First steel diy CNC router build

[JAZZCNC]

Nope - which two terminals you choose out of R,S,T makes no difference as they're just the connections to a 3-phase rectifier, so are equivalent.

Ah but that's where your wrong Numb nuts because I've had several know and infact just fitted 1.5Kw VFD that "S" hasn't been connected so only "T" would work and given this VFD didn't work with "S" then it could have been couldn't it.!!

[ivars211]

OK, on the positive note - all steppers and drivers/ PSU works!
Am I calculating steps per mm right?
I have 1605 ballscrew, so 5mm pitch
So if I want my cnc to move 1mm, it's 0.2revolutions (1/5).
I have set my driver to 200pulse/rev (http://www.longs-motor.com/productin...12_80_131.aspx) by putting SW5-SW8 ON
so thats 0.2*200 = 40 steps per mm, right?
I think I have calculated something wrong as my velocity is set to 31 000 but it moves quite slow compared to CNC's on youtube.
Btw, is there any point having microstepping turned ON on the driver?

[JAZZCNC]

First check in Native units that your running with Metric units not imperial.

Don't set drive to 200 micro steps as that will make the motors run rough. Set the micro stepping to at least 800 for smooth running. Higher MS will give smoother motors but don't go above 2000MS. The down side is that the higher the MS the lower the velocity you will be able to tune the motors for with a Given Kernal setting. Higher Kernal speed will allow higher motor tuning but will need a fast PC with a very good parallel port. Don't set it above 45,000Khz as it will make the system unstable.

Your Steps Per calculations are correct so just do the same for higher MS settings. With a 5mm pitch ballscrew and if running decent amount of volts then you'll top out around 5000-6000mm/min before steppers start losing torque and stall. Obviously this depends on many other factors but 5000 should be achievable.

If running 800MS Start motor tuning with 3000mm/min and 500s/s accel then tweak from there.! Golden rule is you can't have high Velocity and High acceleration. Adjust one then test then do other don't do both at same time. Don't go too high on acceleration otherwise you run the risk of missed steps. when you get above 1500 your in missed step territory.!

[ivars211]

Switched 1600 micro steps, so that's 320 steps per mm. Maximum speed I could achieve smooth and without stalling was 3000 mm/s with 750 accu for all axis, seems good enough for milling aluminium.
Now I have to tie all the wires down properly, get the Y in 90* in respect to X and tie the bolts down... Most importantly, I have to get familiar with Mach3 software which I guess will take a while. Limit switches are on their way, pity that FVD doesn't work, but hey, positive thinking? :D
Thanks all of you, I wouldn't have managed this far if not because of this forum, that's for sure. I added short video filmed from phone (sorry for quality), I will have proper topic with all the build process and photos/videos in few weeks.

[routercnc]

Hi ivars,

3000mm/min on 1605 ballscrews at 36V sounds about right based on the spec and speed of my machine. I'm using 1605 ballscrews with an all-in-one board at 24V and get around 1500mm/min before stalling.

I've just returned a faulty VFD that went bang when I wired it up - it was the same model as yours and the supplier asked me for a photo of my wiring. Once he saw that he was happy to exchange it. I had only connected the mains supply (not the spindle yet) and he never mentioned having to connect a spindle, so if such a lockout feature does exist it's not on that model.

Very pleased for you anyway - I remember how good it feels to finally get some movement out of your first machine. Oh, and do move that monitor back a bit when you are cutting chips!

[ivars211]

That's good news then, hopefully video that I shoot yesterday will prove that it's not my fault, but crappy FVD. The stupid thing is, I cannot even test spindle, as spindle might even be OK.
I will definitely move monitor and all other stuff away, but thanks for heads up anyway :D

[JAZZCNC]

3000mm/min is about all you'll get with 36v and it's actually a good sign you have aligned everything nicely otherwise you'd have struggled to get that speed.

Looks a nice solid little machine well done.

[Neale]

Ah but that's where your wrong Numb nuts because I've had several know and infact just fitted 1.5Kw VFD that "S" hasn't been connected so only "T" would work and given this VFD didn't work with "S" then it could have been couldn't it.!!

Like Jonathan, I thought that the Huanyang VFDs were wired to support single- or three-phase operation as that's how they are described in the manual. I just picked two terminals on mine and it works fine. But if they've changed this, or some of the current crop of Chinese VFDs don't match the usual Huanyang spec, or some are just not wired internally in accordance with the manual, then that's well worth knowing.

[JAZZCNC]

Like Jonathan, I thought that the Huanyang VFDs were wired to support single- or three-phase operation as that's how they are described in the manual. I just picked two terminals on mine and it works fine. But if they've changed this, or some of the current crop of Chinese VFDs don't match the usual Huanyang spec, or some are just not wired internally in accordance with the manual, then that's well worth knowing.

I've fitted 2 inside 5wks 1 x 1.5kw & 1 x 2.2Kw that have been like this and previous to these I've had possibly 3 others same. All but one came from Chai and I didn't ask for single phase only so don't know why.? Maybe they do both and when short of dual and knowing coming to UK provide single phase.? . . . . Can tell you the First one gave me that "Oh NO" sinking feeling.!.

[ivars211]

Hi again, been delayed lately, more or less everything works OK now. Question is, how do you get X-axis right angled in respect to Y? I cannot figure out the most precise method of doing it, any suggestions?

And also, I cannot find anywhere if it's possible to connect breakout board output signal to FVD, so that FVD starts when output signal from breakout board is HIGH...