Olin Lathrop wrote:
>
> > Finally, I finished that propellor clock.
> >
> > ...
> >
> > http://www.bubblesoftonline.com/prop4.gif
>
> Nice job, Tony.  So that's what you meant by a propellor clock!

> If you're willing to tell, how did you do it?  I'm mostly curious whether
> convergent evolution drives these kinds of designs mostly in the same
> direction, or whether there are fundametally different ways of doing this.
> The way I did it seemed obvious to me, but I have no way of knowing what I
> didn't think of.

Hi Olin,

I was going to post it as a project, but unless the builder has a lathe,
it couldn't be done my way. I have since seen a design that is a
rotating PCB with 2 brass fingers rubbing onto a fixed PCB attached to a
motor that has circular tracks as part of the PCB design which would be
easier to make.

I got the original idea from Bob Blick, but I needed a "propellor" theme
for the aero club, so I had to start from scratch.

I use a LED sensor to sense each rotation and this triggers the RB0INT
interrupt. It just outputs each column of pixel data for a few uS from a
RAM buffer which is loaded with new data each time the minutes change.
Here's the simple IRQ code. The LEDs are driven directly from PORTB.

        movlw Digit_HR          ; FSR = 1st display data byte
        movwf FSR
DispLoop        movf INDF,W
        movwf PORTB
        incf FSR

        movlw 0x75              ; pixel delay
        movwf HSCount
HSLoop  decfsz HSCount
        goto HSLoop

        movlw Digit_HR + d'31'  ; 31 bytes per display field
        xorwf FSR,W
        btfss STATUS,Z
        goto DispLoop

        movlw 0xFF              ; display = off
        movwf PORTB

I didn't want to hack a motor for this design like Bob did, so I just
fed AC to the rotating PCB via slip rings. This supplies power as well
as a 50Hz signal for time keeping. The 16F627 is working with it's
internal clock source.

The tricky part was setting the time with external buttons when all I
had was a 2 wire 50Hz AC supply to work with. I ended up switching a
series diode in and out of the input feed which half wave rectifies the
power supply. The PIC detects this drop in frequency and this is used as
a signal that a "switch" is pressed. Half wave is still ok to power the
circuit although the LED brightness drops slightly.

To keep the project free of buttons which could have made the project
look messy, I used Q-Prox sensors to detect finger touches on the glass
panel.

I have seen a rotary display like on you web page and they look quite
good.

The simulator is also interesting, some members of my aero club is
building a sophisticated rig right now.

--
Best regards

Tony

mICros
http://www.bubblesoftonline.com
mailto:sales@bubblesoftonline.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics