Searching \ for 'Speed Capture' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=speed+capture
Search entire site for: 'Speed Capture'.

Truncated match.
PICList Thread
'Speed Capture'
2000\03\08@172220 by Tony Nixon

flavicon
picon face
Hi all,

To find the vehicle speed for my project, I was going to tap into the
speedo wires coming from the gearbox, but after getting a wiring diagram
for the vehicle I changed my mind. It is 97 pages of schematics just for
a 4 cylinder car. That also took ages to get hold of, and I couldn't
believe that the manufacturer will no longer give out any info on this
type of thing. I had to get it from a friend of a friend etc. inside the
business with the assurance that I was not to disclose it.

What a drama!

Gone are the days of self maintainance.

Anyways, as simplicity if often the best solution, I decided to use a $5
opto shmitt trigger detector mounted on the inside of the rear wheel. It
"looks" between the 16 or so slots cut into the wheel rim and gives me
the signal I need.

I'm going to test drive the vehicle (which has a digital speedo) and
press a "button" on the controller when the speed is 64KPH. If I
multiply the accummulated speed pulse result by 64 (easy), I get the
speed at 1KPH. From then on, to get any speed in KPH, I just divide the
1KPH speed by the new capture value.

To do this I need to use the capture module in the 16F877. At 1KPH,
rough calculations suggest a value of C350h will end up captured when
TMR1 prescaler is set at 1:8 and capture on every falling edge.

>From what I gather, after a valid edge is detected, The TMR1 contents
are dumped into the capture registers and an interrupt can be generated.
It seems that TMR1 is not reset as a result. Is this true, or do I do
that from the interrupt routine? That seems to be a little bit
inaccurate to me, although it may not matter much. I figured on a max
frequency from the slots of only 278Hz at 110KPH and I'm only interested
in integer results.


--
Best regards

Tony

http://www.picnpoke.com
spam_OUTsalesTakeThisOuTspampicnpoke.com

2000\03\09@071454 by Russell McMahon

picon face
>To find the vehicle speed for my project, I was going to tap into the
>speedo wires coming from the gearbox, but after getting a wiring diagram
>for the vehicle I changed my mind. It is 97 pages of schematics just for


Tony,

I'd suggest you persevere with the existing speedo signal.
Try approaching firms who fit meters to taxis.
Most cars with electronic speedos have known pickup points that people use
to feed such devices. .

Failing that, you could look at signals that enter the speedometer proper -
there are only so many of them. I did this on one of our cars and found the
correct wire fairly easily.

Another alternative where there is a mechanical speedo cable is a screw on
pickup that goes at the gearbox - these are made to for all common brands of
car. You simply unscrew the speedo cable, screw on the pickup and then screw
the speedo cable to the extension on the rear of the pickup. You can get
opto and hall type pickups (possibly mechanical too). Pulses per kilometre
vary with design and internal gear ratios - easy enough to calibrate by
driving a set distance and seeing how many pulses you get.

I have a friend who makes taxi meters who may have details on a  specific
car but I'll let you ask people at your end first - ask me if you are stuck
and I'll email him.

>I'm going to test drive the vehicle (which has a digital speedo) and
>press a "button" on the controller when the speed is 64KPH. If I
>multiply the accumulated speed pulse result by 64 (easy), I get the
>speed at 1KPH. From then on, to get any speed in KPH, I just divide the
>1KPH speed by the new capture value.

Probably more accurate to find a known distance and simply count the output
pulses over this distance.
Of course, once you have pulses per distance the pulses per time per kph
follow automatically.
If you use your wheel spokes idea you could VERY accurately measure the
distance travelled in one wheel revolution.(about 2 metres) (chalk tyre at
bottom and ground, have person drive car slowly forwards until chalk mark
does one rev, chalk ground, measure distance). You can probably do the
latter to well better than 1% accuracy (1% = about 20mm in one wheel
revolution). This assumes that the tyre's dynamic (road speed) size is about
the same as its static size.












     Russell McMahon
_____________________________

>From other worlds - http://www.easttimor.com
                               http://www.sudan.com

What can one man* do?
Help the hungry at no cost to yourself!
at  http://www.thehungersite.com/

(* - or woman, child or internet enabled intelligent entity :-))


{Original Message removed}

2000\03\09@071458 by Andrew Kunz

flavicon
face
TMR1 is not reset.  The CCP just tells you what TMR1 was when the change (and
interrupt) occured. This works nicely, because you can get both the active and
inactive times if you want, or all kinds of stuff.  And it's only a 16-bit
subtraction (or two) away.

Andy

2000\03\09@072324 by Alan Pearce

face picon face
Rather than doing the sensing on a wheel, it would probably be better to do it
on the drive shaft if using a rear wheel drive vehicle. this would then average
the difference between the inner wheel and outer wheel when turning corners. If
the car did a lot of say left hand corners compared to right hand corners
(especially in city driving) you may find a significant error creeping into your
calculation.

If the vehicle is front wheel drive, then it becomes a lot harder. The ultimate
would be to attach to the speedo takeoff like already mentioned.

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...