Wire winder (Previously Counter)
Jonathan Allison
Posts: 96
My projects requires the application (code on the stamp) to know how many rotations of a motor have been made. Originally my approach on this was to use an old serial mouse. Did some searching on this and it seems as the stamp does't necessarily have that capability built in, and requires other 3rd party devices. Does anyone have any ideas on how I could go about this?
Thanks
Jonathan
Merry Christmas
Post Edited (Jonathan Allison) : 12/28/2004 2:50:11 AM GMT
Thanks
Jonathan
Merry Christmas
Post Edited (Jonathan Allison) : 12/28/2004 2:50:11 AM GMT
Comments
I'm actually using a tab electronics robot jobbie, trying to convert it over to do what I want (not a toy type of project) I want to set it up to do some percision work (less then 1 mm movements at a time). I'm not sure if it has this capability or not...
regardless, now I am unable to get my pc / basic stamp editor to "identify" the stamp.
I know its not the correct forum, but if anyone has a suggestion let me know, in the mean time I am going to look for some troubleshooting info.
Johnny
It's not a stupid question -- it's just more precision than most BS2 user's go for. People have added position encoder's to the wheels of their Parallax Sumobot (a different animal than your Tab SumoBot).
The TAB Sumobot uses a co-processor to control the motors. You can tell it to move the Sumobot in 200 mSec actions, and you can reduce the 'PWM' cycle the Sumobot uses. With a combination of the two, you MIGHT be able to get the precision you're talking about.
Most likely, you'll need an external counter driven by the encoder. The problem with the BS2 is it only does single tasking -- so when it's reading the position encoder, it can't be driving the wheels. On the other hand, with the SumoBot you've already got a co-processor driving the wheels, so the BS2 can be reading the encoder. It will be tricky.
I was kind of leaning towards scrapping the tab sumobot and just getting a standard bs2 and building it up from there. But in that case perhaps I'll keep tinkering with the tab robot.
My project is a very percise instrument for winding spools of wire, the wire if very very fine (I forget the guage, but its a little thinker then thread, but not much). I plan to run at a speed of about 240 RPM which is pretty slow, but its fine by me. For each rotation of the motor which spins the spool, I need to use the other motor to move the wire laying guide about the thinkness of the wire in the direction of the flow of wire. This way, the wire will lay on the spool in the next position, tight to the previous layer. Hope that makes sense, if not I can show ya some images of the drafts.
It is not critical that the rotor motor (spooler) be under constant power, if may take breaks if need be so that the BS2 can do processing for other purposes. It is more critical that the wire guide run properly. I can use gears and what not to make the wire guide more percise, the wire guide obviously will not rotate, the motor will rotate, with a very small gear on the shaft. That gear shall drive another gear, which will be about 15x the size. This will make the ratio very low, IE motor turns once gear moves very little. The larger gear will move a length of plastic with the same cog type pattern, which in the end will be in contact with the wire running to the spooler and used as the guide. Pictures will definitly help if your not following.
This probably doesn't follow the same type of robotics most of you are doing, but I guess it probably still does classify as a robot, just doesn't move [noparse]:)[/noparse]
I am not a robotics engineer by any means [noparse]:)[/noparse] I am a programmer however, so at least thats one thing I have behind me. I just don't really understand all this circuitry stuff
Now if you see any flaws in this, less seeing the pictures first, let me know, I love to get comments, and or critisism.
Hope everyone had a good christmas.
Thanks,
Jonathan
PS - Very excited about this whole project [noparse]:)[/noparse]
Except for the fact that the guide is way too far away from the spool, its pretty much how I plan to actually build it.
This will help my description above.
Again, just looking for comments / critisism etc.
Jonathan
I'd just use gears to move the guide in and out at a fragment of the spool's speed.
Take a look at a reel used on fishing-rods. It'll show you exactly how to do it.
(Who knows, it may be possible to adapt the mechanics from one to do this work for you)
Another consideration is the guide-rod.
While I like the simplicity of it, you will get into trouble as soon as you hit even the slightest defect in the wire as it may snag in the guide.
(And even if that doesn't happen, it'll probably saw right through it sooner or later anyhow)
You will need to use a small wheel with a thin groove to guide the wire correctly.
(As you're working with very thin wire, it should probably be placed very close to the spool, too)
Is the wire coated?
(Like the coper wire used in transformers and coils)
Then you must be extra careful about the construction of the guide and the tensioner.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Knight Designs
324 West Main Street
P.O. Box 97
Montour Falls, NY 14865
(607) 535-6777
Business Page:·· http://www.knightdesigns.com
Personal Page:··· http://www.lightlink.com/dream/chris
Designs Page:··· http://www.lightlink.com/dream/designs
·
This might be a good application for a 'standard' servo. A 'standard' servo comes from the model airplane or racecar hobby industry. It looks like a box with a shaft coming out of it, powered and controlled through a small three-wire cable. The BS2 through its 'PULSOUT' command can control the servo position to divide 180 degrees of travel into 500 'steps'.
Parallax sells the 'Futaba' servo, which is a standard in the industry. If you replace 'H' with a gear mounted on the servo output shaft, and select 'G' appropriately, you may get the resolution you are looking for.
The Parallax 'BOE' board has plugs that fit the servo cable -- so your best bet at this point may be to get the BOE full kit (I think around $130 or so) along with the What's A Microcontroller class parts (which has a servo, I believe) and try it out.
By the way, a 'standard' servo will only sweep 180 degrees. A 'modified' servo is one that has been modified for continuous rotation. In that case, you LOSE positioning, but you gain speed control. Parallax also sells 'modified' controls, which you could use, but you would have to add your own position encoder in that case.
P.S. You could easily use a 'modified' servo for the 'spool' motor -- again, you'd need a position encoder on the spool to know the position of the spool.
Very interesting project.· I'm not sure what combination of 'standard' servos and 'modified' with encoder servo's will get you the right mix of power and precision -- but I think it is do-able.
Post Edited (allanlane5) : 12/28/2004 3:30:03 PM GMT
I esspecially liked the servo idea. I wasn't exactly sure what they where all about, but if I can go 180 degrees in 500 segments, I'm pretty sure that will cover me. The hardest part is going to be getting everything to line up.
Jonathan
Take a look at old sewing machines. Many of these machines have level winders, some of which work very precisely. I know of two basic variations:
The first variation has the thread (wire) feed some distance from the bobbin and square to it. With proper alignment of the feed point, the thread will lay neatly down in a close wind configuration without any additional mechanical parts.
The second technique, found on some all steel Singer home machines from the 1940's and 1950's used a cam arm that directed the thread into place.
There are times when classic mechanical mechanisms are simplier and more precise than can be reasonably developed using digital techniques. For example, a classic sewing machine requires precise timing between a rising/falling needle and a rotating hook assembly. Belts and gears provide the classical timing arrangement. I can't imagine that a microprocessor could more efficiently perform this function on a machine making 30 stitches per second.
This is not a criticism of your attempting to apply digital techniques to a classic mechanical process. But you might find a simpler and more robust mechanical solution to your wire alignment problem while resorting to a microprocessor to only count the turns.
I agree with your comments. Mechanical coiling or winding machines pretty fool proof in their design and percision. This is fine for their purpose of coiling thread or wire etc.
For my project, I have two goals.
My main goal is to design a machine which can replicate what the old machines do, coil wire or whatever in a uniform manner, but also, and perhaps more importantly, the machine needs be able to coil the wire in random patterns, which will vary through out the individual coil. For example, the first "layer" of wire will be uniform, next layer will be in a random pattern, next uniform, etc (This is only an example and isn't exactly the way I'll be doing this). Each pattern might require to be different, I'm not really sure yet what the exact specifications are for the end product. What I am trying to do is automate a currently manual process. Since the process for developing the coils is manual, IE someone stands by the machine to guide the wire the machine is required to be able to duplicate this as closely as possible. To achive the randomness I believe will be easier then to achive the uniformness, but its something which is desirable in the end product.
Last but not least I see this as a challenge and fun little (maybe not so "little" lol) project for myself. I just want to have fun putting this together and Making it work.
Johnny
Consider using a surplus stepper head assembly adjuster like the All Electronics EX-82 which sells for $1. It employs a worm gear on a stepper to move a control arm. See Allelectronics.com