DC voltage drop?
- Thread starter anbm
- Start date
- Status
The NEC is not a standard period. It is a set of minimum safety requirements. Further there is NO requirement in the NEC for voltage drop for either AC or DC.anbm said:
What you are asking is a design specification and practice. The NEC will NOT give you guidance to that respect. For that you will need to turn to IEEE or an industry standard.
I work in the Telecom industry and we have very strict DC voltage drop requirements for 48, 24, and 12 VDC systems.
Last edited:
bphgravity
Senior Member
- Location
- Florida
Voltage drop is more of a design consideration than a safety issue. It's a loss in energy.
It's usually up to designer to determine how much voltage drop is acceptable and/or the manufacturer of the equipment being served will specify min and max connected voltages.
It's usually up to designer to determine how much voltage drop is acceptable and/or the manufacturer of the equipment being served will specify min and max connected voltages.
physis
Senior Member
You need to look at voltage drop in relation to how it affects your circuit.
Mathmatically and conceptually voltage drop is the product of current and resistance:
IR=V
or Current times Resistance equals Voltage. The voltage in this case is being called voltage drop, but there is no difference. The voltage across 130' of feeder, or 260' of conductor is given by IR or Current times Resistance.
Being as how I,R and V are always related, the current will be reduced to V/R, or Voltage divided by Resistance. Or:
I=V/R
The resistance in this case wont change significantly unless you get the conductors particularly hot.
What you need to know is what the allowable margin of error you can have is for current and voltage, (most likely voltage). This is usually called tolerance, except, for some reason, probably not in electrical.
I'm out of practice with the code book but I think the DC resistance per foot or thousand feet of whatever size conductor is in chapter 9. If not, someone will probably correct me in mere seconds. Apply whatever the resistance is of whatever wire size you're thinking of using to your feeder conductors and use the two forms of Ohm's law up there to find how your circuit will be effected.
But none of this will do you any good unless you know how low you can allow the voltage to get.
Mathmatically and conceptually voltage drop is the product of current and resistance:
IR=V
or Current times Resistance equals Voltage. The voltage in this case is being called voltage drop, but there is no difference. The voltage across 130' of feeder, or 260' of conductor is given by IR or Current times Resistance.
Being as how I,R and V are always related, the current will be reduced to V/R, or Voltage divided by Resistance. Or:
I=V/R
The resistance in this case wont change significantly unless you get the conductors particularly hot.
What you need to know is what the allowable margin of error you can have is for current and voltage, (most likely voltage). This is usually called tolerance, except, for some reason, probably not in electrical.
I'm out of practice with the code book but I think the DC resistance per foot or thousand feet of whatever size conductor is in chapter 9. If not, someone will probably correct me in mere seconds. Apply whatever the resistance is of whatever wire size you're thinking of using to your feeder conductors and use the two forms of Ohm's law up there to find how your circuit will be effected.
But none of this will do you any good unless you know how low you can allow the voltage to get.
Again there is no requirement, it is a design issue, All the NEC stipulates the minimum wire size based on the over current protection device size.anbm said:
Well the NEC table 310.16 requires a minimum 12 AWG to be perfectly safe. However your voltage drop using 12 AWG with 20-amps at a 1-way distance of 130 volts is roughly 8.8 volts. Not acceptable IMO on a 12 volt circuit.anbm said:
Now here is what I will tell you, in the Telecom sector the maximum allowable drop on a 12 VDC circuit is 1/2 volt total from battery terminal to the equipment utilizing the power. But here is the catch, a battery plant has 3 components and all three have to total 1/2 volt or less.
1. Battery terminal to discharge bus in power distribution frame (PDF) = .125 volts.
2. PDF to battery distribution breaker bay (BDBB) = .125
3. BDBB to equipment loads = .25
So armed with that info I assume you mean from the BDBB to the equipment using 20-amps @ 130 feet one-way distance you would need to use a 250 MCM cable to achieve a .25 volts or less voltage drop with a 20-amp load.
Off on a complete tangent.....dereckbc said:
Some years ago (some being close to 20) we supplied a few 1200A, 50V chargers to what was then Bermuda Telecoms, now C&W.
From what I recall, the ripple requirements were very stringent - psophometrically weighted with around 2mV in the audible range.
In your experience, is this still a typical Telecoms requirement?
Certainly and that is exactly what is done by using smaller capacity plants called Distributive Power. Catch is two fold it requires VRLA batteries which needs replaced every 5 to 7 years as opposed to 20 to 50 years with a flooded battery, and much higher initial upfront cost.Besoeker said:
Yes they are still stingent. For example Marconi Vortex Phosphometric Noise is less than 1 mV.Besoeker said:
I was replying to anbm's post and I don't know what his application is so I wouldn't know whether his/her application requires batteries.dereckbc said:
On the VLRA topic, we generally use Yuasa ENL series which have a stated life of 15 years. Agreed they're not as good as Plante or ""wet" cells.....but a lot cheaper and maintenance is minimal.
That said, we always use wet cells for HV tripping supplies.
Yes th eVortex line is Switch Mode. These are farly small rectifier units installed in a equipment shelf with self contained distribution. You see a lot of htese in vey small sites like Cell Towers.Besoeker said:
I use a lot of Marconi rectifiers and they make a 200-amp modular unit I and use them in a lor of 4000, 6000, and 8000 amp DC plants at 48 volts.
I have used Yuausa, Excide, C&D plus many others. After years of load test we found after 5 to 7 years their rated capacity drops below 75%. So instead of spending the money on load test, we just replace them every 5 to 7 years.Besoeker said:
On th eother hand we have a lot of Round Cells (pure lead plate flooded) now made by C&D that have been in service for 30+ years that still have 90% capacity.
VRLA are marketed as lower maintenance, but if floated properly and considering how long they last flooded cells are actually lower maintenance IMHO.
That kinda illustrates my point - you'd need 40 units in parrallel to get 8kA.dereckbc said:
We tend to take a different approach. We recently completed a 20kA, 50V system (not for telecoms). The configuration is two 10kA phase-displaced units comprising a step down transformer and controlled (SCR) hexaphase rectifier.
They are, of course, fairly big SCRs but there's just 12 of them for the 20kA system.
There's more than one way of skinning a cat.......
Well to be fully honest for the large DC plants say 6000 amps and higher we use 3-phase transformer SCR rectifiers rated at 800-amps per cabinet made by Marconni. They take up more space, weigh a ton, but practically indestructable and last forever. Switch mode rectifiers have high failure rates so this is why they are small in comparison and if you loose one, the others in parallel are more than capable of picking up the slack.
- Status