Need some advice. We have this donated DIY CNC we’ve been slowly returning to an operational state over the past 6 months. Recently, we started trying some longer jobs on it. We ran a job cutting some shapes out of HDPE, but about an hour into running our X-Axis stepper locked up, seemingly due to overheating.
We’re having a difference of opinion as to whether this is a case of just needing to put heatsinks on the steppers (which I feel is a good idea either way), or if the issue is with the stepper motor at least on the x-axis needing to be replaced.
The system is running off a TinyG, with a PIPO x9 mini hosting the JSON server and running Chilipeppr.
stepper motors can get hot. You may have some binding on an axis that is making the motor work harder than it should? The motor simply may be under powered for the work it has to do too? Drive current may be too high. Those are the obvious things to look into. One thing to be aware of is that steppers draw the most current holding. Which is a bit counterintuitive. Some drives have an automatic current reduction circuit built into them to deal with that. Many do not.
Instead of just a passive heatsink it is not unheard of to mount little fans on stepper motors to keep them cool too. Salvaged from a PC, of course.
Thanks. In looking further into what the steppers currently on this are, they seen to be some ancient OEM ones which put out less than 90oz of torque, so underpowered it may very well be.
@Tony_Roman
Yeah 90 oz/in is not a whole lot. 200 oz/in is about the minimum you really want to use on a spindle machine. Unless you can go dual drive, or something like that. But that is not without its problems either.
With open loop machines the simple fix is to just have a lot more torque than is required, and hope for the best.
Another option is to gear the motor down, and go very slowly. That is a good money after bad call there though.
Stepper motors can be expensive to buy. Ones powerful enough to be used in CNC machines are especially in demand too. But if you can find a scrap commercial copier someplace, you might just get lucky then. You want one of them floor standing models about the size of a washing machine. They’ll have beefy motors in them. Probably a lot of toner too.
Big cities with office buildings are the place to hunt. Look for companies that lease copiers to the corporations. When the machines come back off lease they go right out the back door then too. They’re just not worth refurbishing.
@Paul_Frederick - We’ll probably try and scrape together the money for three of these, as based on what I’ve seen on the TinyG groups, they seem to be well received:
http://www.automationtechnologiesinc.com/products-page/nema-23/kl23h286-20-8b
@Tony_Roman
The high resistance, and inductance of those motors gives me some pause for concern. They have a lot of holding torque, but how much torque do they have at higher RPM?
Examine the torque curve chart on that page. Those motors are half down at 750 RPM. Plus that chart was made at 48, and 80 volts. Which somewhat masks the inductive reluctance issue those motors have. If you’re running lower voltage this will bite you harder.
What kind of drives do you have anyways?
TinyG board controlling it, nothing beyond that.
@Tony_Roman
Yeah TinyGs are pretty tiny. What PSU voltage are you running? 24V?
I had a very quick look at your blog and the specs for the Tiny G so forgive me if I am missing anything.
I am running a cheap 3040 3-axis mill that uses NEMA 23 motors and it easily shears small cutting tools (3mm dia.) if the machine crashes. I am using an Easy CNC controller which runs at 24V and at 75% power it’s output current is not much higher than the Tiny G.
Is your CNC using ball screws or an ACME thread? The first machine I used had an ACME thread and the friction of the thread used up most of the motor torque. Ball screws make a HUGE difference.
@Russell_Cameron
A lot of people ignore linear slides, and actuators. What kind of acme nuts were you using? I use custom anti-backlash ones I made out of HDPE. They’re low friction, being self lubricating. Not quite as good as ball screws, but still not too bad. Balls screws can be incredibly expensive.
@Paul_Frederick TinyG yes, running it at 24v.
@Russell_Cameron Outs is an old x/y table off of some other piece of machinery. It is pretty smooth, but that’s all I know about it.
@Tony_Roman
The PSU is another place DIY CNC often falls short of ideal. The challenge is hitting a specific target voltage with a fair amount of current. Your TinyG board is rated at 30V max, so 24V is a bit below that. You’re losing all of that power between the two values.
It is hard for me to briefly explain the significance of all of this. Suffice to say though that if I could demonstrate it to you, you may be suitably impressed. I’ve seen it myself here. I went from 24V to 28V and even that was noticeable. That netted me about a 25% speed improvement. Which is my case was going from slow, to not as slow. But I still appreciated it.
It all goes back to that inductive reluctance issue I tried to explain earlier. That is the critical factor with stepper motor performance. Voltage is how that is overcome too. Right now you’re in the butter zone where a little improvement could net you substantial gains.
I will add that if you are using an SMPS (which you probably are) they are somewhat adjustable too. But you do have to know what you’re doing to accomplish that.
You have to adjust the reference voltage inside the unit. Then the PSU will output accordingly. So here is a case where performance can be gained through expertise alone. Those opportunities don’t come around every day either.
If that TinyG says it can handle 30V I’d make it my business to feed it 29V. Yehaw. Actually it can handle more. But they used craptacular regulators on the board. Or maybe caps? The drive ICs are rated for 35V. Or maybe they’re just taking back EMF into account? That is another really tricky issue surrounding stepper motors. Something you learn about quickly when things catch fire.