What's new
What's new

Automation newbie question, I am more of a CAD guy

Neil

Hot Rolled
Joined
Jan 1, 2005
Location
Lockport, NY
Dear Forum,
Who are the top 3 suppliers of the vibrating boles, tracks and belts with pulleys I have seen in other factories? I am sure some of it is off the shelf right? I need to design a simple machine that inserts one part into another, plastic, friction fit. Low forces needed and it needs to be a step above a guy with a pair of special pliers doing it. I figure once I see what is out there and standard I can go from there. This is along the line of those nice allen key assembled frameworks that you can use to build machines. Just want to limit customization as much as possible. There is probably a feedback loop in there too, so the top suppliers for that stuff too.
I guess is there a site like McMaster Carr that is all automation components? Thank you so much. I am willing to learn ladder logic to do this (I guess a PLC will be needed) but I have never built anything like this before and want to keep it simple simple simple. Thoughts?
 
Suggest You educate Yourself.

No matter what anything like that will be 10.000$ or multiples thereof, ..
and there are endless critical-path issues.

Load, yerk, positioning, tolerance, lifetime, lots of stuff.

Your expecttions are probably wrong, as is Your expected budget.

Many of us are experts in motion control, and could do such assemblies in 2-4 ways in 10 minutes.
No-one will do Your work for You for free when You think it´s worth less than 100$ vs 8000 $.
 
@hanermo
Ok, thanks for your brutally honest appraisal. I'm just an intermediate CAD operator and mechanical engineer with some experience in a variety of industries designing parts. Its probably a lot harder than it looks is what I hear you saying. The guy asking me to design this as a total rookie is probably a bit too optimistic then. I'll let him know its probably 10k bare minimum. I'm reasonably smart but totally inexperienced when it comes to automation. Glad you experts can do it so fast and easily. I will try to stick to what I actually know. Anybody else like to weigh in?
 
I don't use the "one stop shopping" approach for automation projects because I want to hand pick every part to be the best for the application. Having said that, it may be a good approach for your situation. It's hard to say without more information, but for something simple like this most of what you want (extrusions, conveyors, linear actuators, sensors, etc) can be purchased from Misumi with a cheap PLC from Automation Direct to control it. Coincidentally, I've been using a lot of Misumi conveyors lately. Their website is well laid out and they even have example designs for a lot of their components, showing how you might solve a particular problem.

Misumi would be the simplest approach, but another option is to use a local rep that has a line card including all the big players in automation. The upside of this is you'll get some guidance on selecting and implementing stuff. For example, here in the Midwest we have Allied Automation which reps some of the big companies making extrusion, sensors, pneumatics, actuators, PLCs, etc. They offer phone/e-mail/onsite visit support and training. The downside is they'll push whoever they rep for a particular component and it might not be the best choice for your application.

I'm not sure what your cost expectations are and, again, I don't know your requirements, but I doubt you can do what you're wanting for under $20k. $30-$40k is more reasonable depending on bells and whistles.

Good luck. This sounds like a good introduction to automation, but I don't envy you having to do it without an experienced mentor to show you the ropes.
 
Dear Punkinhead-thanks for the quick and dirty intro. I sent to the Buffalo Public Central Library for some books from the stacks to look up some info on automation. Will keep you all posted. Thanks!
Punkinhead I was thinking of using simple pneumatic automation switches to control everything, or most things to keep the system as simple and cheap as possible. Pnumatic jaws, actuator, rotary actuator motor to turn the belts. Thoughts?
 
Neil,

You are literally a town away from me, anyway, I have not used them so I don't have first hand experience. However, Buffalo Manufacturing Works is a non profit consulting company in the old Niagara Forge building that helps companies gain the knowledge needed to get automated. Perhaps you could try giving them a call.
 
I was thinking of using simple pneumatic automation switches to control everything, or most things to keep the system as simple and cheap as possible. Pnumatic jaws, actuator, rotary actuator motor to turn the belts. Thoughts?
I've never used pneumatics to run a conveyor motor. Without knowing the details of your application I can't really comment other than to say I suspect it would be very hard to control speed, accel, and decel. Personally I'd use an inexpensive conveyor with a 3-phase motor and VFD or steppers for better control. Pallets banging into stops is like fingernails on a chalkboard to me. As far as the rest of the pneumatics for your pick, place, and press application, that's a standard approach when high precision isn't required. My background is using electrically actuated valves so I can't comment on using pneumatic switches. My feeling is if you have even a few signals requiring a controller you might as well control everything using I/O and a cheap PLC. Hell, if timing isn't critical you can do all the controls using a cheap PC and an ethernet or USB based I/O module. Check out offerings from Ontrak or the Advantech ADAM modules. We do that a lot of times for simple applications and write the control software in Visual Basic. That makes the HMI a piece of cake and any highschool kid can write the code.

Don't forget about guarding and a safety circuit. If your pneumatic cylinders are small enough you may get away with not having them, but I don't fool around when it comes to safety. I like to sleep well at night knowing my machines won't hurt anyone.
 
Punkinhead-
1. Who makes decent PLCs and who should I avoid?
2. I have an overarching theory about how to do a budgetary estimate for this-look up all the parts, or as many as possible on McMaster-Carr and understand that since they are usually more expensive, that is the MOST is could possibly cost and it would even take into account a few mistaken purchases, steps.
Thoughts?
 
I think you need to take a step back and draw out your concept on a white board. What are the process steps? Lay out the sequence of events in detail from the time an operator loads a part until a completed part is unloaded. How will you control each step? How will you errorproof each step?

Then you need to make a list of inputs and outputs so you get a count and know how to size your PLC or other controller. For example, you said it will have a conveyor. How will you locate the part? How will you decide when to turn the conveyor on & off? Do you need presence sensors for the parts or pallets? Are there upstream and/or downstream conveyors and, if so, how will you communicate between them (eg. SMEMA)? For your air cylinders, will you have sensors on each end of the stroke so you know where they're at or will you rely on timing? What kind of pneumatic valves will you use, 3/2 which will require an output to control each direction or will you use 5/2 to save an output on each cylinder? Do you need a center blocked valve for your Z axis to prevent it from dropping in the event of an e-stop or do you even care?

Lot of questions to answer ahead of picking a PLC.
 
Post a picture of the parts you want to assemble. Bowl feeders work great and the guys that get them to orient and spit out parts one at a time are magicians. I used to work for a company that made rail feeders that worked well with round or headed parts of any size and were very flexible. However they didn’t handle prismatic parts.
 
Punkinhead I will block it all out first and then get more focused on each process. From general to specific to be sure. That makes sense. Sorry, I am all over the place with this.
 
All other posters mostly reinforced my earlier message.

To OP:
The total cost should be seen as 5-10x the cost of the parts, more or less.
What I tried to indicate, at about 10k$ as a baseline cost, for industrial stuff.

You really want to document aka describe most-all likely errors and then mostly prove/document You have eliminated/reduced them as much as possible.
This requires lots of sensors and cabling at == 200$ each, installed.

You can do all the work for free, if You want.
Professionals will charge about 20K$ for a basic assy line of 100-200 sensors.
Plus parts and install - about double the cost of the kit.
Why would You get paid less ?

The legally required insurance is typically 2500£ - 20.000€, so your fees are of little matter overall.

So:
For me at import wholesale cost it might cost 5k£ in kit to automate a simple line, and maybe 45 days work.
My cost might be == 5k£, as importer, not yours, and not retail cost around maybe 9k in kit.
Most of the work is in docs, certified work (required in the uk, the eu, and generally in the us and oecd countries) so it does not come back to bite me later.
A professional engineer can usually be found to sign and stamp your installation for around 500$ - 1500$, the upper end being for someone from a big business doing that type of thing.


I did lots of automation that was very profitable for my clients, without any docs or liability wavers etc.
20 years ago.
Today I would do the same, for 10x the money - with signed no-liability wavers.

99% of the sensors are to avoid totally silly errors and situations.
That should never happen.
AND ALMOST NEVER HAPPEN.
So what ?

HAAS USA that I know very well, spends more than 1M$ / yr on liability /sw/certs.
I´ve been there, seen the guys.

Where the client expects a 2000$ project he gets a 20-40k$ proposal.
So what ?
Either do it, or do it ?

The client will not get it cheaper and better anywhere.
Industrial-level automation just costs some money, no matter what.
Big firms would charge 100k for the same thing, just with better documentation, powerpoint presentations, and legalese (of no real value).
 
Oh, God, don't do any of that.

Get a $7000 SCARA robot, program it do do whatever you need. Or a cobot if you're made of money, lazy and don't want to learn guarding.

Fixed automation is fine if the product is going to never change, have a long lifecycyle and you know what you're doing.

But a robot,the robot can run the whole show, no PLC needed. Fixing an oops is a code change in five minutes, not days re-machining something.

Plus a SCARA these days is literally cheaper than three nice guided cylinders, sensors, valves the hardware to assemble them and a PLC to run it.

Then when the product changes, you just repurpose the robot cell to make the next thing.

But, if you don't want to listen to me, www.automationdirect.com has most of what you'd want for electrical bits and bobs.
 
Here's an example of a $7000 SCARA. I've used a few and they're fine. In fact, I just added one to a design last week for a simple pick & place. Built in controller, so no PLC needed. It'll give you some I/O and a programming environment, plus there are built in HMI tools. It might be able to do the press fit the OP mentioned if the forces are low enough. They also make one with a longer reach. Epson SCARA T3 Series - Epson
 








 
Back
Top