[merc50]

Thats only a guess and I doubt its really that much but its how they wrote it in the MM&FF article for the lightning. This (pic below) is my new 90mm LMAF. It did what I wanted and it made a difference! Dont know about actual HP#'s but it was enough to move my 2+hundred and my passenger 2+hundred pounds to get 4th gear scratches. SOTP was a big WOW! Will have more numbers later but for now I'm just have to work out a detonation issue. Didnt notice it before the addition but its noticible now. I moved the TA from 15deg to 13 deg and now I only get det between 4200 and 5000, Any ideas?? Oh yea I put in a Triax too. Love the shifter. Got a good deal on it, 150.00 and 100 for the MAF.

_________________
President:
http://www.houstonbullittclub.com

<font size=-1>[ This Message was edited by: black sunshine #161 on 2002-03-06 18:09 ]</font>

<font size=-1>[ This Message was edited by: black sunshine #161 on 2002-03-06 18:15 ]</font>

<font size=-1>[ This Message was edited by: black sunshine #161 on 2002-03-06 18:17 ]</font>

 

[SILVERBULLET]

Black Sunshine I sent you a PM.

 

[mjchip]

Black Sunshine,

Unless you are running 42lb injectors or have a chip programmed to compensate for the different transfer function of that MAF you are going to run SUPER lean. In fact, I would almost bet money that your detonation problem is a direct result of this MAF. BE CAREFUL!

Mark Chiappetta

P.S. If you want to calibrate this MAF yourself, you can check out my new product, the MAFterburnerTM which will allow you to get the benefits of greater flow (on a stock bullitt this will be minimal) without having the a/f problems that you are having. You can see it at: www.seanhylandmotorsport.com

Read the FAQ there.



_________________
BLACK #5434 - Steeda Triax & Pullies & T/A, JBA shorties, Magnaflow X w/cats & cat-back, Alum. Driveshaft, 80mm C&L MAF w/K&N, MAFterburnerTM - 254.0rwhp/ 298.1rwtq SAE (CF=0.96)

<font size=-1>[ This Message was edited by: mjchip on 2002-03-06 19:49 ]</font>

 

[DrivingSimulators]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-06 19:48, mjchip wrote:
Black Sunshine,

Unless you are running 42lb injectors or have a chip programmed to compensate for the different transfer function of that MAF you are going to run SUPER lean. In fact, I would almost bet money that your detonation problem is a direct result of this MAF. BE CAREFUL!

Mark Chiappetta

P.S. If you want to calibrate this MAF yourself, you can check out my new product, the MAFterburnerTM which will allow you to get the benefits of greater flow (on a stock bullitt this will be minimal) without having the a/f problems that you are having. You can see it at: http://www.seanhylandmotorsport.com

Read the FAQ there.



_________________
BLACK #5434 - Steeda Triax & Pullies & T/A, JBA shorties, Magnaflow X w/cats & cat-back, Alum. Driveshaft, 80mm C&L MAF w/K&N, MAFterburnerTM - 254.0rwhp/ 298.1rwtq SAE (CF=0.96)

<font size=-1>[ This Message was edited by: mjchip on 2002-03-06 19:49 ]</font>
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Mark,

That's some pretty cool tech. Some questions:

1. Why not interpolate the datapoints with a cubic interpolating spline such as a Catmull-Rom spline? It can easily be implemented in fixed point (if your DSP has no floating point).

2. Have you tested your hardware+software using a USB to serial adapter? (many modern laptops have no serial port (such as my 2805-S402 Toshiba)).

3. Can your product eliminate "tip-in ping" and WOT ping on a stock vehicle?

 

[mjchip]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-07 00:29, DrivingSimulators wrote:
Mark,

That's some pretty cool tech. Some questions:

1. Why not interpolate the datapoints with a cubic interpolating spline such as a Catmull-Rom spline? It can easily be implemented in fixed point (if your DSP has no floating point).

</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Based on my investigations which included lengthy discussions and even tuning sessions with one of the top Modular tuners (I'll give you one guess who that is) I came to the conclusion that any interpolation scheme that was more complex than a linear one (I was originally going to use a fourth-order polynomial) was overkill.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>

2. Have you tested your hardware+software using a USB to serial adapter? (many modern laptops have no serial port (such as my 2805-S402 Toshiba)).

</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

I'm currently working on this and have an iogear USB-serial adapter but am running into trouble with it. To top it off they, iogear, provide no support for developers and claim that their adapter is guaranteed to work only for PDA's, digital cameras, and modems. My software should work and it's only a matter of time before I get it to work with one of these adapters. So, I guess the answer is, YES we intend to support PC's/laptops that only have USB ports.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>

3. Can your product eliminate "tip-in ping" and WOT ping on a stock vehicle?
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

If the predetonation is due to improper a/f mixture then I'd venture to say that it is a very good possibility that this would help.

Mark Chiappetta
Zone-5 Performance Products

 

[DrivingSimulators]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-07 08:10, mjchip wrote:
<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-07 00:29, DrivingSimulators wrote:
Mark,

That's some pretty cool tech. Some questions:

1. Why not interpolate the datapoints with a cubic interpolating spline such as a Catmull-Rom spline? It can easily be implemented in fixed point (if your DSP has no floating point).

</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Based on my investigations which included lengthy discussions and even tuning sessions with one of the top Modular tuners (I'll give you one guess who that is) I came to the conclusion that any interpolation scheme that was more complex than a linear one (I was originally going to use a fourth-order polynomial) was overkill.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>

2. Have you tested your hardware+software using a USB to serial adapter? (many modern laptops have no serial port (such as my 2805-S402 Toshiba)).

</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

I'm currently working on this and have an iogear USB-serial adapter but am running into trouble with it. To top it off they, iogear, provide no support for developers and claim that their adapter is guaranteed to work only for PDA's, digital cameras, and modems. My software should work and it's only a matter of time before I get it to work with one of these adapters. So, I guess the answer is, YES we intend to support PC's/laptops that only have USB ports.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>

3. Can your product eliminate "tip-in ping" and WOT ping on a stock vehicle?
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

If the predetonation is due to improper a/f mixture then I'd venture to say that it is a very good possibility that this would help.

Mark Chiappetta
Zone-5 Performance Products

</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

If you have enough entries in your lookup table, linear interpolation is fine (is that the case for you?).

To enable easier editing of the table, you could still use a cubic (all you need) and then sample the curve to build your lookup table.

I'm a big fan of curves (in all dimensions) :wink:

 

[mjchip]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
If you have enough entries in your lookup table, linear interpolation is fine (is that the case for you?).
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

*You* tell me. Each data table is 256 elements long. Each element is 8-bit signed. Therefore, the resolutions are <1% air-fuel adjustment and 40rpm intervals. In contrast, Ford tables for the MAF transfer function in the PCM are less than 20 entries (exact number varies) and I believe that they also use linear interpolation.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
To enable easier editing of the table, you could still use a cubic (all you need) and then sample the curve to build your lookup table.
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Did you check out the user interface? Editing the tables is done graphically and is pretty simple. I would be hard pressed to make it simpler. Also, we are not directly modifying the MAF transfer function (volts vs. kg/hr) but rather modifying a table that is (% a-f correction vs. RPM). A normalized function with the same general shape of the MAF transfer function is stored in memory and used to translate flow correction to voltage. (this is a fourth order function).

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
I'm a big fan of curves (in all dimensions) :wink:
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Me too!!! :smile:

 

[Mus408]

Geeezzzzz,I have been building and tuning Musclecars for 30 years,and never heard of this lingo????? I'll stick with my timing light,degree wheel and carb jets. Old tech still lives!

 

[DrivingSimulators]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-07 18:42, mjchip wrote:
<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
If you have enough entries in your lookup table, linear interpolation is fine (is that the case for you?).
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

*You* tell me. Each data table is 256 elements long. Each element is 8-bit signed. Therefore, the resolutions are <1% air-fuel adjustment and 40rpm intervals. In contrast, Ford tables for the MAF transfer function in the PCM are less than 20 entries (exact number varies) and I believe that they also use linear interpolation.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
To enable easier editing of the table, you could still use a cubic (all you need) and then sample the curve to build your lookup table.
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Did you check out the user interface? Editing the tables is done graphically and is pretty simple. I would be hard pressed to make it simpler. Also, we are not directly modifying the MAF transfer function (volts vs. kg/hr) but rather modifying a table that is (% a-f correction vs. RPM). A normalized function with the same general shape of the MAF transfer function is stored in memory and used to translate flow correction to voltage. (this is a fourth order function).

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
I'm a big fan of curves (in all dimensions) :wink:
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Me too!!! :smile:

</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

I don't understand the big picture of how all the elements work together, so I can't say if you have a big enough lookup table or enough bits of resolution in the table. If what you are doing gives optimal output as measured on a dyno with wideband 02, etc., then you're good enough :wink: .

As for GUI editing with curves, if you have say 50 sliders, and the end result always looks fairly smooth (slider positions don't have discontinuities), then you could use far less controls, and allow moving of curve control points with tangent handles. This would also allow you to abstract (hide) the internal rep of you algorithm(s). Also, it would allow you to change table length and never have to change the UI (it will then be table resolution independent). Thus, you start out with a straight line, then as you click on the line, a new control point is created. You move it up and down, adjust the tangent curve, and click again precisely where you need to make a change, and edit again. Should be very fast and easy to use...

Are you saying that your system effectively increases the resolution of the adaptability of the MAF sensor to flow conditions? It seems to me the optimal solution is to get the Ford computer's tables optimized as much as possible, or chuck the Ford computer for a programmable aftermarket control system (if they exist). Maybe something off an F1 car :wink: .

Fourth order transfer function (quartic): Why? Perhaps I'm misunderstaning something, but if they are using a small lerp'd table, why would they bother with a quartic elsewhere? Seems like you could get way better results with an interpolating cubic and a lookup table (define just about any smooth function you want, without the wiggle problems of a quartic). Memory is cheap... Am I missing something?

 

[DrivingSimulators]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-07 21:03, Mus408 wrote:
Geeezzzzz,I have been building and tuning Musclecars for 30 years,and never heard of this lingo????? I'll stick with my timing light,degree wheel and carb jets. Old tech still lives!
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

I wish it were still that easy! Swap a carb, try different jet sizes, change accelerator pump cam(s), rotate the distributor, and bingo, a well tuned car. I'm just learning about what must be done to "tune" a modern car. I would not have as much of an interest if the cars came well tuned from the factory (maybe they are and my car has some defective components which cause detonation/hesitation, etc., on my stock Bullitt). It would be cool to be able to diagnose and fix it myself.

<font size=-1>[ This Message was edited by: DrivingSimulators on 2002-03-07 21:48 ]</font>

 

[Mus408]

Yes,to me it is a MUST,to know what is going on with the engine and it's state of tune,ignition,A/F ratio. I use a fairly accurate A/F gauge,on my 85 LX,to fine tune the carb. I believe one of these could be tied in to the O2 sensor,to do the same. I picked up some good HP,using a much modified annular booster Holley carb,I had been working on,for my 85 LX. It took a bunch of carefull tuning of high speed air bleeds to get it just right. It is great to know what is happening under the hood! With my Bullitt,I have had a very slight ping only under full load,in 3rd gear, at WOT, at the track. I will see if the swap to the 3.73 gear will help with the problem.

 

[mjchip]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
I don't understand the big picture of how all the elements work together, so I can't say if you have a big enough lookup table or enough bits of resolution in the table. If what you are doing gives optimal output as measured on a dyno with wideband 02, etc., then you're good enough :wink: .
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Yes, the dyno results that we're getting are totally awesome. Very precise control of air-fuel over the entire rpm range (on many different vehicles). Feedback from a wide-band and a scan-tool on both a chassis dyno and the open road verify our results.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
As for GUI editing with curves, if you have say 50 sliders, and the end result always looks fairly smooth (slider positions don't have discontinuities), then you could use far less controls, and allow moving of curve control points with tangent handles. This would also allow you to abstract (hide) the internal rep of you algorithm(s). Also, it would allow you to change table length and never have to change the UI (it will then be table resolution independent). Thus, you start out with a straight line, then as you click on the line, a new control point is created. You move it up and down, adjust the tangent curve, and click again precisely where you need to make a change, and edit again. Should be very fast and easy to use...
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

OK, I understand now. Sounds like interesting concepts.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
Are you saying that your system effectively increases the resolution of the adaptability of the MAF sensor to flow conditions? It seems to me the optimal solution is to get the Ford computer's tables optimized as much as possible, or chuck the Ford computer for a programmable aftermarket control system (if they exist). Maybe something off an F1 car :wink: .
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

You can't really do better resolution-wise with the Ford tables in the picture. They act as sort of a crude filter. Very few bins.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
Fourth order transfer function (quartic): Why?
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Physical properties of the sensor itself. In a nutshell, a platinum resistor is heated and the rate of cooling is measured and compared with another (unheated) platinum resistor. There is effectively a square-law relationship between voltage and power and another square-law relationship between electrical power put into the resistor and mass-air flow that it takes to cool it. Hence the fourth order relationship. For more information, check out www.pro-flow.com and read their tech section. It's pretty comprehensive.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>

Perhaps I'm misunderstaning something, but if they are using a small lerp'd table, why would they bother with a quartic elsewhere? Seems like you could get way better results with an interpolating cubic and a lookup table (define just about any smooth function you want, without the wiggle problems of a quartic). Memory is cheap... Am I missing something?
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

See above. Thanks for the interesting conversation!

Mark Chiappetta
Zone-5 Performance Products

 

[DrivingSimulators]

Mark,

Is this the equation you are referring to?

Q = heat or power
E = voltage

Q^2 = flow and E^2 = Q
therefore : E^4 = flow

If so, that's actually quite a bit simpler than using an interpolating cubic (I was expecting coefficients (and associated tuning) and more terms: ax^4 + bx^3 + cx^2 + dx + e, etc.). Computing the cubic takes more MACs (multiply-accumulates), but tuning is trivial.

I also read about them (Pro-M) smoothing the signal flow to get better performance. While this may be true with full constant air velocity, on delta velocity events (stabbing the throttle), the associated lag from the (effectively) low-pass filter will allow the car to go lean briefly (maybe that's what's happening with our stock Bullitt's with tip-in ping: overfiltering and associated lag).

What DSP/microcontroller do they use in the Ford ECU? Motorola? Intel? Using something like a Kalman predictive filter (requires more cpu cycles to compute) allows the smoothing of input data while predicting the output, minimizing lag. A neural network can also produce similar results. If you are curious, do a www.google.com search on Kalman Filter. Cool (and useful) information.

 

[merc50]

Im lost

 

[mjchip]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-08 00:38, DrivingSimulators wrote:
Mark,

Is this the equation you are referring to?

Q = heat or power
E = voltage

Q^2 = flow and E^2 = Q
therefore : E^4 = flow
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

This is an oversimplification of the actual function. In reality it is more like you describe below (there are more coefficients). I compiled a database of existing MAF curves and fit functions to all of the data to come up with single function that had the closest shape. I then reduced and normalized the function and then tuned the coefficients to get a final transfer function that was most representative of my existing data set.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
If so, that's actually quite a bit simpler than using an interpolating cubic (I was expecting coefficients (and associated tuning) and more terms: ax^4 + bx^3 + cx^2 + dx + e, etc.). Computing the cubic takes more MACs (multiply-accumulates), but tuning is trivial.
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

The utility of this function in the MAFterburner is only to allow the user to make intuitive air-fuel (really air flow) corrections and have them properly translated into voltage corrections to the PCM.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
I also read about them (Pro-M) smoothing the signal flow to get better performance. While this may be true with full constant air velocity, on delta velocity events (stabbing the throttle), the associated lag from the (effectively) low-pass filter will allow the car to go lean briefly (maybe that's what's happening with our stock Bullitt's with tip-in ping: overfiltering and associated lag).
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

I'm in agreement with you. That is why the MAFterburner's (analog and digital) filtering parameters were calculated very carefully in order to not affect the possible fast (with respect to the system) transient response of the system.

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
What DSP/microcontroller do they use in the Ford ECU? Motorola? Intel? Using something like a Kalman predictive filter (requires more cpu cycles to compute) allows the smoothing of input data while predicting the output, minimizing lag. A neural network can also produce similar results. If you are curious, do a http://www.google.com search on Kalman Filter. Cool (and useful) information.
</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

EEC-IV = 8061 (intel)
EEC-V = 8065

From the intel website:

"Intel's history in engine control electronics is lengthy, from the first 1983 Ford EEC-IV based on the 8061 microcontroller to today's modern 8065 chip set, also known as the EEC-V. Today, Ford is still using the 8065 in many of their new vehicles produced through the start of the next decade. The 16-bit architectures, such as the 8065 and MCS(R) 96 controllers, have been widely accepted in Europe and the U.S in ABS, engine control, and networking applications. From the introduction of the 8061, Intel has continued to provide more innovative and highly integrated microcontrollers in order to help their customers with evolving powertrain applications. "

Mark Chiappetta
Zone-5 Performance Products

 

[mjchip]

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font size=-1>Quote:</font><HR></TD></TR><TR><TD><FONT SIZE=-1><BLOCKQUOTE>
On 2002-03-08 07:54, Black Sunshine #161 wrote:
Im lost

</BLOCKQUOTE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

Sorry! We kind of got off of your subject. I guess that what happens when two techies bump brains (cerebral contusion). :smile:

Anyway, I gave you my recommendations earlier in the thread. If you want to use the 90mm LMAF, then you are going to have to make adjustments to your system. Of course, you can do this with the MAFterburner (shameless plug) but you can also do this with a chip. You are most likely running VERY lean and you can cause serious engine damage to occur should you push the car to the point that the cylinder temps go up too high.

Mark Chiappetta
Zone-5 Performance Products