280ZX ECU in '78 BMW

WERNER_HAUSSMANN at HP-Loveland-om2.om.hp.com WERNER_HAUSSMANN at HP-Loveland-om2.om.hp.com
Tue Jun 25 18:32:54 GMT 1996


* Why doesn't the closed loop system compensate for any mismatch in the 
EFI system?

There are two problems. 
1. The range for the correction is only so much, maybe 10% or so.  I really 
don't know.  But there are limits for the amount of correction the system 
can make, especially an anolog system like the L-Jetronic.

2. the other problem is that the closed loop is only operational when the
   car is cruising and hot.  During acceleration or high power cruising
   like going up a hill the closed loop system is not operating. The 
   ECU (electroninc control unit, computer) will read the Air meter etc and
   directly control the fuel injection. Under these conditions you want 
   the Air meter, injector and ECU to match



* Other Z cars do a lot of mods with injectors and air meters

You can make all kinds of mods if you can measure the results.  You need as 
a minimum a CO meter and and a fuel pressure meter to see what the mods do.
Increasing fuel pressure or adding bigger injectors will make the mixture 
more rich when running open loop.  This will give a little more power, let 
you run with a little more advance or more compression etc. You can see 
there are so many variables.


* MW is 180, Z car is 135-150

Using all the Z stuff in the BMW may cause you to run lean at max power. 
This is because the air meter 'might' run out of range at about 160hp. This 
is important if you use the all the matched components from one car. You 
can of course use larger injectors, or a different air meter.  The problem 
is which ones?  You can calculate the injectors, but you will still need to 
trim the result.


* The existing system is open-loop and runs rich

See first item above.  Very few systems run closed loop when burning valves 
etc is a concern.  I don't know of any system that truly runs closed loop 
when under max power. Some digital systems calculate the mixture during 
cruise closed loop running, and then apply enrichment for max power. 

I think your best bet is to back off the enrichment (change the tune on you 
existing car), run a custom system like the TechII, or get a system from a 
car witht the same injectors and engine size.  Otherwise it is too 
difficult unless you have lots of equipment.


Werner



  


 



______________________________ Reply Separator _________________________________
Subject: Re: 280ZX ECU in '78 BMW
Author:  Non-HP-owner-diy-efi (owner-diy_efi at coulomb.eng.ohio-state.edu) at 
HP-ColSprings,shargw2
Date:    6/25/96 6:51 AM


> Item Subject: 280ZX ECU in '78 BMW 
> 
> To get the correct calibration from your 'new' L-Jetronic, you will 
> need to get the AFM, the injectors and the ECU to match. 

OK, I'm kind of a "why" guy, so why?  If there is a closed-loop 
system, the assumption is that as long as I can get the mixture in 
the ballpark with it running open-loop (by checking the plugs, 
adjusting the AFM, etc), then once I add the O2 sensor to the mix it 
should take care of any fuel mix problems on an instantaneous (sort 
of) basis, right?

> Unless you
> know that they are calibrated the same, that means you need to get 
> all of them from the same car. 

Well, now I know it is a fairly common mod on Z cars to upgrade to 
bigger injectors or to add a rising rate pressure regulator.  Also, 
upgrades to bigger AFMs are fairly common.  Do the guys doing these 
mods just imagine that their cars are running better (I'm not being 
sarcastic, many people do imagine that mods improve things evven if 
they don't), or is there some way the FI compensates for these types 
of mods?

> Also the HP from the donor car needs
> to about the same as the receiving car, but no less. 

Well, in full emissions trim, the hp of the 280ZX NA that this ECU 
came from is about 135-150, and stock my BMW is about 180.

> 
> If you are lucky the injectors will be the same. The Japanese AFM is 
> similar to the Bosch (licensed) but it may not be the same in
> calibration.
> 
> 
> You might have better luck by changing the calibration of the AFM or 
> ECU on the existing car.  I assume you don't have a O2 sensor?

No.  The existing system is open-loop and runs really rich.  Rather 
than play the yo-yo mix game, and risk burning valves, etc., in the 
process, I thought I'd go straight to a closed-loop system to 
eliminate variables.  I am assuming that the ECU will take a reading 
from the O2 sensor, the AFM, and the throttle position sensor, and 
then calculate a injector pulse width.  It should take some very 
short time to get good feedback from the O2 sensor, so the ECU can 
adjust the pulse width accordingly.  If this is in fact the way it 
will work, then the injectors, AFM, etc., all should be completely 
irrelevant as long as they are operated in the same manner.  IOW, if 
you put injectors from one car, an ECU from another and an AFM and a 
throttle from yet another, then it should all still work, regardless 
of the engine, tuning, etc.

Am I wrong about this?  I know that the old L-Jet computer in my BMW 
calculates the pulse width above about 3000 rpm based on some 
circuitry completely independent of the AFM or any other sensor.  So, 
without modifying the analog ECU, there is no way to lean out the 
mixture in the higher rpms except to increase airflow.  Really, this 
process is not adaptive at all.  The digital ECU should be responsive 
over the wide band of rpms regardless of the state of tune, air 
intake flow, or any other variable, as long as everything is within 
some required spec.

I know the injectors are bigger in the BMW than they are in a 280ZX, 
and so is the throttle body and AFM (and, coincedentally, the engine 
:).  As far as the suggestion to just ditch the thermal reactors and 
adjust the car to run lean, the main problem with this is that for 
me, this is very expensive.  Old manifolds are not cheap.  I'd havve 
to have a headpipe made and installed at a muffler shop, and swap out 
all the parts, etc., not worth it, especially if I can diisable the 
hot-function of the thermal reactors by denying them air and unburnt 
fuel.

Has anyone actually tries something like this?

Thanks.


_____________________________________________________________________________ 
Josh Karnes                                                 joshk at tanisys.com 
Renaissance Man                        http://www.tanisys.com/~joshk/home.htm 
Tanisys Technology                                     http://www.tanisys.com 
Austin, Texas                      '78 BMW 530i | '72 Datsun 240Z | IZCC #308 
_____________________________________________________________________________
           *** opinions expressed herein are MINE, ALL MINE!! ***



More information about the Diy_efi mailing list