Reasonably Rigid Router (RRR) - a cautionary tale (likely) in the making.

[peteeng]

Hi - an emi filter such as this? Peter

[luke peter]

Thanks Craig!

That is exactly the kind of answer I was hoping for, I really appreciate the time you took to type it.

I have a cheap red usb mach breakout board, it has very limited IO, so limit switches all need to be bundled. I intend to move to a linuxCNC mesa 7i96 based control later, when I put servos onto my axes, and want to add ATC functionality. The plan was a solid mech build coupled with childlike electronics, with the elec being easier to upgrade (at least in my mind) at a later date.

[luke peter]

Hi - an emi filter such as this? Peter

Thats what is in my head when I say EMI filter...

I do notice it doesnt have an earth line on the output, how do I maintain earth integrity to my PSU's?

[joeavaerage]

Hi,

Hi - an emi filter such as this? Peter

No, that's a bit lightweight. For a 2.2kW spindle you should be looking for at least 20A and two stage, not a single stage like the one you linked to.

https://nz.element14.com/multicomp-p...t=line+filters

Cheaper 6A and/or 10A but still two stage filters would be adequate to cover the PC and 65V and 24V supplies, but he bigger one for the VFD will cost a penny or two.
Three stage would be nicer still, but becoming cost prohibitive. If you need three stage filters then you should have a line reactor, or if your VFD is wired for it a DC choke
instead.

Craig

[joeavaerage]

Hi,

I have a cheap red usb mach breakout board, it has very limited IO, so limit switches all need to be bundled.

What on earth for??? You've spent a lot of time and money on this project only to throw it away one one of those little BS controllers.
Buy an Ethernet SmoothStepper from Peter Homann at Homann Designs in Melbourne and be done with it....or get straight into LinuxCNC,
don't mess around with that Chinese rubbish.

Craig

[luke peter]

You make a good point Joe...

though, in my defence (however feebly), I have every intention to upgrade the control, and I can buy a ESS or mesa set now, or I can get a suitably thick base plate(there is not enough disposable income for both), and have a working, but somewhat control limited machine in the next month... South Africa is not spoiled for choice when it comes to CNC controllers or shipping

[luke peter]

Further questions - does a servo motor drive create the same noise as a typical chinese 1-3phase vsd?

I found a suitable 20A filter for the VSD.

The pc and driver PSU's will se about 1.7kw (there are 2 13A(if i remember correctly) 65v psus). should I get a big filter for them also?

[joeavaerage]

Hi,

Further questions - does a servo motor drive create the same noise as a typical chinese 1-3phase vsd?

Yes. VFDs, direct offline servo drives, switch mode power supplies all have the same basic input circuits, a bridge rectifier followed by a smoothing capacitor.
They all draw serious current in a distorted waveform which can and does affect other nearby devices, PCs for example. Its for this reason that we use line filters,
or in extremis, a line reactor or DC choke to prevent that distorted current waveform from polluting the AC power supply.

If I understand arights you want to use a servo as a spindle motor? I use the same strategy. My most often used spindle is a VFD driven 24000rpm asynchronous
spindle, but I also have a servo driven spindle for high torque low speed use. If so you'll only be using one at a time so have just one line filter for the two
of them.

The pc and driver PSU's will se about 1.7kw (there are 2 13A(if i remember correctly) 65v psus). should I get a big filter for them also?

It maybe correct that the total could reach 13A, but it so very unlikely. It wouldn't hurt to have another big line filter....except your budget. I think a 10A line filter
would be big enough.

Craig

[luke peter]

Noted with thanks.

Correct, pretty much the same spindle split that you use, a bt30 cartridge for low rpm work, and a 24krpm router spindle for the rest, they wont be on the machine at the same time, so I will switch them off the same supply as you suggest.

[joeavaerage]

Hi,

I found a suitable 20A filter for the VSD.

Make sure its a two stage filer.

Craig

[luke peter]

cost about $35 us

[luke peter]

Some movement - but not enough.

Painting all done, and the wheel scan go on:




All the little bits


And assembly has moved along a little - only this far since I have been helping my son with his science project, but its nice to see stuff happening.

Parts look much more red in person than on screen

[LorenBMS]

Looking really good!

[luke peter]

alignment and assembly progressing,

The gantry assembly is nice and rigid, I can back drive the slave screw on the x axis by pushing on the master side, no twisting or shifting off 90 deg. All nuts can travel to either extremity without binding.

Question - how do you all tighten the lock nut on your ACB mounts? I have tried vise grips with leather on the screw, but it always spins in the grips.



I printed some solid couplers to help hold the motors and their brackets axially aligned with the screws for shimming. Motor brackets will go on next, along with the shims for the bracing struts.
I have unpacked the electronics and cleaned up the panel, will start with layout and generate a basic cut list for the cables tomorrow.

[luke peter]

Looking really good!

Thanks Loren!

[luke peter]

Its going to be a tight fit...

[luke peter]

Does anyone know what the power supply requirement is for the 24v psu? I can save some space by using a little 2A 24v psu in place of the bigger 10A one.

I have been fiddling around with component placement that will make sense w.r.t. airflow and cable management, and this is the best I have come up with - possibly I can replace the smaller PSU with an even smaller one (about 1/10th the size) as per the above.
The tape shows where the EMI filters will go, and the 3 position selector switch and two contactors for spindle drive switching are on their way also.


I also printed a stand for the PC, that allows me to mount the controller upright on its side.

[luke peter]

220 v circuit done (mostly - fuses still on their way)

Switch engages the contactors for either spindle drive - EMI filters on the inputs of the drives, and the psus.

[luke peter]

Spindle control question time. I want to output the same 0-10v signal to both spindle drives at the same time, only the one which is powered up will do anything with the info though...

The normal VSD I think I will manage fine - there is plenty of info available for it.

The servo is a little more tricky since there are very few examples of them being used before, and the manual is a translation...

From what I can gather, I put the drive into speed control mode, and feed it 0-10v on pins 10 and 26 (bridged according to the second image first table point) and a common gnd input on pin 11.

I can change direction using relay output on pins 2 and 17.

Is that roughly correct?

[joeavaerage]

Hi,
this is the problem with cheap Chinese servos.....the manuals are crap.

The idea is correct however, set the servo up in velocity mode and the velocity of the servo will be determined by the analogue inputs at 26,10,11.

The conventional idea is that 0 to10V will cause rotation in one direction, say CCW and 0 to -10V will cause rotation in the opposite direction. Referring to my own Allen Bradley and Delta servos
the primary direction (CCW OR CW with positive input) is programmable via parameters.. I imagine your servo is the same...good luck trying to find and program the parameters however.

Do you need counter-rotation?. In the early days I had Mach programmed to run in either direction, but then hardly, if ever, used it.

Start by doing some simple experiments with a simple 9V battery and a variable resistor and a couple of bits of wire on inputs 26,10, 11 until you can control the speed and direction to your
satisfaction....which will inform you as to how to wire the analogue inputs into your machine.

Craig