VFD Cable - Worth It?

Status
Not open for further replies.
A

adamscb

Senior Member
Location
USA
Occupation
EE
I'm doing a project where we're installing a 300hp vfd on an existing motor. The length of the entire run is about 250-300 feet. I'm buying load and line side 3% reactors - my question is do I need this special cable for VFDs? I've been told that it is a pain to pull (to the point of having to take apart the conduit), as well as much more expensive. Does anyone have any experience with this? Thanks
 
Opinions on this will vary, but in MY opinion (and I am in the Drives business), VFD cable in STEEL conduit is redundant, so your comment on "having to take apart the conduit to pull it" leads me to believe that this is the case and really, the benefits are not there. There is a slight benefit in that the 3 cables and (typically) 3 nested ground conductors will provide a consistent geometry in the cable formation, which aids in minimizing cable capacitance and standing wave generation, a LITTLE. But like I said, that benefit is likely not worth the hassle factor when it comes time to try to pull that inside of conduit.

My suggestion if it is steel conduit is to use XHHW or RHW-2 insulated cable (not THHN) and this is the only circuit in that conduit. If it is PVC conduit (or aluminum, fiberglass etc, anything non magnetic), or you are planning on pulling multiple circuits of any sort in the same conduit (which should be avoided if possible) then you must use the shielded VFD cable.

You will see most VFD mfrs (including mine) state in their installation instructions that you MUST use it, but that's because most of them are selling their drives all over the world and outside of North America, it's often a codified requirement anyway as they tend to not use steel conduit like we do. So the VFD mfrs do a little CYA statement about always using it. And of course, you will find that EVERY cable mfr suggests that you use it...

Now let the s___ storm begin on my comment about not using THHN... :roll:
 
This is inside RMC and is the only circuit, so that leads me to believe I should be fine. I'm under the impression that this special VFD cable is used in place of multi-conductor regular THHN cable, but we don't use multiconductor cables for motors, we use separate runs of THHN inside the same conduit.

I am curious as to why you recommend RHW and not THHN. Is it because of the tougher insulation, or is it because I'm using a VFD? This 300hp has been running off of THHN for some time now, and we haven't had any issues. If you could, could you share your thought processes behind that recommendation?

Now let the s___ storm begin on my comment about not using THHN... :roll:
Click to expand...

Ha! I saw this after I typed the above response...you called it.
 
Last edited:
There are several characteristics of wire insulation that describe its performance when exposed to voltage stress.
The nominal maximum applied voltage (either RMS AC at 50-60HZ or DC) is the one which dominates the NEC use descriptions.

But another is the response of insulation to rapid changes in applied voltage (dV/dt), and the output of a VFD is high voltage DC pulses with much faster rise times than normal AC power. In addition to that the possibility of standing waves on VFD output wires can cause even the pulse voltages seen to be up to twice as high as the nominal DC bus voltage in the VFD. So that characteristic becomes important.

Yet another characteristic is how the insulation responds to repeated overvoltage voltage stress. It can degrade slowly or fail catastrophically.

Taking all of these (and probably others) into consideration, there are those who feel that THHW is not the best choice for VFD output wiring.
 
GoldDigger said:
There are several characteristics of wire insulation that describe its performance when exposed to voltage stress.
The nominal maximum applied voltage (either RMS AC at 50-60HZ or DC) is the one which dominates the NEC use descriptions.

But another is the response of insulation to rapid changes in applied voltage (dV/dt), and the output of a VFD is high voltage DC pulses with much faster rise times than normal AC power. In addition to that the possibility of standing waves on VFD output wires can cause even the pulse voltages seen to be up to twice as high as the nominal DC bus voltage in the VFD. So that characteristic becomes important.

Yet another characteristic is how the insulation responds to repeated overvoltage voltage stress. It can degrade slowly or fail catastrophically.

Taking all of these (and probably others) into consideration, there are those who feel that THHW is not the best choice for VFD output wiring.
Click to expand...
All that, plus this bag of chips;

Thermoset plastic PVC insulation, which is what is on THHN (Thermoset High Heat Nylon jacketed), is injected as a liquid into a heated mold as the wire passes through and the heat sets it up to a more solid form. But as a liquid flowing through the wire making machine it has bubbles in it. Those bubbles are generally microscopic and in a standard sine wave environment are basically irrelevant. But in the VFD output conditions GD just described so well, those bubbles become weaknesses in the insulation that loom large under the higher "corona discharge" stresses that can be created by the VFD PWM output and cable capacitance. Secondly, the wire can move around a little inside the mold before the PVC sets up, so some areas of the wire may end up with thinner insulation than others and again, no big deal on sine wave power, but not good with PWM. The better type of insulation is Cross Linked Polyethylene or "XLPE", which is what you find in RHH/RHW-2 or now more commonly, XHHW. The "X" is for XLPE ; then High Heat Water resistant. XLPE is basically heat shrink tubing that the wire is inserted into, then when heated it shrinks onto it. No bubbles in the material, better resistance to corona discharge.

No need to run out and yank your THHN, but we should all cease using it now on VFD outputs (input cables are not subjected to the same issues). I've been involved in dozens of somewhat spectacular cable insulation failures over the last 6 years or so, it's getting worse. I believe that's partly because as VFDs get smaller/faster/cheaper, one of the reasons is that they have reduced the switching losses in the transistors by using faster transistors, meaning the turn-on time is now MUCH faster, buy a factor of more than 2, than the previous generation, which was already faster than it's predecessor. So VFD cables that have been installed and running fine with THHN for a decode or more were not being subjected to the same thing that cables are seeing from the latest generation of VFDs. THHN is beyond it's usefulness now.
 
JMNSHO.

It seems silly to me to run VFD cable in metal conduit. What has happened is that the Europeans use little in the way of conduit so everything they do is mostly oriented around cabling methods. Because everyone wants to build equipment that can go anywhere in the world, pretty much the VFD manufacturers have accepted the idea of cabling, even though it has been a hard sell in the US, just due to tradition if nothing else.

Then they got in bed (metaphorically speaking) with the VFD cable manufacturers and started suggesting it is somehow an improvement over traditional wiring methods.

It used to be that it was really common for machine tool builders to wire up motors that had drives on them with SO type cord run through RMC, as that was considered to be a better option standard wire. Some machine tool builders decided this was a code violation and so started taking individual strands of wire and twisting them together to form their own make shift cables, since they were not otherwise available. My personal opinion is that the SO cord was probably a superior wiring method, but that is just an opinion.

Then the VFD manufacturers went crazy selling VFDs to every Tom, Dick, and Harry. They no longer wanted to provide good pre-sale advice on how to use install them so chose to go with recommendations that are highly conservative, presumably to reduce tech support calls, and provide an out when stuff doesn't work right. This is understandable as VFDs became more of a commodity type item, it became impossible to pay for tech support people.

My suspicion is that only a small percentage of motors that have drives on them use VFD cable. It is becoming more common, but it adds considerable cost and complexity to the picture. Just terminating the cable itself correctly is not as easy as one might think. A lot of motors are not well setup to do so, especially if wired in conduit.
 
zbang said:
Jraef, I'm curious whether, or how much, a load-side reactor close to the VFD mitigates the need for either specific VFD cable or for using XHHW instead of THHN.
Click to expand...
Load reactor alone, not much of a difference in helping to protect the cables. A DV/DT filter is better for that and helps with other issues such as bearing damage; it's a load reactor plus an RC trap filter for any high voltage spikes that are created anyway. If the cable length is 2,000 ft or more, you need to step up to a Sine a Wave filter.

ptonsparky said:
X2

XHHW is SO in this area. :(

What are the failures like? The load side failures of conductors I've been privy to, the drives shut down. My exposure is limited for sure.
Click to expand...
I'm at home so I don't have it here, but later when I get to work I'll share a photo I have of THHN pulled out of a project where you can see the burn marks. Unfortunately the shot is close up to see the marks, which doesn't show the more interesting aspect; the equal distances between them along the entire length, which is the wave length of the standing wave that was in them. The burns were phase-to-phase and the conduit was dry, so they never saw anything with a megger test to ground, you could only see it after pulling the wire. The VFD was seeing it however, it kept tripping on what's called "Hardware Over Current", meaning it was a current spike on the output side that was too fast for the microprocessor base OC trip functions to see and this drive had a backup protection circuit in the transistor hardware that doesn't wait for calculations, it just acts first and tells you later.
 
We use SWAPVC or XPLE for power cables - input, output and fixed speed. And pretty much everything else.
No conduit involved.
 
Status
Not open for further replies.
Top