Printable View
I've been wanting to do this project for a while. I wondered if it would be possible to build a spindle monitor to check if it has stalled out and then e-stop. Hopefully this would stop it breaking end mills. Sounds easy in principle . . .
https://youtu.be/QuVHUmiW8xo
I would exxpect the VFD to have a motor stall function in it's parameters, the ABB one I use certainly has although I have not tested it in practice.
Maybe but I've not found it in the 'documentation'. Would simplify things if it did.
Episode 2, just making it better all round . . .
https://youtu.be/nSVjT4hpeaE
These inverter parameters look interesting;
PD118, PD119, PD120, PD121, PD122, PD123, PD124, PD125
Reading them now. Looks interesting! Worth experimenting with to see if it really can detect a stall through current monitoring and then trip the onboard relay (although I think that is omitted on my model, although I might be mixing it up with the inverter inspection I did for the drill press project)Quote:
Originally Posted by EddyCurrent
External current monitoring did not seem to show much but maybe internally it is reading a different part of the circuit.
Might turn out to be the biggest waste of time project (except for the fun and learning)
I decided to look into the inverter parameters . . .
https://youtu.be/rUjTy3utnsc
Prototype PCB!
https://youtu.be/cuSpk5dGZ4c
More developments, some hardware but mostly software including storing user values in non-volatile memory. Also a look ahead to the next PCB revision . . .
https://youtu.be/O_N773pcSwo
New PCB boards to split the LCD display and the processor. Went smoothly apart from the bits that didn't.
https://youtu.be/uu9BClLFsGY
I finally get the controller installed . . . after more tinkering!
https://youtu.be/hDzTCiMsE7Y
Thanks for sharing this @routercnc, love seeing this kind of stuff come through on the forum, special hats off to you for always going the distance.
Time for episode 8 - Hall effect sensors. Should be fairly straightforward . . .
Can't seem to paste video direct, forum buttons have disappeared!
Youtube link below:
https://youtu.be/gKwJzVpfJPk
I had another idea, seems so obvious now . . .
https://youtu.be/bcbtGN6m8Rs
Thanks for embedding the video links Lee!
Time to test this thing out for real !
https://youtu.be/3iZA39cHo2g
Without going back over your design decisions, was there a reason not to use a "spindle at speed" or "spindle stalled" signal from the inverter, rather than have to create your own detection system? Sorry if I misunderstood the process.
On my large milling machine, nothing's going to stall the spindle unless I'm running a large face mill, in which case the axis drives would stall before the cutter broke. Smaller cutters go ping if they get gummed up.
That's an impressive machine you've built there, with great attention to detail!
I looked into the inverter controls in one of the earlier videos and it doesn’t seem to support any of those features. But if I’ve missed something let me know. Inverter is the usual Huanyang 1.5kW models. ThanksQuote:
Originally Posted by Muzzer
Yes, those are the sort of features that get missed off some of the clone VFDs. You can get a fully featured 2.2kW Yaskawa GA500 with in-built EMC filtering and a plethora of features such as spindle at speed for a little over £200, which would have saved a lot of messing about, albeit without the challenge and fulfilment of developing something like this yourself.
More development, more testing . . .
https://youtu.be/H4SR3w2m5jw
Finally finished. We needed to strip down some of the machine to route the wiring, then it was time to upload all the information to GitHub.
https://youtu.be5aAXnm29t5g
I'm a bit bemused by this. In my experience of breaking cutters the spindle has never stalled, if it had the cutter might not have broken! The last one I broke I made a mistake setting Z height and the machine made a rapid move straight down into the work, fortunately not through the spoil plate. The spindle carried on regardless! Am I missing something?
Cutting 1000 grade thin sheet tend to gum up the cutter. It’s almost pure aluminium. Then it stops cutting and the spindle stalls and the end mill breaks. 6061 alum alloy is better but if the bit gets hot again it can gum up. Mist or flood cooling minimises this risk but it gets messy.Quote:
Originally Posted by JohnHaine
Well, I'm still bemused. If it's the cutter that stalls the spindle then it has to take all the momentum of the spindle to stop it so the instantaneous torque on the cutter must be much larger than the motor torque. As I said The motor hasn't stalled any time I've broken a cutter. Anyway it will be good to hear of your experience with the system - will you be doing some controlled experiments?Quote:
Originally Posted by routercnc
It's going to be very dependant on your machine. If you have a low powered spindle, it seems it will stall before snapping. Presumably on such a machine, the axis drives are likely to bog down if the spindle stalls, in which case(s) you appear to be able to kill the drives in time to save the cutter. It wouldn't be much use to me, based on my own unfortunate experiences, as it's the axis motion that causes breakages pretty much instantaneously when the cutter stops cutting - usually due to recutting / chip welding in my case. The answer to that is chip evacuation by air and/or coolant.
It would also be redundant if your VFD had a "spindle at speed" output but again, it's only going to be relevant in some applications. Although the Yaskawa spindle VFDs I've used on my machines have that feature, it would be of little use to prevent cutter breakages. Instead, its used to allow the spindle rpm to reach the programmed speed before the next G or M code is executed.
It's only a 1.5kW motor and so once the cutter is clogged with aluminium it turns into a cylinder being pushed through the material as it is not longer cutting. This brings the motor to a halt fairly quickly but the axes keep moving and so the end mill breaks. The inertia of these Chinese WC motors is pretty small as the rotating core is quite small in diameter. Tends to happen in thin materials as they are almost pure aluminium, not a high alloy content, although cutting 6061 dry can have the same effect but usually a lot less often. Mist or flood cooling helps significantly. I'll use the system and see how it performs over time, rather than specific experiments. The few test runs I've carried out (some were in the video) all work OK. My main concern is how the PLA will perform over time with heat build up - it is only a friction fit so it could slide down, same thing with the sensor disc.Quote:
Originally Posted by JohnHaine