[Diy_efi] Making a fake O2

Wed Jun 1 03:28:19 UTC 2005

Dan,
    Yes, that is exactly where I am.  The stock computers mapping is in no
way set up for the engine that is now connected to it.  Everything you
said is right on.  We have a mismatch.  The stock ecm and its mapping
isn't going to work.
     I have kept reading hoping the bickering would stop.  I agree with
both sides of the debate.  GM computers are terrible when their O2s
fail.  And run so rich I'm couldn't understand why anyone would drive
it.  Now the ECM I'm running in my Landcruiser is a GM with TBI setup
out of 1989 C series 5.7l GM truck.  And it will actually use data
gathered during closed loop to effect WOT.  And it can do damage. 
Now, the Honda/Acura setup we're working on right now will run so
rich with the o2 disconnected it will barely idle.  Now these are two
cars.  I have also seen setups where you could remove half the
sensors on the car, and the computer will keep it running like
nothing ever changed.  And it won't set a light.  Dodge Caravan
OBDII.  And the emissions will be perfect.  I believe it is smart
enough to use other data to make a good guess (kinda like
Speed-density) at what the sensor should be reading.  You would be
hard pressed to get one of these engines to hurt itself.  And if
there was misfire that could cause damage it will kill cylinders to
wake the driver up (Thats if they don't see the flashing MIL lamp)
and protect the engine.
     There are many computers out there, and many different approaches
used.  Now I have never personally seen a motor go boom because of a
bad o2 sensor, but I have seen a lot of damage done.  The o2 goes
bad.  The cat plugs up, the exhuast is restricted, and the temps at
the engine exhuast go steadily upward.  Exhuast valves damaged, seals
damaged, exhuast manifolds damaged, cylinder heads damaged.  etc,.. 
I believe that it all depends on the computer you are using.  There
are computers that go both ways....  I also don't see the problem
with an open loop car.  The Carburetor analogy is a good one.  I've
run vehicles 100,000 of miles with a carburetor without any problem. 
And I can think of one setup or two aftermarket setups that don't use
o2 sensors unless you add them.  Both Eldebrock, and Holley have
aftermarket EFI setups that don't intially use an O2.  I think Accel
has one two.  And these things have been used in many cars, and
trucks without any problems.
    And this all started with a simple question about faking out the o2...

Marcello


> I've been watching this thread over the last few days.  People are
> debating
> the safety of the stock maps with 02 faulty but no one has mentioned that
> the engine in question is no longer stock.  Marcello mentioned that his
> friend "only made a few changes" (eye-roll) meaning to me that probably
> the
> friend put together a parts combination far from stock.  Most likely
> higher
> compression and a more aggressive cam.  Those two possibilities alone
> probably would make the stock maps irrelevant or at least "not safe".
> Considering how much $$ most people typically throw at a motor when they
> "make a few changes" I would proceed slowly and work to develop new maps.
> I
> think Marcello realizes this and hopes his friend will keep his foot out
> of
> it until he gets home to work at the problem logically.
>
> My experience is with Ford EEC's.  I know when I "made a few changes" the
> EEC was totally confused because my parts changes had taken the range of
> operation well beyond the stock limits of compensation.
>
> Good discussion.
>
> Dan Nicoson
>
>
>> -----Original Message-----
>> From: diy_efi-bounces at diy-efi.org [mailto:diy_efi-bounces at diy-efi.org]On
>> Behalf Of Steve Ravet
>> Sent: Tuesday, May 31, 2005 12:36 PM
>> To: diy_efi at diy-efi.org
>> Subject: RE: [Diy_efi] Making a fake O2
>>
>>
>> I was going to refer to the EFI testbench that Bruce built.  My
>> recollection is that he used a 555 type circuit to generate an O2
>> signal.  There's not really any mention of it in the article, but I'm
>> sure the details are in the archive.  I think for some computers this
>> worked OK, for others that expected to see more correlation between
>> pulsewidth and O2, it didn't.  Anyway, go to the gmecm page, projects,
>> then ECM testbench for the ascii version of the article.  The MS word
>> version is probably gone for good.
>>
>> --steve
>>
>>
>> ________________________________
>>
>> 	From: diy_efi-bounces at diy-efi.org
>> [mailto:diy_efi-bounces at diy-efi.org] On Behalf Of Bret Levandowski
>> 	Sent: Sunday, May 29, 2005 11:16 AM
>> 	To: diy_efi at diy-efi.org
>> 	Subject: Re: [Diy_efi] Making a fake O2
>>
>>
>> 	Actually, it is not 'impossible to fool an ECU with a constant
>> or switching voltage source'. There are two companies I know of (in the
>> states) that have developed a plug-in replacement for the O2 sensor on
>> numerous vehicles (mostly OBD II) that generates a 'proper' signal.
>> These are used in instances where inj size was increased or a turbo or
>> screw-charger was installed. Some also did require a flash of the ecu.
>>
>> _______________________________________________
>> Diy_efi mailing list
>> Diy_efi at diy-efi.org
>> http://lists.diy-efi.org/mailman/listinfo/diy_efi
>>
>
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> http://lists.diy-efi.org/mailman/listinfo/diy_efi
>