I’ve asked this question at other forums to folks that are using big CNCs such as CAMaster and ShopBot; pros. Most of them scoff at the power loss effects to their projects. I’m not convinced that a power drop out during a 3D route is not damaging.
So, the question is “Can one obtain a UPS with sufficient capacity to allow a controlled shutdown of a project being routed?”
I don’t have any electronic equipment that isn’t running through a UPS. I’m not going to neglect my CNC router system. Since the “devil is in the details”, some further questions come to mind.
How can the computer running the gcode detect a significant power outage and “withdraw” the router from the cutting surface?
Can a typical Dewalt router be operated for a short time on a UPS so that it can be withdrawn either by command or manually?
We typically have short power “blips” that are often so quick that the UPS alarms barely beep, but for electronics that aren’t prepared that could mean a complete reset.
You have to find a UPS that can handle the amp draw of the router. I looked into this due to a few power blinks ruining my projects. Due to cost I settleded on a small UPS to keep the machine from dropping out completely and I can run over and pause the job. Typically this happens before the router spins completely down and minor power blinks no issue.
I know have limit switches but I primarily did this before those switches were available or installed.
Yes, the power draw on a hobby CNC can be carried by a UPS — not sure about the spindle — the problem is there are apparently electrical considerations which make this a bad idea, and if I recall correctly, result in a very short lifespan for the UPS due to a constant degradation. That said, there are a number of folks who do this and it solves EMI and problems such as yours.
Unfortunately, a power fault is too brief and quick an event, and the machine doesn’t have sensors for it.
I’m purty sure that most UPS devices don’t actually provide power until the line power drops out, but yeah that brings up questions about “true sine wave” and etc.
The power draw is probably less than 10 amps on these trim routers, and if you can keep one going then the startup currents shouldn’t be an issue. All I’m saying is to have time to shut the equipment down before damage to the part occurs.
Then what @ShallowCreekHill says begs the question, “Is there a problem with restarting the job a short time before the outage occurs?” Seems to me, you would have to restart at “zero” everything.
So, it wouldn’t be very difficult to place a relay in the router power circuit that would “drop out” or remove the power to the router even with short power faults. Can the Z axis drive the router off the part in some way? Is there some kind of way to test an input?
I was given an APC UPS by a friend when his work surplused a bunch because of bad batteries. I put in (I believe) two 12V 6AH lead acid batteris and it has been running just fine for the last two or so years. When I was running my DeWalt DWP611 router, the current LEDs on the UPS (there are 5 indicating the percent of load capacity) would rarely go above two of the 5 LEDs lit up. On startup, it would spike to 3 or 4, but once cutting, the UPS could easily handle the load. I was also running a SuperPID, so that may also have an effect on how much power the router consumes compared to running it without.
I have been fortunate enough to not find out what happens if the power goes out completely, but I have experienced a few brown-outs. When that happens, the UPS quickly switches over to battery and my SO3 kept on cutting just fine. Then the UPS switches back to wall power - the brown out usually lasts only a second or maybe two.
And I would definitely go out and find out, but I gave my dad my SuperPID controller I have been using for the last three years, and I have not assembled the new one yet. And I am not going to most likely because I bought a 1.5KW spindle and VFD for mt SO3 XXL that I am slowly assembling.
I guess I could just hook up my spare 611 to the UPS, along with the SO3 and laptop and see what happens when I pull the power on it. The router would be running without a cutting load, but it would probably be close enough for a test.
EDIT: Okay, I did it. Here’s the video. It’s short, but it gets the point across.
“battery on bus” and “Load on line” - “battery on bus” puts the load on the battery side of the circuit. The load circuit is large enough to charge the battery and feed the load at the same time (it’s more complicated than that, but…work with me). These are pretty expensive, but they never glitch. Load on line is the type you are most familiar with (they’re much cheaper) because the load is usually on the line side, and it switches (which takes a minimum of 3 AC cycles, and usually more) to the battery side when power fails. This causes a glitch, but most power supplies on gear can handle that without any real difficulty (they happen all the time on a regular line). You generally only see load on battery type ups’s in data centers (per rack, per row, or for clusters of racks).
“Probably less” than 10 Amps…look at the router, it says right on it by law. Edit: It’s 7A - about what a laser printer uses running it’s fuser, I don’t know what initial inrush is though
Here’s your problem…if you don’t have the spindle on the ups, it’ll spin down, the computer has no idea that’s happened, and you’ll ruin a project. If you do have the spindle on ups, you’re going to need to have a fairly large ups just to keep up that load long enough to start shutting things down (which could be automatic). In the end, it’s a call on what your project is worth and how bad your power is. Is a few hundred dollars for a UPS just more hassle, or is it going to save a multi-hundred dollar project more than once? The big guys, running a 4x8 sheet of unobtainium through their huge cnc can easily have that sort of material cost sunk on the table when power fails. You an me? …Not me.
I run my dust collector separately though, too many total amps. I’ve purposely unplugged during a test run and the machine kept going. I never have EMI issues either, but I can’t prove this is why. Did a bunch of research a couple years ago on this subject, but sadly I’ve forgotten most of it (because it works, so I don’t mess with it).
