Load analysis - opinions needed

[PetrosA]

Last summer I posted about a customer of mine who has a building with what appears to be a 400A service (feeders are sized for such) with parallel fuses in the main disconnect with a combined total rating of 800A per leg (2x400A fuses per leg).

We decided to put a logger on the service in the wintertime, when the owner feels that the heaviest load is present from electric heat in the apartments. I set up an Amprobe DM III last Wednesday evening and it ran until this morning. Based on those readings, it looks like the load peaked at L1 - 371A and L2 - 379A.

I'm not very familiar with how fuses as mains react to loads, so my question is, would replacing the parallel 400A fuses with 200A fuses put the residents in danger of blowing a fuse with those loads?

Any thoughts?

BTW, based on a recommendation from someone here on the forum, I rented the unit through Dave Gibson with Intellirent. Excellent support and service!

 

[Smart $]

...would replacing the parallel 400A fuses with 200A fuses put the residents in danger of blowing a fuse with those loads?

Any thoughts?...

Yeah. Why are you asking? You should know without looking it up that fusing higher than the ampacity of the supply conductors [other than what 240.4(B) permits, and where permitted for a large motor load] is a violation.

As to answering your question without the cynicism blink, just look up the fuse TCC and make an evaluation.

 

[Jraef]

Last summer I posted about a customer of mine who has a building with what appears to be a 400A service (feeders are sized for such) with parallel fuses in the main disconnect with a combined total rating of 800A per leg (2x400A fuses per leg).

What? Is this a hack of some sort? Aside from the likely illegality of running parallel fuses (only allowed if it was designed and tested that way as a FACTORY assembly), they would be acting as an 800A fuse (at best), on a 400A service? WTF?

We decided to put a logger on the service in the wintertime, when the owner feels that the heaviest load is present from electric heat in the apartments. I set up an Amprobe DM III last Wednesday evening and it ran until this morning. Based on those readings, it looks like the load peaked at L1 - 371A and L2 - 379A.

I'm not very familiar with how fuses as mains react to loads, so my question is, would replacing the parallel 400A fuses with 200A fuses put the residents in danger of blowing a fuse with those loads?

Any thoughts?

BTW, based on a recommendation from someone here on the forum, I rented the unit through Dave Gibson with Intellirent. Excellent support and service!

So you are thinking of PARALLEL 200A fuses? Again, not legal unless it is a FACTORY assembled, tested and listed assembly, you CANNOT run parallel fuses in the field. So if you put in SINGLE 400A fuses, you can load them to 320A continuously, 400A for short periods. From your graph I'd say that was fine.

But again, YOU cannot take it upon yourself to use PARALLEL 200A fuses. Article 240.8

 

[PetrosA]

I'll forgive you guys since I know you're latecomers This is a 50 year-old, factory assembled, obsolete main disconnect with paralleled fuses that I researched last summer and can not get any information or specs on. Someone, at some time installed 400A fuses in it, although the service lateral and feeders out of it are rated for 400A. This whole monitoring job is to establish whether it will be possible to fuse it properly at 400A (instead of the 800A worth of fuses in there now) without issues, since no one knows why 400A fuses were installed in the first place (an overloaded service or professional incompetence? Your guess is as good as mine). Anyway, we're not going to just slap the 200A fuses in until we can establish with reasonable certainty that the load is within the limits for a 400A rating because this feeds 16 apartments in a complex for seniors. NOT the kind of place we want to be having unnecessary power outages in

There was a previous discussion here as well as consultations with local inspectors (and calls to some bigwig inspector people in Texas) about this service and switchgear, and it was pretty much determined that since the gear was designed for parallel fuses, it would likely be a violation of the listing and possibly a dangerous idea to try and install single fuses in the holders. Short of changing out the whole service, I don't see a better option.

 

[zbang]

... and it was pretty much determined that since the gear was designed for parallel fuses, it would likely be a violation of the listing and possibly a dangerous idea to try and install single fuses in the holders.

Listings aside, I could see a problem with installing a single larger fuse, but can't really imagine how just removing one 400a fuse creates an unsafe situation. Heck, if one fuse failed open, you could run for years without knowing.

To the original question- with the peak load (how long a peak? how often) of call it 380a/leg, I'd be nervous about fusing down to 400a. OTOH if the feeders are sized for that, I don't see much choice.

 

[Jraef]

I'll forgive you guys since I know you're latecomers This is a 50 year-old, factory assembled, obsolete main disconnect with paralleled fuses that I researched last summer and can not get any information or specs on. Someone, at some time installed 400A fuses in it, although the service lateral and feeders out of it are rated for 400A. This whole monitoring job is to establish whether it will be possible to fuse it properly at 400A (instead of the 800A worth of fuses in there now) without issues, since no one knows why 400A fuses were installed in the first place (an overloaded service or professional incompetence? Your guess is as good as mine). Anyway, we're not going to just slap the 200A fuses in until we can establish with reasonable certainty that the load is within the limits for a 400A rating because this feeds 16 apartments in a complex for seniors. NOT the kind of place we want to be having unnecessary power outages in

There was a previous discussion here as well as consultations with local inspectors (and calls to some bigwig inspector people in Texas) about this service and switchgear, and it was pretty much determined that since the gear was designed for parallel fuses, it would likely be a violation of the listing and possibly a dangerous idea to try and install single fuses in the holders. Short of changing out the whole service, I don't see a better option.

Oh well that's a relevant twist. I'm not sure I would agree with the assessment of it being a "violation of the listing and possibly a dangerous idea to try and install single fuses in the holders" because if you cannot now document what the original design was intended for, even though you can see what it appears to be now, then re-purposing it on your own with parallel 200A fuses would be no different than putting in new fuse holders and 400A fuses. But that's just an opinion, which counts for zilch. If your AHJ is OK with it, that's what matters.

So from a technical standpoint then, the PROBLEM with paralleling fuses is that there is no way they can EXACTLY share load. Close, but no cigar. So as a gross general rule, you are going to have a LOWER threshold of nuisance clearing using parallel 200s vs 1 x 400. How much lower is impossible to tell. So if you and your customer have confidence that the data chart is representative of a typical peak use, then I see no exceptional risk: 209A average with 380A peaks for short duration still leaves you 5% of fudge factor in reserve for any imbalance between the fuses.

But...

Is the place air conditioned? If so, might that not be a better representation of the worst case scenario? Because if so, with almost all the AC is running on the hottest day of the year, then the utility voltage drops because of the strain, your margin of error can evaporate with one final compressor motor coming on. If it is air conditioned, make sure you look at the peak power draw with every AC unit and air handler running full-tilt-boogie while the kitchen is cooking lunch and the maids are doing laundry. That may be how they ended up being over fused.

 

[PetrosA]

Listings aside, I could see a problem with installing a single larger fuse, but can't really imagine how just removing one 400a fuse creates an unsafe situation. Heck, if one fuse failed open, you could run for years without knowing.

To the original question- with the peak load (how long a peak? how often) of call it 380a/leg, I'd be nervous about fusing down to 400a. OTOH if the feeders are sized for that, I don't see much choice.

I've gone over my notes again, and actually the feeders out of the disconnect are 2x 350 MCM so I could use 250A fuses and be alright by code (well, other than the fact that I'd be paralleling fuses...). I think treating it as a 500A service based on the wire size with those peak loads looks pretty safe. So the issue with using a single fuse is that there's no specs on the switchgear available and no way to determine what the ratings of the individual fuseholders is.

 

[mayanees]

parallel fuses

parallel fuses

Given the history of this situation, I like your proposed parallel 250-amp fuses protecting the cable at 500-amp capacity. I would plot it out on a TCC to examine how the fuse curve protects the cable damage curves, in parallel and singly, for selection of the actual fuse type. I would also make sure the fuses were all brand new and identical, unlike what's shown in the picture.

I'm probably the one who recommended Dave Gibson at Intellirents to you, and am glad to hear of his continued first-rate support.

 

[iwire]

Given the history of this situation, I like your proposed parallel 250-amp fuses protecting the cable at 500-amp capacity.

Seems like the logical thing to do and in my area would meet our local rules just fine.

 

[Bugman1400]

Isn't it typical for fuses to carry at least 200% of their rating? How about checking the TCC curves!

I'm not sure what the fuss is over parallel fuses, I see it all the time on large current applications.

 

[charlie b]

So the issue with using a single fuse is that there's no specs on the switchgear available and no way to determine what the ratings of the individual fuseholders is.

I don't disagree with the parallel 250 solution. However, I am of the opinion that if you can't determine what you have in front of you, then the less you touch it the better. Since the existing condition has a pair of parallel 400 amp fuses on each leg, removing one of the fuses on each side seems to me to have the least impact on the existing installation. If you are concerned that you don't know whether the fuse holders are really rated for a 400 amp fuse, then I would ask how you resolve the question of whether the fuse holders are rated for a 250 amp fuse.

 

[Cow]

Given the history of this situation, I like your proposed parallel 250-amp fuses protecting the cable at 500-amp capacity.

Seems like the logical thing to do and in my area would meet our local rules just fine.

I agree with these guys, I wouldn't think twice about using the 250 amp fuses if the wire is rated for it. There are times when you can't make everything perfect, but you can make it better OR safer than it was.

 

[Smart $]

... There are times when you can't make everything perfect, but you can make it better OR safer than it was.

More often than not, speculation prevents perfection. :happyyes:

 

[JRW 70]

Perfection

Perfection

Isn't speculation an iterative process
with the failures leading to new avenues
of discovery possibly leading to near
perfection or perfection itself?

JR

 

[topgone]

Isn't speculation an iterative process
with the failures leading to new avenues
of discovery possibly leading to near
perfection or perfection itself?

JR

Yep. In my other life, my boss calls it "error after error" method!

 

[Smart $]

Isn't speculation an iterative process
with the failures leading to new avenues
of discovery possibly leading to near
perfection or perfection itself?

JR

But any failure is imperfection in and of itself... which can easily be achieved without speculation.

I'm not trying to say speculation doesn't have its merits, but true perfection is only attained without speculating.

 

[PetrosA]

But any failure is imperfection in and of itself... which can easily be achieved without speculation.

I'm not trying to say speculation doesn't have its merits, but true perfection is only attained without speculating.

But other than for a supreme being, perfection can only come by speculation, then iterations of hypothesis, trial and error.

 

[PetrosA]

I don't disagree with the parallel 250 solution. However, I am of the opinion that if you can't determine what you have in front of you, then the less you touch it the better. Since the existing condition has a pair of parallel 400 amp fuses on each leg, removing one of the fuses on each side seems to me to have the least impact on the existing installation. If you are concerned that you don't know whether the fuse holders are really rated for a 400 amp fuse, then I would ask how you resolve the question of whether the fuse holders are rated for a 250 amp fuse.

I agree with you to a point. I think that in this case "less is better" in the sense that the lower the ampacity of the fuses, the better the system will be protected. Since the fuseholders are designed for paralleled fuses, using the smallest, appropriately sized fuses in parallel will protect the wire and give me the best chance at not overloading a fuseholder. Using a single 500A fuse puts me into unknown territory as the system has not been operating with that size fuse and I have no way of verifying the rating of the fuseholder. Leaving a single 400A fuse in while knowing that the service can see peak loads just 20A under that seems very risky from a reliability standpoint.

I think I am justified in assuming that the fuseholders are rated for at least the 250A fuses in parallel based on the physical dimensions of the disconnect itself (approximately 48x48x10), but because of the original parallel fuse design, I would feel less comfortable altering the layout to a single fuse one as that introduces operating conditions that we can infer were not originally taken under consideration.

 

[iwire]

I have no way to back this up it is just my own thinking.

Considering manufacturers still use fuses in parallel inside of equipment I propose there is zero inherent danger in using fuses in parallel.

My guess as to why the NEC prohibits it is simply because in most cases that would mean two separate disconnecting means would be supplying a circuit. This to me could be a safety issue.

In your case you have a single disconnecting means specifically designed to supply a pair of fuses per phase. If you open the disconnecting means all power to the load and fuses is killed.

 

[Smart $]

But other than for a supreme being, perfection can only come by speculation, then iterations of hypothesis, trial and error.

You have to state it accurately to understand:

Perfecting involves speculation, iterations of hypothesis, trial and error.

Perfection is hoped to be the end result of the process. When and if that result occurs, there should be no speculation. If there is speculation at that stage or thereafter, the achievement of perfection is rendered null and void in one form or another.