Main > Main Forum
motor advice for rotating crt setup
<< < (11/12) > >>
DaOld Man:
Hey that looks pretty cool Ed..
What is it? A robot?
I love watching those battling robots tear each other up..
Please tell me you are into that!
I see a different forum coming up..
Got a bigger pic?
koz319:

I did a motorized rotating monitor setup a few years ago.   I used the standard monitor 'wheel', with a vbelt (fanbelt) encircling the wheel and a sheave (pulley) on the motor shaft.   You can pickup sheaves and vbelts from grainger or emotorstore.com 

The motor is controlled by an h-bridge connected to the lpt port on the computer.  An hbridge is just is motor controller that lets the low level signal on the lpt port control a higher power circuit to the motor.  If pin one on the hbridge is high, the monitor rotates clockwise.  If pin two is high, the polarity to the motor is reversed and the monitor rotates counter clockwise.  The lpt port is also connected to the degauss circuit of the monitor, so when the rotation is complete, the monitor is degaussed.  In a previous post somewhere I put a link to the schematic for the hbridge I used if anyone is interested.

The motor is a 24v dc motor that I drive at 12v with an old PC power supply.  If you want to use an old supply, just check its power rating on the 12v rail.  It should list either watts or amps for the 12v supply.  The one I used was rated for about 170w @ 12v.  170w/12v = just over 14 amps for the motor.  I measured current draw during rotation, and this was plenty.  The motor is only running for about 4 seconds at a time, so there is not really enough time to generate any heat.  The power supply/motor/hbridge have been working fine for about 3 years now.

There are two limit switches in the cab to signal orientation - I just used basic happs buttons I had left over.  I cut two small lengths of coathanger and screwed them to the back of the monitor 'wheel', these activate the limit switches.  The limits serve two functions, they interrupt the low level signal to the hbridge when hit, stopping the rotation (the normally closed side of the switch is opened when pressed), and a signal is sent to the lpt port when limit is reached (the normally open side of the switch is closed when pressed, and is connected to the lpt port).

I wrote a command line util that will rotate the monitor based on whatever paramaters you send, rotation direction, whether to degauss, and a timeout in seconds.  The timeout just shuts the motor off after 'x' seconds, so the monitor wont spin forever if the limit switch fails for some reason.    This utility can be added called from a front end, but is pretty clunky in its current form.  I'm about halfway through a much better version that should work with most frontends, but it seems like I  just haven't  had the time to finish it.   I actually use a slightly modified version of mame,  just added a few lines to the source code and recompiled so mame rotates the monitor automatically based on the orientation of the game.

Here is a link to a video of the monitor rotating & degaussing when a game is selected:





Here are a few pics of the cab, plexi, bezel, motor & belt, and hbridge.  I couldn't easily get to the back of the cab to take pics of the limits or better pics of the belt, sorry.  Maybe this weekend.


Koz


DaOld Man:
Thanks Coz.. This is what ive been envisioning.
To have the monitor automatically turn by the game that was being started.
I was just thinking of using relays driven off the lpt port, instead of the H bridge.
I was thinking of simplicity and ease of repair. If a relay fails, just unplug it and plug in a new one.
But an H bridge would be more efficient.
Can you send me your program that you wrote for this? Id like to experiment with GLaunch.
danny_galaga:

--- Quote from: Anubis_au on October 10, 2007, 07:38:21 am ---I never wrote explicitly that you only need 90 degrees or so of chain on the circumference, but its obvious to all.

Seems this solution is gaining popularity... now we just need a tester...

The only problem I see with this solution over others, is that the bike chain adds to the circumference of the circular plate, which means your cab is even wider than it was with just a circular plate. the friction-based caster-roller solution, for example, avoids this.

Thinking out loud though, you could do a two-layer solution like Jimbo, with the second layer slightly smaller than the first, and the chain connected there, thus not adding extra to the circumference... that would mean one part of the plate would not have casters holding the plate up.... would that introduce instability to the plate mechanism?

--- End quote ---

you could also just route a channel for the chain to sit in. i was thinking about that last night. route or a sandwich of three disks to create the same effect...
telengard:
I'm watching this thread closely...

This is the last thing I'd like to do to my cab.  I don't mind manually rotating (only takes a sec or two) but it'd be nice to have it auto rotate and degauss w/ the push of a button or even when starting a game.

~telengard
Navigation
Message Index
Next page
Previous page

Go to full version