Very high inrush current 4

[MetalworkerMike]

I've got an old granulator with what I believe to be around a 20 horse motor on it, 575V 3 phase, 20 or 30 years old, or older maybe. The ID plate is no longer legible.
I put a fuse block in for 30A CC time-delay fuses, midget size, and this motor blows them like popcorn.
The motor only draws 11A when running, and the fuses are rated for 150A for at least 10 seconds, so I was amazed when the fuses blew. I put a meter on the motor lines and found that the inrush current is 165A.
I did a megger test and the motor passed without trouble. The amperage isn't perfectly even, ranging from 9.7A to 11.2QA, but this amount of inrush has me baffled.
I can switch to RK5 fuses so I can put 40A fuses on there, but then I have to change the wire gage, etc.
Can anyone suggest why the inrush current is so high for this motor?

Mike

 

[itsmoked]

Are you sure there are no mechanical hangups? A motor that after umpteen years changes like this is likely pointing at a mechanical problem with the drive-train stuff. Can you hand turn the machinery?

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[waross]

If the motor is 20 or 25 HP then 11 amps is lightly loaded. 20 to 25 amps would be closer to FLC and the 165A is then only a little high.
respectfully

 

[RogO]

If you use 40 ampere RK-5 fuses does not necessarily mean you need to increase you wire size. Code requires a minimum of 125% ampacity for wire size, but you may put fuses sized up to 175% FLA. This is for short current and locked rotor protection. You still need to have overload protection sized for the FLA.

 

[RogO]

Looking at your post again the Class CC fuse that you are using is not designed use in other then small motors. If you motor is a 20 HP 575-volt motor the FLA would be 22 amperes. If it is a 25 HP 575-volt motor the FLA would be 27 amperes. I would suggest a 30 to 35 ampere or a 35 to 45 ampere RK5 (Bussmann FRS-R or Shawmutt TRS-R)respectfully.

 

[itsmoked]

Welcome to the forums Rog0.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[rawelk]

Current inrush looks in the ballpark for a 20 HP motor. I think RogO has it nailed; CC fuses aren't good choices for high horsepower motor protection.

RK5's like Bussman FRS or LPS will work, and would be my preferred choice as well. I've seen type J fuses (Bussman JKS, for instance) hold up well in centrifugal pump motor service, and would probably work on a granulator.

I prefer physically larger fuses on motors such as on granulators where load normally surges around a lot. Larger fuses -> more surface area -> better heat dissipation.

One thing you might find interesting to do is assay and record inrush currents vs. nameplate FLA (and grab the 'KVA code' as well) for a number of different motor applications in your plant, and you'll probably see most of the inrush currents 3x to 6x nameplate FLA.

Most NEMA motor nameplates have an entry called "KVA Code", and this can be used to give a rough estimation of expected inrush currents.

A good resource with a table of KVA codes, and formulas for calculating inrush current from them is at

http://www.motorsanddrives.com/cowern/motorterms6.html

 

[itsmoked]

Your undersized fuses could also have heated the holder clips and replacing them often causes the clips to distort and not make good contact. This then runs the too-small fuses even hotter which raises their resistance making them hotter yet, and reduces their current trip point, making them blow sooner than otherwise.

Go with bigger fuses.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[MetalworkerMike]

This motor had run 'flawlessly' on 30A 250V RK-5 fuses. These fuses never blew for the apparently 7 years that they were installed. When I found 250V fuses in a 600V application I decided to swap them out for 600V fuses. I replaced the 30A 250V RK-5's with a new fuse holder and 30A 600V CC fuses. Although smaller, they should have been closer to properly rated for the application. These 600V fuses blew every other startup. The CC fuses were selected partly because there are a lot of them in my plant, and partly because the new fuse holder would fit in the location that the old fuseholder fit in. To use the larger 600V RK-5 fuse holder I had to put it in another box which made the situation a bit more irritating.
It made no sense to me that the 250V 30A fuses worked flawlessly but the 600V 30A fuses blew all the time, but I replaced the new 30A CC fuse holder with a new 60A 600V RK-5 fuse holder and new 40A 600V RK-5 fuses, and they have been working just fine ever since.
So I have my solution, but the fact that the 600V 30A CC fuses didn't work at 600V where the 250V 30A RK-5 fuses _did_, really weirds me out. Both fuses have the same overload rating.

Since the CC fuse holder was new, distortion of the clips shouldn't be an issue.
The physical size of the CC fuses didn't occur to me as a potential problem. Maybe that was the key.
The inrush current on this motor was 165A, which still seems pretty high to me for a 20HP motor.

The wiring I replaced partly because I was there, anyway, and it was easy to do, and partly because I didn't like the sloppy job that was done originally. The #10 wire may have worked just fine, but I _have_ had #10 wire melt when protected by a 30A CC fuse, though it was in a heater control box and the ambient temperature is probably 30-40C in there, anyway. It seemed like cheap insurance and made me feel better.

Mike

 

[itsmoked]

Thanks Mike for the come back.
Glad you got it working.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[rawelk]

Thanks for letting us know how everything worked out.

... Both fuses have the same overload rating.

This didn't seem quite right, so went looking through Bussman's time-current characteristic curves
(from the "Low Voltage, Branch Circuit Rated Fuses" PDF) at their catalog download page

http://www.bussmann.com/2/FullLineCatalogPDFDownloads.html

to see if I could answer why the 250V/30A fuses held up while the 600V CC fuses cleared immediately.

This manual had all of the relevant curves except for higher amp FNQ fuses (they're in data sheet 1012), and figuring for 150A inrush working on a 30A rated fuse the following melt times correspond with fuse type.

LPJ SS - 20 seconds
LPS - 15 seconds
FRS - 12 seconds
FRN - 12 seconds
KTK - 1.2 seconds
FNQ - 1.0 seconds (data sheet 1012)
LP-CC - 1.0 seconds
KTS - 0.5 seconds

I was surprised to see an LPS fuse's curve shows it holds together longer than an FRS, but then looked a current let-thru after 0.05 seconds, and saw the LPS has a lower let-thru of ~ 320A versus 500A for an FRS fuse.

In any case, if the 250V/RK5 30A fuse you indicated is equivalent to a Bussman FRN then it has roughly the same time-current curve as the 30A/600V FRS, and will clear after ~12 seconds with 150 amps flowing through it.

All of the midget fuses (KTK, FNQ, and LP-CC) clear in between 1.0 and 1.2 seconds at the same current, so you'd have been running a tight race bringing the motor to speed before they cut loose.

 

[MetalworkerMike]

Rawelk: I'm having some trouble with your data. I don't have contrary data, but I'm seeing that you're listing a KTK fuse, which is a fast-acting fuse, as having a longer melt time than a LP-CC which is a time-delay fuse. This just doesn't make sense to me. I had looked up the data before changing the fuses to make sure that the CC time-delay fuses had the same 500% for 10 second rating as the RK5s, but maybe I screwed that up. Or maybe the FNQRs have it but LP-CCs don't. Or maybe something is not right in the published specifications.
Actually, I just found a reference to NEC 430 which _seems_ to imply that all CC fuses are considered fast-acting for the purposes of the code as it applies to protecting motor loads, and by that code I can use a fuse rating up to 300% FLA with CC fuses. Of course, no such fuse exists in this case.

Mike

 

[rawelk]

MetalWorkerMike: I'm wondering about the 500%/10 second rating you mention for the LP-CC fuse, because it shows up in Bussman's spec sheet #1023 (page 19 of the aforementioned PDF catalog) as a time-delay, current-limiting, rejection type fuse with 12 seconds (minimum) clearing time at 200% rated amps. That is, a 30A fuse of this type will hold together for at least 12 seconds at 60A. Per the curve it'll want to clear at ~ 1 seconds at 500% (30A*5=150A).

Those are the right numbers, though (500%/10sec), for a Bussman FRS, RK5 fuse.

Fuse clearing curves are funny things. Take a look at the one for the LP-CC fuse, for instance ... over part of it's range (between roughly 1 and 10 seconds) the 15A fuse actually withstands slightly higher current than the 20A fuse, then abruptly the curve kicks over towards lower current.

Thats why I tend to compare several different fuse curves. It is difficult to be very accurate since they are log-log charts, but here are my eyeball estimates for fuse element current clearing at the 0.01, 0.1, 1, 10, 100, and 200 second marks.

KTK-R = 500A, 220A, 150A, 90A, 50A, 40A
LP-CC-R = 450A, 220A, 150A, 95A, 65A, 55A
FNQ = 550A, 280A, 150A, 80A, 55A, ??A
(FNQ trend chart only goes to 100 seconds).

The KTK does clear faster, so to speak, but only once you get into longer time scales.

In the 0.01 second range the LP-CC (a current limiting fuse, after all) clears at a lower current than even the fast-blow KTK, and, at one second, all three types clear at about the same current. At 100 seconds the KTK will have cleared with ~50A, the FNQ at ~55A, and the LP-CC at ~65A.

 

[MetalworkerMike]

Okay, I think I've got it now. Thanks in large part to the information kindly provided by rawelk, I now know where I went wrong.
Firstly, I assumed that a CC time-delay fuse had the same time delay characteristics as an RK time delay fuse. That was incorrect.
Secondly, I double-checked this assumption and saw what I wanted to see, rather than what was there.
Thirdly, when I checked the charts after rawelk's message on October the 7th, I was looking at the performance of a 30A fuse under an inrush load of 150A, which is far outside of the envelope of a 30A CC fuse, so the time delay and fast acting CC fuses acted about the same, increasing my confusion. If you check the behaviour under 200% load instead of 500% load, the time delay fuses act appropriately.

So, I must remember that CC class time delay fuses are only rated for 200% amperage. To offset this the NEC allows for using a CC fuse at 300% of the FLA of the motor. No such fuse exists for an FLA of 20A, so a CC class fuse was simply not available, let alone appropriate, for this situation.

As it turns out, of course, I probably could have gotten away with 30A 600V RK-5 fuses and not bought the 60A holder to use the 40A fuses, but it works, it's safe, and I'm not changing it again if I don't have to.

This new bit of knowledge (the 200% ceiling for CC fuses) is going to change a lot of my thinking when selecting and modifying fusing systems at work. It has already been suggested in this thread, by RogO, that CC fuses are not the best choice for motor starting, and this 200% ceiling certainly puts the last nail in the coffin for me, personally. For small motors where the wiring is more than heavy enough to handle 300% of the FLA, then using CC fuses of up to 300% of the FLA would work, and they're dead easy to deal with. For a motor of over, say, 3HP, though, you're getting into an area where you would need to intentionally run more robust wire to use CC fuses, and that is simply not economical.

So, in the end the situation cost me some irritation, and a few bucks in fuses before it got figured out, but I learned something, so it was worth it.

Knowledge is power!

Mike

 

[itsmoked]

Nice summary Mike!

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[MetalworkerMike]

Thank you for the kind words, itsmoked.
I think that I have decided that my personal threshold with using CC fuses on motors will be 5 horsepower (at 480V or higher). Above that I will use RK-5s or larger. This will avoid wiring issues since I very rarely use thinner than #10 wire for primary power to a motor, even when such a thickness is not necessary. At the same time, the use of CC fuses is a convenience in small control boxes, so I will continue to use them wherever they are usable, for the 5hp and lower motors.

Mike

 

[rawelk]

MetalWorkerMike: Thanks for your words as well, and indeed, you've written an excellent summary, but, in re-reading the thread, find I blew a bit of errata powder into it earlier on.

I'd said Bussman 'JKS' fuses were protecting several of our pumps, but this is bad information - a 'JKS' is a fast-acting, current limiting fuse, and wouldn't be a good choice for motor protection. It should have read 'LPJ'.

 

[MetalworkerMike]

rawelk,
I spent some time yesterday with a Bussman catalogue and pondered the difference between the class J fuses and the RKs and CCs that are in wide use in our plant. The class Js seem to have the benefits of the relatively compact size of a 250V RK, while protecting to 600V, while keeping the full 10s/500% overload rating of the RKs, unlike the CCs which have a diminished overload protection. They seem like a good choice in fuses. Better than the RKs. They have a higher 300,000A interrupting rating, even. Now, they don't come in a 'rejection' version, but if they've never made a fuse that size that doesn't have at least 200,000A interrupting rating then that complaint is not going to hold up under scrutiny. It's a minor issue that you have to have a bit more of your brain turned on when inspecting the systems to make sure it has appropriate fuses... right now I just look to make sure every fuse is a rejection-type fuse and then I don't have to worry about it. As I said, a petulant complaint at best. I will definitely keep the class J's in mind for future use. Right now I'm trying to standardize to reduce the number of fuses I need to stock, but maybe that standard needs to be updated.
More to mull over.

Mike