I don't think that's fair. Why would say it would take months? I've tackled several projects with arduino and it was easy peasy in my book. And all I did was copy/paste the widely available example code into my sketch with a few modifications.

It's not the Arduino per se, just the many details of motion control that are non-obvious and complex, if you are not well versed.
Don't get me wrong- the Arduino is an excellent platform.

 

Look I keep trying to guide you back to arduino because my impression is that you're just a guy trying to make a living selling rain spouts and you need a machine to get the job done. I'm trying to get you toward the simplest and cheapest solution that works.

If I'm wrong about that; If you are part of a large organization and you've been given a substantial budget to complete this task, then by all means, put a PLC in it.

But this is not a cheap PLC application. It is an expensive PLC application (that a $30 ArduinoMega is fully capable of doing). If you can afford to spend money (I'd budget at least $1500 USD) on this and do it the proper industrial way, let me know and I'll shut up about the Arduino. But if you're willing to give it another shot, below is a reference of how to control stepper drivers with arduino. It describes a "easydriver" board but that can be substituted for any step/dir control like the gecko drivers or chinese drivers.

http://www.schmalzhaus.com/EasyDriver/Examples/EasyDriverExamples.html

Your spot on .I'm just a guy. From new Zealand with limited funds. Spent the last two yrs building a tool and now I need to build the machine that will supply the tool it's components. Not to mention I love this stuff. It's a hobby for me.
So in short. Please don't shut up.
Would you agree that it doesn't matter what way I go, I still need the same size steppers?
I already own an arduino. So it's just power supply and drivers and relays.
What I'm getting at is maybe I try it the cheap way with arduino first and if it's not cutting the mustard I can upgrade the drivers etc and retain the power supply and steppers.
I'd like yours and Max's thoughts on that.
Cheers
Mark
Ps.bring on the Americas cup! We've got peddle power this time lol

 

I don't think that's fair. Why would say it would take months? I've tackled several projects with arduino and it was easy peasy in my book. And all I did was copy/paste the widely available example code into my sketch with a few modifications.

My only reasons for staying away from arduino in this application would be :
1. If it is to be sold. I would not want to buy an arduino-controlled machine for the sake of serviceability, and I doubt anyone else would either.
2. If I expect anyone other than myself to be able to troubleshoot it ( I guess that's just a rewording of #1)

This is the only machine for new zealand market. I will run it. Anything I sell offshore will be industrial .but that's a couple or more yrs away.
Cheers Mark

 

This is the only machine for new zealand market. I will run it. Anything I sell offshore will be industrial .but that's a couple or more yrs away.
Cheers Mark

Your spot on .I'm just a guy. From new Zealand with limited funds. Spent the last two yrs building a tool and now I need to build the machine that will supply the tool it's components. Not to mention I love this stuff. It's a hobby for me.
So in short. Please don't shut up.
Would you agree that it doesn't matter what way I go, I still need the same size steppers?
I already own an arduino. So it's just power supply and drivers and relays.
What I'm getting at is maybe I try it the cheap way with arduino first and if it's not cutting the mustard I can upgrade the drivers etc and retain the power supply and steppers.
I'd like yours and Max's thoughts on that.
Cheers
Mark
Ps.bring on the Americas cup! We've got peddle power this time lol

Ok then I would definitely put in the effort to get the arduino solution off the ground before resorting to industrial stuff. Cut your teeth on that and if you decide to sell the machines in the future, setting them up with industrial components will be a breeze.

What @Sensacell alludes to is the bane of anyone trying to build a CNC machine with hobbyist grade stuff (arduino, etc.) And it's a real problem. The problem in my experience shows its ugly face when you are trying to synchronize multiple axes; trying to maintain accurate position in 3 dimensions with 3 or more different motors, all who have their own position error, lag, inertia, etc.

But I suspect your machine will be much simpler than that, and lend itself well to arduino control. You will be working with sequences; one motor turning at a time. Maybe more than one at a time, but not needing to moving in perfect choreography with one another.

As far as your question about motor sizes, it does not matter what the "brain" is; PLC, arduino, motion controller, etc. As far as I know, we have sized the main bending motor properly per the info you provided. As for the other motors, I have no idea. We haven't talked much about them, but you stated tbe sizes you plan to use, and I assume you did your homework about the torque required for each axis. If you want to discuss those motors as well, we can discuss them.

 

Ok then I would definitely put in the effort to get the arduino solution off the ground before resorting to industrial stuff. Cut your teeth on that and if you decide to sell the machines in the future, setting them up with industrial components will be a breeze.

What @Sensacell alludes to is the bane of anyone trying to build a CNC machine with hobbyist grade stuff (arduino, etc.) And it's a real problem. The problem in my experience shows its ugly face when you are trying to synchronize multiple axes; trying to maintain accurate position in 3 dimensions with 3 or more different motors, all who have their own position error, lag, inertia, etc.

But I suspect your machine will be much simpler than that, and lend itself well to arduino control. You will be working with sequences; one motor turning at a time. Maybe more than one at a time, but not needing to moving in perfect choreography with one another.

As far as your question about motor sizes, it does not matter what the "brain" is; PLC, arduino, motion controller, etc. As far as I know, we have sized the main bending motor properly per the info you provided. As for the other motors, I have no idea. We haven't talked much about them, but you stated tbe sizes you plan to use, and I assume you did your homework about the torque required for each axis. If you want to discuss those motors as well, we can discuss them.

All of the other motors are for 10mm trapezoidal lead screws and I'm trying to wrap my head round the linear thrust nm provided. I'm thinking it would be comparable to worm gear calculations, but I'm unsure.
Also,Max mentioned a basic feedback of incorporating limit switches i think. If i slow the machine down( have one motor finish its job before the next starts) do you think it would help make it all a little simpler? This way i could play with the code by removing the switches and start overlapping some of the operations as my competence grows.

 

The idea of L.S.'s was in order to use open loop DC or BLDC motors etc, this may only be feasible if the motors are geared and stop in sufficient time when the LS trips, the fact that you are using point to point motion might make this a feasible alternative.
Although many use stepper motors I traditionally do not like them and prefer the DC/BLDC versions.
It is difficult enough to design a machine on site, even harder when trying to give advice remotely.
There is a motor sizing programs out there that may help, such as Kollmorgen graphic design.
Max.

 

The idea of L.S.'s was in order to use open loop DC or BLDC motors etc, this may only be feasible if the motors are geared and stop in sufficient time when the LS trips, the fact that you are using point to point motion might make this a feasible alternative.
Although many use stepper motors I traditionally do not like them and prefer the DC/BLDC versions.
It is difficult enough to design a machine on site, even harder when trying to give advice remotely.
There is a motor sizing programs out there that may help, such as Kollmorgen graphic design.
Max.

http://www.omc-stepperonline.com/ge...ue-nema-23-stepper-23hs302804spg47-p-282.html

http://www.omc-stepperonline.com/bi...-40vdc-input-16-subdivision-st6600-p-246.html

http://www.omc-stepperonline.com/sw...04a-for-cnc-router-115v230v-s50048-p-168.html

In conjunction with the arduino? What do you guys reckon. I figure if the stepper is not quite powerful enough, i can use it on another part. but its a start. to get me underway. less than a couple of hundred $ . i would like to know i can add more steppers though. Is this all i need at this stage?
cheers mark

 

Sorry, who or what is Kook?

 

In conjunction with the arduino? What do you guys reckon. I figure if the stepper is not quite powerful enough, i can use it on another part. but its a start. to get me underway. less than a couple of hundred $ . i would like to know i can add more steppers though. Is this all i need at this stage?
cheers mark

You could always bring in one of each for assessment, I have never used the Arduino.
Max.

 

I have never used the $19 chinese stepper driver so I can't actually vouch for it. That being said, if it were my machine and my money, I would try the cheap driver first. If this were a CNC application I would use a gecko driver, but I think for this application chinese would be fine. As for the power supply, others have recommended against a switching power supply like the one you chose. I've never had issues with switching P/S but others have, so if you're going to buy something, might as well be a linear supply like this.

 

After looking at this thread since the beginning, I have one big question about the direction it's going. And that is, is this a real machine to make real things or is it a 'school' assignment? After making my living in metal work and related industry, this whole concept isn't how it would be done. "Drill motors" and steppers to do this is not a 'real world' concept for a manufacturing machine to make this type of product. More like a college engineering assignment.

 

After looking at this thread since the beginning, I have one big question about the direction it's going. And that is, is this a real machine to make real things or is it a 'school' assignment? After making my living in metal work and related industry, this whole concept isn't how it would be done. "Drill motors" and steppers to do this is not a 'real world' concept for a manufacturing machine to make this type of product. More like a college engineering assignment.

It's a real machine to make real things. What motors would you use? The reason the thread looks as you say , is because these guys are basically educating (or trying too)a blank canvas. Your idea of cutting blanks then doing multiple steps sounds a little industrial age sorry. The prototype works, needs automation.
Cheers

 

I have never used the $19 chinese stepper driver so I can't actually vouch for it. That being said, if it were my machine and my money, I would try the cheap driver first. If this were a CNC application I would use a gecko driver, but I think for this application chinese would be fine. As for the power supply, others have recommended against a switching power supply like the one you chose. I've never had issues with switching P/S but others have, so if you're going to buy something, might as well be a linear supply like this.

Done, max reckons linear too so that's what I get. shortbus seems to think steppers are no good at all, but no suggestions yet for alternatives, have I gone down the wrong rabbit hole?

 

You could always bring in one of each for assessment, I have never used the Arduino.
Max.

Yea, I will, but I think I'll go with the linear power supply and make it a good one. And a good stepper too.

 

Done, max reckons linear too so that's what I get. shortbus seems to think steppers are no good at all, but no suggestions yet for alternatives, have I gone down the wrong rabbit hole?

The alternative using DC or BLDC motors for motion control require some kind of feed back, usually an encoder, the alternative as I mentioned earlier is open loop with L.S. positioning, as long as any slight overshoot is acceptable IF it occurs.
Max.

 

The alternative using DC or BLDC motors for motion control require some kind of feed back, usually an encoder, the alternative as I mentioned earlier is open loop with L.S. positioning, as long as any slight overshoot is acceptable IF it occurs.
Max.

After some more thought, Im positive I can get away with limit switches on 5 of the motors, the feed,cut and extract . Still need high degree of accuracy on the main rolling motor. Would this simplify things? And can I still easily incorporate the large stepper ? Do I still need the same drives for bldc?

 

No servo drives are different for servo's than steppers, you can get drives for DC or BLDC that have either step/dir or ±10vdc analogue command, many BLDC drives are also capable of driving DC brushed motors.
Max.

 

It's a real machine to make real things. What motors would you use? The reason the thread looks as you say , is because these guys are basically educating (or trying too)a blank canvas. Your idea of cutting blanks then doing multiple steps sounds a little industrial age sorry. The prototype works, needs automation.
Cheers

shortbus seems to think steppers are no good at all,

Never said steppers were no good, just not for something like this.

Have you ever seen a sheetmetal shear? I assume that's what your calling a 'guillotine' Never saw one that used a lead screw to move the blade. Part of shearing is speed, don't think you will get that with a lead screw and stepper.

Where in this is your seam(to join the tube together) getting done? That in production sheetmetal work is done in a separate seam machine, by a series of rollers/wheels to form it.

Tubes like this are also ,in production, done by rollers/wheels in a linear fashion. Not in a 'slip roll' like they do one at a time. Never saw a 'slip roll' done like your picture. The rolls are usually two rolls that "pinch" the sheet, and a third one to form it to go around the top roll of the pinch pair.

There's a reason why it is still done "industrial age" as you said. In all the years it's what works the best. People that design stuff like this haven't figured out a better way.

 

Never said steppers were no good, just not for something like this.

Have you ever seen a sheetmetal shear? I assume that's what your calling a 'guillotine' Never saw one that used a lead screw to move the blade. Part of shearing is speed, don't think you will get that with a lead screw and stepper.

Where in this is your seam(to join the tube together) getting done? That in production sheetmetal work is done in a separate seam machine, by a series of rollers/wheels to form it.

Tubes like this are also ,in production, done by rollers/wheels in a linear fashion. Not in a 'slip roll' like they do one at a time. Never saw a 'slip roll' done like your picture. The rolls are usually two rolls that "pinch" the sheet, and a third one to form it to go around the top roll of the pinch pair.

There's a reason why it is still done "industrial age" as you said. In all the years it's what works the best. People that design stuff like this haven't figured out a better way.

 

OK. So. For the last 30 yrs I've been working with sheetmetal. Pipes are manufactured using both methods of slip rollers and/or roll formers( rollformer is what my bizuness is). As my product is a 50mm long pipe 76 dia roll forming is not an option. The method is called lockseaming and I'm doing it with small rollers which have a slot and said slot grabs and folds .a sheer is different than a guilotine, it has a blade that comes down evenly, guilotines are pivoted one end, so with a guillotine, it will depend on leverage and thrust from a lead screw, I do agree however that I may not get enough speed from a stepper. But I'm trying to keep it simple and not use hydrolic or air.. maybe a power hammer scenario could help here? A cam maybe? Or wheel slitters like on the front of a set of folders. Either way, that's the easy stuff, im trying to keep everything working from rotational force, weather chain,gear or linear. I figure it's going to be easier to program or code if it's all rpm x steps to adjust.