Date: Thu, 30 Sep 1999 14:53:32 -0400
From: Mark Sirota wrote:
<< "Mark J. Andy" wrote:
> On Thu, 30 Sep 1999, Mark Sirota wrote:
>> I wonder if John Carriere or any of the other standalone timer folks
>> might be interested in developing such a card specifically for
>> autocross timing applications. That would be a good way to hedge
>> against the trend towards computer-based timing systems.
>
> Having it be a card could kinda suck though... Can't operate
> standalone, can't make it talk to a laptop (and who wants to setup a
> desktop PC in a van?), etc.
>
> If I were a timer manufacturer and I was convinced the market was good
> enough (hah!), I'd develop a box with the card you mention above in
> it, and have it speak serially to my win32 app. I'd also put the most
> basic rudimentary controls on the box itself so that if forced to, it
> could work without the other computer around.
>
> But that's just me.
<< I agree, that makes good sense. I was thinking of it from this point of
view:
(1) There seem to be a lot of folks out there willing to use PC
architectures and Microsoft OS combos for timing, living with the
liabilities of poor accuracy and repeatability.
(2) These people have already invested in the hardware needed for
*scoring*, and are using that same hardware for *timing*.
(3) On the assumption that scoring is largely a software function and
timing is a hardware function, these people may be willing to invest a
SMALL amount of money into proper hardware for timing that could
interface with their existing or new PC/Microsoft scoring system.
Such a system would need to be a lot less expensive than a fully-
functional standalone box, such as the very popular offerings from
JACircuits. I'm envisioning something to do the timing and only the
timing, plus an interface to scoring. Any add-on do-dads that you
include, such as displays, buttons, etc. add both to the cost and the
footprint. >>
<<<SNIP>>>
Okay, I think I can step in here with an idea/plan.
I have built and programmed several items using a Motorola 68HC11
microcontroller chip. It includes an internal hardware timer system that can
take 4 inputs and freeze the internal time accurate to .0000005 seconds. It
then triggers an interupt and the service routine reads the result frozen in
a register. Using dividers this can be slowed to .000008 sec, so it still has
125 times the resolution needed. I can make a very simple box with this chip
that would take the photo heads and freeze the time, and then send it out a
serial port to the PC. If this will sell at least 20 units I will write the
program and test it. The boards can be made for under $200, maybe even with a
box and connectors.
For the scoring software people out there, I have a few questions.
(1) do you want "the box" to give a complete time for each car, OR give an
absolute time of every start and finish?
The earlier way requires that the timer count the starts and stops and keep
them in sync, a false trip would need a way to reset it in "the box"
requiring a button at the least. If I just make it send an absolute time with
a start or stop code, then the PC can track the number of cars and do the
subtraction when it gets the final stop time. The advantage of this is if you
get false trips, you can take the data from the last few inputs and line up
the starts and stops till you get proper times, a false trip on start gives a
real long time, and on a stop gives a very short time, it would be easy to
move the times to find which were the right trips and delete the erroneos
ones. This way the box does not need any buttons, just 2 photo inputs, a
power connection, and a serial port.
If we do the absolute time, we do need to over wrap from time to time and the
PC will have to take care of it, like Y2K but every minute or so. How long
should the over wrap interval be? I can software track more digits, maybe
overflow at 100 seconds? In this case, the box should send a flag to the PC
that the over flow happened so it know to add the 100 before the subtraction.
Whoever wants to write the PC side, let me know what would be best for you
and I can do this. If there is a market for it, I am game. Our season is
almost over here, so I can work on this instead of my car for a bit.
The new JAC timer we use at TSSCC is very nice with great features and can
send times to a PC, but it is expensive and to keep it in sync with false
trips requires the displays and buttons. We use a PC for scoring, and the PC
operator manually types in the times from the JAC timer. I think having the
timer put the data in the scoring would be much better. I think the software
we have can do it, but it is very new to our club so I think it is a matter
of familiarity. And trust.
Gary M.
|