Homemade Robotic Arms for CD Duplication? 68
LA Kings Fan writes "I have this current gig of a job which requires me to make numerous copies of CDRs (in the thousands). Since it would be ludicrous to sit in front of my computer to remove a burned CD and put a fresh one in everytime, I've looked around for better, more sensible solutions. There are two alternatives: CD Duplication Towers and Automated Duplicators. They both have their advantages but are very costly. The cool thing about the automated duplicators is that the burning process is automated by the use of a robotic arm which replaces the burned CD with a fresh one. This is neat, so I was wondering if anyone has attempted to take this concept a step further by essentially building their own robotic arm for their burner on their personal computer. Is this feasible? Can a robotic arm like this be created from off the shelf parts? I'm clueless when it comes to this engineering stuff, so any help would be appreciated."
Third-party? (Score:1)
Re:Third-party? (Score:1)
If it's just in the tens buy a goddam burn station. If your work is mission critical they'll have the money to buy a station, complete with warranty and support. If it's not, well then go ahead and build your own from parts, although I guess it's gonna cost as much as buying one - don't forget to count the many hours you need to put the things together, write the software, etc.
Re:Third-party? (Score:2)
Re:Third-party? (Score:1)
Re:Third-party? (Score:1)
Lego! (Score:2)
Re:Lego! (Score:1)
Re:Lego! - up dated link (Score:1)
Amazing Lego DAT Tape Changer (Score:2, Interesting)
Someone built an Amazing Lego DAT Tape Changer [slashdot.org] out of Mindstorms. I imagine the same could be done for a CDR changer. Here's your excuse to go buy yourself a cool toy!
2 words (Score:1, Redundant)
if your clueless about engineering... (Score:3, Interesting)
hire a college intern, tell them they will be working with multimedia on a daily basis and have them burn CDs all day. it might be cheaper than a robot and the intern is more mobile.
Re:if your clueless about engineering... (Score:2)
My bet is that that's exactly what the company did. Now the intern wants to be lazy and is asking
mod a commodity audio changer? (Score:2, Interesting)
say one could marry the cd changing mechanics with an off-the-shelf cd writer. serious hack indeed, but perhaps doable. has anyone disassembled such a changer already?
this would rock.
--dzsino
Re:mod a commodity audio changer? (Score:1)
Re:mod a commodity audio changer? (Score:1)
Hello! (Score:3, Insightful)
He has a job that requires him to manually copy "thousands of CDs"? Get real. Any company that is putting out that much product can afford a $500 copying machine instead of paying, at minimum, $5.50/hr to some clueless punk.
Re:Hello! (Score:2)
But if you've got "thousands of CDs" you should be having them pressed, not burned. Better quality, shorter turn-around.
Re:Hello! (Score:1)
Complication (Score:5, Informative)
By the time you finish planing and building you robotic arm the whole project could be done. If this is a "gig" as you say your getting paid to do a service. Do the job you were hired to do. Don't go on a tangent just to impress us here.
I find doing a good job well and fast much more impressive than flashy unnecessary extras.
Re:Complication (Score:1)
Besides, it would be fun if someone built it (mistakes and all) and I just copied it...I seriously need one for my Linux burner at home, I could do my burning stuff sitting at work.
Re:Complication (Score:4, Funny)
You'll never make it into upper management with that attitude.
Hey! You're stealing my synaptic impulses! (Score:1)
Building the mechanism is relatively simple. The hard part is writing software that will 'talk' to your CD burning software to figure out when to swap discs.
Alternately you could hire a minimum wage student to do it for you
when to swap disks (Score:1)
Re:when to swap disks (Score:1)
Lego Mind storms (Score:2)
But if you did not think about that, then you probably do not have the technical expertise to design, build, & program a robotic arm.
Just go buy a duplicator or pay a stupid kid $5.50/hr + Caffiene + a few cdr's
-- Tim
like a picnic spinner (Score:1)
My idea was two mechanisms: an injector or loader, and an ejector or unloader. They'd be driven by the cd tray popping out.
The ejector would probably just be some sort of lever arrangement to dump discs into a chute to stack in a box. The ejector would finish and trigger the injector.
The injector would be an escapement of similar design to the one on an old record player where you'd stack the discs on a spindle, and it would drop one at a time.
Consulting with a fellow maker of silly things, he reminded me that I could never inflict such a solution on my employer, having lived with other people's painful hacks far too often.
Now maybe I'll get back to it, now that you mention it. It's more fun to make stuff than build servers or study perl & regexps, or work late.
This is bigger than burning CDs (Score:1)
The implications of this robotic thingy (if someone finds a way to build it well) are bigger than just replacing CDRs with fresh ones...thing of being able to put all of your CDs(drivers, MP3s, etc) into your CD drive remotely !
I could use one of these for my Linux burner at home which is on DSL all the time...all my CDs available to me all the time !
Re:This is bigger than burning CDs (Score:1)
thing = think
Re:This is bigger than burning CDs (Score:2)
You want the Plextor MegaPlex 200 [dvsystems.com].
That's assuming that you have the $5500 you need to buy one, of course.
Mindstorms (Score:2)
Basically, you set up the CDR drive s.t. it burns whatever you want it to when the device closes. When it's done burning, eject it. The ejection nudges a light or touch sensor on the legoBot. The legoBot picks up the freshly burned CD, and drops it onto a spindle, then gets a new blank, inserts it into the drive, and closes the drive door.
The tough part would be getting the "pick up the new blank" part, since you could only pick up one, and the height of the stack of CDs would differ. I dunno, maybe something like "the Claw" from Toy Story would work.
Add a rooster and some fire, and Rube Goldberg would be proud.
Re:Mindstorms (Score:2)
Two words: suction cups.
What a coincidence! (Score:1)
These kooks going on about having it done by a "pro" are so full of shit. $1.50 per disc is a great deal? Ehr, what if these are promo discs?
Blanks are twenty cents in my neighborood?Besides, I very strongly suspect that most of these outfits do indeed simply hire someone to change the discs off of regular burners rather than using these way overpriced mechanized solutions. A buck fifty! Fuck that.
But enough ranting, what have I come up with?
Well first of all, I think an arm is not the way to go. The simplest thing I can come up with for the input is a conveyer belt that drops the loading disc into a shoot which lays it in place. Every time the player ejects, it hits a switch that advances the conveyer belt the length of a CD laying flat on the belt thus dumping the next CD into the shoot.
Of course before that can happen the old disc needs to be dumped. I think the simplest mechanical approach to that is a spinning wheel like a rubber RC car wheel that comes up from the bottom of the feeder tray on a little hydraulic jack. The wheel is offset from the center of the CD a bit, so it lifts the side of the CD and pulls it across the tray into a padded hopper.
I think that's a simple as it gets. Adapting an existing changer is probably an awesome idea, but I'm not in the States so E-Bay isn't an option for me and I haven't had a chance to try it.
I think a conveyer belt is going to be about as low tech as you can get and there's no reason you couldn't line up a hundred before you took off for work. It would be funky looking, but I think it's got a real chance of working. If I get something rigged together, I'll submit it as a story.
Personally, I'm quite into something like this. It's essential for people who have real live small businesses that need to cut costs wherever they can. Customers want to know where the costs come from and when you throw away money on bullshit services in order to get it done quick you've got to pass this along to the customer. When you're a small business and you've got lots of anaccountable expenses, you're toast.
Kooks? (Score:3, Interesting)
At 500 quantity, Furnace CD (furnacecd.com) will do them for $0.89 each. Less than $500 bucks! That's a hell of a lot cheaper than me trying to build some crazy machine with a conveyor belt. Best thing is, I can give them my master on CD-R and get my CDs in less than a week (3 day turnaround).
Now, if I was getting more done, it would make sense to get them done as pressed CDs. I don't know the price break down for those, but they said that it becomes cost effective to do that at around 1,000 copies, because there are up front fees they make you pay for setting up the pressing machines.
Get them done somewhere, don't waste time trying to "engineer" some solution with rubber bands and legos...
Re:Kooks? (Score:1)
But! The topic at hand is people who ARE interested in a home automated rubberband and bailing wire solutions.
And, as a publisher of book/CD sets, I'm fairly sure that the point at which it actually becomes cheaper to get a stamped master CD is no less than ten thousand once you're buying CDRs for less than twenty cents a piece. I think I said I pay 20cents in my first post, but in fact I think my actual price is around 16cents when I buy them by the thousand.
If I'd agree with any of the nay sayers on automation, it would be simply to do what those "pro" CD dup scam artists are going to do you for --pay someone else to sit there and change CDs. Those motherfuckers are worse than used car salesman.
Allow me to help you out with some math.
At six bucks an hour operating two 32X CDRs, the ridiculously simple labor cost of a dedicated disc changing empolyee is going to add a little less than ten cents per disc. If you're getting blanks for 20 cents and paying some dork 10 cents a disc at minumum wage that's a grand fucking total of thirty big fat pennies. In that case, how in the hell is 89 cents a piece a good deal? Compared to being straight up robbed as you're walking to your car in a parking lot, it's a fabulous deal, othewise it sounds like shit.
And, if you think you break even on a stamped disk at 1000 pieces I don't think you know what you're talking about. That might have been true five years ago when CDRs cost more than a buck. But guess what . .
And furthermore, did you know that pressed CDs are more likely to fail and have errors than CDRs? I bet you didn't know that. Well now you do and it was a free as in beer tip from your bud Steve in Taiwan.
Re:Kooks? (Score:3, Informative)
What I said was that it was crazy to waste the time and money on a "rubber band" solution. I think it's stupid to try and spend time and effort on developing your own automated system that it will end up costing you as much or more as a purchased one. Especially for the original poster, since he flat out states he has no engineering experience.
He should either get them done professionally, or buy a 10 disc burnstation. They aren't THAT expensive and you can quintuple the performance your $6 an hour drone gets, or even better, make it easy enough that an employee you already have can do it - no new employee at all. It would take someone about 30 minutes of their day to change the 10 disc burner 8 times. Why pay someone to sit there and burn cds if you already have an employee? Or hell, do it yourself. The money you save in ease of use and wages will pay off the burnstation in a few weeks/months.
(If the original poster is in a Uni environment, it might be worth it to get a student worker to do it, because, if student workers are anything like they were when I were in school, they are basically free labor and have a minimum amont of hours they must put in for their work-study, anyway.)
And furthermore, did you know that pressed CDs are more likely to fail and have errors than CDRs? I bet you didn't know that. Well now you do and it was a free as in beer tip from your bud Steve in Taiwan.
The place I go through error checks all CDs. They have a 10% either way margin of error on the size of your run because of this.
I've never had a CD fail on me, ever. Never had anyone complain to me, either. Or one of my associates mention any of their discs failing. All my cd's are audio, so maybe that makes a difference. I don't deal with software, can't much about that. When I worked at a CD store, I also don't recall anyone ever returning a CD because it didn't work. Once someone did return a alternative cd because it actually had opera on it, though.
And, if you think you break even on a stamped disk at 1000 pieces I don't think you know what you're talking about. That might have been true five years ago when CDRs cost more than a buck. But guess what . .
The thing is, I'm putting out music CDs. Not software. I can't have them be CD-Rs, because that looks cheap. They have to be silver backed, they have to have silkscreened faces. Think about it, you plunk down ten or twelve bucks for a cd, open it and it is a cd-r with one of those sticker labels - are you gonna think you got your money's worth? No.
I don't care if it costs me 20-30 cents extra to look professional (at 1000 the price dips down to about 0.60 per disc)- it HAS to. If looking professional means the difference between me selling out a run and only selling half, or me being able to charge $6 for something that looks shoddy and $10 for something that looks good, I'm gonna go with professional-looking. I'd imagine it is good to look professional in whatever business you're in, too, but maybe not - I don't know your business.
I also like going through a company because, well, I do not have a steady employee. I don't put out 5,000 cds every week. I put out 5,000 one month, 1,000 two months later, 2,000 a week after that. It's not steady. I can take my master to a place on monday and have my 5,000 cds by friday. I couldn't get that turnover with an employee burning them in my office, and I don't want to pay someone to sit around when I don't need them, or have expensive equipment (and 1,000s of blank CDs lying around to go along with the regular CDs). I also don't wanna have to hire someone just for a weeks worth of work, nor do I wanna deal with a temp who is just going to mess things up somehow.
For _me_, getting them done is the way to go. They also handle getting the lyric inserts printed, shrinkwrapping them, etc. That way I don't have to sit there and put the inserts in a few thousand CDs, then shrink wrap them, nor do I have to pay someone an hourly wage to do that. This changes the dynamic of your estimates BIG TIME. How much more would it cost me to pay someone to do this, and how much longer would it take? It already would take me longer to burn them than have them made.
If you can do CD-Rs, and you're doing small run or have plenty of time to wait around, I'd say a burnstation is they way to go. Building some sort of contraption seems like a recipe for disaster, and if it's not, please post details about the contraption you built and how much time/effort/money you spent building it, compared to how much it saves you and how much it would have cost you to just buy an off-the-shelf automation solution.
Re:Kooks? (Score:1)
For me since I am doing them in the thousands and every penny counts, I silkscreen them by hand so that doesn't go into my costs.
And I was being a little snotty in my reply. You know, it was after work and I was playing Mr. Tough Guy. Didn't mean to be shitty and I was being a bit crass.
But I'm still gonna take a crack at this conveyer belt. My wife is gonna hate it, but I'm going to do it anyway.
Re:Kooks? (Score:1)
How does hand silk-screening them work for you? Do you find you mess up many? Do you use waterbased inks or oil? It seems like it must be oil.
Now that they have the silver-backed cd-rs, it's something I've considered for small runs.
Re:Kooks? (Score:1)
Alright, you're forcing me to reveal just how far my impractical DIY streak goes.
In fact, I'm still in the process of perfecting the silkscreen thing and I'll explain that in a bit of detail, but first I'll tell you what I do for the moment that isn't actually quite silkscreening, but is a kind of hand printing that looks cool enough that I'm proud to sell them.
For the moment, I use a rubber stamp dipped in standard art shop dayglo acrylic paint for a logo on the top and then I use flashy mirrored sticker with some text printed on it on the bottom. It looks pretty slick and any little imperfection still looks sweet with the mirrored background, but this technique doesn't enable me to cover the whole disc with the elaborate patterns that I've printed out as transparencies on my ink jet printer just for such use.
Oh, and by the way, I do use those all-silver CD-Rs you refer to. In fact, I've screwed up and printed the wrong sides of those things quite a few times so they really do look a lot like pressed CDs and I get them so cheap I can't afford not to use them.
As for the current state of the silkscreening adventure, it's a long story but I'll keep it condensed.
Apparently in the States you can get this stuff called Liquid Light which is a trade name for a prepared silkscren emulsion. From what I've read, that is the way to go and I probably should just buy some of this stuff and get on with it. Using that, you can find lots of directions from Google as it's a standard for T-shirts and what-not.
But myself being a DIY freak from hell, I decided to make my own emulsion from silver nitrate that I tracked down at a chemist shop here in Taipei. I got some okay screens using gelatin and albumin with some table salt. Well, by okay I mean they looked awesome when I developed them. The detail was crispy, Unfortunately, they didn't last long or resist the ink/paint well because I apparently needed to add this stuff called Chromium Tartrate which was a bit tougher to find although I have a good lead on it that I will pursue.
So, basically I must confess that wherever there's an easy and a hard way to go things, I'm likely to take the very roundabout way, which is one of the reasons I was being very self conscious about trying to look at the whole mechanized CDR thing from the simplest procedural vantage because I know one of my own weaknesses is always making things too complicated.
But not to fear, I love chemistry and elctronics and all such trivia so it makes sense in a twisted way. Any excuse to go cruising around the industrial chemical facilities outside of Taipei works for me. And as for makin a CD conveyer belt --hell, I relish the thought. I'll make the belt out of old bicycle inner-tubes. I love the details.
Thanks for asking though after I was so shitty up front.
Distributed Solution (Score:2, Funny)
Earn extra style points.... (Score:1)
if you can integrate this little guy [tomheroes.com] into your CD dupe project.
the only thing i have a problem with is... (Score:1)
im not looking at writing cd's so much as making like a juke box to switch through cd's, almost like a cd server. this is for home use so speed plays no role in this what so ever.
Re:the only thing i have a problem with is... (Score:1)
Re:the only thing i have a problem with is... (Score:2)
- a shopvac has more than enough power, even if the hose is sucking through the hole of the CDrom.
Actually, too much power, too much noise, too big, and to control it, you need to control a 120V motor.
- A wimpy CPU fan doesn't have enough power to hold a CDrom. Much less to pick one up if things aren't in perfect alignment. (This particular CPU fan wasn't powerfull enough to keep my Athlon cool, that's why it was available for experiments.)
So a question for those who know pneumatics:What is a good cheap, small, relatively quiet fan/blower/pump for a robotic pick and place CD changer?
I think I'll probably take apart a cordless vacuum cleaner and use its blower.
My application is unattended archiving of data to CDroms. 2 copies of each CDrom is enough for me. One copy of 10 Gbytes of data takes about 15 CDroms. Pretty cheap if there isn't labor involved in swapping the CDs.
Three words (Score:2)
Re:the only thing i have a problem with is... (Score:1)
vacuum sucker - What some industrial machines use
Low tack adhesive, like on post it notes
try to design it so that gravity does most of the work, i.e. drop them
off of the bottom of a stack onto the tray. - What some other industrial
machines use
Re:the only thing i have a problem with is... (Score:1)
Re:the only thing i have a problem with is... (Score:1)
PIAB vacuum pumps [piab.com]
I have used them on automation equipment before.
A small home air compressor is all you need.
Re:the only thing i have a problem with is... (Score:1)
im thinking something that drops down just the right height and expands to fill the cd hole. a circle cut in half perhaps with a rubberband around it and then have the half split or slid, as to break the circle shape and put pressure on the cd through the rubberband
what do you folks think? any other good ideas?
Re:the only thing i have a problem with is... (Score:1)
Re:the only thing i have a problem with is... (Score:2)
Who says you have to pick it up from the middle? Just use two suction cups on either side of the hole, or four for a pattern around the hole.
x
xox
x
or
xox
where o = the hole and x = suction cups
Re:the only thing i have a problem with is... (Score:1)
Re:the only thing i have a problem with is... (Score:1)
Re:the only thing i have a problem with is... (Score:2)
Once lowered they close in to grip the top CD. The angle should enable them to close in on only the top CD and lift that off the spindle leaving the others behind.
The main problem I can think of is that they will need to close a different amount for the top CD compared to the lowest CD. I don't know enough about robotics to know whether a pressure feedback sensor can be rigged up to steer this. When it "feels" that it has gripped the top CD it stops closing.
The other problem is that in my experience, CDs on a spindle won't separate from gravity alone. Lift the top one and at least one more will follow with it. Perhaps one would need to go through the spindle manually first and separate all the CDs.
I work in this field... (Score:1)
Consider the cost (Score:4, Insightful)
You have a fast CD burner (24x) with which you can make perhaps 10 CDs per hour. That means 1000 hours for 10,000 CDs, or about 6 months. Assuming a wage of $6/hr, to pay someone to babysit this machine would be in the range of $6000. Plus your media cost of around $5000 if you're paying $0.50 per, including label. You are looking at $11,000. Ouch!
On the other hand, you could purchase an automated duplicator for $2500. Yup, that's a lot of money to lay down in one chunk. Now you're down to a month and a half (your duplicator can crank away 24/7, your schmuck 8/5). Your cost? $7500.
Of course, you could build a changer out of Lego Mindstorms for a hundred bucks plus your labor and have it up and running by April... 2003.
Or, you could just pay a replication house to press the CD's, print a fancy label on them, and get them to you in a week for probably $5,000 or less (wild guess). Remember, as quantity goes up, price per goes down. Way down. Don't let the "setup charge" scare you; consider the total cost and compare that with the total cost of one-at-a-timing it.
My boss has a saying: "A poor man can't afford cheap tools." You don't save money by buying cheap. If you skimp now, you'll spend a lot more later. If you do it right the first time, you won't have to do it right the second time.
Re:Consider the cost (Score:2)
And the printing sucks too. You have to buy specially coated CDs to print on. The ink takes forever to dry if you put a lot on (which you would do to make the image look good), so we don't put a lot of ink on and you get this faded type image. If you put a lot of ink on, then the back side of the cd that gets placed on top of a cd in the out bin blots the ink. You get the image on the front and the back of the cd!
Robots are cool, but they are in no way low maintenance. Now if someone could make one that could be made smarter...that would be cool.
One word... (Score:2)
How to build... (Score:3, Informative)
CD Duplication is much easier.
Think of device as an "arm" that can move linearly on one axis, and travel up and down by a small amount on another orthogonal axis. So, basically a 2 axis pick-and-place arm.
Place two spindles on either side of the burning drive. One spindle is full of blanks, the other is empty (to hold burned blanks). Line the centers of the holes in the CDs up with each other, as well as with a CD in the drive tray with the tray ejected, so the all the holes fall perfectly in line. Mount the drive and the spindles down in some manner (screws, glue, something).
Now, you need to build an arm - a couple of cheap RC servos and some aluminium square tubing, maybe some threaded rod, so that it can move up and down, and move in and out along the line of the holes. Build a forked picking appendage out of aluminium tubing, with the ends of the fork bent down at a 90 degree angle - the clearance between the two "tines" of the forked tubing should be wide enough to clear the spindle. At the ends of the tines, attach cheap suction cups drilled through - seal them well to the tube ends. The fork needs to have a tee split off of it that will connect to a piece of silicone tubing that runs to an aquarium air pump - this tube will connect to the pumps air inlet (the pump may need modifications for this) to form a cheap, low cost, but efficient "vacuum pump".
The arms servos can be connected to a BASIC Stamp with appropriate driver software and hardware - the stamp can be programmed to simply accept commands to move the servos properly to certain amounts as sent over the serial port (via a MAX232). The Stamp will also need to be connected to some 120VAC relay (12 volt coil, 120VAC contacts), or a 120VAC solid state "relay" to allow control of the pump. Then code would have to be written to do the following:
1. Move the arm to the full spindle. Turn on vacuum pump, and lower fork to "suck up" the CD.
2. Lift arm, eject tray, position the CD, turn off pump to drop CD, inject tray.
3. Burn CD.
4. Eject CD, turn on vacuum pump, lower arm, "suck up" CD.
5. Move arm to second spindle, turn off pump and drop the CD - goto step 1.
Remember that at step one, your pile shrinks, and at step 5, the pile is growing, so you would need coding to account for that.
At any rate, such an arm could be easily built, probably for under $200.00 if you shopped carefully (and already had the burner).
Of course, if you don't have any experience building such devices (basically, homebrew robotics and electronics, plus coding) - you won't get very far, and I would have to concede that it would be more worth your time and money to purchase a machine, as other posters have reccommended...
Re:How to build... (Score:2)
*must* it be CDRs? (Score:2)
suction (Score:1)
The caviat is that supposedly this arm cost $20,000. I'd don't know what you would have to do to get a reasonable piece of equipment cheaper. If you used a vision system, you wouldn't need an arm that was quite a precise or stuck to it's calibration as well. But cheap arms (like the Radio Shack armitron line) aren't going to be as easy to get suction grippers for.
mindstorms news thread (Score:1)
I personally like the idea of the suction cups
idea for picking up the cd's..
http://groups.google.com/groups?hl=en&threadm=7