153 replies to this topic

[PyroSam]

Interesting device, though it won't handle the large motors we need to use.

http://www.pc-contro...perbee_info.htm

Now to find something similar that will.

[digger]

gareth

what made you opt for pic as a micro controller for your work as opposed to other mcu's?

No real reason other than there is a bewildering number of variants with all sorts of built in interfaces. They have been around for ages so they are not likely to disappear anytime soon. The assembler instruction set is pretty simple, I learned assembler programming over 25 years ago when I was a kid, so though I should be able to pick it up again. There are a number of great development kits out there with well developed high level programming and testing interfaces (these are what I use, and only look at the assembler when I need to optimise code).

Yes there are other chips available which have on board high level languages such as basic. However I wanted to go for something I can optimise myself without relying on interpreted operating code.

But that is just my choice. There is no reason why any of the others aren't perfectly adequate for this sort of job, such as BASIC Stamp chips (A very good choice for a beginner).

[digger]

Interesting device, though it won't handle the large motors we need to use.

http://www.pc-contro...perbee_info.htm

Now to find something similar that will.

Interesting device that should make it easy to program for (could do it in excel!)

I have not looked at the design too closely, is there any way to have a slave high current output stage?

P.S. when I have unpacked all of my stuff after moving house I will dig out my notes (this may take a few weeks at least).

Edited by digger, 13 July 2012 - 07:16 AM.

[PyroSam]

Hi Digger (& all),

I'm not well versed in stepper/servo motors and I've been trying to learn a bit but my work schedule is horrendous right now (rotating shifts and lots of hours) so it seems I'm always tired when I'm trying to read up and that doesn't bode well for retention or understanding.

I may be way off here but it's appearing to me there are a couple ways to control stepper motors.

The first is to control the motor directly. This can be kind of tricky as the methods used to control the motor tend to vary with the design of the motors - things like the inductance of the coils, how much power it takes to get the motor moving from a dead stop, etc. come into play so the controller you build may not work with all the different brands of motors out there.

The second is to purchase a controller with the motors. My understanding is that those controllers receive a more or less standardized set of input signals and then convert those into something the motors can use. The standardized input signals are sent via specilized control software such as Mach3, LinuxCNC, etc.. That software is programmed via a programming convention called "G Code". The G code can be programmed directly or, more often, it's ouputted from a piece CAM (Computer Aided Manufacturing) software.

Am I understanding the whole process correctly? If so, then I would think we would want to create something along the lines of option 2 - by supplying G code output one wouldn't be locked to a certain motor - just get a motor and matching controller than can be instructed by G code and your good to go.

This doesn't negate PICs and stand alone control solutions at all, in fact I would think it may make things easier as the messy motor controlling is the job of the motor controller and the PIC would just need to send G code output based on the variables selected or entered by the user.

Thoughts, suggestions, corrections?

[PyroSam]

Well heck, it looks like this version of the Stepper Bee may just fill the bill!

http://www.pc-contro...e_plus_info.htm

[PyroSam]

Wow, just been reading the manual for the stepper bee and I think it'll work nicely for our application. I'm going to do a little more research, but I think it's time to step up the CAD drawings and get a machine built so I can play with this software!

[barnsley-mark]

Nice find, in the past I have used pc to drive a board to control the stepper board in order to get higher current handling, this makes it easier and 7 a/ph is a massive stepper motor.

[PyroSam]

The caveat with the Stepper Bee + is that it only controls unipolar motors, not bipolar. Bipolars seem to be more powerful for their size so one may have to step up to a NEMA 34 motor to get the torque they want. Still, for the simplicity of the device it's probably a small trade off.

[cooperman435]

sensing would be easier to do on the machine side if the simalar setup is used as per WASP does now, then the sensor would be part of the wired machine too rather than an extention. Definately a good idea though.

Flow chart is pretty easy to do and I suggest a set of variables that can alter the program that is written to then run per shell, pretty much as they do already. that way everyone can tweak their program to suit their needs and shell sizes?

If youd be hapy to let me join in the discussions youve had by PM Id really like to be involved?


Ive always liked the idea of a gluing station rather than wetting, you could use MUCH cheaper paper rolls and apply glue to them rather than glued tape

[PyroSam]

Hi Cooperman,

What do you want to sense? The end of the tape? The amount of tape laid down? Or??? The original machine doesn't have any external sensors, though we could add some if we wish. The laying down of the tape is simply done by algorithm with no feedback.

I agree a flow chart would be nice and I think ultimately the taping algorithm will be distilled down to a formula with variables plugged in - users could then easily modify the variables much like they do on the original.

I would be happy to involve you - I'm not trying to hide anything from anyone, it's just that I was contacted privately and since it was obviously intended to be a private conversation I've honored their intent.

I've wondered about a gluing station also as it seems that it depends on the humidity in one's climate as to whether you can wet the tape near the machine or you need to pull the wetting station back. Also, it seems wheat paste makes a bit more brittle casing to me. I thought that would be a subject for later tweaking though - let's just get one built and working and then work on improvements. In this area though, I'm thinking one could apply glue and then use a serrated screat like you do when you lay down glue for floor tile to get the proper amount of glue down without complicated metering equipment? Something to think about.

I'm actually getting kind of excited here - I wish the Stepper Bee Plus would do bipolar motors, but I can step up to a NEMA 34 and make do.

[cooperman435]

Sensing the end of the tape yep, anything such as a lever switch or simalar woud do it, just would be easier on the machine its self as its alrady cot wires n connections to it.

Theres LOTS of glue to choose from, if neceserry use dextrin as they do now its cheap and strong, PVA is a good one but I by choice moved the station a LONG way back every time, doing this allowed the glue to get very sticky and as the paper had absorbed plenty of water it had relaxed, when it dried out the shells were much tighter and smother. Experimenting is the joy :-)

[PyroSam]

Hmmm, this unit allows you to talk the motors through a serial port by just issuing simple commands to it. Heck, I bet a batch file could be used! (not pretty, but I bet it would work!). It's also a bipolar driver so one can take advantage of the cheaper higher torque bipolar motors. I even read in one of the forums where a guy was able to use a usb to serial adapter on a lap top that didn't have a traditional serial port - spiffy! Though it's listed as a kit, it apparently isn't as I've seen it on serval sites and it's always listed as fully assembled and tested. There seems to be some question on it's current handling capability though as many sites list it as capable of 8 amps but the pdf documentation says 6. You'd probably need heat sinks and a fan for 6 amps but that's a minor cost. It doesn't report inputs back for external things like the stepper bee does but that capability could be added through kits that use the parallel or serial ports.

I've ordered a copy of Visual Basic and a basic learning book for it as I've never used Visual Basic, just QBasic and PowerBasic (yeah, I'm old! LOL!).

http://store.qkits.c...fo.cfm/KT-5197A

Not really related (yet), but I ran across some stuff on a new tiny little computer called a Raspberry Pi that looks really spiffy for use in experimenter projects. It's not a mature product yet, but I'm sure it will be soon so it may be worth keeping on eye on.

[parabolic]

Pyrosam,

Forget about G codes, they are not whats needed. G codes are canned cycles used in CNC milling machines or lathes ect.. They are there to aid part programing to shorten the length of program.

Again PIC chip and the relevant program is all you need

[PyroSam]

Hi Parabolic,

I agree a PIC would be a much more elegant solution, I just don't have the time to learn it.

I'm thinking the KT-5197A controller looks like the most expedient path. You just write ACII to the serial port to control the motors - I'm sure I can whoop a Visual Basic program up to do that in a pretty package without too much trouble.

Doesn't mean that's the final word on it - it just appears to be the fastest path for now, if myself or someone else takes the time to learn PIC programming and gets a controller up and running it'd be a simple retrofit.

I'm slowly working on the CADs now so I can hash out a design I want to build. While that's going on I've ordered a copy of Visual Basic and will start brushing up on my basic programming skills. I'm holding off on ordering motors and stuff until I have a design more firmed up.

Stay tuned, more to come!

[parabolic]

Good luck friend!,

If you need help with any cad, when I get back from turkey I can whip out the solid works.