|
|
|
|
|
|
Registered
Join Date: Feb 2003
Location: chicago
Posts: 1,077
|
stupid fuel injection idea
If a carburator basically works by metering fuel depending on volume of airflow through the venturi, why can't a fuel injector with no control unit or sensors meter the fuel based on throttle position? i know without air temp, O2, MAP etc.. sensors the efficentcy in different altitudes and temps would not be the best but i would think it would be better than a carb. the fuel injected at idle to open throttle is not linear but that could be adjusted with a series of potentiometers at different throttle positions. it seems that in studing mfi and electronic efi systems they are really complex and carbs really don't use most of the variables but still work - sometimes very well. any thoughts?
-matt |
||
|
|
|
|
Registered
Join Date: Jan 2007
Posts: 11,758
|
Quote:
|
||
|
|
|
|
Registered
Join Date: May 2004
Location: Encinitas (San Diego CA)
Posts: 4,495
|
Matt-
carbs have various circuits and are not as simple as one would think: idle circuit accelerator pump / circuit cruise circuit jets for WOT, etc., etc. most carbs also have a choke for cold idle. Yes, they don't account for barometric pressure changes and respond very simply to temperature changes. I don't think they are so simple. Automotive carburetors aren't like the one on your lawn mower. Even that has a choke and a governor of some sort to keep speeds appropriate. My understanding is you can build a simple EFI system that doesn't utilize tons of sensors. I added a holley system to a V-8 in a boat years ago. That was throttle body injection (bold on in place of carburetor) and had VERY few sensors as inputs. The throttle body had a potentiometer (throttle position sensor), no O2 sensor and a temp sensor. In the CPU box, it might have had a pressure sensor. The user "tuned" rich and lean at idle, cruise and wide open. EXTREMELY SIMPLE. That boat ran *much* better with the minmalist EFI than vacuum secondary carburetor. Doug
__________________
1971 RSR - interpretation Last edited by DW SD; 02-04-2008 at 05:15 PM.. |
||
|
|
|
|
Registered
Join Date: May 2003
Posts: 3,346
|
You could make one that used throttle position and RPM. This would be exactly the equivalent of MFI. Your electronic map would be the equivalent of the spacer cam in the MFI pump. Of course even MFI uses a temperature and pressure correction.
-Andy
__________________
72 Carrera RS replica, Spec 911 racer |
||
|
|
|
|
Diss Member
Join Date: Jul 2002
Location: SC - (Aiken in the 'other' SC)
Posts: 5,022
|
I have seen a number of really smart people try to "reinvent the wheel" in relation to fuel injection. I have seen some really well thought out schemes go down in flames. If you look at a good fuel map for an injection system you will find that it has hills and valleys that make no sense at all. (until you run it for real) It is just too complicated. (Read read the section about the fuel injection for the turbo 917 in Mark Donahue's "Unfair Advantage". Really smart engineers at Porsche can miss the point too!)
If you truly want something really simplified you have to reduce a lot of variables. Take a look at an old constant flow injection system for a drag car. It only has 2 places where it runs. Idle and full throttle. You run it real rich to cover the lean parts of the range but fouling a plug and fuel consumption isn't something they worry about. They basically threw everything else away for simplicity. It also throws away power in various spots across the RPM range but it is simple and works ok (not great) with an automatic transmission (less RPM change). Basically the gist of this is you can't get a silk purse out of that sows ear. You need to pay close attention to what an aircooled sportscar engine is doing. That requires a number of sensors or other compensation devices.
__________________
- "Speed kills! How fast do you want to go?" - anon. - "If More is better then Too Much is just right!!!" - Mad Mac Durgeloh -- Wayne - 87 Carrera coupe -> The pooch. |
||
|
|
|
|
Diss Member
Join Date: Jul 2002
Location: SC - (Aiken in the 'other' SC)
Posts: 5,022
|
Quote:
As mentioned before there are a whole bunch of additional bits in the carb to adjust the mixture across the operating range. In the race carbs I used to build there are 13 specific mechanical systems/devices that could be adjusted besides what is in the venturis. As complex as people think EFI is, it is actually a lot simpler then all the junk that goes into a good working carb.
__________________
- "Speed kills! How fast do you want to go?" - anon. - "If More is better then Too Much is just right!!!" - Mad Mac Durgeloh -- Wayne - 87 Carrera coupe -> The pooch. |
||
|
|
|
|
|
Registered
Join Date: Feb 2003
Location: chicago
Posts: 1,077
|
thanks guys, i knew it was more complex than my simplified idea. i'm well aware that carbs and fuel injection require plenty of tuning to work properly in varied conditions. by taking away sensors or carb circuit adjustments in my simple efi it likely would never be able to be adjusted enough to work at all, and even if i could get it to run it would only work on my engine on that day. it just seems that with all the different possible types of sensors on an engine many of them are redundant or only used in special situations. i originally was thinking that a system could run closed loop with just an O2 sensor to adjust the AF ratio, however i think the O2 sensor is limited in its adjustment and slow to react. oh well.
|
||
|
|
|