PDA

View Full Version : VFD cable and shielding info request



caveBob
10-09-2014, 10:32 AM
In the process of tear down and cleaning of my first milling machine (8x30 knee mill). It came with a 1.5hp motor that would draw more than my available 15A service could provide so am replacing that first thing. What I ended up with:

1 HP, 1800 RPM, Brook 3 Phase Motor Package with 1 HP, 115 Volts, Teco Variable Frequency Drive
Item#: 1 HP 115 Volt Input Brook Motor VFD Package
http://dealerselectric.com/1HP115VoltInputBrookMotorVFDPackage.asp

VFD: Teco Westinghouse Model #: JNEV-101-H1, 1 HP 115V Single Phase Input-230V Three Phase Output
Motor: Brook Crompton Model #: WA4N001-4

My plan is to mount the VFD in an enclosure on the wall behind the machine with a separate button enclosure mounted on the machine for easy access. From what I've read so far, I need to use twisted pair cable (Belden linked below?) for the control cabling from switches to the VFD and shielded cable (what type/brand/whatever) from VFD to motor, right?

One link I've found so far took me here:

Variable Frequency Drive (VFD) Cable
http://www.houwire.com/products/belden.asp

...downloaded the "Belden VFD Specifications" PDF on that page, lotsa choices... what/which one do I need? Get it from them or somewhere else?

Plan on the machine mounted button control box is:

Three position switch (twist left for forward, center for stop and right for reverse)
Potentiometer for vari-speed
E-Stop

What else do I need to know? What guage size do I need for this setup?

thanks

George Bulliss
10-09-2014, 10:56 AM
I just hooked up my new mill with the same VFD and same control setup (minus the E-stop). The manual calls for 14 ga. wire for the single phase input and the three phase output and 18 ga. for the control wires. It says nothing about any special requirements for the control wiring and it's only 12 volts at very low amps. I ran the vfd for several years on my last mill with 22 ga. control wire and had no problems. For this mill I ran some 6 conductor cable I had on hand – not sure the size but its smaller than 18 ga.

I used 14 ga. for the high voltage stuff, as suggested. Don't be tempted to use 12 ga., you won't be able to stuff it into the VFD's screw terminals.

MaxHeadRoom
10-09-2014, 11:00 AM
I have installed many VFD's over time and never yet used any of the specialized cables, I usually ensure the conductors to the motor are tightly wound if using MTW or TEW etc, and usually use either conduit or liquid seal metalic flex,
IMO it is more of an advantage to put the money in inductor or three phase choke on the motor side.
14G conductors. Make sure the ground wire is connected to the VFD and continue on to the motor frame.
Max.

lakeside53
10-09-2014, 11:30 AM
You don't need shielded cables to the motor. if your worried about rf interference, just put them in metallic flexile conduit. Input wiring -The only place I use shielding is the potentiometer, and only if it's a run over 3-4 feet AND not in metallic conduit.

Paul Alciatore
10-09-2014, 01:49 PM
To pick a cable just go down the list of specs.:

1. Number of conductors
2. Wire Gauge
3. Voltage rating
4. Stranded or Solid
5. Shield or none
6. Environment: determines outer jacket material
7. Are pairs needed?

1. Number of conductors: just count them from the hookup diagram or table

2. Wire gauge: should be specified in your installation docs. If not, 18 gauge should be more than enough. Most control circuits carry very little current so even a 20 or 24 gauge wire would probably work. Larger wire sizes will work, but may be difficult to fit in the connectors. Of course, they will cost more.

3. Voltage rating: should be specified in your installation docs. If not, use cable rated for control circuit Voltage used. If you can not determine that, use cable rated at line Voltage. Higher Voltage ratings will work, but probably be more bulky and expensive.

4. Stranded or Solid: Solid can be used in areas where there is no flexing or movement. ALL other places, use stranded. Stranded can be used in almost any application so it is the more universal choice.

5. Shield or none: Again from your specs. I would NOT use a foil shield in a shop environment. Foil shields are like solid wire, intended for installation where there is no flexing or movement. I have seen foil shields literally turned to confetti when used in cable that are subject to movement. If you use a shield, DO use a braid shield.

6. Environment / outer jacket material: Pick a jacket material that can resist the rigors of shop environment: oil, abrasion, etc.

7. Pairs? Some situations need paired wires. Some cables are made this way. You can even get cables with individually shielded pairs. I doubt that you need a paired cable for this application but it is OK to use one in any case.

A good solid ground between the main chassis, the machine and motor, and the control box is a very good idea. Be sure to include this ground in both the control cable and the motor cable. Also in any cables to sensors or switches used by the control. This will save many problems.

As for a shielded cable to the motor, this would be more for preventing any radiation from that cable rather than from shielding it from any pickup. Many speed control circuits use switching techniques that produce high frequency harmonics in the 60 Hertz power line frequency on in a variable frequency used by the drive. These harmonics can radiate out into your shop, home, and neighbor's homes and cause problems there in any number of different devices (radio, TV, computers, etc.) If your installation instructions call for a shielded cable to the motor, it is probably a good idea to use one.

gzig5
10-09-2014, 02:14 PM
You probably don't need shielded cable, but they do make it for a reason. Really depends on the design of the drive. I will say to that if you do use a shielded cable to be sure to bond one end or the other (or both) of the shield to earth and/or the frame of the drive. The PWM switching can charge the cable capacitance formed by the cable shield and a surprising amount of voltage will be waiting to greet you if the shield is left floating.

caveBob
10-09-2014, 07:24 PM
Great, thanks for all the info! I was in a hurry to try to get the ball running info-wise before heading out today. Main concern besides hooking up correctly is from what I've read regarding Yuriy's TouchDRO app in that ground loops and RF noise can play havoc with some scales. TouchDRO is planned to be added once everything is up and running peachy, so want to try to get it right here. Atm, not sure whether I'll use magnetic scales for all 4 axis (1 for spindle, 1 for knee, X, Y) or go mixed with a couple iGaging remote digital readout scales for the knee and spindle.

Never dabbled with VFDs before so google has been my only source for obtaining "what I think I need to know...". I'll paste a few texts that I've saved so far (sorry, just grabbed text no URL or author), maybe that'll help you see why I was/am concerned/confused:



I STILL SEE NO SHIELDED CONTROL WIRING... This is REQUIRED on low voltage controls...

IT MUST BE USED.. Stray RF can get into VFD... and cause all kinds of fun.. unswitched starts, stops, speed changes, or even (extreme cases) possible damage to VFD itself..

Stray RF no problem... NOT ... A mobile HF rig often runs 500 watts or more output.. I have seen fairly simply power supplies, go crazy when transmitter is keyed..

Since you cannot control mobile transmitters, or even know when one is around... You never know what can happen..

The MODERN DC low voltage control wiring, is easily overloaded/interfered with, with fairly low RF levels.. Not a problem at all, with OLD 110/24v ac controls..

The VFD manuals clearly show shielded wiring, and specify it..

Still no mention of the fused/circuit breaker on input to VFD.. Not only just handy to shut off/protect VFD. You can LOCK the power OFF for service..

Using pretty 18 ga wire, to carry ~12 v at ~20 milliamps, is a waste of copper, AND IS NOT SHIELDED.

Just because someone (or many) does not use shielded cable, is no reason to even suggest... that is the right, SAFE, or code approved way of doing it..

Do what you want, I see people driving cars with the cord showing on tires too... Just don't tell others its fine... It's NOT..




I'd use shielded, twisted-pair cable for the control cabling all the way from switches to the VFD, and shielded cable from VFD to motor, a rule I follow is that the twisted-pair is carrying the two switch contacts back to the VFD, the shield is for electrostatic (capacitive) field rejection and earthed only at the VFD to prevent magnetic pickup inducing hum onto the control cables (twisted pairs are more immune to magnetic pickup too so these work together to prevent false signalling and.unintended operation).

Bond the motor casing to the lathe castings for a safety earth and tie that "star" connected to the VFD PE terminal along with the incoming power, the motor cable shield and the control wiring shields, leave the cable shield disconnected at the motor (and sleeve to prevent contact).

If you're using a VFD, or just working with electrics, you need a multimeter of.some.sort to check which contacts are open and closed in which positions, that's going to dictate how use your existing switches with the VFD - see which colours go to each switch (some dismantling required) and measure *on the lowest ohms range* the resistance to determine open/closed. A meter that beeps for continuity makes.this a lot easier!




The wiring from your VFD to the motor will be, as Raef noted, will be a wonderful source of RF noise, and is best kept in a shield of some sort. It is possible, but not easy to find armored-shielded power cable for such applications, but not easy or cheap. Using flexible metallic conduit is usually the best route... but you probably won't find stranded wire, so plan on pushing through the proper type. Again, the current rule prevails- use at least 14awg for 15A loads, but with the VFD running, you'll have propensity for what's called 'skin effect' (where electric current tends to flow over the outer surface of the conductor, rather than the middle)... so running the next-size-larger is (as always) a good thing.

Your control wiring... when referred to as 'twisted'... it's not just twisted at the ends... it's twisted a certain number of turns-per-foot over the whole length. What this does, is contain any radio-frequency interferance as 'common mode' (it appears equally on both wires), hence, the receiving end doesn't see it as a 'differential' signal. Most twisted-pair wiring you'll see, is solid conductor. I use some 3-pair twisted stranded 16awg stuff that my company uses for intercom and instrument communications cable, and it also happens to be jacketed in braided stainless-steel armor. It's not common off-the-shelf stuff, nor is it cheap, so again, the best advise is to put it in metallic conduit as much as you can.

IF you're willing to scrounge, you might find old computer cabling that'll do the job- they're usually full of stranded twisted pairs, and since the controls are carrying a very small current, small sizes are just fine.


I really appreciate and thanks for everyone's input, with special kudos to Paul Alciatore for taking the time to "break the problem down into bite sized chunks". Further digestion on my part I'm sure will be required... :) before I begin hooking everything up.

MaxHeadRoom
10-09-2014, 07:53 PM
IMO these do far more good on the motor side,http://www.automationdirect.com/adc/Overview/Catalog/Drives/AC_Drive_(VFD)_Spare_Parts_-a-_Accessories/AC_Line_Reactors than 'special cable'
I have fitted up to three VFD's in one CNC cabinet, I used these reactors on the output, they make it easier on a non-vector motor, and also the drive.
Max.

macona
10-09-2014, 08:49 PM
Meh, I have used non shielded cable on vfds and servo drives and have never had a problem. Mostly it is to help stop the drive from radiating power into other devices, not to keep it out of the VFD. VFD and servo drives are very noisy. On the little servo drives for my telescope the motor power wires are shielded and the grounds were loose, the amount of electricity gathered on the shield was quite shocking, literally. I measured over 100v. Low current but enough to tingle.

Paul Alciatore
10-09-2014, 09:06 PM
I have to agree with others about the need for shielding. In a system like this, it is primarily to keep it from radiating electrical noise, not to shield it from receiving it.

Considerations like ground loops and pick-up of ambient electrical noise are usually relevant to audio, video, data, and other low level signal transmission systems. In these systems the desired signal is often at levels of several Volts, a single Volt, or even tenths or hundreths of a Volt. Also, these signal transmission systems may have a high input impedance and that high impedance allows a weak E-M field to generate a rather large Voltage in the wires connected to it. This combination of factors makes these circuits sensitive to any interference that may be present.

However, control circuits are usually operated at larger Voltages, 5V, 12V, 24V, and even 120V. I have even seen 240 Volt control circuits. And they usually have relatively low input impedances which tend to limit the level of any Voltages that can be picked up by the wires. So, such control circuits are relatively immune to such problems. I am sure there are some (poor) designs that do have problems of this sort, but I am talking about the general situation. So the shielding is most likely to protect other devices from the control system, not the opposite.

jlevie
10-09-2014, 09:21 PM
The control signals to the VFD are low level DC and thus probably need nothing larger than 22ga unshielded wire. But to be on the safe side I'd use multiple twisted pair shielded wire or a cable containing same. Ground the shields only at the VFD. It is a plus if you run the cables in a metallic conduit.

gzig5
10-09-2014, 09:23 PM
Meh, I have used non shielded cable on vfds and servo drives and have never had a problem. Mostly it is to help stop the drive from radiating power into other devices, not to keep it out of the VFD. VFD and servo drives are very noisy. On the little servo drives for my telescope the motor power wires are shielded and the grounds were loose, the amount of electricity gathered on the shield was quite shocking, literally. I measured over 100v. Low current but enough to tingle.

That's exactly why I mentioned bonding the shield. Seven years ago I was doing HALT testing on a new servo product family. The new motor cables we used did not automatically connect the case of the motor to the case of the drive, giving the energy a path to ground like the ones we normally used. I was repeatedly shocked by 160+/- VAC before I figured out it wasn't stray strands of wire poking my arm when I reached across the cable bundle, but stray electrons. Not my most shining moment, but it did provide an opportunity to improve the product before release.

Because simple VFDs typically run open loop I don't think they are as susceptible, but I've seen several brands of servos choke (fault) on the the "noise" energy they produce, due to poor design and/or component choices. Proper shielding and bonding would get customers through in most cases, but in others it was back to the drawing board.

J Tiers
10-09-2014, 09:35 PM
Very timely, since at work we are in the process of eliminating emissions from a military system including a VFD (for one of our clients). We so far have successfully dealt with what they told us came up at the lab. Most of it was due to wire harness problems. (we don't have a large enough shield room to do the RE102 tests on this product)

A shielded cable from the VFD to the motor is good, it does indeed reduce emissions quite a bit. The client had one, but included temp sensor wires inside it, which naturally was a not-so-good idea.

That said, a good twisted or tightly braided set of wires will do a very good job also. Not quite up to the performance of a shield, but quite good so long as other wires are not bundled together with the VFD output.

However, ONE thing is really very important. RUN A GROUND WIRE WITH THE VFD OUTPUT WIRES, AND CONNECT IT TO *BOTH* THE VFD CHASSIS AND THE MOTOR CHASSIS.

Someone may pop up and talk about "ground loops"..... save that for the airport.... It is very important to give the current that flows through the wire capacitance inside the motor, some way to return directly to the VFD, where it originates. That "common mode" or "longitudinal" current is at fault in many , even most, cases of interference.

Because the current is high frequency, it will tend to take the path that is most direct, and that will be right back along the ground wire close to the VFD output wires. Any other path will be higher impedance, usually.

Why do you care? because a VFD can wipe out local radio and non-cable TV reception, and may even interfere with its own control signals, if you don't take some care with wiring.

MaxHeadRoom
10-09-2014, 10:54 PM
I still maintain the benefits of a 3 ph inductor can't be disregarded and does good, in three areas, eliminates or reduces the high frequency switching noise, removes any stress on the motor insulation, and is easier on the drive itself.
Max.

caveBob
10-09-2014, 11:39 PM
I still maintain the benefits of a 3 ph inductor can't be disregarded and does good, in three areas, eliminates or reduces the high frequency switching noise, removes any stress on the motor insulation, and is easier on the drive itself.
Max.

Max, thanks for the info and link. I'm just beginning to learn about VFDs so have image searched many examples used on machines. I remember seeing line reactors in a few of them but didn't know what they were or did. Is this the one I would need for my Teco JNEV-101-H1 (http://dealerselectric.com/JNEV-101-H1.asp)?:

LR-11P0-1PH
http://www.automationdirect.com/adc/Shopping/Catalog/Drives/AC_Drive_(VFD)_Spare_Parts_-a-_Accessories/AC_Line_Reactors/LR-11P0-1PH

malbenbut
10-10-2014, 04:15 AM
I use Computer Ethernet cable for remote control wires, very cheap £1 for 5 metres works great never had any trouble using them usually 8 wires per cable.
MBB

J Tiers
10-10-2014, 07:47 AM
An output inductor is a very good thing... reduces stress on the insulation, and MAY reduce EMI.... MAY do so, not guaranteed to.

Most of the inductors are iron core (literally) and are wound in such a way that high frequencies can simply bypass them. The pictured ones in the link are of that general type. Some are better than that. But they are not designed to be EMI reduction devices primarily.

caveBob
10-10-2014, 09:28 AM
...still digging for more info... J Tiers or MaxHeadRoom - so the line reactor can be likened to a better power supply on the PC where it conditions/regulates the power? Turns the "bad stuff" into heat which dissipates? Trying to get my head around what it does...

So, if I can find say, a 12x12x8 Hoffman box to enclose the VFD and now line reactor, will/might need to use a fan to ventilate and cool?

George Bulliss
10-10-2014, 09:56 AM
...still digging for more info...

The manual is actually pretty good and readable for this unit. I picked up a pdf somewhere and would be happy to send you a copy if you like. It covers pretty much all the questions you have, wiring specs, mounting and cooling, filters, etc. PM me if you would like a copy.

BTW, I see the manual recommends shielded cable for runs over 330'. Not really a problem in my shop!

MaxHeadRoom
10-10-2014, 10:00 AM
Max, thanks for the info and link. I'm just beginning to learn about VFDs so have image searched many examples used on machines. I remember seeing line reactors in a few of them but didn't know what they were or did.
LR-11P0-1PH
http://www.automationdirect.com/adc/Shopping/Catalog/Drives/AC_Drive_(VFD)_Spare_Parts_-a-_Accessories/AC_Line_Reactors/LR-11P0-1PH

That link is for a input side reactor, Some makes differentiate by also including MOV's and small Caps on the input side version, but you need a plain motor side version.
It reduces or eliminates the H.F. PWM switching component which is mainly the cause of EMI or radiation, this also reduces the stress on motor windings, especially if you have an older non-vector rated motor, it is also somewhat of a buffer between motor and VFD.
If you use it totally enclosed you may need a cooling fan, there are some VFD's that allow fully sealed enclosure by allowing the Heat sink to protrude through the rear of the enclosure allowing cooling while retaining the seal.
Max.

J Tiers
10-10-2014, 08:16 PM
Line side reactors are used to limit the input current surges as the diodes conduct. Many VFDs require them.

Output side inductors reduce the dv/dt of the output waveform, basically "slowing down" the wave so it consists of lower frequencies when it reaches the motor. The added inductance also tends to cause the output to be more sine-wave-like (current waveform, mostly).

SGW
10-11-2014, 07:25 AM
Re: Reply #16
I used a piece of Ethernet cable for the low-voltage control wiring on my mill as well. Haven't had any trouble.

Glug
10-11-2014, 09:33 AM
When I put a chinese spindle and Hitachi VFD on my CNC router, I used Igus cable. They shipped very quickly, prices were reasonable, and they did not have have the large minimum order that many places do. I like the cable a lot - very flexible and still robust. Being able to just go online and order it without getting quotes, etc, was also key.

I ended up going with CF31-15-04. The website is .DE but it ships from the USA.

http://www.igus.de/iPro/iPro_01_0013_0016_USen.htm?ArtNr=CF31-15-04&c=US&l=en

caveBob
10-11-2014, 10:06 AM
When I put a chinese spindle and Hitachi VFD on my CNC router, I used Igus cable. They shipped very quickly, prices were reasonable, and they did not have have the large minimum order that many places do. I like the cable a lot - very flexible and still robust. Being able to just go online and order it without getting quotes, etc, was also key.

I ended up going with CF31-15-04. The website is .DE but it ships from the USA.

http://www.igus.de/iPro/iPro_01_0013_0016_USen.htm?ArtNr=CF31-15-04&c=US&l=en

Glug, for the same reason (wanting to stifle electrical noise in the system), I've saved a few to check out again later:

IGUS Chainflex® Shielded PVC Control Cable - (IGU-CF6-07-04) $1.98/ft - 18AWG
http://www.cableorganizer.com/igus/pvc-control-cable/

IGUS® Chainflex® CF10 Shielded TPE Control Cables - (IGU-CF10-07-04) $2.62/ft - 18AWG
http://www.cableorganizer.com/igus/cf10-tpe-control-cables/

IGUS® Chainflex® PVC Power Cables - (IGU-CF31-25-04) $3.74/ft - 14AWG
http://www.cableorganizer.com/igus/pvc-power-cables/

I've ordered plastic bearings from IGUS/USA website before, they shipped quickly, very good experience. Anyway, CableOrganizer carries some of their stuff as well and have ~9' left of this:

IGU-CFBUS-066 IGUS Chainflex TPE USB Cable, 20 AWG, 2 Conductors / 20 AWG, 1 Twisted Pair, Per Foot
http://gcs.cableorganizer.com/?q=IGU-CFBUS-066

...might work well for the control cable part, only 2 conductors though...?

That's been the goal with special cabling, enclosures and as yet un-found: specific "plain motor side version reactor", keeping it quiet, electrically.

MaxHeadRoom
10-11-2014, 11:11 AM
That's been the goal with special cabling, enclosures and as yet un-found: specific "plain motor side version reactor", keeping it quiet, electrically.

If you look at the chart the column of the link you posted it shows Phase-Use has I O or I/O for the Input Output or Dual use types.
http://www.automationdirect.com/static/specs/aclr.pdf
Max.

caveBob
10-11-2014, 02:21 PM
Max, thanks for you patience with my ignorance... I'm trying... but still a little cornfused. From the aclr PDF it says, emphasis mine:



LR Series Line Reactors Input line reactors protect the AC drive from transient overvoltage conditions typically caused by utility capacitor switching. Input line reactors also reduce the harmonics associated with AC drives, and are recommended for all installations.

Output line (load) reactors protect the motor insulation against AC drive short circuits and IGBT reflective wave damage, and also allow the motor to run cooler by “smoothing” the motor current waveform. They are recommended for operating “non-inverter-duty” motors, and for any motors where the length of wiring between the AC drive and motor exceeds 75 feet.

Input line reactor recommended for all installations, the Teco is 120V in, so I would need to block/insulate B1 and B2 right? From the same PDF:



Single phase applications Some of the line reactors are listed for use with single-phase input power. Make sure that terminals B1 and B2, if present, are properly insulated before any connections are made.

If I'm understanding you and J Tiers correctly, you guys are saying I should be running a motor side reactor as well? The Brooks motor is "inverter duty rated" and the line from Teco to motor will be ~5'... still need one here too? Just trying to understand and get only what I need... :)

MaxHeadRoom
10-11-2014, 03:07 PM
Although I have fitted both to VFD installations, I tend to use just the motor side only now, the input reactor is more for preventing line interference and in the case of a number of VFD's, the power factor, not so much a benefit for the VFD itself, apart from maybe cushion the input bridge surge, unlike the output side, which has multiple benefits, even with a vector rated motor.
There are many that run with neither, just depends on what chance you want to take?
On grounding, bonding and shielding http://www.automation.siemens.com/doconweb/pdf/840C_1101_E/emv_r.pdf?p=1
Max.

lakeside53
10-11-2014, 08:17 PM
+1 on the above. forget the input filter or your small vfd's.