SAFC... What exactly does it do?

Thread Tools
 
Search this Thread
 
Old Aug 7, 2004 | 03:19 PM
  #1  
beepy's Avatar
Thread Starter
Member
 
Joined: Feb 2004
Posts: 2,451
Likes: 1
From: Pearl City, HI, USA
Default SAFC... What exactly does it do?

I have a basic understanding of the SAFC and VAFC units, and I own one and will soon install it, but after reading a lot of posts on them, I am a little confused.

It is my understanding that the AFC intercepts MAF or MAP signals, modifies them ±50% (for the SAFCII) and then sends these modified signals to the ECU, which uses its fuel and timing maps to operate the engine. For the majority of us, we are using the negative side of the adjustment in order to install larger injectors, so that theoretically, one could use an injector twice the stock size with the -50% adjustment, and have the engine run the same as before.

Now this is what is confusing me. I have a MAF sensor.

1) If you use a -50% adjustment, what will happen if the stock fuel map doesn't have an entry that low? Can I just assume that the stock fuel maps go all the way down to 0 air flow values, and that the ECU can extrapolate the correct injector timing?

2) What if my airflow values exceed what the RPM/airflow table have entries for? IOW, if I am producing 100% more than the stock airflow would be at 3000 RPM because I hit full boost, will I hit a "hole" in my fuel map? Another way of looking at this: Can I assume that my RPM/airflow table is "square", or that there are entries comparing all possible values of RPM with all possible values of airflow, even though the stock engine would only use a small part of these under normal conditions? For example: The stock engine is not going to use 500 CFM at idle. But would there still be an entry for this value? (The Hondata/Uberdata gurus should know the answer)

Fortunately I know I don't have to worry about exceeding the airflow tables themselves, since with the -50% adjustment I could theoretically move twice as much air within those tables, and I am only making 7 PSI, or about 50% more than stock airflow.

3) How is timing really affected? The AFC does not directly alter ignition timing, but how does the change in the airflow alter the timing (in, say, a generic ECU). IOW, would reducing the airflow values advance or retard my ignition timing? I have seen posts where timing was shown for a certain Honda motor, and it was dangerously advanced, but this was a MAP vehicle and not MAF.

4) If the timing is advanced, can I counter this advancement with the boost retard available in the MSD DIS-2 ignition? (I have coil packs, not a distributor)

5) Last question. I read one reference to the SAFC that said if higher RPMs were reached without boost (say, at high cruise), that the engine would go excessively rich. I don't understand this. Shouldn't the airflow (or MAP) values differ between high-speed cruise and WOT at the same RPM, and therefore shouldn't the fuel injected differ?

Or is this just a case where he had to use a lot more fuel at the upper RPMs because of making a lot of boost, in order to avoid detonation, and it was probably rich in both situations? Again, do I have to worry about this effect with only 7 PSI or less?

The reason for all of these questions is because I want the car to work properly when just using it as a daily driver, and I want it to work properly when making boost. IOW, I want it to work like an OEM turbo car. I am tuning with a wideband. For the past 6 months or so, I have been driving without fuel management, but rather just the largest injectors that my ECU could "learn" to use, and relying on the short-term fuel adjustment built into the OBDII ECU to add more fuel for boost. It has worked, but now I want to tune it better!
Reply
Old Aug 7, 2004 | 10:08 PM
  #2  
beepy's Avatar
Thread Starter
Member
 
Joined: Feb 2004
Posts: 2,451
Likes: 1
From: Pearl City, HI, USA
Default Re: SAFC... What exactly does it do? (beepy)

Bump for the MAF-users!

I've read the entire instruction booklet, and it only raised more questions than answers. Does the fact that I can set a "full throttle" and a "part throttle" allow the SAFCII to be used for a daily driver, or do I have to make seperate files for communting and performance driving?
Reply
Old Aug 8, 2004 | 12:59 AM
  #3  
Just Checking In's Avatar
 
Joined: Mar 2003
Posts: 233
Likes: 0
Default Re: SAFC... What exactly does it do? (beepy)

1) If you use a -50% adjustment, what will happen if the stock fuel map doesn't have an entry that low?

Remember that when even at idle the map sensor is reading a small voltage in the tenths decimal place, well your taking half the value of a value.....it can never be 0. A percentage of the begining spark and fuel values on the table will be used.

For your other questions, I just wrote this in another thread titled VAFC info. I don't want to pick it apart in here again for each question.....here is my whole post.
:
It looks like some of you guys have the concept down to a degree but not really why its neccessary.
With the vafc hack, maximum amount of boost before 11 psi when the map sensor errors, is dictated by the injector size you choose to run. 450 cc injectors are popular with the vafc hack because they will allow you to scale the map sensor voltage nearest 3.09 volts with room to increase the amount of fuel needed above the percent differance between the injectors. This lets you clear the greatest timing advance section of the ignition table and lets you create a close 14:1 air/fuel ratio per psi of boost.

Mathematically heres an example.

Say you want to run 10 pounds of boost and you bought the correct 450cc injectors for this.
The differance of flow is 46% because (240/450)-1= .46
Then because larger injectors have to be compensated for latency we subtract about 5% for 450cc's leaving us with a 41% fuel difference
This means at -41% setting in your VAFC your 450's are giving you the same fuel your 240's were.
Now the map sensor goes into limp mode at 3.1 volts so you need to stay below that, and 10 psi is about 4.7 volts.......... so 4.7-(4.7x.41)=2.77
This shows us that 450's give us enough room to reduce the map signal below 3.1 volts with enough differance in volts between 2.77 and 3.1 to increase the fuel %cuts to accomodate for boost.

So what we do in terms of setting the vafc is this. We know the differance in flow between 450's and 240's is -41% and at idle air/fuel ration does not change with the turbo. So at idle, if you run 450's you will set your VAFC to -41%. As your rpms increase you want to add fuel to compensate for boost you could be making at those rpms. So on average, people add 1% per 1,000rpm's above 3,000rpm where boost begins to build with average sized compressor housings and wheels. You don't really want to go past a 5-6% overall increase though because you will be increasing your flow differential at the same time the ecu is increasing pulse duration. This can make for a really rich combustion.

Here is a good example of settings for a b16 running 450's and VTEC set at 4,500 rpm:
Same settings for wide and narrow with hi and lo
1000 -41%
1500 -41%
2000 -41%
2500 -41%
3000 -40%
3500 -40%
4000 -39%
4500 -38%
5000 -37%
5500 -36%
6000 -36%
7000 -35%
8300 -35%

Do you see how you want to transition into the air fuel ratio gradually? If you just set it at -35% across the board you are running entirely too rich at idle and anything below 4k roughly. Different setups will build boost at different rpms so this is just an example of something average.

Once you grasp the concept you'll notice some issues. You will see why the VAFC is a piece of **** for spirited city driving. It works for cruising below 3,500rpm or full throttle runs and thats about it. Otherwise all your settings that you input so that you have a correct a/f mixture when flooring it......creates entirely too rich a/f's when you're just trying to cruise around and you just happen to go past 3,500rpm with anything less then half throttle on the average sized compressor housing, on a small displacement motor. You will also see why to run safe with the vafc you should run 10 psi, so that your fuel correction values aren't putting you dead center of the highest ignition advance on the table. You also should slightly compensate your idle and low end power by retarding your base TDC timing down to about 12 degrees instead of stock 16-18.

:

If you look at a spark bin on ghettodyne or zdyne or whatever you choose, you will understand why your timing is advanced for boost. Because at 5 psi for instance, a -38% cut in fuel will move you into the advanced middle area timing section of the stock table. This is why people suggest running 10psi, because a -38% cut of 10 is a higher number.....thus not moving you so far back into the advanced area in the middle columns. When looking at those value bins you can regard the vacuum/boost column headings as being linear and directly related to your map sensor voltage....which is the only thing the VAFC is altering.

If theres something your still having trouble grasping just ask and myself or someone else will hopefully be able to answer it.
Reply
Old Aug 8, 2004 | 01:23 AM
  #4  
Just Checking In's Avatar
 
Joined: Mar 2003
Posts: 233
Likes: 0
Default Re: SAFC... What exactly does it do? (beepy)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beepy &raquo;</TD></TR><TR><TD CLASS="quote">Bump for the MAF-users!

I've read the entire instruction booklet, and it only raised more questions than answers. Does the fact that I can set a "full throttle" and a "part throttle" allow the SAFCII to be used for a daily driver, or do I have to make seperate files for communting and performance driving?</TD></TR></TABLE>

Making your narrow throttle settings different from your wide leads to driveability issues when using the VAFC as a turbocharger fuel management computer. For instance, say your driving along and you decide to go up to 5k rpm at 1/4 throttle, if you then decide to floor it.....the instant your pedal goes over your wide throttle point you will be violently changing fuel values into one that expects full boost.....which you have not hit because only half a second ago did you decide to floor the pedal. So you would be cruising along at -40% and then when flooring it at 5k rpm go straight to -36% without full boost. This would be a harsh transition and it would become annoying.

VAFC boost "hack" users almost always use the same values for both wide and narrow throttle, as well as making sure the high and lo cam settings within those match for whatever couple of points are on both hi and lo cam selections. These are defined by you but usually are two rpm values around 4-5k rpm since thats where most people set VTEC engagement.
Reply
Old Aug 8, 2004 | 09:11 AM
  #5  
beepy's Avatar
Thread Starter
Member
 
Joined: Feb 2004
Posts: 2,451
Likes: 1
From: Pearl City, HI, USA
Default Re: SAFC... What exactly does it do? (Just Checking In)

Thank you, that is a good explanation.

I guess I will have to see how it works with my MAF sensor and OBDII ECU. Since the FO2S is independant of the SAFC, I am hoping my short-term fuel correction will smooth out some of the inconsistancies with part-throttle cruising.
Reply
Old Aug 8, 2004 | 11:35 AM
  #6  
mekkis's Avatar
 
Joined: Apr 2004
Posts: 151
Likes: 0
From: I smoke crack rocks
Default Re: SAFC... What exactly does it do? (beepy)

MAF ignition tables are typically in RPM X TPS format, and AFC intervention doesn't do much to hinder.

You have to have a very mature understanding of MAF theory to truly understand how/what an AFC does... essentially, the biggest problem (among many) is the volume of intake tract between MAF and engine throw it's reading off... and AFC manipulation only skews the manner in which intake tract volume is already misread. Less dangerous than the AFC hack in a SD system, and you can fiddle around the problem for most streetcar apps if you understand completely how MAF + AFC works.

Here is a very good paper on the inner workings of MAF and SD setups, and thier failings in accurately reading air flow through an engine. Read, digest, ponder, I've had a dialogue with Per Andersson in the recent past, sad to say most of their papers are not available for consumption due to copyright/patent issues.

http://www.fs.isy.liu.se/Publi...A.pdf
Reply
Old Aug 8, 2004 | 12:22 PM
  #7  
beepy's Avatar
Thread Starter
Member
 
Joined: Feb 2004
Posts: 2,451
Likes: 1
From: Pearl City, HI, USA
Default Re: SAFC... What exactly does it do? (mekkis)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">MAF ignition tables are typically in RPM X TPS format, and AFC intervention doesn't do much to hinder.</TD></TR></TABLE>

Excellent.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">You have to have a very mature understanding of MAF theory to truly understand how/what an AFC does... essentially, the biggest problem (among many) is the volume of intake tract between MAF and engine throw it's reading off... and AFC manipulation only skews the manner in which intake tract volume is already misread.</TD></TR></TABLE>

I read the paper you provided me, and it stated only a 5% error in the readings when using the MAF sensor near the air filter. (page 35 showed a 95% confidance interval for measured vs. calculated volumetric efficiency that fit within 6% of the assumption) Furthermore, it is a constant error and can be accounted for, can it not?

Besides that, the paper mostly concerned how the wastegate opening affects the exhaust manifold pressure and its subsequent effects on the remainder of the engine, as well as an experimental setup to better account for such changes. I appreciate the fine reading, but I didn't see much about how an AFC affects a MAF-equipped engine. Regardless, you speak of a condition which occurs in ALL MAF-equipped vehicles, not just ones with AFC's.
Reply
Old Aug 9, 2004 | 05:43 AM
  #8  
mekkis's Avatar
 
Joined: Apr 2004
Posts: 151
Likes: 0
From: I smoke crack rocks
Default Re: SAFC... What exactly does it do? (beepy)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beepy &raquo;</TD></TR><TR><TD CLASS="quote">I read the paper you provided me, and it stated only a 5% error in the readings</TD></TR></TABLE>

Not quite; it stated that MAF error was directly related to the volume of the intake tract - in the case of their test vehicle, it was 5%. Subtle differences.


<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beepy &raquo;</TD></TR><TR><TD CLASS="quote">Furthermore, it is a constant error and can be accounted for, can it not?</TD></TR></TABLE>

AFC is going to skew things slightly. That makes it an inconstant error, eh?

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beepy &raquo;</TD></TR><TR><TD CLASS="quote">Besides that, the paper mostly concerned how the wastegate opening affects the exhaust manifold pressure and its subsequent effects on the remainder of the engine</TD></TR></TABLE>

Fascinating stuff.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by beepy &raquo;</TD></TR><TR><TD CLASS="quote"> I appreciate the fine reading, but I didn't see much about how an AFC affects a MAF-equipped engine. Regardless, you speak of a condition which occurs in ALL MAF-equipped vehicles, not just ones with AFC's.</TD></TR></TABLE>

Your question was a product of lack of MAF theory - the AFC's effect is quite simple. So, I gave you one single solitary decent overview of how MAF systems work. There's definitely a lot more to learn on the subject while you ponder how AFCs work in conjunction with MAFs. Long story short, you're asking a question that requires I type a short novel to answer for you, which I'm not.

Truthfully, I've fiddled maybe four MAF vehicles in my life, and could possibly lead you astray due to my own ignorance. I have some general theory, and very little real world applicability learned by hands on, when it comes to MAF.
Reply
Old Aug 9, 2004 | 06:40 PM
  #9  
quik sol's Avatar
Honda-Tech Member
 
Joined: Aug 2003
Posts: 2,607
Likes: 0
From: Sandusky, Ohio, USA
Default

bookmarking information.
Reply
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
Fnix
Forced Induction
12
May 17, 2012 03:04 PM
Joseph Davis
Forced Induction
154
Feb 7, 2007 12:18 PM
jdm_ef9_b16
All Motor / Naturally Aspirated
6
Jan 24, 2007 03:27 PM
black98gsr-t
Forced Induction
5
Oct 29, 2004 10:23 AM
95gisser
Acura Integra
1
Feb 29, 2004 08:26 PM




All times are GMT -8. The time now is 05:01 PM.