I am a millwright for a local company. We have this burn table. The controller went out and the foreman wants an upgrade. I came across open build and looks to be a good solution for a few reasons. I don't think I understand nema ratings enough to spec out stepper size. The existing drives and motors on the burn table are servos (2 motors in the y and 1 in the x) and the specifications do not directly transfer over to steppers. Esab Shadow 2 Questions: Is a high torque NEMA 23 stepper a good motor for this project? Do you think open build is a good program for a semi-automatic short batch run production environment? Is there a way to interface the existing servos with the open build? Thanks!
Post some pics of the machine, should give us a better idea of what you are expecting. Is it belt, chain or leadscrew driven? Yes No For further reading, see start [OpenBuilds Documentation] and in particular https://docs.openbuilds.com/blackbox
Based on holding torque and output shaft dia. The nema 23 high torque steppers seem too small. I can't find the specs on the servos to find a comparable stepper. I am sorta stuck without knowing how to size a motor for this. It seems the black box controller is rated for 4 amps per channel, do I probably shouldn't go bigger than the ones sold here too. The mechanical drive is a rack and pinion style. I'm going to go to that plant tomorrow and see if I can push the gantry by hand maybe I can guesstimate on how much horsepower needs and maybe work my way backwards from there. I really want to be able to use openbuild on this. It looks badass. But need some help. I am in over my head on this one Thanks Albert j
The motors and drives look pretty good. CMC have changed hands, but Advanced Motion Controls still exist. They have manuals and app notes for their stuff available. Moving to stepper drive will slow the machine down considerably- if you mainly cut thicker stuff, that might not be an issue other than losing some rapids. Don't generally consider steppers to be usable over about 400rpm- look for that point in their torque curves, etc. If that works for your rack and pinions at the machine/cable/hose weight required and gets you to your cutting speed, a stepper swap is definitely going to be the easiest way to go and opens up basically all your grbl hardware options. You don't need to worry too much about extra overhead power since you're not using physical cutting tools. However, if you have to go NEMA34- which wouldn't surprise me, that's a beefy-looking machine- then your easiest option there is to go closed loop (gets you a touch more power at any given rpm) because the kits include the matched drivers and are widely available fairly inexpensively. You can still use grbl, but you'd need hardware that easily allows for breaking out the STP/DIR pins to the stepper drivers. Doable with BlackBox, but not convenient and definitely not cost-effective if you don't already have one. Though you could still use OpenBuilds CONTROL with any other grbl hardware, even a plain Arduino Uno. If those are analog servo drives running in velocity mode- which I'm guessing, with all the "signal"s and "ref"s and "tachometer"s on there- and you determine it's best to keep those for high-speed torque requirements or whatever, you'll need a control that can deliver 0-10V analog output for all three axes. Which to my knowledge off-hand, is only LinuxCNC with a Mesa 6i25 and probably a 7i77, though I'm sure there are actually other options out there. That's a... Pretty complex setup (I have a 6i25/7i76 setup on a mill), I wouldn't recommend it for a beginner on a time budget, though once it's actually done it's very powerful and pretty bulletproof for a working environment. In an ideal world, you can get away with a NEMA23 stepper swap, if that'll give you the speed/torque requirements you need.
You definitely understand my conundrum. You seem to be more experienced with servos than I am. I remember some stuff from school, but it was 100% plc controlled. I am not sure how to make this happen in this instance. I stayed up late last night looking at steppers and specifics for each one. 400rpm isn't bad for rapid. Each stepper seems to be max 1000rpm. With the size of the sprocket. It comes out to be 600~ ipm which I think is plenty. But, something seems way wrong with that calc considering the 1/4" output shaft. I am afraid of snapping it. Especially with 1.5~ ft pounds of torque. I am headed to that plant this afternoon and I am going to pull a servo and move the gantry by hand to est HP needed. I am having serious trepidations about undergoing this conversation. I don't know enough about servos and steppers to really be confident in my actions here. I'm not sure if I want to monkey around to with this for days. If I could simply slap a couple steppers on this and buy the openbuild controller or hook a laptop up for under $2000 I would do this. But, I am not sure.
They all say they are. They're not. Figure 400rpm max under load. Acceleration settings matter here too- they're fairly easy to stall, especially open-loop steppers. They're not like servos where they know the motor is struggling and can dump more current to overpower the load- the step commutations are electronic. Think of them kinda like 200-pole brushless motors. So on an open-loop stepper, where the driver has no idea where the motor is, it just keeps commutating the current and the motor has stalled way back because the magnetic field has moved further on than it can recover. Makes a terrible grinding noise too, but it's pretty harmless. Yeah, not sure about that. 1/4" shafts are pretty puny but they do take a fair bit of force, plus you have the mechanical advantage of the pinion gearing. That said, NEMA34s come with either 1/2" or 14mm shafts, usually, for the 3-9 ftlb range. If you can find a NEMA24, they often come with 8mm shafts which is a little better. Not as common a size though, usually run around 2ftlb. A sensible mindset to take when going in semi-blind. It's a very large topic and takes years to really fully start to grasp, many decades to become an expert, I'm sure. It's not too hard to get up to speed with the basics though. Others have definitely done similar work to this in the past that you can probably use to help. Any of the solutions I offered would be comfortably under $1000- even the LinuxCNC one. Of course you can add time and additional materials (cutting motor flange mount adapter plates, say) but it's not an enormously expensive- or even complex- project, just takes some specialist knowledge. Which of course isn't your friend when you're on a deadline.
As for the trepidation, perhaps, checkout the other end of the scale, a hobbyist orientated plasma build, by Dr-D Flo (videos embedded below), to see how it goes together "from scratch" - to give you some idea of the territory ahead: and corrections to some statements in part 1, see the followup: His videos talk about the details a lot. This is a machine based on the LEAD1010 platform, leadscrew driven, etc - ie different, but still, i think, worth a watch to give you some in depth overview
Great information! Thanks for the videos. I will check them out tomorrow! I took a servo off to inspect it. It's a custom motor. I can't find information on it because of that point. The pn# is 3528t00m2000tac-ag. I think the 2000 maybe oz inches. So maybe 10 ft lbs of holding force? That seems reasonable. The output shaft is 1/2" not keyed. It has a 30:1 gear box on it. A 1.625~ sprocket on the output of the gear box. I took the take up spring off and was able to push it with about 4~ ft lbs of force. So maybe your high torque options will work if I use them in tandem in the y axis. If you can recommend 3 steppers with above considerations and a 1/2" output shaft. I can make a simple adapter plate if holes don't align. I am ready to present this as an option to the foreman. Likely the best option to get this junk pile working again. I attached a picture of the motor and gear box. Thanks for your consideration.
