VFD Recommended Fuse Dictates Input Conductor Size

[Inrush Jeff]

Hello, I believe this should be pretty straight forward. We have a 40hp, 460VAC, 3 phase motor and a nice ABB ACS880 drive rated for 52 input amps. The input conductors would then be 52 x 1.25 is 65A, so 6awg at 75C. Well, the ABB manual lists 80A Class T fuses and 4awg "Power Cables".

I am a bit confused about scenarios where only the input fuse is specified by the manufacturer but perhaps not the conductors. The ABB manual is a bit confusing since they simply say "Power Cables" so it could mean input or output or both, although I would lean towards it meaning only input conductors. If an 80A Class T fuse is specified I don't see how you could use 6awg conductors. Then it makes sense that they list 4 awg "Power Cables". This leaves it looking good to use 6awg for the output conductors to the motor, sized at 125% of the motor FLAs, which is 52A from 430.250. I am way off base here?

Thanks a lot! Jeff

 

[Jraef]

The fuses are mostly there in order to get a decent SCCR (Short Circuit Current Rating) listing from UL. It's virtually impossible to get more than a 5kA SCCR on any VFD without using high speed fuses, so VFD mfrs specify fuses that will not blow unless there is a major meltdown in the drive front-end. The VFD itself will trip on any fault on the load side far far faster than any fuse. If you look more carefully, the fuse recommendation is for MAXIMUM fuse size, you can use a fuse that is lower if you want to, but it increases the chances of it clearing and having to be replaced.

If that fuse is going to be the Branch OCPD, then yes, you would have to size the conductors to the fuse that you use. But you can use a lower Branch OCPD up stream that protects your #6 and also have the 80A fuses ahead of the drive to satisfy the UL requirements.

 

[Inrush Jeff]

Ah yes, that makes more sense. They are moving from a normal starter for this motor to a drive so the old bucket will likely keep the breaker since it is also the disconnect. The 6 awg meets 430.122(A) sizing and protection is provided for the existing 6awg conductors that they will shorten and reroute to the much closer drive. New 6 awg VFD cable will then be installed from drive to motor. Overload protection for that comes from the drive.

If the breaker is replaced with a simple disconnect then the fuses would be the only short circuit and ground fault protection and the conductors would need to increase to 4 awg. Although now that I think about it, 6 awg looks fine with 80A non-delay fuses per 430.52, allowing 150 to 300% of the motor FLAs?

 

[Inrush Jeff]

Here is my understanding now, let me know if this makes sense:

40hp 3 phase, 460VAC motor is 52FLA per [430.250]
This ABB drive also has an input rated current of 52A listed on it

[430.122 (B)] Conductor sizes from drive to motor at least 125% of motor FLAs so 52A x 1.25 = 65A so 6 awg at 75C is OK. Overload protection is provided by the drive.

[430.122 (A)] Conductor sizes from disconnect to drive at least 125% of drive rated input current so 52A x 1.25 = 65A so 6 awg at 75C is OK. Short Circuit and Ground Fault protection is provided by the fuses or breaker after the disconnect.

If only Class T fuses will be used for Short Circuit and Ground Fault protection then per [430.52] 150% to 300% of motor FLAs so 52A x 1.5 = 78A so 80A non-delay fuse (Class T) is OK to protect 6 awg.

 

[paulengr]

Drive to motor, 125% of motor FLA. Drive protects.

System to drive, use DRIVE FLA x 1.25. Motor side has nothing to do with it. Can use in this case fuse size but need bigger cables.

 

[Inrush Jeff]

Drive to motor, 125% of motor FLA. Drive protects.

Sounds great, pretty clear in the code.

System to drive, use DRIVE FLA x 1.25. Motor side has nothing to do with it. Can use in this case fuse size but need bigger cables.

I don't quite know what that means. Please refer to NEC code sections and use more of a description.

I will also clarify that this is a wall mounted VFD being fed from an existing motor bucket so it would not be considered a listed single assembly or combination motor controller. The final design will be a simple disconnect feeding a set of fuses feeding the wall mounted drive that feeds the motor. The motor and drive data are in the first post.

I am using the 2020 NEC and it looks like the SC and GF protection for a VFD comes from 430.130(A)(1) and that says to refer back to the classic 430.52 where we do appear to use the motor amps but per 430.130(A)(1) (Exception to 1), we get to use the drive amps. Since this drive lists a maximum Class T fuse of 80A it looks like we move to 430.130(A)(2) where it says we cannot exceed that 80A protection. So I believe that makes it our maximum value for the motor branch circuit SC and GF protection. It also looks like we can correctly use a non-delay fuse for that type of protection. I believe that if the drive didn't list a maximum fuse then we could size it 150% to 300% of the drive input rating.

Now to figure out the size of the conductors that feed the drive has already been mentioned and made clear and is shown in 430.122(A). It appears to simply be at least 125% of the drive input current, not the motor FLAs. The input conductor size perhaps has nothing to do with the size of the SC and GF protection? I know we often end up with very large circuit breakers protecting very small conductors with a more normal motor using a starter setup so I feel that also makes sense here with a VFD?

So I suppose this leaves the question: if the SC and GF aren't listed by the manufacturer then do you simply size the SC and GF protection up to 300% (if using non-delay fuses) of the drive rated current and size the drive input conductors to at least 125% of the drive input rating? I don't think I would ever upsize the input conductors because of the size of the SC and GF protection. Thanks a lot! Jeff

 

[Eddie702]

Sounds great, pretty clear in the code.


I don't quite know what that means. Please refer to NEC code sections and use more of a description.

I will also clarify that this is a wall mounted VFD being fed from an existing motor bucket so it would not be considered a listed single assembly or combination motor controller. The final design will be a simple disconnect feeding a set of fuses feeding the wall mounted drive that feeds the motor. The motor and drive data are in the first post.

I am using the 2020 NEC and it looks like the SC and GF protection for a VFD comes from 430.130(A)(1) and that says to refer back to the classic 430.52 where we do appear to use the motor amps but per 430.130(A)(1) (Exception to 1), we get to use the drive amps. Since this drive lists a maximum Class T fuse of 80A it looks like we move to 430.130(A)(2) where it says we cannot exceed that 80A protection. So I believe that makes it our maximum value for the motor branch circuit SC and GF protection. It also looks like we can correctly use a non-delay fuse for that type of protection. I believe that if the drive didn't list a maximum fuse then we could size it 150% to 300% of the drive input rating.

Now to figure out the size of the conductors that feed the drive has already been mentioned and made clear and is shown in 430.122(A). It appears to simply be at least 125% of the drive input current, not the motor FLAs. The input conductor size perhaps has nothing to do with the size of the SC and GF protection? I know we often end up with very large circuit breakers protecting very small conductors with a more normal motor using a starter setup so I feel that also makes sense here with a VFD?

So I suppose this leaves the question: if the SC and GF aren't listed by the manufacturer then do you simply size the SC and GF protection up to 300% (if using non-delay fuses) of the drive rated current and size the drive input conductors to at least 125% of the drive input rating? I don't think I would ever upsize the input conductors because of the size of the SC and GF protection. Thanks a lot! Jeff

To size the line side conductors you have to use Drive Input Amps x 1.25 not code motor amps. The drive uses some power that is the reason. In your case you say both the drive and the motor are rated at 52 amps so you should be fine in this example. The drive usually adds a couple of amps

 

[paulengr]

With a motor the issue with circuit breakers and fuses (breakers more so) is avoiding motor starting characteristics without nuisance tripping. With los peak fuses (some J, RK1, some CC) you can size to 125% without nuisance tripping. With others it’s 175%. With standard thermal-magnetic breakers at 250% of FLA inrush is still often an issue so adjustable magnetic trip is necessary. So what we end up with is that overcurrent and GF are provided through the overload relay. The breaker/fuse provides some backup protection (particularly with low peak fuses) but mostly only provides short circuit protection. Among other things inrush is as high as 22-23x FLA with motors intended for European markets (so-called world motors) although NEC limits short circuit settings to 17x FLA. Locked rotor currents generally around 6x FLA but with some high torque motors can get up to 8-10x FLA.

In contrast a VFD is flat out incapable of delivering these levels of starting current but we don’t need it. Although many drives don’t actually state peak current draw it is usually no more than 2x input rated current.

Treat a VFD like any other load for ampacity. Sure there are special rules available but it is a continuous load so rated as 125% of inout current for minimum ampacity for sizing. Larger won’t hurt. I have no idea where ABB gets their data from. They frequently have goofy information that does not match any Code or standard. It isn’t a motor load.

As to short circuit protection pay attention to rated SCCR which with most VFDs is very low. The solution is either to use a small isolation transformer, add impedance (line reactors) which is the same idea but cheaper, or use high speed low energy let through fuses so usually “semiconductor” fuses.

That leaves feeder protection. There is nothing wrong with just treating it as a normal branch circuit device which would use an overcurrent device sized to ampacity (125% of continuous load). The high speed fuses are then supplemental and trigger only during a large short circuit. If you only use the drive fuses then the 430 rules apply. I don’t normally do this though and I put in a lot of drives because high speed fuses are stupidly long lead time and expensive. Adding impedance has other benefits. It’s just a 3% line reactor.

 

[Inrush Jeff]

Very interesting, in this case the Class T fuses required by ABB are allowing them to get the high SCCR (100kA) and I thought that since they are already needed to do that then they can also provide the SC and GF protection. They certainly should only come into play during a very bad situation and they need to stay out of the way during startup, much easier with a drive. I know many times with normal motors we can have the SC, GF and Overcurrent protection in one device and the SC and GF trip is set well beyond the ampacity of the conductors.

Treat a VFD like any other load for ampacity. Sure there are special rules available but it is a continuous load so rated as 125% of inout current for minimum ampacity for sizing.

That leaves feeder protection. There is nothing wrong with just treating it as a normal branch circuit device which would use an overcurrent device sized to ampacity (125% of continuous load).

I don't know if that is actually code compliant if the SC and GF protection is only 125% of rated drive current. Of course working fine long term with no problems is quite likely but I am definitely trying to sort out this as purely a code compliance question. With no code section saying to not treat it as a motor branch circuit I still believe I will need to size the SC and GF protection for a motor branch circuit using a VFD as I had suspected and use 150% to 300% of the rated drive input current if using non-delay fuses. If using an Inverse Time Breaker then it is 150% to 250%. This is shown in 430.52. Since ABB recommended 80A fuses those do end up being in that range at 153%.

This is a real good discussion. Jeff

 

[Inrush Jeff]

To size the line side conductors you have to use Drive Input Amps x 1.25 not code motor amps. The drive uses some power that is the reason. In your case you say both the drive and the motor are rated at 52 amps so you should be fine in this example. The drive usually adds a couple of amps

Yes, seems straight forward and I agree. Now what are your thoughts on sizing the Short Circuit and Ground Fault protection for the drive?

 

[Jraef]

Don't make this harder than it needs to be.

You size the SCPD per the manufacturer's instructions as per their UL listing, that is what it says 430.130 A.2 says.
Their instruction only give you the MAXIMUM size, you can do what you want so long as it is never MORE than that. By the way, 450.52 is also the MAXIMUM size, not the absolute size. Its a rule, not a design guide.

I believe that if the drive didn't list a maximum fuse then we could size it 150% to 300% of the drive input rating.

I guarantee to you that you that there is no way you can use that drive without fuses if they only list fuses as the SCPD. The fuses will be part and parcel to their SCCR listing, so without using their specified fuses, you will only get a 5kA SCCR and in most industrial environs, you will ahve way more than that available.

 

[paulengr]

Very interesting, in this case the Class T fuses required by ABB are allowing them to get the high SCCR (100kA) and I thought that since they are already needed to do that then they can also provide the SC and GF protection. They certainly should only come into play during a very bad situation and they need to stay out of the way during startup, much easier with a drive. I know many times with normal motors we can have the SC, GF and Overcurrent protection in one device and the SC and GF trip is set well beyond the ampacity of the conductors.




I don't know if that is actually code compliant if the SC and GF protection is only 125% of rated drive current. Of course working fine long term with no problems is quite likely but I am definitely trying to sort out this as purely a code compliance question. With no code section saying to not treat it as a motor branch circuit I still believe I will need to size the SC and GF protection for a motor branch circuit using a VFD as I had suspected and use 150% to 300% of the rated drive input current if using non-delay fuses. If using an Inverse Time Breaker then it is 150% to 250%. This is shown in 430.52. Since ABB recommended 80A fuses those do end up being in that range at 153%.

This is a real good discussion. Jeff

When I stated 125% of continuous load short circuit is 6-10x that value on say a thermal magnetic circuit breaker or a fuse. Same as a general load.

SCCR is the issue just as it is with an MMS. The standard technique is to use line reactors to increase impedance thus reducing short circuit. Done VFDs though are much better than others. 5kA is a pretty common spec. It’s awful but not uncommon.

 

[Inrush Jeff]

No problem at all just trying to get to the bottom of this rabbit hole so I can be confident in the method.

Jraef

You size the SCPD per the manufacturer's instructions as per their UL listing, that is what it says 430.130 A.2 says.

Inrush Jeff

I am using the 2020 NEC and it looks like the SC and GF protection for a VFD comes from 430.130(A)(1) and that says to refer back to the classic 430.52 where we do appear to use the motor amps but per 430.130(A)(1) (Exception to 1), we get to use the drive amps. Since this drive lists a maximum Class T fuse of 80A it looks like we move to 430.130(A)(2) where it says we cannot exceed that 80A protection. So I believe that makes it our maximum value for the motor branch circuit SC and GF protection. It also looks like we can correctly use a non-delay fuse for that type of protection. I believe that if the drive didn't list a maximum fuse then we could size it 150% to 300% of the drive input rating.

This was from my earlier posts. I was a bit confusing at the end where the 150% to 300% is actually a maximum depending on the type of motor you have and not a range.

Jraef

Their instruction only give you the MAXIMUM size, you can do what you want so long as it is never MORE than that. By the way, 450.52 is also the MAXIMUM size, not the absolute size. Its a rule, not a design guide.

That is an excellent point that those are maximums. So PaulEngr mentioning 125% for SCPD was indeed code compliant. It is as simple as that.

Jraef

I guarantee to you that you that there is no way you can use that drive without fuses if they only list fuses as the SCPD. The fuses will be part and parcel to their SCCR listing, so without using their specified fuses, you will only get a 5kA SCCR and in most industrial environs, you will ahve way more than that available.

I totally agree and it is true that fuses are a big part of the SCCR for drives and that is no different for this particular one. If one wasn't specified by a manufacturer, which would be very rare indeed, then is it true that it would fall to 430.130 A.1 to select the protection from 430.52? If this is true that puts that question to bed.

So that leaves the situation of the input conductors sized for at least 125% of the drive rated current with no need to take into account the SCPD that can be much higher. This is a question that has been very hard to track down an answer for VFDs, very common for normal motors without drives.

Jraef

If that fuse is going to be the Branch OCPD, then yes, you would have to size the conductors to the fuse that you use. But you can use a lower Branch OCPD up stream that protects your #6 and also have the 80A fuses ahead of the drive to satisfy the UL requirements.

Your original reply is what made me confused about this but now I see that you say OCPD and not SCPD. If the 80A fuses specified by the manufacturer are also used to provide ONLY the short circuit and ground fault protection for the motor branch circuit, then do you need to size the input conductors to take that into account? I don't think that is true and they can simply stay at at least 125% of the drive input current. I believe you can use the 6 awg conductors with the 80A fuse and if, for whatever reason, there were no fuse or breaker specified by the manufacturer, you could protect the 6 awg conductors with up to 156A non-delay fuses (52A x 300%) or 130A Inverse Time Breaker (52x250%). There is no need for overcurrent protection from these fuses since that is done in the drive and protects the input and output conductors as well as the motor from long term high loads. Is this correct?