AFM measurement/backpressure (was Turbo speed sensor)

[Mike (Perth, Western Australia)]

At 09:38 AM 8/03/2000 -0500, you wrote:

>You ain't gonna overspeed one on any street car.   Overheat, underlubricate,
>yes.

OK - hope so - but I was interested in injecting water in the exh manifold
as an anti-lag measure, without some feedback and flakey bleed valve I'd
just be unsure where I'm heading etc.

>> Regardless of the practical need/outcome - I'd still be interested to
>> see what sort of revs the damn thing actually does
>
>60K to 120K rpm

mmm OK :)

>> Yeah true, though with 1GigaHertz DSP's from TI and other smarts, ones
>> wish lists of avoiding compromises gets longer and longer but at the
>> expense of our pockets - we can still speculate (fancy name for dream ;-)
>
>Processor speed ain't an issue.

Well I think it is - for a reliable realtime o/s and scheduler its helpful
to have lots of grunt - given the typical overhead a really good o/s
introduces into the overall responsiveness. With a faster processor there
are so many more exception conditions you can operate on in dynamic
conditions. Sure lazy AFM to injector translation with correction from
engine temp etc doesn't need a 10Mhz cpu - a simple 1Mhz 68705R3 will
do nicely - but once you start adding high speed sensors or at least
expect to provide cleverer filtering algorithms - such as kalman filters
and all the related conditions then the faster the processor the reduced
chance you have of a scheduling bind at really high rpms.

>  An engine is sooo slowww in operating
>speed.  Even slower is the human mechanism operating the whole deal.

Yeah sure, I also had in mind that such a high speed processor might
provide extensions not normally found in embedded EFI's or not at all
practical at moment due to such lack of memory and cpu speed.

>Wounldn't it be more productive thou to think about what can be improved
>upon, rather then worring about say making something with some something
>that doesn't exist?.

Keh ?
a.	I'm not actually embarking on any design to use a 1Gig DSP CPU
	If however you were looking at a high end commercial product then
	now would be the time to consider what *will* be available in mass
	production within your development timeframe 6 to 12months.

b.	Actually theres a 2400 mips engine (1.5v) available (samples now)
	rising to 5000 mips (at least the roadmap says so;) by year end.

Besides there's nothing wrong with thinking ahead.

You and others should start getting worried if I ask about stripline
transmission from my AFM to individual injector CPUs and Firewire
to the dash instruments ;-)

*grin*

:) Mike


----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org