XHHW and RHH Insulation

[mbrooke]

Where would XHHW-2 and RHH be used over THHN/THWN-2 in building wiring and underground conduit runs? I can't imagine a specific scenario or advantage.

 

[Jraef]

XLPE insulation, as is used in XHH and RHH, is basically heat shrink tubing shrunk down onto the wire, as opposed to THH being liquid PVC being injected onto the wire, then “baked” to cure it. The PVC can get microscopic bubbles in it while liquid, which increases the risk of corona discharge failure, hence the limitation of 600V (XLP is usually 1kV or higher). This often is no issue with standard sine wave power, but looms ugly when used on the output of an inverter, like a VFD, UPS or solar inverter.

 

[mbrooke]

Good to know this. Are you saying that the output of a VFD should be XLPE insulation?

 

[Jraef]

Yes, I’ve been recommending that for years now. I heard about it but ignored it, then I started to see older installations where I could notice burn marks at regular intervals along the motor leads where the corona discharge burned right through from phase to phase. In one case, the leafs passed a megger test because there was no leakage to ground, but the VFD was picking up the p-p fault.

 

[mbrooke]

Yes, I’ve been recommending that for years now. I heard about it but ignored it, then I started to see older installations where I could notice burn marks at regular intervals along the motor leads where the corona discharge burned right through from phase to phase. In one case, the leafs passed a megger test because there was no leakage to ground, but the VFD was picking up the p-p fault.

Never knew this. Any idea what the science behind it is if the voltage isn't going over 600 volts?

 

[Jraef]

With PWM outputs, there is a capacitive coupling effect in the adjacent conductors created by the high speed pulses that creates a "standing wave" of higher voltage on the conductors, which then travels to a point of change in impedance (the motor terminals), then reflects some back, until it hits another change in impedance (drive terminals), and reflects back again, building up all along, like ripples in a pond building on each other to make higher peaks. This "reflected wave phenomenon" (look up that term) can result in voltage peaks that can reach 2-3x the RMS voltage.

Decades ago when I first started working woth VFDs, we worried about motor lead lengths in excess of 100ft., believing that it took that much cable capacitance to cause it. Then it became 75ft after alot of people started seeing damage, then 50ft, then 25ft. Now, it's known to not be totally dependent upon length, although under 25ft it does seem to be rare.

 

[MD Automation]

If I recall correctly, the nature of PWM to control motor speeds can result in short transients where the voltage in all those little square waves can overshoot at (I think) 2x the DC bus. These transients are short but they can exist. Theory states the cable from VFD to the motor is a transmission line and if there is an impedance mismatch between the cable and the load at the end (the motor) it can result in a reflected wave heading back along the line. This reflected wave can combine with the source to create short ringing peaks which are much higher than anything you would have ever seen with a smooth old sine wave driving that motor.

Quick math would show a 480VAC mains VFD should create a DC bus just shy of 700 volts. So the motor cable might see short peaks ~ 1400 volts.

Looks like Jraef beat me to it by a minute. And I trust him when he says it's 2x to 3x transients. So those little overshoots could approach 2000 volts from a 480 VAC main.

 

[mbrooke]

With PWM outputs, there is a capacitive coupling effect in the adjacent conductors created by the high speed pulses that creates a "standing wave" of higher voltage on the conductors, which then travels to a point of change in impedance (the motor terminals), then reflects some back, until it hits another change in impedance (drive terminals), and reflects back again, building up all along, like ripples in a pond building on each other to make higher peaks. This "reflected wave phenomenon" (look up that term) can result in voltage peaks that can reach 2-3x the RMS voltage.

Did not know this, thank you!

Can XC=2pifc also be a culprit?

 

[LarryFine]

Sounds like the electrical version of engine exhaust pressure waves.

 

[synchro]

Decades ago when I first started working woth VFDs, we worried about motor lead lengths in excess of 100ft., believing that it took that much cable capacitance to cause it. Then it became 75ft after alot of people started seeing damage, then 50ft, then 25ft. Now, it's known to not be totally dependent upon length, although under 25ft it does seem to be rare.

Also during the evolution of VFDs, IGBTs replaced earlier GTO and bipolar devices because its faster risetimes enabled higher carrier frequencies and improved efficiency allowing smaller heat sinks. But faster risetimes also made signal reflections become an issue at shorter cable lengths. When the propagation time through the cable (which is proportional to its length) becomes more than about 1/2 the risetime of the VFD output waveform, then the cable starts behaving as a transmission line and reflections can become a problem.

 

[PhenixFord]

Good to know this. Are you saying that the output of a VFD should be XLPE insulation?

Interesting read fellows. But I've never seen this in practice?