FI performance
-
- Posts: 1278
- Joined: Sun Aug 24, 2008 9:20 am
- Your car is a: 1978 Spider [1979 2 ltr engine]
- Location: Aiken, SC
FI performance
The conventional wisdom seems to be that we are limited as to how much we can increase the performance of the FI cars since the Bosch L-tronic ECU isn't programmable. Sooner or later you flow to much air for the injectors to keep up with and you go lean and bad things happen. To overcome this you need a programmable ECU and cost rapidly goes up (although the DIY units are cheaper). In addition, I don't know of any aftermarket ECU supplier currently supplying a unit with maps pre-programmed for our cars so there is a lot of tuning / mapping to do if you go that route. BUT:
The amount of the fuel from the injectors is a three variable formula: pulse duration, injector size, and fuel rail pressure. Even if we assume a fixed pulse duration, if we increase either injector size or fuel pressure that same pulse will provide more fuel. Can't we solve the lean out problem by simply putting in larger injectors and / or raising fuel pressure?
Or is my logic still flawed in that the ECU pulse duration is still controlled by air flow as sensed by the AFM and you would get more fuel for a given air flow BUT still have a problem at higher air flows? After all the leaning is in theory because we increased the efficiency of the engine with cams / headers / porting etc. so it will flow more air. The hardwired "map" in the ECU still only varies pulse duration with sensed air flow.
BUT, we can also vary the spring pressure in the AMF to vary the flapper deflection for a given air flow SO in theory you could use larger injectors and / or increased fuel pressure while STIFFENING the AMF spring to retard the pulse duration such that you ended up with the correct stochimetric ratio across the desired rev band? Maybe? I know it would take a lot of trial and error and you'd need at a minimum a wide band O2 sensor to get it set up. Thoughts?
The amount of the fuel from the injectors is a three variable formula: pulse duration, injector size, and fuel rail pressure. Even if we assume a fixed pulse duration, if we increase either injector size or fuel pressure that same pulse will provide more fuel. Can't we solve the lean out problem by simply putting in larger injectors and / or raising fuel pressure?
Or is my logic still flawed in that the ECU pulse duration is still controlled by air flow as sensed by the AFM and you would get more fuel for a given air flow BUT still have a problem at higher air flows? After all the leaning is in theory because we increased the efficiency of the engine with cams / headers / porting etc. so it will flow more air. The hardwired "map" in the ECU still only varies pulse duration with sensed air flow.
BUT, we can also vary the spring pressure in the AMF to vary the flapper deflection for a given air flow SO in theory you could use larger injectors and / or increased fuel pressure while STIFFENING the AMF spring to retard the pulse duration such that you ended up with the correct stochimetric ratio across the desired rev band? Maybe? I know it would take a lot of trial and error and you'd need at a minimum a wide band O2 sensor to get it set up. Thoughts?
Jeff Klein, Aiken, SC
1980 FI Spider, Veridian with Tan (sold about a year ago), in the market for another project
1989 Spider, sold
2008 Mercedes SL65
2008 S600 Mercedes V12
1980 FI Spider, Veridian with Tan (sold about a year ago), in the market for another project
1989 Spider, sold
2008 Mercedes SL65
2008 S600 Mercedes V12
-
- Posts: 1278
- Joined: Sun Aug 24, 2008 9:20 am
- Your car is a: 1978 Spider [1979 2 ltr engine]
- Location: Aiken, SC
Re: FI performance
mark, have you ever mapped O2 / air fuel mixture with a set of your FI cams and a header to see where it puts you with a stock ECU and injectors?
Jeff Klein, Aiken, SC
1980 FI Spider, Veridian with Tan (sold about a year ago), in the market for another project
1989 Spider, sold
2008 Mercedes SL65
2008 S600 Mercedes V12
1980 FI Spider, Veridian with Tan (sold about a year ago), in the market for another project
1989 Spider, sold
2008 Mercedes SL65
2008 S600 Mercedes V12
Re: FI performance
You've touched on a couple of different issues with the stock fi. Performance cams are an issue at lower rpms because of overlap. At lower rpms you get dilution in the cylinders from exhaust gasses since both valves are open at the same time. And it's worse if you have an exhaust system with back pressure since the dilution in the cylinders is worse. That's why we developed the FI cams with reduced overlap. You end up sacrificing high end power to get some driveability at lower speeds. But since the stock fi doesn't do much above 6k rpms, it works out so the big gains are in the midrange, where it's really usable for street driving.
Adding a larger afm is tricky, too large and there isn't enough air flow to move the flap properly. I'd advise against changing spring pressure in the afm; it won't accomplish what you're thinking of. We've tweaked many afms' to provide a slightly richer mixture, but once the flap is wide open it ends up where the stock setting was. Again, it's low and midrange gains there. A slightly larger afm and higher flow injectors would help. There are several cars out there that use injectors that will interchange that have much higher flow rates, so it's an easy upgrade.
It comes back to measuring merely air volume to base your injection on. Systems now measure air density for the most part. A more modern system could be adapted, but the cost means it will be aimed at a small percentage of the owners.
Adding a larger afm is tricky, too large and there isn't enough air flow to move the flap properly. I'd advise against changing spring pressure in the afm; it won't accomplish what you're thinking of. We've tweaked many afms' to provide a slightly richer mixture, but once the flap is wide open it ends up where the stock setting was. Again, it's low and midrange gains there. A slightly larger afm and higher flow injectors would help. There are several cars out there that use injectors that will interchange that have much higher flow rates, so it's an easy upgrade.
It comes back to measuring merely air volume to base your injection on. Systems now measure air density for the most part. A more modern system could be adapted, but the cost means it will be aimed at a small percentage of the owners.
Re: FI performance
i bet there are different ways to trick the bosch ecu... if you put potentiometers on certain sensors, you could trick the system into giving it more fuel. a good try might be the coolant sensor.
Re: FI performance
it still maxes out at about 6k and can't flow any more. You gotta figure out how to get more air and fuel; larger afm, throttle body, ported intake, larger injectors
-
- Posts: 1833
- Joined: Wed Mar 24, 2010 10:45 pm
- Your car is a: 1975 Spider
- Location: clermont fl
Re: FI performance
ive played with some toyota/nissan afm's with pretty good results
Automotive Service Technology Instructor (34 year Fiat mechanic)
75 spider , 6 Lancia Scorpions, 2018 Abarth Spider, 500X wifes, 500L 3 82 Zagatos. 82 spider 34k original miles, 83 pininfarina, 8 fiat spider parts cars
son has 78 spider
75 spider , 6 Lancia Scorpions, 2018 Abarth Spider, 500X wifes, 500L 3 82 Zagatos. 82 spider 34k original miles, 83 pininfarina, 8 fiat spider parts cars
son has 78 spider
Re: FI performance
yes, there are a couple of afms out there that provide an improvement, but still have limitations
Re: FI performance
haha Mark im guessing youre a carburetor guy?
you must see the light my friend!
you must see the light my friend!
-
- Posts: 1278
- Joined: Sun Aug 24, 2008 9:20 am
- Your car is a: 1978 Spider [1979 2 ltr engine]
- Location: Aiken, SC
Re: FI performance
Shark,ClarkTheShark wrote:haha Mark im guessing youre a carburetor guy? you must see the light my friend!
Mark can defend himself but I am quite certain he is very well versed on the FI system. he simply recognizes the limitations of a first generation Bosch system that operates on minimal inputs and has all hard components in the ECU.
Jeff Klein, Aiken, SC
1980 FI Spider, Veridian with Tan (sold about a year ago), in the market for another project
1989 Spider, sold
2008 Mercedes SL65
2008 S600 Mercedes V12
1980 FI Spider, Veridian with Tan (sold about a year ago), in the market for another project
1989 Spider, sold
2008 Mercedes SL65
2008 S600 Mercedes V12
Re: FI performance
If you search this site, you'll find several posts I've made stating that no carburetor can ever work as efficiently as FI. Both systems exist in my world so I deal with both. The L-jet has limitations, just as carbs do. I believe that redesigning the FI system is not something most owners want to deal with so we've developed some methods of improving performance within certain parameters
Re: FI performance
i was not implying mark doesn't know his stuff.zachmac wrote:Shark,ClarkTheShark wrote:haha Mark im guessing youre a carburetor guy? you must see the light my friend!
Mark can defend himself but I am quite certain he is very well versed on the FI system. he simply recognizes the limitations of a first generation Bosch system that operates on minimal inputs and has all hard components in the ECU.
But anyways to the guy that started this thread, you should investigate aftermarket FI systems. Megasquirt for me has been a very valuable learning experience. It brings gobs of technology into our cars. The newest version of MS allows for sequential MPFI... something that some new cars don't even have. for about 5 or 6 hundred dollars you can get all the parts you need for a standalone setup. This includes complete fuel and spark control. granted its not a bolt-on "voila" unit, but it certainly is worth it once its completed. I converted my car from carburetors so your conversion should be much simpler. It actually would be possible to just swap computers with megasquirt. If you do enough research you can actually place your megasquirt inside your L-jet box and jumper the old connections to your new computer! This has been done with other cars but i don't think anyone has ever done it with the L-jet. And the cool factor of being able to tune your car from a laptop is awesome. definitely something to investigate if you're into fuel injection.
- kmead
- Posts: 1069
- Joined: Sat Aug 02, 2008 9:24 pm
- Your car is a: 1969 850 SC 1970 124 SC 85 X19
- Location: Grand Rapids, MI
Re: FI performance
"It actually would be possible to just swap computers with megasquirt. If you do enough research you can actually place your megasquirt inside your L-jet box and jumper the old connections to your new computer! This has been done with other cars but i don't think anyone has ever done it with the L-jet."
This has been done on quite a few X1/9s details can be found over on the Xweb forum: http://www.xwebforums.org/
You will likely need to search the older forums to find it as they changed web platforms last year.
This has been done on quite a few X1/9s details can be found over on the Xweb forum: http://www.xwebforums.org/
You will likely need to search the older forums to find it as they changed web platforms last year.
Karl
1969 Fiat 850 Sports Coupe
1970 Fiat 124 Sports Coupe
1985 Bertone X1/9
1969 Fiat 850 Sports Coupe
1970 Fiat 124 Sports Coupe
1985 Bertone X1/9
Re: FI performance
ahh here we go... i was wrong. just like this guy said... if i had it all over to do again, This is what i would do.
http://www.alfagtv6.com/MegaSquirt/Megasquirt.htm
http://www.alfagtv6.com/MegaSquirt/Megasquirt.htm
-
- Posts: 141
- Joined: Tue Mar 24, 2009 6:52 pm
- Your car is a: 1976 124
Re: FI performance
pry off the black cover on the air flow meter and take about 3 to 4 notches of spring tension out , lessen the tension on the sensor flap this will give you more fuel, i have done this on many afm meter cars all with good results. don't forget to replace the cap and mark the origional notch wheel position before moving it