Matrix or Column-Row Light Chaser

[jukingeo]

Hello all!

Back with a semi-quick question.   As you know I have been on a light chaser kick with the Basic Stamp 2.   So far I have been pretty much satisfied with messing around with 6 & 8 channel chasers, but I have the Basic Stamp 2 which has 16 channels and that got my wheels turning.

I remembered a carnival ride in which the lights were set up on bars and chased within the bars AND across the bars.  I never fully figured out how to do this, but I believe the light control has to be done with some sort of matrix.  I gave up on the idea for a while, but when I saw how you can reverse the 'polarity' of a Red & Green LED using the Basic Stamp...it got me thinking about the matrix light chaser again.  If anything...I can see how the BS2 should be able to do this.

For those of you that don't know what I am referring to.  Here is the carnival ride:

The Gravitron (one of my favorites):

http://www.youtube.com/watch?v=AxODpCfrFOE

Look at the lights on the middle of the spinning part of the ride.  They are chasing against the ride rotation right now (column chase), but towards the last two seconds of the clip, the pattern switches to row chasing in which the lights chase up the ride instead of across it.

My best guess is that being you are dealing with AC currents that you would need to have a solid state relay in each leg (one on the hot side and instead of normally returning the neutrals to the AC source, you would have another solid state relay in this leg).  So the question is, would one set this up as an 6 by 6 (or 8 by 8) matrix using 12 (or 16) solid state relays, or would each bulb string in the matrices would need a channel (that would mean 36 or 64 solid state relays (OUCH!! on the wallet)).

Now unlike a regular chaser, this would make hook up a bit of a challenge.  I have tried to envision how something like this is done.  However, I don't think it should be too difficult to set up because I do see this type of chaser quite often.  Granted, I mostly come across it on amusement park rides.

So some assistance with code and hookup would be appreciated.

(Jon you don't have to go out of your way with the program on this one...I can wait, I know you are busy helping haunters right now.  I have learned a lot already on my own thanks to your guidance).

Thanx in advance,
Geo

[JonnyMac]

"I don't think this should be too difficult because I see this type of chaser quite often."  Well... open heart surgery is performed every day in various parts of the world but that doesn't make it easy.   ;)

You're going to be somewhat limited with the Prop-2 because of its operational speed.  You could use OUT0..OUT7 to select a row (only one at a time) and then perform the chase sequence on P8..P15.  You'll also have to run the P8..P15 outputs through a non-inverting driver like an L293D.  So, the low side (ground) comes from OUT0..OUT7, the high side from the drivers connected to P8..P15.  You'll need steering diodes unless you're lighting LEDs or driving SSRs.

I created a 8x8 LED matrix with animated graphics for my Nuts & Volts column, but I used the SX to multiplex the LEDs:
-- http://www.parallax.com/dl/docs/cols/nv/vol8/col/nv142.pdf

[menehune]

I don't think you need to hook every light "module" to it's own relay.
My feeling is that there are four vertical segments in each column and possibly four columns wired together.  That would give you a 4x4 matrix which could be easily controlled.  Just gang all the extra light together until the whole set ride is connected.
ABCD
1111
2222
3333
4444

You would have the verticals columns pan anti-rotation (D1,2,3,4 then C1,2,3,4 etc.).
The ending sequence would be a little harder, but easy to do in software-(A1, B1, C1, D1 then down to A2, B2, C2, D2, etc.)

Once you figure out the smallest "block" it's a relatively simple matter of writing the code to make the block do what you want it to do.  You might be able to use a simple pin addressing scheme like the police chaser light demo program that ships with the prop.

[jukingeo]

You would have the verticals columns pan anti-rotation (D1,2,3,4 then C1,2,3,4 etc.).
The ending sequence would be a little harder, but easy to do in software-(A1, B1, C1, D1 then down to A2, B2, C2, D2, etc.)

Once you figure out the smallest "block" it's a relatively simple matter of writing the code to make the block do what you want it to do.  You might be able to use a simple pin addressing scheme like the police chaser light demo program that ships with the prop.

True this is what I was thinking as well.  Usually these lighting systems cannot produce dot matrix type patterns.  If that were the case then most definitely there would be a need for individual outputs within the matrix.  Usually either direction has the same chaser speed, so my guess is that the same trigger source is used as well.  For the most part It could very well be that the actual chaser uses the same outputs and then a master relay just changes the output banks around.   I almost never see one light traveling both up AND across (meaning diagonal movement in the chase pattern).  It is either horizontally or vertically.  So my take on it is that there is a master relay that switches between row and column chase and that there is a single set of output devices that does the actual chasing.  Thus all in all there would be 8 connection points and each 4 section branch would have to have two wires going back to the control box.  When one leg is neutral it chases one way, when the relay clicks over it makes the other side neutral and then it chases the other way.   At least that is the way I envision it.  The fact the system does have built in limitations does suggest that it is not a true 4 by 4 (16 output) grid.  It is definitely switching banks.

I truly wish I could take one of these things apart to see what they do.

[menehune]

Personally, I would never run both the neutral and hot wires from a load back to a relay and alternate applying power between the lines.  Too much potential for a dead short if the relay contacts stick or fuse together.  I also think switching the neutral could be a National Electric Code violation.  I need to check with my electrician friend.

I wonder...Maybe a rotating drum with cutouts that turn multiple roller switches on and off.  Like washing machine timers or mechanical music boxes.  Change the pattern by changing the drum.  If the drum were large and had a sufficiently slow rotation, you could have a pretty long pattern programmed in the controller.  I've re-invented the analog computer! :)

The up/down, right/left pattern switch could also just be a simple shorting device across contacts like how vehicle distributor caps are wired-the rotor spins distributing power to the spark plugs in a pattern.

I would assume the rides are made a simple as possible for maximum life and ease of maintenance.  You don't want to have a rocket scientist on site to troubleshoot the ride if it goes down.  The operator is probably someone without an electrical degree.

[menehune]

Although there is one special circumstance where you are permitted to switch the load and neutral wires of a circuit*, you never want to switch the neutral conductor of a incandescent socket.  The neutral is always the "outside" of the bulb socket and it is the "grounded" part of the circuit. 

A worker changing a bulb is likely to contact the metal threaded part of the bulb when removing/replacing the burnt out bulb and you never want the worker to be in a situation where he can be electrocuted by completing a circuit between a hot lead and the grounded metal body of the ride.

*you are allowed to switch all non-ground conductors in a gasoline pump emergency cutoff circuit provided the switch are turns off all conductors at the same time-i.e. a 4 pole, ganged switch for the three phase conductors plus the neutral conductor going to the pump.  The safety ground conductor is never  permitted to be switched or interrupted at any time for any reason.

[jukingeo]

Personally, I would never run both the neutral and hot wires from a load back to a relay and alternate applying power between the lines.  Too much potential for a dead short if the relay contacts stick or fuse together.  I also think switching the neutral could be a National Electric Code violation.  I need to check with my electrician friend.

I wonder...Maybe a rotating drum with cutouts that turn multiple roller switches on and off.  Like washing machine timers or mechanical music boxes.  Change the pattern by changing the drum.  If the drum were large and had a sufficiently slow rotation, you could have a pretty long pattern programmed in the controller.  I've re-invented the analog computer! :)

The up/down, right/left pattern switch could also just be a simple shorting device across contacts like how vehicle distributor caps are wired-the rotor spins distributing power to the spark plugs in a pattern.

I would assume the rides are made a simple as possible for maximum life and ease of maintenance.  You don't want to have a rocket scientist on site to troubleshoot the ride if it goes down.  The operator is probably someone without an electrical degree.

True that switching power could be done with separate relays.  I am just at a disadvantage because I don't know 'How' it is done.  But I do have an idea.  As for national code...well, that does go pretty much out the window when you are talking about items made overseas.  Even many US companies are no longer requiring items to be UL listed.  Is it a scary thing?  Yes it is.  I have seen many items come from China with many violations.  I service sound systems for a living and I even have the dubious honor of shutting down production of a particular item because I found something that could be a fire hazard.  Granted the company wasn't too happy with me because they had to redo the product...but I sure saved them from a huge lawsuit!

Yes, most amusement rides are designed very simple.  Most of the lighting control systems still use regular relays instead of solid state devices.   Many European rides have lighting systems that are optically isolated from the computer and use a transistor drive to the relay bank.  Then the relay coils themselves have diodes and filter caps on them.  It is very rare a channel goes on one of these European systems.  But then again we are talking about systems that are made in England, Germany, and Italy....not China.

Anyway I will stop here because this is potentially getting off topic.

Thanx for the input, but I believe I am going to err on the safe side and stick with standard chasers for now. 

[menehune]

Good luck!  Maybe you can vary the start times of the chasers so they can make different patterns.

If it was just me in the prototyping stage, I would be more flexible with the rules and regs, but once the public is involved (and the fire marshal and the lawyers) I prefer to be super conservative so I can sleep soundly.

[jukingeo]

Good luck!  Maybe you can vary the start times of the chasers so they can make different patterns.

If it was just me in the prototyping stage, I would be more flexible with the rules and regs, but once the public is involved (and the fire marshal and the lawyers) I prefer to be super conservative so I can sleep soundly.

Hello menehune,

I believe I finally figured this out after studying the operation of a matrix chaser closely.  The one I was looking at was from an old Styx concert I was watching this afternoon on YouTube.  Low and behold they used a matrix chaser on a mock up theatre marquee for their Paradise Theatre album Tour. 

Anyway, examining the detail on the chaser (by countless rewinding and forwarding of the video footage) I could clearly see the chase patterns and I followed them closely.  In this case the marquee was set up as an 8 by 8 and it is exactly how you described it as a chaser bank set up with each bank as a selectable group.  Both group and chase timing were identical. 

In keeping with what you said about keeping all the neutrals tied together, I envisioned the hot side chase relay and a group relay has a huge AND gate.  So the group relay HAS to be set on in order to provide power to the normal chase relays.  Now the converse will also become true, by setting all the chase relays to the 'on' position, you can then chase down the groups.

All in all with the example presented above, I would need 16 channels set up in two groups of 8...not 64.  Guess what?  The Prop-2 is already set up in two groups of 8 channels each.  One is a low byte and the other a high byte.

Hmmmm, So I would say the Prop-2 is a natural for this...but of course I probably wouldn't go full 8 channels.  But a 6 by 6 is tempting.

Also I learned from Jon how to do some fake multi-tasking with the Prop-2 and I could set up two different timing patterns for the group and the chase.

I can also envision how to write the program for this in the Prop-2...but I think in terms of loading the data...I might need the assistance of Vixen.  We are talking quite a bit more READs from a data table with a chaser like this.  But I believe the Prop-2 can do it.

Looks like I have to buy some more solid state relays!