salient points to date randomly included within what has now become a rather convoluted thread.

Myxalplyx 08-20-2007 08:00 PM

Teh Newb Posteth
 
Remind me again......

Why is it that the maf sensor is significant if you can push as much airflow through it as you want and add the amount of fuel needed by measuring your O2 sensor (or Wideband O2 sensor)?

I mean, if you can add the necessary fuel needed by measuring your exhaust O2 readings (Or EGTs which are my preference with O2 readings), you really would not have much to worry about. As long as the ECU isn't pulling timing based on the erratic readings. An Apex ITC could be used possibly but that is for another discussion.

OK...I'm scrolling back to read some more. Must've missed something.


All times are GMT -6. The time now is 11:35 AM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
© 2001-2015 SVX World Network
(208)-906-1122