Progress1: Software Dyno

[Dave Zug]

The part about the software delaying the read frequency is what I 
was wondering about. I'd sure like to put up a dynamic graph but this 
takes ticks. probably not very useful while you are driving anyway.

I dont know what the vss looks like yet. scoping it this weekend.

the simplest software routine will catch the present value, store it 
in ram and catch the next one. then later store to disk. the next 
slowest will store each value to disk immediately. I might be woried 
about nothing.. probably plenty fast enough. One thing though.. 
windows is unpredictable in the time slices it gives its programs. I 
have a watchdog on this though.. VB allows you to get a milisecond 
timer reading whenever you want. I'll grab it after each read and all 
the time differentials should be the same. If not.. a minor 
adjustment in the software to account for it and a little bit of 
error.

If  I would just learn C I'd be all set and windoze could do the 
graphing and interpolation alone. would be really simple I think for 
someone to write a C routine to capture parallel pin one continuously 
and stop with a keypress and store the recording to disk. then I'd 
take it from there. My input would just be a loong file of zeros 
and ones.

adjusting the voltage from vss to parallel is the circuit I need yes. 
I'll do it myself after I chack the VSS with the scope and get off 
the software end but if someone who already knows the signal wants to 
help and make a diagram... ~

> From:          "Bartz, Jamison" <Jamison.Bartz at COMPAQ.com>
> To:            "'gmecm at efi332.eng.ohio-state.edu'" <gmecm at efi332.eng.ohio-state.edu>
> Subject:       RE: Progress1: Software Dyno
> Date:          Fri, 20 Aug 1999 13:35:07 -0500
> Reply-to:      gmecm at efi332.eng.ohio-state.edu

> What does the VSS signal look like?  If you are sampling through the
> parallel port and your software isnt too complex, you shouldn't have a
> problem capturing each pulse. Depending on the format of the VSS signal, you
> could possibly be able to just use a comparator to adjust the voltages and
> do the timing stuff in your program, or if you want to be able to read in a
> value, a freq to volt converter will do the work in hardware.
> 
> As far as the parallel port's sampling rate goes, the bus behind it inside
> your box usually runs at 8Mhz.  Newer boxes may have that changed.  If you
> read a sample on every clock you get a sample time of 125 nanoseconds.  Your
> software could hinder you from sampling every clock.  It depends on the
> complexity (and possibly language) of your software.  If the VSS sender
> spits out 2002 pulses every mile then at 100mph you get roughly 56 pulses
> per second.  In computer terms, this is an eternity.  You should have no
> problem catching every one of them.
> 
> -----Original Message-----
> From: Dave Zug [mailto:dzug at delanet.com]
> Sent: Friday, August 20, 1999 8:54 AM
> To: gmecm at efi332.eng.ohio-state.edu
> Subject: Progress1: Software Dyno
> 
> 
> Okay I've been working and I have the software reading the parallel 
> port bits directly and logging them. I'm not sure of how quick the 
> sample rate is.. still a test to do. What I need now is a very simple 
> circuit to turn vss into what the parallel port considers a 0 and 1. 
> 
> another circuit could turn ABS signal or even the spikes from 
> the ground noise from the distrib (another responder idea) into 
> parallel voltage transitions it does not matter to the software. 1 
> setting maybe.
> 
> If a more complex circuit is used.. all 8 bits can be read 
> representing the number of transitions sensed in a 0.1 second period 
> or something (the basis for this idea is from one of the responces-- 
> thanks). 
> 
> I think "simple" is a good way to start.. if the electronics portion 
> can be left simple and very cheap all the better IMO.
> 
> The circuit can be made more complex later.. my goal is to see data 
> with the PC as simply as possible as a first step
> 
> more to come.
> ~~~
> Dave Z. www.delanet.com/~tgp
> 
> 
~~~
Dave Z. www.delanet.com/~tgp