[Diy_efi] Re: Minimalist micro based EFI ?

[Mike]

At 12:04 AM 3/12/03 +0800, you wrote:

<cough> On further thought, given most basic 8 bit ecus offer
10 bits or more of a/d then its reasonable to expect any
mapped AFM signal at higher loads to provide a coarse
setting with the O2 used as the final trim might well be
readily achievable,

In any case the scenario you paint where one has heavy overfueling
or conversely very lean operation is one experienced during
tuning only not normal operation - ie. given a basic 10 bits for MAF
would be an acceptable numeric and practical (cheap) minimum
with trim suported by Lambda and only after a tuning run, assuming
of course the fuel timing figure isnt too granular but, this is
not that directly related to MAF in a tuning operation...

Yes I should have discussed the issue of dynamic range though
(also) in any case there is no need to convert log to linear
to map - the MAF to fuel table might as well have log entries
on a one to one basis with each bit of measured MAF,

rgds

mike


>Yes true :)
>
>Well articulated,
>
><sigh>
>
>rgds
>
>mike
>
>
>
>At 09:08 AM 3/11/03 -0700, you wrote:
>>At 11:29 PM 3/11/03, Mike wrote:
>>
>>The basic control theory element you are skirting around here is "dynamic
>>range".
>>
>>The ratio between an engine's minimum and maximum output power is a pretty
>>fair definition of the dynamic range which ANY fuel delivery system must
>>cover.
>>
>>Next question is, how much granularity in fuel flow control rate can the
>>engine tolerate without getting into either emission problems (an
>>artificial constraint), excessive fuel consumption, including plug fouling
>>and/or bore washing (valid concerns) , or detonation (a very real and
>>immediately expensive concern).
>>
>>With a boosted engine, the engine's dynamic range is generally wider, and
>>the higher the boost level goes, the wider the dynamic range gets--so more
>>precision (more bits) MAY well be necessary in measurement of inputs
>>critical to controlling the fuel flow rate fpr boosted engines without
>>getting into excessive fuel control granularity.
>>
>>In fact--the need for more precision in fuel metering may well increase
>>_geometrically_ with boost, since boosted engines are so much more
>>sensitive to detonation than NA engines are !
>>
>>Greg
>>
>>>Yes Mos, but not from the perspective of a design decision to
>>>run an engine over a certain power come boost range...
>>>
>>>ie. A Gen III 5.7L V8 uses a MAF sensor 5v max output, 300Kw
>>>
>>>A RB30 N/A puts out 120 Kw and uses a different MAF sensor also
>>>with a max 5v output.
>>>
>>>Case 1 above at say 100Kw flow has less MAF voltage than the
>>>case 2 with 100Kw flow as its a sensor sizing throat/voltage issue.
>>>
>>>So clearly the sensor is chosen for the range and the a/d resolution
>>>is chosen under some other metric, other than the difference in
>>>vintage between the 2 samples above, is there a compelling reason
>>>the 5.7L V8 *needs* higher resolution if emissions is not an issue,
>>>
>>>rgds
>>>
>>>mike
>>>
>>>
>>>
>>>At 02:06 AM 3/12/03 +1100, you wrote:
>>>>On Tue, 11 Mar 2003, Mike wrote:
>>>>
>>>>> At 01:38 AM 3/12/03 +1100, you wrote:
>>>>> >> <cough> I dont see how boost has anything to do with the number of
>bits.
>>>>>
>>>>> So now you are saying air flow voltage is related to boost pressure
>>>>> rather intimately ?
>>>>
>>>>There's no standard (that I was referring to). I was referring to MAP and
>>>>atmo - hence 5V/1bar (yes, very generally). As another poster stated,
>>>>there's 5V/3bar (and also 5V/2bar).
>>>>
>>>>It doesn't matter what you're measuring, but if your range increases, your
>>>>resolution will decrease if you use the same number of bits, period. Your
>>>>10 bits will not give you the same resolution with 2 bar as they will with
>>>>1 bar.
>>>>
>>>>I was in no way saying that it was insufficient - merely stating that
>>>>boost *can* affect the number of required bits - so yes, boost does have a
>>>>lot to do with the number of bits.
>>>>
>>>>Air flow voltage is related to air flow (obviously), and air flow is very
>>>>intimately related to boost pressure.
>>>>
>>>>Mos.
>>>>
>>>>--
>>>>84 AE86, 90 ST185GrpA, 91 MX83Gr, Sydney, Oz.
>>>>
>>>>
>>>>
>>>>_______________________________________________
>>>>Diy_efi mailing list
>>>>Diy_efi at diy-efi.org
>>>>http://www.diy-efi.org/mailman/listinfo/diy_efi
>>>>
>>>>
>>>
>>>_______________________________________________
>>>Diy_efi mailing list
>>>Diy_efi at diy-efi.org
>>>http://www.diy-efi.org/mailman/listinfo/diy_efi
>>
>>
>>
>>_______________________________________________
>>Diy_efi mailing list
>>Diy_efi at diy-efi.org
>>http://www.diy-efi.org/mailman/listinfo/diy_efi
>>
>>
>
>_______________________________________________
>Diy_efi mailing list
>Diy_efi at diy-efi.org
>http://www.diy-efi.org/mailman/listinfo/diy_efi
>
>

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi