Fire Alarm Voltage Drop
- Thread starter Eng
- Start date
- Status
dhalleron
Senior Member
- Location
- Louisville, KY
- Occupation
- Master Electrician/Senior Fire Alarm Technician
We used to use various percentages for max voltage drop. Most of the time I used 10% as the maximum drop. The city of Los Angeles used to make us use 3% because it was in a fine print note in the NEC.
UL changed the way they rated fire alarm devices. A 24 volt device is required to work between 16 and 33 volts. As long as you have 16 volts at the end of a circuit the device should work. Still you should always leave room for error.
Fire alarm code 2010 edition:
10.14.1 Voltage, Temperature, and Humidity Variation. Equipment
shall be designed so that it is capable of performing its
intended functions under the following conditions:
(1)*At 85 percent and at 110 percent of the nameplate primary
(main) and secondary (standby) input voltage(s)
So the device can work down to 16 volts but code would require you design so not to go below 20.4 volts.
I?d say that is a maximum 15% drop per code, but still leave more room for spare.
UL changed the way they rated fire alarm devices. A 24 volt device is required to work between 16 and 33 volts. As long as you have 16 volts at the end of a circuit the device should work. Still you should always leave room for error.
Fire alarm code 2010 edition:
10.14.1 Voltage, Temperature, and Humidity Variation. Equipment
shall be designed so that it is capable of performing its
intended functions under the following conditions:
(1)*At 85 percent and at 110 percent of the nameplate primary
(main) and secondary (standby) input voltage(s)
So the device can work down to 16 volts but code would require you design so not to go below 20.4 volts.
I?d say that is a maximum 15% drop per code, but still leave more room for spare.
MichaelGP3
Senior Member
- Location
- San Francisco bay area
- Occupation
- Fire Alarm Technician
This may be familiar to a lot of you here; when projects I was involved in were required to have no more than 10% voltage drop, we would measure the voltage at the terminals of the power supply and see a typical 26.7 volts. We would use this figure in our calculations to give us a little more headroom (2.67 volts vs. 2.40 volts).
gadfly56
Senior Member
- Location
- New Jersey
- Occupation
- Professional Engineer, Fire & Life Safety
dhalleron
Senior Member
- Location
- Louisville, KY
- Occupation
- Master Electrician/Senior Fire Alarm Technician
Funny you should mention that. I had a guy call me yesterday about a Firelite dialer programmer that wasn't working right. It was being powered from the panel and Firelite told him it would not work on anything higher than 24 vdc. said he needed another power supply. I think any power supply is going to measure more than 24 volts. The tech was measuring 25.6 volts. I can't believe it wouldn't work on that voltage. I think something else was going on.
DM2-Inc
Senior Member
- Location
- Houston, Texas
As previously referenced, the fire alarm system is required to perform at 85% of the name plate rated voltage. If the transformer used to power the fire alarm devices is a 24 VDC transformer, the starting voltage for the voltage drop calc's is 24VDC x 85%, or 20.4 VDC. This requirement has been in the standards for some time now. It is better explained in the handbook.
A good reference is http://www.rpsa-fire.com/documents_open/SampleFireAlarmSystemCalculations.pdf
A good reference is http://www.rpsa-fire.com/documents_open/SampleFireAlarmSystemCalculations.pdf
had to repost,, time limits and all prevented me from doing so ,,ignore above, see below
Yes, and you wont find your answer until you find a """"very good"""" DC power guy (ie- me
the formula is
11.1 x LL x A / VD
You must look to the manuf spec or calculate the OCP size in amps.
once that is done, you inject that amperage or 80% of the amperage into the formula to determine the circular millimeters of copper required to power the device.
Lets say your brand x fire panel has a 400w, 24v name plate rating (some of these can operate at -48vdc as well as a ups backed 120vAC so double check your options)
inject ohms law and you arrive at 16.6A, 16.6A / .8 = approx a 20A dc fuse size.
next find out how far away in cable feet the power run is.
lets say its 20' one way or 40' loop (you have to account for both the + and - legs because its dc)
Now we can calculate the wire size required for a 40'Loop 20A dc circuit with a constant of 11.1 for copper in the DC power industry , and account for no less than a .875 volt drop (you could allow 1v if you you want)
so, here we go,,
11.1 x 40 x 20 / .875 = 10,148 circular millimeters of copper required to properly engineer this circuit per Telcordia and Bellcore standards (both of which are well known in the DC Power industry).
the same calculation applies if you de-rate to 80% OCP but you come up with 8118cmil of copper required which is still pulling you to #10.
you could get the wire size down in my example "if" this device is the only DC power item and nothing downstream of the fire panel requires the 24v dc source. If so, you could use 80% ocp and a 1v or even a 2v allowable loss but you have to explain your engineering if questioned.
In my honest opinion, if your 24v is "not" coming from a 24v battery back up DC power plant be it small or large, there is no reason to engineer as aggressively as i have here, i figure worse case all the time.
SO, i would power your new panel in this situation with a #10 1pair cobra copflex non halogen CT rated wire and using the 75degC column.
I have such a hard time with my AC guys getting them to understand why on the last project the 100A -48V circuit got a #2 +/- wire but yet on this other project i had them cable with 2/0.
again, the formula is
11.1 x loop length x amps / allowable voltage drop
you wont find any of this in the NEC, they barely have just begun to define anything DC power
Yes, and you wont find your answer until you find a """"very good"""" DC power guy (ie- me
the formula is
11.1 x LL x A / VD
You must look to the manuf spec or calculate the OCP size in amps.
once that is done, you inject that amperage or 80% of the amperage into the formula to determine the circular millimeters of copper required to power the device.
Lets say your brand x fire panel has a 400w, 24v name plate rating (some of these can operate at -48vdc as well as a ups backed 120vAC so double check your options)
inject ohms law and you arrive at 16.6A, 16.6A / .8 = approx a 20A dc fuse size.
next find out how far away in cable feet the power run is.
lets say its 20' one way or 40' loop (you have to account for both the + and - legs because its dc)
Now we can calculate the wire size required for a 40'Loop 20A dc circuit with a constant of 11.1 for copper in the DC power industry , and account for no less than a .875 volt drop (you could allow 1v if you you want)
so, here we go,,
11.1 x 40 x 20 / .875 = 10,148 circular millimeters of copper required to properly engineer this circuit per Telcordia and Bellcore standards (both of which are well known in the DC Power industry).
the same calculation applies if you de-rate to 80% OCP but you come up with 8118cmil of copper required which is still pulling you to #10.
you could get the wire size down in my example "if" this device is the only DC power item and nothing downstream of the fire panel requires the 24v dc source. If so, you could use 80% ocp and a 1v or even a 2v allowable loss but you have to explain your engineering if questioned.
In my honest opinion, if your 24v is "not" coming from a 24v battery back up DC power plant be it small or large, there is no reason to engineer as aggressively as i have here, i figure worse case all the time.
SO, i would power your new panel in this situation with a #10 1pair cobra copflex non halogen CT rated wire and using the 75degC column.
I have such a hard time with my AC guys getting them to understand why on the last project the 100A -48V circuit got a #2 +/- wire but yet on this other project i had them cable with 2/0.
again, the formula is
11.1 x loop length x amps / allowable voltage drop
you wont find any of this in the NEC, they barely have just begun to define anything DC power
- Location
- Massachusetts
had to repost,, time limits and all prevented me from doing so ,,ignore above, see below
Yes, and you wont find your answer until you find a """"very good"""" DC power guy (ie- me
the formula is
11.1 x LL x A / VD
You must look to the manuf spec or calculate the OCP size in amps.
once that is done, you inject that amperage or 80% of the amperage into the formula to determine the circular millimeters of copper required to power the device.
Lets say your brand x fire panel has a 400w, 24v name plate rating (some of these can operate at -48vdc as well as a ups backed 120vAC so double check your options)
inject ohms law and you arrive at 16.6A, 16.6A / .8 = approx a 20A dc fuse size.
next find out how far away in cable feet the power run is.
lets say its 20' one way or 40' loop (you have to account for both the + and - legs because its dc)
Now we can calculate the wire size required for a 40'Loop 20A dc circuit with a constant of 11.1 for copper in the DC power industry , and account for no less than a .875 volt drop (you could allow 1v if you you want)
so, here we go,,
11.1 x 40 x 20 / .875 = 10,148 circular millimeters of copper required to properly engineer this circuit per Telcordia and Bellcore standards (both of which are well known in the DC Power industry).
the same calculation applies if you de-rate to 80% OCP but you come up with 8118cmil of copper required which is still pulling you to #10.
you could get the wire size down in my example "if" this device is the only DC power item and nothing downstream of the fire panel requires the 24v dc source. If so, you could use 80% ocp and a 1v or even a 2v allowable loss but you have to explain your engineering if questioned.
In my honest opinion, if your 24v is "not" coming from a 24v battery back up DC power plant be it small or large, there is no reason to engineer as aggressively as i have here, i figure worse case all the time.
SO, i would power your new panel in this situation with a #10 1pair cobra copflex non halogen CT rated wire and using the 75degC column.
I have such a hard time with my AC guys getting them to understand why on the last project the 100A -48V circuit got a #2 +/- wire but yet on this other project i had them cable with 2/0.
again, the formula is
11.1 x loop length x amps / allowable voltage drop
Are you sure you are talking about fire alarm systems as the OP is?
We don't install fuses in fire alarm circuits nor would we have a circuit with 16 amps of current on it.
Not sure what you mean, the NEC applies to DC as well as AC.
You won't find voltage drop calculations in the NEC for either as it is a safety code not a design guide or instruction manual.
Last edited:
the op supplied a question and i had to inject some knowns in order to explain the standard DC power math.
I do not know what the 20v device is,, it could be a wall wart powering the fire panel, it could be a downstream supervisory circuit the fire panel is controlling.
there were too many unknowns so i just started with some made up assumptions to support the OP so it could be applied in the same way where he is delivering the dc power from and to, thats all. We dont know the loop footage or the ampacity demand on what is being powered. We see a hint that the data sheet implies the system will operate down to 16v. when voltage goes down, current goes up. I wonder if anyone has commissioned a full blow fire alarm panel and engaged all possible alarm conditions in order to load down the unit and then take clamp on readings of amp draw,, i never have, im no fire panel expert but if i had to guess, no more than 5 or 10A at the most. Its the distance that kills you in DC so the wire size has to increase to compensate. Just because you wired in #10 wire on a 16A circuit has nothing to do with amperage, it has everything to do with voltage drop only.
if he is powering up a Fire Alarm Panel off a 24v source, by all means allow for no more than a 1v drop especially if the 24v source is a battery back up 24v source. Could he get away with more of an allowable voltage drop, sure,, just be able to explain the engineering like we all have to for liability sake.
As for fusing,, i would assume we all make our best attempt to fuse each DC circuit no matter how small it is. We fuse right down to the 1/8A and even lower. In DC you want to fuse any power lead possible but most people think little loads will be fine because the wire becomes the fuse link,,, then the fire suppression dumps ect ect.
im probably just a little too cautious but its good practice in the dc industry.
if i knew more exactly of what the op's application was, then we all might a better answer.
I do not know what the 20v device is,, it could be a wall wart powering the fire panel, it could be a downstream supervisory circuit the fire panel is controlling.
there were too many unknowns so i just started with some made up assumptions to support the OP so it could be applied in the same way where he is delivering the dc power from and to, thats all. We dont know the loop footage or the ampacity demand on what is being powered. We see a hint that the data sheet implies the system will operate down to 16v. when voltage goes down, current goes up. I wonder if anyone has commissioned a full blow fire alarm panel and engaged all possible alarm conditions in order to load down the unit and then take clamp on readings of amp draw,, i never have, im no fire panel expert but if i had to guess, no more than 5 or 10A at the most. Its the distance that kills you in DC so the wire size has to increase to compensate. Just because you wired in #10 wire on a 16A circuit has nothing to do with amperage, it has everything to do with voltage drop only.
if he is powering up a Fire Alarm Panel off a 24v source, by all means allow for no more than a 1v drop especially if the 24v source is a battery back up 24v source. Could he get away with more of an allowable voltage drop, sure,, just be able to explain the engineering like we all have to for liability sake.
As for fusing,, i would assume we all make our best attempt to fuse each DC circuit no matter how small it is. We fuse right down to the 1/8A and even lower. In DC you want to fuse any power lead possible but most people think little loads will be fine because the wire becomes the fuse link,,, then the fire suppression dumps ect ect.
im probably just a little too cautious but its good practice in the dc industry.
if i knew more exactly of what the op's application was, then we all might a better answer.
Last edited:
- Location
- Massachusetts
I think you are trying to force fit what you aprently do at telecom places into a standard fire alarm system. Other than both being fire protection equipment they are as different as night and day.
The panel will be supplied with 120 VAC, the circuits leaving the panel are 24 VDC or less current limited by the design of the fire panel to 2 amps or less. Not only are fuses unneeded they would be a violation.
As far as keeping VD to less than 1 volt that is unneeded, as pointed out the devices are designed with a large voltage tolerance.
You know nothing of fire alarms, but you are certainly well versed in DC. Fire alarm DC circuits are designed, rated and used in a prescriptive manner within the manufacturers guidelines, in concurrence with UL 864, NPFA, 1, 70 and 72 (and others to a lessor extent).
We don't have to explain our engineering, because its written into the code, as long as you follow it, you are covered. The code only applies to most installations. Some require engineering supervision, but its few and far between that its mandatory.
Yes you are too cautious, in the FA industry 24 volts is nominal. Under battery load panels are rated to function at an input voltage of 20.4VDC on batteries, this is assumed as the worst case scenario for calculations
Fuses are not used because every circuit is power limited, pretty much all circuits are PTC protected. Some panels employ fuses on line voltage inputs, but are not field replaceable. The exception to fuses in fire alarms are the batteries, they are not power limited and do often have fuses, but not always. Generally, remotely located batteries are fuse, but not always. Some outputs invert when overload, they will then measure the load and when the resistance increases sufficiently, it will reset. (Silent Knight 5495 Aux power)
Voltage drop threshold is 10% for most manufacturers on the notification circuits.
Batteries are required (in NPFA 72, 2010 ed.) to be derated by 20%.
You 'best practices' specs seem foreign to us, its not that they are bad, just fanatical for us. They are practices that I am sure are commonplace in systems you are dealing with.
Last edited:
MichaelGP3
Senior Member
- Location
- San Francisco bay area
- Occupation
- Fire Alarm Technician
Since this was mentioned earlier, one cannot power a fire alarm panel from a wall wart.
Well......
Bosch fire rated GV3 and GV4 series technically are, with the wall wart in a metal enclosure (D8004) securing and dedicating the receptacle. Personally, I wouldnt use them for a fire panel beyond a couple of devices. I have worked with them on large 100 point systems, but not optimal in my opinion.
Admittedly, just about all other panels are not that way.
- Status