QuickStart POV ?
cavelamb
Posts: 720
POV, or Persistence of Vision.
Quick and dirty demo using the QuickStart LED string...
About an hour to design and code.
It worked well enough to read numbers in the big bathroom mirror
just moving it by hand. Timing matters, of course.
How fast you move determines how much of the string you may see.
But it really does (almost!) work.
I never could see the entire string in one pass.
But I saw 2 3 4 or 5 6 7 several times.
Even saw an 8 once!
The trick is to lock your eyes straight ahead and NOT FOLLOW the board.
If you follow the motion all you see is the single stack of lights...
Some weights - if anybody is bored enough to follow up with a motorized version.
QuickStart board 18.9 grams
AA battery pack (with switch) 122.7 g.
9 volt transistor radio battery 46.0 g
It would also need a hall effect transistor to index the starting position.
I run with a 4 cell AA pack connected to exp pins 39/40, so that end of the board
would be "up" for the display to read right-side-up. USB connector down.
Dorothy hid the duct tape when I started measuring the ceiling fan blades.
Quick and dirty demo using the QuickStart LED string...
About an hour to design and code.
It worked well enough to read numbers in the big bathroom mirror
just moving it by hand. Timing matters, of course.
How fast you move determines how much of the string you may see.
But it really does (almost!) work.
I never could see the entire string in one pass.
But I saw 2 3 4 or 5 6 7 several times.
Even saw an 8 once!
The trick is to lock your eyes straight ahead and NOT FOLLOW the board.
If you follow the motion all you see is the single stack of lights...
Some weights - if anybody is bored enough to follow up with a motorized version.
QuickStart board 18.9 grams
AA battery pack (with switch) 122.7 g.
9 volt transistor radio battery 46.0 g
It would also need a hall effect transistor to index the starting position.
I run with a 4 cell AA pack connected to exp pins 39/40, so that end of the board
would be "up" for the display to read right-side-up. USB connector down.
Dorothy hid the duct tape when I started measuring the ceiling fan blades.
CON
{ Propeller Propeller display for the QuickStart board }
_CLKMODE=XTAL2
_xinfreq = 5_000_000
ColTime = 10
VAR
LONG MS001
PUB SetUp
MS001 := CLKFREQ / 1_000 ' define 1 millisec
outa[ 16..23 ] := $00 ' %00000000 for LED string
dira[ 16..23 ] := $FF ' %11111111
Repeat
D0
Kern
D1
Kern
D2
Kern
D3
Kern
D4
Kern
D5
Kern
D6
Kern
D7
Kern
D8
Kern
D9
Kern
WaitMS(500) ' wait a sec before serving again
PUB D0
outa [16..23] := 127
WaitMS(ColTime)
outa [16..23] := 65
WaitMS(ColTime)
outa [16..23] := 65
WaitMS(ColTime)
outa [16..23] := 127
WaitMS(ColTime)
PUB D1
outa [16..23] := 127
WaitMS(ColTime)
PUB D2
outa [16..23] := 121
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 79
WaitMS(ColTime)
PUB D3
outa [16..23] := 65
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 127
WaitMS(ColTime)
PUB D4
outa [16..23] := 15
WaitMS(ColTime)
outa [16..23] := 8
WaitMS(ColTime)
outa [16..23] := 8
WaitMS(ColTime)
outa [16..23] := 127
WaitMS(ColTime)
PUB D5
outa [16..23] := 79
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 121
WaitMS(ColTime)
PUB D6
outa [16..23] := 127
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 121
WaitMS(ColTime)
PUB D7
outa [16..23] := 1
WaitMS(ColTime)
outa [16..23] := 1
WaitMS(ColTime)
outa [16..23] := 1
WaitMS(ColTime)
outa [16..23] := 127
WaitMS(ColTime)
PUB D8
outa [16..23] := 127
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 127
WaitMS(ColTime)
PUB D9
outa [16..23] := 79
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 73
WaitMS(ColTime)
outa [16..23] := 127
WaitMS(ColTime)
PUB Kern
outa [16..23] := 0
WaitMS(ColTime*2)
PUB WaitMS(W) 'wait for W milliseconds
WaitCNT (W*MS001+cnt)
640 x 424 - 50K
Comments
program: 536 longs
variable: 1 long.
That's 25 longs over budget, so obviously that won't fly for a full blown spinner.
I suppose it could be broken down so that one cog handles alpha, another handles
numeric and special characters.
Then the main cog could parse the message text and call whoever was needed.
But it seems to me inelegant. There is probably a better way...
you just poaered the quickstart with 4aas wired to a usb connector, then taped it to your celing fan? how bout a couple of pics?
Yes and no.
Yes, I usually power the QS from 4 cell AA pack. But it doesn't attach to the USB connector.
Instead it connects to the expansion connector via pins 39 and 40.
See the other thread about powering the QuickStart board)
But a long USB cable / powered hub might be pressed into service for a quick peek.
But no, I didn't get to tape it to the ceiling fan (yet).
(Dorothy hid the duct tape! She's as bad as my Mom used to be!)
And... there are some real issues to solve first.
Balance...
The AA pack weighs 122.7 grams.
The QuickStart board weighs 18.9.
My ceiling fans have about 8" centers that can't be used.
That puts the battery pack 8" from center.
8" x 122.7 grams = 9816 moments (in gram-inches, of all things)
9616 moments / 18.9 grams = 52.9 inches to balance.
But my fan blades are only 25 inches long.
So no way that combination will balance!.
So, going the other way...
Putting the QS board at 24" from center yields 453.6 moments.
9 volt battery weighed 46.0 grams (without connector).
453.6 / 46 = 9.8 inches from center for the 9v.
That's a lot more workable.
So next trip into Radio Shack I'll grab a 9v connector and give it a try.
There are a lot o minor details to overcome to build a real POV display though.
This was just a play-with-it-and-see-if-it-might-work test.
Like timing (who knows how fast the fan turns, or how fast the characters should be "printed"?),
And a start index (don't have one, so the numbers will probably move until the timing is right.).
A start pulse is the way to go. Hall effect transistor and a strong magnet would do that nicely.
HOWEVER!
It's a cheap experiment.
It cost you nothing to load the program and wave it by hand.
(Except maybe feeling silly at first?)
My best results were standing in front of the mirror, staring myself in the eyes and
moving the board far left to far right.
Don't look at the lights!
Just let it sweep across your field of view.
Yeah, it was a mess trying to get the timing anywhere in the ball park.
And without an index, the display wanders around the circle at random.
Then there is the problem of photographing something in motion - without
the flash stopping it.
None the less, with a little tweaking (both software and fan speed) we did
finally see some numbers. Believe it or not, that's 8 and 9, and a 2.
I haven't dug into it yet, but this seems like a perfect application for an
infra red remote control! Just press up or doen to change the timing?!?
Maybe you can add something to detect the g-force change and automatically sync...
I've seem some Chineese toy that did this using a piezo disk...
Could use an accelerometer...
-Phil
I think all the "propeller" displays I've read about use a magnet for indexing.
And they know haw fast the thing in\s spinning, so they can actually calculate the
timing parameters and get it really exact.
I kinda like the accelerometer idea for hand waived. Then it could key off of the
acceleration to start writing.
But the best setup would be if one of our board layout gurus took an interest and
laid out a propeller based propeller board. Long skinny arm with 16 to 24 leds
on one end and the prop and stuff on the other for counter balance. Use a slip ring/
armature to provide power to the board...
A Hall effect (magnetic sensor) transistor to index and an IR transistor for an IR
remote control.
I was just playing around with this because it's been 107+ outside and I've been
bored outta my mind...