[peteeng]

Hi Luke - Round rails have a convex to convex contact eg round rail to round ball. This means the Hertzian stresses at the surface can be very big. Hertzian stresses fatigue the material and this is the limiting factor for life of the rails. Overload limit is brinelling and again this is made greater by the convex contacts. Square rails have balls but the rail side is concave. Some have round grooves (single contact) others have gothic arc shape so it has two contacts in the groove reducing local stresses considerably. So the square rails have greater life and greater strength due to these factors. Then there are 4 grooves providing moment resistance which can be good or bad depending on the application... Peter

https://en.wikipedia.org/wiki/Contact_mechanics

[peteeng]

Evening all - Well Matt put in a good day today on Scoot-S3-01. He levelled and squared the base, installed the walls and the back board. I checked it out for rough level and square and it was acceptable. He struggled to get it right with the engineers level but was close enough for first fit. He's not used to working to small figures...The rail foundations are within rough spec so will be straightforward to get the rails right. The motion parts arrive this week so he'll be able to get up to the gantry stage very soon. The main thing now is to sort the electronics "drawer" so he is starting a breadboard with all the components. Next week he builds the drawer and we start wiring a mock up enclosure. Aim to get the motors turning on the bench asap... Gantry parts arrive in 2 weeks so then its all GO to the finish line... Peter

[peteeng]

Hi all you scruffy lot - Today was busy. I drilled a few holes in some cutting boards at the Menshed on Brevis. I found out that the last sign I did was spelt wrong. The person who gave me the art left an E out so I redid the sign on Scoot when I got home. Then a friend dropped in and we dragged the tank up the paddock back to its correct spot. It had floated away in the flood... Now I have to rebuild the workshop subfloor plumbing that was damaged as well and the washroom is back on track.... Peter

[peteeng]

Hi all - I have a question as I am not an electronics expert. To date I have used discarded wires from commercial projects that electricians have put in the bin and I have recovered over the years. For my projects that has been fine but for Scoot-S3-01 I'd like the wiring to be a bit more pro. I usually use 4 core speaker wire of AWG18 for the driver to motor wiring. I'm happy with that. At 0.82mm2 cross section it has heaps of ampacity for the motors. My question lies on the signal side of things. The driver specs say 10mA or100mA so the wire could be quite small. I have two choices at present 1) 4 core twisted pairs 4x0.5mm2 used for telephone wiring or 2) microphone cable 4xawg22 0.34mm2 which is shielded what do you think out there? which one is the better? plus any thoughts on emf filters would be good. Peter

[joeavaerage]

Hi,
either will work fine. If you are careful with your design you may well find you don't need the shield. I'm careful with impedance levels which improves natural noise immunity
and none of my signal cables are shielded and yet no noise problems.

The bigger question is flexibility, the finer the copper strands making up the wire the more flexible it is and the longer it will last in an environment where its required to move.
Microphone cables are usually pretty good in that regard. For fixed wiring or wiring that moves but occasionally then telephone cable would be fine.

Craig

[peteeng]

Thanks Craig my main concern is getting the vsd signal into the control wires. Seems they are the main culprit so I shall use a good separation for them in the box. I have not used a vsd before only DC spindles or 240V trimmers and have had no interference issues. Even with all things going down a single multiflex cable. I always have things well earthed. We'll know soon the parts left Singapore yesterday so will be here soon!! Peter

[luke peter]

Following this electrical discussion closely, lots of black magic occurs inside wires and controllers, and unlike hydraulics and mechanical drives, they are sneaky about what's going on and you cant see where they leak

[peteeng]

Hi all - I think the microphone cable or similar will be better thru the cable chain as it flexs. For the static control wires I'll use the telephone cable as its $1/m vs nearly $4/m for the microphone cable. Now I have to size the vsd cables. I'll look up the manual for the drive. Cheers Peter

[mactec54]

Hi all - I think the microphone cable or similar will be better thru the cable chain as it flexs. For the static control wires I'll use the telephone cable as its $1/m vs nearly $4/m for the microphone cable. Now I have to size the vsd cables. I'll look up the manual for the drive. Cheers Peter

You won't be using a VSD it will be a VFD, mic cable and phone cable is not suitable for any cable that flexes may be ok for short term testing but not suitable for machine wiring in general

What size is the Kw of your Spindle??

[joeavaerage]

Hi,
a VFD is commonly fixed in your equipment panel and does not move. Any control wiring to it will likewise not move. The connection to the spindle however will move as the
spindle goes up and down relative to the equipment box, but the VFD/spindle cable will be required to deliver many Amps so microphone cable is not even in contention.

Eight or nine years ago I bought my first VFD and 24000 rpm spindle. As a temporary measure to get it going I used 3 phase cable, multi-stranded 2.5mm² 4 core
(3 phases and an earth), the sort of cable you would make a 3 phase extension lead from. It worked perfectly, and I'm still using it 81/2 years later. About a year ago I re-terminated the
plug that goes into the spindle, as that is where the most of the flex occurs, but it was precautionary measure only.

As no doubt Matec will tell you there is cable made for this type of service. It has a large number of very fine strands to make it survive repeated flexing and a woven shield.
Its not cheap...although Matec has a source of competitively priced material. I always meant to get some but have just never got around to it.

The cable I used does not have any shielding. You can get 'extension cable' with a woven shield, but if you are going to that expense then just get the real deal. You might imagine
that because the VFD/spindle cable is unshielded that I would suffer a lot of electrical noise issues....but no. Firstly a I keep the VFD/spindle cable as far separated as possible
from any control wiring and I pay close attention to impedance levels within my control wiring such that it has good natural noise immunity without shielding. It works pretty well.

There is microphone cable and there is microphone cable. The genuine Pro stuff is very flexible well shielded but eye-wateringly expensive, $10-$20 per meter.
Belden is a known good brand:
https://www.ebay.com/itm/26561825010...sAAOSwNshiQdwC

The one area that I did encounter trouble was that the 230VAC input to the VFD suffers from harmonic distortion as a result of the VFDs operation and that distortion can be
carried into other 230VAC supplied devices like electronic power supplies and PCs. I installed a line reactor to the input of the VFD and a two stage EMI line filter to the input of
the sensitive devices, solved. But note that this was on the input side not the un-shielded output side.

Craig

[peteeng]

Hi Mactec - Yes its a 1.5kW VFD haven't played with them before. Finding the manual today. Looking at the wiring, the cable in the cable chain will be multicore flex last time I used 12corex0.75mm2. Was an offcut from a commercial machine project. My ideas about routing changed last night. Wiring is always a bit of a journey I find. Matt would like the electronics in a drawer with connectors and that in itself has brought up some deep thought.... I want to keep the power circuits, signal circuits and VFC trunking as separate as possible. But they will be together in the cable chain. Attched is the wire suppliers data.

craig you mention impedance levels how do you check these? Peter

[peteeng]

Hi Luke- They don't leak but I have let the smoke out on occasions. The smoke is impossible to put back in Peter

[joeavaerage]

Hi,

craig you mention impedance levels how do you check these? Peter

You probably can measure the impedance but its really a design concept, its not easy to change once built.

The idea is that Electromagnetic interference transfers energy from one circuit to another as an electromagnetic wave. The electromagnetic energy passed out into the 'ether'
is determined by the emanating circuit, and there's damn all you can do about it. Likewise the receiving circuit is pretty much fixed as well. Shielding is one of the few things that you can do to
prevent the electromagnetic energy coupling into your control circuit.

Lets say a certain pulse of energy is radiated into your circuit, lets say 1mW. If the resistance of your circuit is very high, lets say 100kOhm, then that energy will generate a significant noise voltage:

V_noise=(P x R)^0.5
=(0.001 x 100,000)^0.5
=10V

So if your control signal is 24V then the noise voltage is nearly 1/2 of it, you might expect trouble.

If you design your control circuit to have a resistance of 5kOhm then:
V_noise=(0.001 x 5000)^0.5
=2.24V

So the noise voltage is now a little less than 1/10th of the signal voltage, and you might expect that the control signal will not be affected as badly.

So you can see that choosing the resistance (same, or nearly the same as impedance) correctly you can make your control signal MUCH less subject to noise.
Note that your circuit is not 'noise proof', there is no such thing, but with intelligent design you can make it 'nearly immune'.

If you do so, then the extra shielding and filtering that you do will be really effective, failure to do so may well mean that all the shielding and filtering won't be enough.

Craig

[peteeng]

Hi Craig - I appreciate that interference is the result of voltages being induced between adjacent wires via AC inductance. If I use a shielded VFC cable is the shield grounded to the electrical ground or left as open? If grounded this can create voltage loops? I'll look for the VFD manual... Peter

[joeavaerage]

Hi,
a shielded VFD/spindle cable is highly recommended, it should reduce the amount of electromagnetic noise from being radiated in the first place.

The shield should be earthed at one end only, the VFD end, and it should go to the common ground point for all other earths in your machine. Most VFDs provide a clamp
that connects the shield direct to the VFD case, and that's not bad, but better that it go to the common earth. It's commonly called the 'star point'. If you make the mistake
of grounding your shield at the spindle end AND the VFD end, then you would have a loop which could introduce noise currents even worse. If you earth the shield at the
spindle end only, it may reduce the electromagnetic radiation but would cause noise currents in the earth wire from the frame of the machine
to the star point and you may have a worse problem than no shield at all.

As I posted earlier my spindle cable it not shielded at all and I have no insurmountable noise problems, shielding is nice and recommended not absolutely required.

Craig

[peteeng]

Thanks Craig - smallest 3core+earth shielded is a 1.5mm2 conductor rated at 16A should be fine for a 1.5kW spindle? 1500/230= 6.5A rough calc ? I usually use star point so all good there. Peter

[mactec54]

Hi Mactec - Yes its a 1.5kW VFD haven't played with them before. Finding the manual today. Looking at the wiring, the cable in the cable chain will be multicore flex last time I used 12corex0.75mm2. Was an offcut from a commercial machine project. My ideas about routing changed last night. Wiring is always a bit of a journey I find. Matt would like the electronics in a drawer with connectors and that in itself has brought up some deep thought.... I want to keep the power circuits, signal circuits and VFC trunking as separate as possible. But they will be together in the cable chain. Attched is the wire suppliers data.

craig you mention impedance levels how do you check these? Peter

IGUS is your best cable source https://www.igus.com.au/

This cable is designed for high flex nothing else comes close to their quality, that being said there are some other cable manufacturers that make suitable quality cable

The main cable that needs to be shielded is from the VFD Drive to the Spindle, this must be 4 cores, shielded and correctly terminated including the shield

This is the IGUS cable part number (You can get samples from this company of anything they sell)

CF6.10.04 (17AWG) $11.92 per metre

CF6.07.04 (18AWG) $10.84 per metre

This is not the most expensive cable that they have, you could use for this same application, both of these cables are suitable for your spindle, one being just a heaver AWG your choice as to what you use

Multi-core cables can be useful, just remember any unused wires have to be Grounded at both ends or they can cause interference.

[mactec54]

Thanks Craig - smallest 3core+earth shielded is a 1.5mm2 conductor rated at 16A should be fine for a 1.5kW spindle? 1500/230= 6.5A rough calc ? I usually use star point so all good there. Peter

1.5 is too big for your 1.5Kw spindle 1.5 is used for 2.2Kw spindles if that cable you were looking at is only rated at 16A it is of poor quality, just a note the outside diameter of the cable is important also as if you get the wrong cable, it will not fit in the spindle plug, (not all cables are created equal)

[mactec54]

Hi Craig - I appreciate that interference is the result of voltages being induced between adjacent wires via AC inductance. If I use a shielded VFC cable is the shield grounded to the electrical ground or left as open? If grounded this can create voltage loops? I'll look for the VFD manual... Peter

The shield has to be directly Grounded to the Ground plane, Ground plane being the metal plate that everything is mounted on, your Ground plane is where your Ground Bus is mounted (Star Ground)

So, No Shield should be connected to a Ground directly

[mactec54]

Hi,
a shielded VFD/spindle cable is highly recommended, it should reduce the amount of electromagnetic noise from being radiated in the first place.

The shield should be earthed at one end only, the VFD end, and it should go to the common ground point for all other earths in your machine. Most VFDs provide a clamp
that connects the shield direct to the VFD case, and that's not bad, but better that it go to the common earth. It's commonly called the 'star point'. If you make the mistake
of grounding your shield at the spindle end AND the VFD end, then you would have a loop which could introduce noise currents even worse. If you earth the shield at the
spindle end only, it may reduce the electromagnetic radiation but would cause noise currents in the earth wire from the frame of the machine
to the star point and you may have a worse problem than no shield at all.

As I posted earlier my spindle cable it not shielded at all and I have no insurmountable noise problems, shielding is nice and recommended not absolutely required.

Craig

The VFD Drive Cable must have the Shield Terminated at Both Ends, this does not cause a loop as you say check any quality VFD Drive manual for how to install a VFD Drive.

It is Required by Code in NZ/Australia to use a Shielded cable for all VFD Drive installations