Hi everyone, I'm looking to lift a ~5 Kg weight vertically, with a speed of min 20/30 cm/sec. I was thinking to use a C-beam actuator lead screw, but I'm not sure if it will be fast enough. What would be the best solution? I also have to build a frame to support the actuator vertically, but I never had to work with these stuff, so I'm not sure what pieces to buy. But first, I need to know what actuator can do the job and then think about the frame. Furthermore, I was trying with a belt actuator I already have, but the belt skips and the whole solution seems a bit shaky. Any suggestion for a newbie?
An 8mm-lead screw can move 30cm in about 40 rotations, which per second ends up being 2400rpm. Anything over 800-1000rpm is an ask for a stepper motor, and when you're trying to move a load, it's better to stay below 400rpm- the torque curves are brutal. Remember you can use a counterweight to take some of the load off of the drive motor. If you do this, is the belt still impossible? This kind of speed- 12,000-18,000mm/min- is generally reserved for belt drive machines like lasers, so I'd try to stay in that area if possible. Upgrading from a GT2 belt to an HTD 5M belt might be a good idea, for example. This is a non-trivial project for hobby-grade components, so just be aware of what you're trying to get into. What are you trying to build?
Thanks for your reply. My build has to be as simple as stated. A vertical motion with a certain speed. I will need it to make perceptual experiments. From what I understand, a lead screw actuator won't be fast enough. I thought about mounting a counterweight for the belt one, but, I'm not sure how much I can stress my 3 mm with belt, and also if there is a power outage, my weight will collapse to the ground without any other breaking system (I'm not an expert and trying to keep things simple). I also thought of using two belt actuators that shares the same gantry plate, but not sure how to daisy chain the two uStepper drivers so that the actuators will move simultaneously without errors to avoid breaking the system. I liked the screw lead because it seems sturdier and in case of outage there is no risk for the load. So, what's the fastest speed I can reach in cm/sec with a lead screw actuator?
a 5kg load will easily backspin an 8mm pitch leadscrew The weight = massive loads at the speeds you expect, it is quite complicated in terms of the physics
I see. Thank you for your replies. I have a few more questions then: I need to see if I can reduce the weight to 3 Kg, but maybe it would still be not enough for a fast screw lead actuator? What about daisy-chaining my motors? Is there a guide somewhere I can use? Or is there a belt driven actuator that is more solid? At the moment, I have a 2GT-3 mm belt. Is there anything sturdier for my aim? Any safety measure in case of outage (breaking system or any other trick)? How do I mount a counterweight on my belt?
The direction of these questions suggests that not only did you not thoroughly read the replies already in this thread, but that you're simply substantially out of your depth in designing a system like this. It's not impossible, but I'd strongly recommend you take 3-4 months and research in detail how actuators work, how different drive systems (belt, screw, etc) work as force transformers and their limitations, the engineering design of machine frames in general, and Z-axes in particular, methods of stabilizing Z-axis loads, how CNC controllers interact with stepper drivers and motors and how to build the electronics of a CNC system, how stepper motors work in general, and how motors develop power and torque and the tradeoffs therein. That's just a start, you'll quickly find out how expansive the topic can get, but this is not an introductory beginner's project. It is, however, possible to gain a somewhat expansive topic knowledge in just a few months- I did it myself a few years ago. Having a project in mind is good encouragement to keep going as well.
3mm??? The belt is literally 3mm wide? On my 3d printer, which has no substantial loads, the belt is 6mm. I wouldn't be surprised if hanging a 5kg weight on the belt will stretch it. On my CNC router I have a NEMA 23 stepper on the X axis and the the weight is about 4kg-5kg. I can go as fast as 2450 mm/min before I get into problems with the lead screw.
OK, thanks. True, it was 6 mm as well (it was 3 mm pitch), but anyway it would be a bit slow for my needs indeed. I'll find an alternative solution.
Yeah, that was somewhat obvious by reading my first message thoroughly (hence, I think it's understandable I might not grasp 100% the meaning of the answers). For sure that is what I'd do, and what I usually do when I can, but this time I don't have the time for that. Not denying the field is complex, was looking for pointers at least, but no problem. I'll ask some colleague. Many times I discover it's just a matter of being face to face and discuss things.
This sounds like a great project @Davidino and really (at the moment), as the guys mentioned in the eco system we have here in order to keep cost down, we do not have an actuator that is capable of the scope of the project your working on. You could look into a more industrial automation system as there you may find the solution your after.
This will inevitably be time-consuming. It's the nature of the beast. I specifically gave you a list of 10 pointers to start your research. And there it is. These secretive-but-specific-and-complex types of questions are always for an organization, to avoid hiring consultants or engineers, or pulling them from other projects. People are happy to be paid for their labor and expertise, but asking the internet to do one's homework is generally frowned upon. If someone demonstrates that they've made significant headway on the project, or have done extensive research and are just stuck on some point of confusion, people love to share their knowledge, but asking for an entire solution from fundamentals that just looks like you've been tasked with as a stretch project... Not a great look. For safety, and to simply do the right thing, hire (or barter with!) someone with the requisite knowledge. I've done it before, probably a few people on this forum have. It's not unusual to do some small-project consulting for research teams, startups and the like.
No harm done , thank you guys, everyone did a good job working to help Dave and through the process if someone learns something along the way that helps them that's what counts.
Not really, we have lots of engineers, but I was trying a solo project, as I often did successfully. I didn't ask anyone the code or anything, maybe a few links were all I was thinking, and some yes/no answers. I like the satisfaction of doing it myself, even if usually, I have more time for it. Often too, online it's hard to talk, and experts say its impossible or complex, but, at times, there is a solution not that overthought around the corner. Maybe not in this case, but just saying (there's no harm in trying). I'll look into master slave system, that for my specific case, doesn't sound too complex, but I'll see. I'm happy if a direct solution implies a lot of background work, but at least it is useful to know what is the name of the work I have to do, less abstract. If too hard, I'll ask some ME. Thanks for your support guys! And I hope I didn't waste too much of your time.
@David the swarfer Love the auto tuning of these bad boys! Gets a little too deep for the average user needs but nice to know there's this option out there.
