Solid or stranded wire for motors on VFD's 2

[unwiredo]

Does it matter if the wire feeding a motor from a VFD is stranded or not. I would think the more strands the better.
This is relative to the same wire gauge for comparison.

The stranding gives more paths for the current to flow so the theory goes.

thanks

 

[dpc]

The VFD is irrelevant, but stranded wire should always be used for all motor connections, due to vibration. Stranded wire will generally give a much better, more reliable connection than solid wire.

Basically, we specify stranded wire for everything except 120 V receptacle and lighting circuits.

Who is trying talk you into solid wire?

 

[ScottyUK]

For the stranding to have any effect on current carrying capacity you would need Litz wire where each strand is insulated from its neighbours, and you don't want to be using that without good reason: very expensive and distinctly awkward to terminate.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[jraef]

Look up a wire ampacity chart, the current ratings don't specify solid or stranded. If you find one that does, the ratings for a given gauge will be the same. The only thing that makes a difference from one type of wire to another of the same gauge is the insulation rating.

http://www.allaboutcircuits.com/vol_1/chpt_12/3.html

It's all about flexibility. Motors are used because they make things move. Even though the inherent soft starting capabilities of a VFD will reduce torsional stresses on the conductors, movement means vibration. Stranded conductors flex with it, solid wire can transmit the vibration to the device terminals, eventually loosening them. I don't know of any industrial plants (with engineering standards) that would allow solid wire on anything other than lighting and office outlets.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[davidbeach]

Come on guys, the VFD is relevant and always needs stranded wire even if it were possible to anchor the motor in such a manner as to allow solid for that connection. Finer strands will help as the high frequency components of the VFD output are much more subject to skin effect and increased resistance when compared to the fundamental components.

 

[DickDV]

davidbeach, I doubt your claim that stranded wire has a beneficial effect at high frequencies simply due to the fact that, for skin effect to operate, the skin can't be shorted to anything else over its length. Clearly, a strand in the middle of a bundle has no "skin" since it is all shorted to adjacent conductors.

Or am I missing something here?

 

[davidbeach]

I don't know, but my understanding was that in each strand the higher frequency components more closer to the surface of the strand. If that isn't the way it works, there is a lot of wasted money on finely stranded braid for high frequency grounding.

 

[ScottyUK]

David,

Check out Litz wire. It exists specifically because ordinary stranded wire doesn't bring any benefit for HF usage. Laminated busbars with insulation between layers exists for the same reason. The wiki entry isn't bad:

http://en.wikipedia.org/wiki/Litz_wire

----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[davidbeach]

OK. I haven't had to worry much about skin effect, so what I thought I knew may have been overly simplified.

 

[ozmosis]

Braided shielded cable (as in the screen surrounding the cores) is an important factor when using VFD's to ensure the radiated high frequency interference created has a low impedance path to ground. This is the benefit of the strands in respect to your point David, minimising the impact of skin effect within the screen.

 

[davidbeach]

Thanks ozmosis, glad to know I wasn't completely out to lunch.

 

[DickDV]

ozmosis, if that is why braided shielding is used, then Pirelli is in big trouble. Their motor lead is shielded with sheet copper spiral-wrapped in a single layer. The strips are at least 1/2 inch wide.

Rather, I suspect that the wire braiding is to make the cable more flexible. The Pirelli stuff almost requires a conduit bender to get it around a corner but Belden and Olflex braided shield cable is much easier to bend.

Many drive manufacturers suggest grounded metallic conduit as a first step in containing drive/motor lead noise. That's certainly not stranded!

 

[ozmosis]

Braided shield would provide a low impedance path to ground (assuming the shield is bonded 360Deg at both ends to ground) than a metal conduit. The pirelli cable is also a good path for high frequency interference assuming the screen is bonded 360deg and assuming the cable is not kinked as this increases the impedance.
This point is a slight tangent to the original post but my response was to highlight the simple benefit of braided screen.

 

[ScottyUK]

The benefit or a braided screen is primarily that the braid doesn't open up in the way that a lapped screen can. Both the braid and foil screens are thin enough that skin effect isn't an issue at typical drive switching frequencies, unlike a typical conductor which often has a diameter large enough to be affected by both skin effect and proximity effect. If the strands of a braid aren't insulated from each other then the braid offers minimal electrical benefit, only mechanical benefits such as flexibility. It could be argued that a braid has a larger physical 'diameter' than a solid conductor of equivalent conductor cross section, in the same way as a tubular conductor, but this is not because it is stranded per se but because the copper is physically distributed over a larger area. The same effect could be achieved using a crumpled foil, although that would be mechancially less desireable.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

I was under the impression that the changing magnetic field around a conductor caused the skin effect. This is both current dependant and frequency dependant. I would expect that insulating the individual strands would be effective only if the insulation provided enough separation to avoid interaction of the magnetic fields around the individual strands. I have seen badly corroded microphone cables with enough copper oxide that as the current transferred from the inner strands to the outer strands it detected and fed to the amplifier nearby marine radio transmissions. Individually insulated strands would avoid this.
I was under the impression that braided ground cables for HF grounding used a braid to avoid the increased inductance of a conventional cable with twisted strands. Bare cable may develop a surface corrosion that will impede current flow between strands, forcing the current to spiral down the individual strands increasing the inductive reaction to a high frequency signal.
I have suggested to audiophiles that instead of outrageously expensive special stranded speaker wires they would be better served by solid conductors (no oxidation impeding current transfer between strands after time) or even better served by insulating lengths of copper tubing.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[toygasm4u]

In my experience, mostly with low HP and long lead lengths, solid core is a no-no. HF cable charging currents are always higher on solid; the higher currents are due to the "skin effect" or the fact that your dealing purely with surface current. The surface current is directly underneath your insulation, so your capacitance values go up considerably over the length of the cable, just because of the HF component of the VFD.

Even though most people will recommend a shielded VFD cable for any drive motor combination, it should be noted that shielded is the worst performer for cable charging and common mode currents. The only true benefit of shielded is RF immunity for nearby equipment, and the reduction of ground circulating current in the VFD motor system. The shield, when terminated on both sides provides a low impedance return for the HF vs. the basic bond lead.

It has been my experience simply from a VFD and motor performance perspective, that the best performer is good ol' 600V stranded in conduit. The lack of symmetry in the run keeps capacitive effects low, and if the conduit is up to snuff, your RF mitigation will be minimal.

The secondary concern for solid core is at the motor termination end; solid core has a tendency to suffer insulation scraps at the entry point and in your bends. While your average 460V RMS won't find this pin-hole, the VFD's HF Switched DC output... will. At longer lead lengths, this becomes a greater concern, as the impedance mismatch in the cable will sometimes produce twice buss at the motor terminals (reflected wave). For a 460V drive, this is approaching the 1400Vp mark. Insulation pin-hole "found".

 

[unwiredo]

Thanks to all for the help.
My understanding now is that the stranded wire is much better because it carries the RF current better and requires less current to drive the wire and therefore is more efficient.

Wow who would have though there is so much detail in just type of wire.


Appreciate the feedback

Thanks

 

[ScottyUK]

I'd really appreciate someone posting the reason, with references and / or some analysis of the underlying physics, why stranded wire comprised of non-insulated strands should be better than solid wire of equivalent cross section. Without inter-strand insulation the skin effect affects stranded and solid cored conductors of the same diameter equally, save for the slightly larger diameter of the stranded conductor because of the inter-strand air spaces. The reasons for skin and prox effects are well understood, and solutions exist in the form of Litz wire, laminated busbars, Roebel windings for machines: all these solutions rely on insulation between layers or strands to reduce skin effect.

So how does stranded wire without inter-strand insulation perform better than solid wire? Let's have some facts and analysis, not speculation and hearsay!


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[2dye4]

ScottyUK
I think you are right to be suspicious of this reasoning.
The main reason skin effect is not significantly different is that the circulating currents flow around the outside strands as they are in contact with one another.
There is some higher resistance between strands but this is not very significant at VFD frequencies.

And I also question the idea of "charging current" being different for the two. The surface area may be slightly higher with stranded but those surfaces face each other and their impact on capacitance would be less.

What all this has to do with VFD's i am not sure of. They operate far down in the frequency region where most of this stuff is unimportant.

 

[toygasm4u]

What I stated is in no part speculation or hearsay; in my business, we rely on long lead low HP VFD and Motor systems. It's our bread and butter, and the primary function of my position with the company is to make sure this stuff works.

I have worked closely with Rockwell Automation in Mequon Wisconsin, Delta, ABB, and others regarding low HP applications of this type. Without releasing customer service records, I can tell you that VFD installations which utilize solid core wire for motor leads yields us an 80~90% failure rate. The most common mode of failure is a buss overvoltage and or motor overload due to HF feedback on the buss, and motor insulation breakdown. I get on planes and fly around the country to resolve these kinds of issues. It is all purely a matter of DV/DT effects on the total system impedance which includes motor and motor lead set. While there are output filters that resolve these issues, the price point is an uncomfortable figure to look at, thus, we avoid solid core at all costs.

While my "in a nutshell" explanation may have been a little rough around the edges, my company would literally be out of business due to warranty issues if it weren't true. I would recommend for anyone to have a conversation with a VFD design engineer or project manager to discuss the nature of these phenomenon.

I'm supposed to be working, but here are a few good reads:

http://www.amercable.com/doc/catalogs/vfd/Cable Selection Guide.pdf

http://www.ab.com/drives/techpapers/ieee/pcic.pdf

http://www.belden.com/pdfs/Techpprs/capwmtp.pdf

No speculation required.

~Mark