Voltage Drop limited to 3%

[cdcengineer]

I got one for you.. I tried posting this in an existing thread, but nor responses..

We have a run that's about 800'. 225A/3P feeder at 208V.

The feeder is from a generator to a distribution panel at 535' away and than we continue one with the 2nd run at 265'.

The most significant load is a 13.6HP elevator. The mfg. recommends sizing feeder at 97 Amps which I assume is full load running. The other loads are some ventilation fans, EM lighting and other EM loads (dampers, etc.).

I calculate the connected load for the combined first and second loads at 190A, the 2nd run drops 15A - so we continue on with 175A.

I am proposing 2 sets of 600AL which is huge in the eyes of the contractor. I want to limit the drop to 3ish% because the 2nd feeder than continues on to secondary distribution panels and than on to branch circuits.

The elevator is used as a means of egress (re: 4 or more stories above grade, IBC..)

I am taking PF at 95% which I think is fair. The reality is that this may never draw the 190A, but I don't know if I'm being too conservative. Life safety makes me err to the conservative side.

Anyone have any input?

 

[bob]

Using the information you provided, I get 3% using 2 - 600 AL as you did.

 

[skeshesh]

I don't think there can be much input other than opinion. I totally understand the contractor's position as far as the load probably never getting close to that, but the fact is when you design you have to go by calculated load. Unless you have an existing load you're adding too and you provide a 30 day reading, in which case you will still need to add the "calculated load" to it, you pretty much have to design the way you've done (correctly so).

 

[cdcengineer]

Thanks for the responses.

Any opinion on whether there is a real and definite need to use 0.95 PF in calculations? I always have because most utilities require it when calculating lateral sizes. I originally considered going 2 sets of 500AL but would sleep better using the 600's. It's difficult to know how bad voltage drop could be at the farthest branch circuit without calculating worst case. I don't want the EC to have to worry about running #10's to lighting branch circuits on the top floor (farthest point), and yet I also don't want to have to calculate branch circuit VDR. I usually try to limit VDR to 3% to the 1st point of distribution. Never had problems going that route.

I imagine that with a VDR over 5% at the farthest point could wreak havoc on CF lighting ballasts. We have no equipment loads too far from the distribution to which I've calculated. I don't want premature failure of equipment due to VDR, but the reality is that this equipment should seldom be running on this generator feeder (only in the event of normal power loss) - only in short duration's I would assume (dangerous word)..

I really appreciate the comments and insight.

 

[charlie b]

Any opinion on whether there is a real and definite need to use 0.95 PF in calculations?

My default value of pf (i.e., for load calculations, in the absence of specific nameplate data) is 85%.

 

[Cleveland Apprentice]

I'm a little confused when to use or not to use PF in load calculations and voltage drop. Anyone care to show an example? Thank you.

 

[weressl]

My default value of pf (i.e., for load calculations, in the absence of specific nameplate data) is 85%.

Get's my vote too.

 

[kingpb]

I'd use 500KCMIL AL for the first 535', then switch to #4/0 AWG AL from the panel to the next 265 feet, and sleep just fine.

 

[bob]

I'd use 500KCMIL AL for the first 535', then switch to #4/0 AWG AL from the panel to the next 265 feet, and sleep just fine.

You might sleep ok but you will not have met the requirements of 3%. Better Check those numbers again.

 

[broadgage]

Is this a circuit supplied ONLY from the generator, or is it supplied also with utility power, via a changeover switch ?

If ONLY generator power is to be used, then IMHO a larger voltage drop than 3% would be acceptable.
With utility power, the consumer has no control over the actual voltage supplied, except that it must be within the legal limit of plus or minus 5%. Since the utility power could be 5% low, additional loses in the wires should be limited so as to ensure sufficient voltage at the load. For example a 208 volt utility supply might be only 198 volts actual, with 3% lost in the wires that would give 192 volts at the load, under worst case conditions.

However with a generator supply, the customer DOES have control over the actual voltage supplied.
If the AVR on the generator was set to give 218 volts, that would give far more of a margin.
Even 10% voltage drop in the wires would still give 198 volts at the load.

 

[jdsmith]

I'm an owner's engineer who enforces the 3% rule that we have in our specs. I only require engineering firms to use the maximum designed "real" load for voltage drop calcs. For motor feeders that are sized using FLA from the NEC tables, the motor nameplate FLA will always be lower because we only use IEEE 841 premium efficiency motors. Large feeders to MCCs, panels, etc. are treated as "real" load = the bus rating of the panel or the main, whichever is less. We are a process plant with fully redundant loads at the 480V level and many redundant loads at 4160V, so the typical loading is less than half of the design load to allow for one of the two sources to be taken out of service without overloading the other source. When one source is out of service for maintenance or a failure the voltage drop on the remaining, heavily loaded source still needs to be under 3%, which leads to some very large feeders sometimes.

Contractors complaining about oversized cables is a fact of life. Aside from voltage drop, cable tray derating will also force cables up a size or two. I've yet to convince a contractor to double check our tray fill and derating calcs, they just grumble about the sizes.

The most extreme example I've heard of was due to voltage drop and fault current avaialbility - they ran approx 800 ft of 500 MCM copper for a single 20A, 120V UPS circuit. The goal was to have sufficient fault current available at the far end to blow a fuse and clear a fault in a timely fashion.

I also vote for 0.85 PF as a default PF value. At the distribution level my plant runs about 0.9 or 0.91 with the synchronous motors factored in, but keep in mind that a 0.95 or 0.90 PF device running at reduced load almost always has a lower PF. Since we're doin voltage drop calcs where reduced load = reduced amps = less drop, 0.85 is a good compromise.

 

[kingpb]

You might sleep ok but you will not have met the requirements of 3%. Better Check those numbers again.

There is no requirement limiting a design's voltage drop to 3%, that is a fallacy.

Besides, based on mathematical calculations using significant figures, I'll bet there is essentially no difference between 3% and 3.4% for the calculation.

Based on my experience, letting the VD rise a little to save a whole bunch of money on installation is generally an acceptable practice especially when your talking about 10ths of a %. However, each case must still be investigated on it's own merit.

 

[cdcengineer]

Thx for the responses. I usually try to limit VDR to the 1st point of distribution to 3%. Allowing another 2% drop in the branch. Code only requires that we limit to 5% overall. I think that's what the groups consensus is. I may start using 85% PF in the future.

Thx again.

CDC

 

[bob]

There is no requirement limiting a design's voltage drop to 3%, that is a fallacy.Besides, based on mathematical calculations using significant figures, I'll bet there is essentially no difference between 3% and 3.4% for the calculation.
Based on my experience, letting the VD rise a little to save a whole bunch of money on installation is generally an acceptable practice especially when your talking about 10ths of a %. However, each case must still be investigated on it's own merit.

The 3% was the specs required by the OP and his question was related to the wire size to limit the VD to 3%. The answer to his question is 2-600 kcm per phase.
He did not ask anything about saving money. VD using 2-500's and 2-4/0's results in a VD of 4.5%. Apparently that is to high for the OP's design.

 

[tesi1]

power factor ?

power factor ?

: most generators in which we installed had a power factor of around 80 %, unless yours had a higher power factor you vd. calc's at 95% would be off.

 

[suemarkp]

The 3% was the specs required by the OP and his question was related to the wire size to limit the VD to 3%.

I never saw that directly stated. It was just a question about a 3% voltage drop. Didn't know if he was assuming that was code, or if it was in his specs. If in the specs, you have no recourse. The Vd in the code is a FPN, so it can go as low as you want and still be compliant.

 

[bob]

I never saw that directly stated. It was just a question about a 3% voltage drop. Didn't know if he was assuming that was code, or if it was in his specs. If in the specs, you have no recourse. The Vd in the code is a FPN, so it can go as low as you want and still be compliant.

Look at the title of the thread.

 

[suemarkp]

It says "Voltage drop limited to 3%". Again, it doesn't says "Specs say voltage drop....". Was afraid he thought the code required a 3% drop limit.

 

[bob]

Thanks for the responses.

It's difficult to know how bad voltage drop could be at the farthest branch circuit without calculating worst case. I don't want the EC to have to worry about running #10's to lighting branch circuits on the top floor (farthest point), and yet I also don't want to have to calculate branch circuit VDR. I usually try to limit VDR to 3% to the 1st point of distribution. Never had problems going that route.I imagine that with a VDR over 5% at the farthest point could wreak havoc on CF lighting ballasts. We have no equipment loads too far from the distribution to which I've calculated. I don't want premature failure of equipment due to VDR, but the reality is that this equipment should seldom be running on this generator feeder (only in the event of normal power loss) - only in short duration's I would assume (dangerous word)..

I really appreciate the comments and insight.

The OP set the limit to 3%. It had nothing to do with the code

 

[kingpb]

Well, if you want to get technical, stating that you usually try to limit VD to 3% versus limiting it to 3% are two different things.

The first would mean that you try whenever possible, but not absolute, could mean that 3.4% is still close enough to 3% and therefore good enough especially for cost implications. The second, would be that it cannot be greater than 3.0% no matter what.

But that's my interpretation, and everybody has their own opinion. That's what keeps lawyers in business.

My experience tells me that after the system is up and running normally, the VD using the 500KCMIL AL will be at, or less than 3%. But that's just my opinion.