How to Calculate Circuit Breakers for RGB LED Strips with CV + CC Power Supply?

DIYLED

Active Member
As many people here know, connected load can't be above 80% of the circuit breaker rating in the United States. Sometimes this is straightforward, but with LED strips there are sometimes several ways to calculate this. I'm wondering which is correct for most city codes, and if there's a way to read/interpret building codes to clarify which method to use.

For example, suppose there are a bunch of lightup boxes, and each box has a Meanwell HLG-240H-24A power supply (240W max output) with 300 Watts of RGB LED strip connected, but the program only uses primary colors (100W) and secondary colors (200W).

Here's the Meanwell datasheet:
http://www.meanwellusa.com/webapp/product/search.aspx?prod=HLG-240H

which gives the following info:

- Page 2 OUTPUT: 240W. (Maximum output power is 240W. 24V and current-limited to 10A, voltage will drop if load doesn't reach 24V at 10A as seen on page 3.)

- Page 2 INPUT: AC CURRENT (Typ.): 4A / 115VAC

- Page 6 chart: PFC is around 0.99 in CC mode. (100% load is where CC mode meets CV mode.)

- Page 6 chart: Efficiency is around 90% at 100% load.

Now, I don't know how it could possibly draw 4A at 115V when the max DC output is only 240W. It would either generate a lot of heat, or the PFC would be terrible, both of which contradict the datasheet charts. But this is what is listed in the official data sheet.

Alternatively, the line current can be calculated by:

AC current = DC power / efficiency / power factor correction / AC voltage

Calculating using maximum power output:
240W/.90/.99/120V = 2.24A

Calculating with secondary colors:
200W/.90/.99/120V = 1.87A

Or the fictional scenario of all-white 100%:
300W/.90/.99/120V = 2.81A
Sounds silly, but I've seen somebody use the connected LED strips as the max load, even though the power supply output is current-limited.

Any thoughts on which method to use? With a large number of boxes, the calculation method can greatly affect the number of breakers required to meet code.

It looks like calculations usually use the nominal voltage of 120V, although I thought 115V is nominal and the range is 110 to 125V.
 
As many people here know, connected load can't be above 80% of the circuit breaker rating in the United States. Sometimes this is straightforward, but with LED strips there are sometimes several ways to calculate this. I'm wondering which is correct for most city codes, and if there's a way to read/interpret building codes to clarify which method to use.

For example, suppose there are a bunch of lightup boxes, and each box has a Meanwell HLG-240H-24A power supply (240W max output) with 300 Watts of RGB LED strip connected, but the program only uses primary colors (100W) and secondary colors (200W).

Here's the Meanwell datasheet:
http://www.meanwellusa.com/webapp/product/search.aspx?prod=HLG-240H

which gives the following info:

- Page 2 OUTPUT: 240W. (Maximum output power is 240W. 24V and current-limited to 10A, voltage will drop if load doesn't reach 24V at 10A as seen on page 3.)

- Page 2 INPUT: AC CURRENT (Typ.): 4A / 115VAC

- Page 6 chart: PFC is around 0.99 in CC mode. (100% load is where CC mode meets CV mode.)

- Page 6 chart: Efficiency is around 90% at 100% load.

Now, I don't know how it could possibly draw 4A at 115V when the max DC output is only 240W. It would either generate a lot of heat, or the PFC would be terrible, both of which contradict the datasheet charts. But this is what is listed in the official data sheet.

Alternatively, the line current can be calculated by:

AC current = DC power / efficiency / power factor correction / AC voltage

Calculating using maximum power output:
240W/.90/.99/120V = 2.24A

Calculating with secondary colors:
200W/.90/.99/120V = 1.87A

Or the fictional scenario of all-white 100%:
300W/.90/.99/120V = 2.81A
Sounds silly, but I've seen somebody use the connected LED strips as the max load, even though the power supply output is current-limited.

Any thoughts on which method to use? With a large number of boxes, the calculation method can greatly affect the number of breakers required to meet code.

It looks like calculations usually use the nominal voltage of 120V, although I thought 115V is nominal and the range is 110 to 125V.
@DIYLED Permit me to offer what MAY be your bottom line answer and quote one non-sensical illustrative example:
Go with whatever your Authority Having Jurisdiction (AHJ) decrees.
Here's an illustrative example of AHJ interpretations from the bad old days of dimmers powering exclusively incandescent loads:
I'm writing from Hamilton, Ontario, Canda, thus Canadian rules and interpretations but I'm confident you'll find similar AHJ's and interpretations on your side of the line with your NEC.
The theater was installing a brand new, fully loaded, Strand CD80 48 slot rack loaded with 48 x dual 2.4 Kw dimmer modules to augment their existing FOH transfer patch to better accommodate touring productions.
The AHJ had little to no experience with theaters and theatrical lighting loads thus he applied his heavy industrial motor load experience and came up with the following:
48 slots loaded with two x 2.4 Kw per slot = 96 dimmers.
96 dimmers across 3 phases = 32 dimmers per phase.
32 dimmers x 20 Amp breakers = 640 Amps per phase.
640 Amps per phase x the 125% rule to allow for motor starting in-rush currents = an 800 Amp 3 phase / 5 wire 120 / 208 Volt supply.
This was his decree and a 3 pole 800 Amp breaker was installed and fed to power this single rack.
Approximately one year later, the same theater chose to replace their 5 huge racks in their basement dimmer room. Each rack housed 20 x 6 Kw Century Strand dimmers for a total of 100 x 6 Kw dimmers each with 50 Amp breakers and fed by a motor operated 1200 Amp three pole breaker.
With the aid of a Sawz-All, the five huge racks came out in pieces since they'd been originally installed when the theater was built and were each larger than the steel frame of the fire door.
Three new Strand Century CD80 racks, pretty much fully loaded with dual 2.4 Kw modules like the FOH rack installed less than a year previously only one floor up but including 17 6.0 Kw modules to bring the entire installation up to dimmer per circuit including the 17 - 6.0 Kw circuits, were installed looking absolutely miniscule in the middle of the exposed unpainted area of the dimmer room floor.
A different AHJ this time. This fellow's decree was: Power the new racks with the existing 1200 Amp motorized 3 pole main breaker and he arrived at his decision as follows: We took out the old and put in the new but these were only dimmers, ran quite efficiently and produced very little heat. He reasoned that even though we'd bought a greater number of dimmers we HAD NOT purchased any new incandescent lights thus there'd be no additional current / power consumption thus re-use the existing 3 pole 1200 Amp 120 / 208 Volt feed and get on with it.
SAME BUILDING. ONE FLOOR LEVEL APART. LESS THAN A YEAR LATER. DIFFERENT AHJ. Go figure, I rest my case. @TimMc Care to comment?
Toodleoo!
Ron Hebbard.
 
Note the peak inrush current of 75 A, (at 230 VAC). The maximum number of units per 16 A breaker is given as four. What that shows is that, for short periods of time, they act like a 4 amp load at 115 VAC, even though the efficiency is much higher most of the time. If you want the system to be reliable, put no more than four on a 20 Amp circuit, and you'll sleep well at night knowing there is adequate headroom.
 
AHJ, meaning a person's opinion. That is funny. :) I might have to go digging through the NEC docs and ask a lawyer to get an answer but thought somebody here might know. I'm guessing that not too many electricians or electrical engineers know the legal answer in this situation. There's potential for the 1st AHJ to say "this is okay" then later on somebody else says "that's not okay." Best to get the "this is okay" in writing I think.

Going 4 per 20A breaker really drives up the number of breakers when you have a large number of power supplies. I used to worry about inrush currents, but I've never been able to trip a breaker that way with these units. For a temporary setup, I had 6 of the HLG-600H-12 (only running at about 50% power) on a single 20A breaker, and the breaker never blew even at 3 different installation sites. The backup plan was to turn on 3 at a time or buy really long extension cords and make figure 8 piles. Never had to do it.
http://www.meanwellusa.com/webapp/product/search.aspx?prod=HLG-600H
 
I generally just use a kill-a-watt or clamp meter and turn the device on to see what kind of power it's drawing. I've run into to many electronic devices that ended up being nowhere close to their rated power. Some well under, and some over. Usually someone in the marketing department gets in the middle of things and all of a sudden numbers become theoretical. The only way to really know how much power something is using it to measure it yourself.
 
AHJ, meaning a person's opinion. That is funny. :) I might have to go digging through the NEC docs and ask a lawyer to get an answer but thought somebody here might know. I'm guessing that not too many electricians or electrical engineers know the legal answer in this situation. There's potential for the 1st AHJ to say "this is okay" then later on somebody else says "that's not okay." Best to get the "this is okay" in writing I think.

Going 4 per 20A breaker really drives up the number of breakers when you have a large number of power supplies. I used to worry about inrush currents, but I've never been able to trip a breaker that way with these units. For a temporary setup, I had 6 of the HLG-600H-12 (only running at about 50% power) on a single 20A breaker, and the breaker never blew even at 3 different installation sites. The backup plan was to turn on 3 at a time or buy really long extension cords and make figure 8 piles. Never had to do it.
http://www.meanwellusa.com/webapp/product/search.aspx?prod=HLG-600H
@DIYLED Quoting you: "AHJ, meaning a person's opinion. That is funny." EXACTLY! AHJ = Someone's opinion and interpretation who is licensed by our province and earns his living purely by decreeing his interpretation / opinion and officially signing off on it. We had a case of exactly what you're citing in nearby Burlington, Ontario, Canada, literally across a highway from Hamilton, Ontario, Canada. We installed an 800 Amp, 3 Phase, 120 / 208 Volt service as the main service in a community center powering an 800 Amp splitter trough powering two 400 Amp fused disconnects into two metering cabinets with their metering transformers being monitored by one Ontario Hydro modem sharing a common telco line via two distinct phone numbers. All per code, ALL inspected, approved and SIGNED OFF by a bona-fide Ontario Hydro inspector only to have to rip out and change part of the installation AT OUR COST when the AJH, an experienced, career, inspector in his sixties had his approval overridden by someone in Toronto with less experience but at a higher salary level and with more interpretation clout.
As I've previously stated, in my experience, it ALWAYS comes down to your AHJ. (And her / his bosses and superiors)
@TimMc Care to comment on AHJ's on your side of Trump's wall?
Toodleoo!
Ron Hebbard
 
Note the peak inrush current of 75 A, (at 230 VAC). The maximum number of units per 16 A breaker is given as four. What that shows is that, for short periods of time, they act like a 4 amp load at 115 VAC, even though the efficiency is much higher most of the time. If you want the system to be reliable, put no more than four on a 20 Amp circuit, and you'll sleep well at night knowing there is adequate headroom.

Exactly right. The limiting factor in determining how many LED fixtures can be fed from a single branch circuit is the aggregate inrush current of their switch mode power supplies, NOT their steady-state current. Once you consider the maximum inrush current (which happens only at energization of the power supply) and adjust the fixture count for that, you will be far below the steady-state rating of the circuit, even after 80% derating for continuous loads.

ST
 
Exactly right. The limiting factor in determining how many LED fixtures can be fed from a single branch circuit is the aggregate inrush current of their switch mode power supplies, NOT their steady-state current. Once you consider the maximum inrush current (which happens only at energization of the power supply) and adjust the fixture count for that, you will be far below the steady-state rating of the circuit, even after 80% derating for continuous loads.

ST

That works fine if the client just says yes we'll give you a larger panel to use, but this can create a problem if the client doesn't want to do that.

Much could be possible with a small panel, just by turning on each power supply one at a time and running white at 33%, secondary colors at 50%, primary colors at 100%.

There are also breakers that are specifically designed for the inrush of switching power supplies. Maybe somebody here knows how those affect panel size v. a regular breaker?
 
Its hard but sometimes you have to tell the client the truth, not what they want to hear.

Yes you could power up the drivers one by one, but that will require relays and a panel probably. That approach will cost more than more circuit breakers.

Even I have understood that the connected LEDs or what they are set at is irrelevant to the inrush - its all about the driver - period.

I'm not an expert on inrush current limiters and breakers with different curves, but my instinct says if they exist, it will be much more expensive than a larger or additional breaker panel.
 
@DIYLED
You are sandwiched between the legal authority of the AHJ and the manufactures stated 2/4 per breaker. (Show the AHJ the data sheet and they will likely approve that without problems.) No matter what engineering argument you might come up with they are the official experts, anything else will be direct defiance. Those cases go very poorly before judges and insurance adjusters.

Just to learn, I looked up prices on some D-curve breakers. (for very high impedance loads) They seem entirely reasonable, but I would still want someone to assemble the whole panel for me. And I'd call Meanwell to ask how many more drivers they can handle. It's an entirely reasonable engineering solution. It just that I don't like taking on the life-safety responsibility for the system for next several decades.
 
Thanks. That's some good stuff to think about. I suppose if the common practice is to increase the panel size, then it's likely that's often the most economical and/or safest legal option.

The larger power supplies seem to have a less inrush in proportion to their steady-state loads. Just something I noticed that may make things easier.

Sounds like the datasheet recommendation for breakers at 230V is also a safe number for 115V, perhaps too safe, but looks way better in court than using the 80% method and trial-and-error for inrush.
 
That works fine if the client just says yes we'll give you a larger panel to use, but this can create a problem if the client doesn't want to do that.

Much could be possible with a small panel, just by turning on each power supply one at a time and running white at 33%, secondary colors at 50%, primary colors at 100%.

There are also breakers that are specifically designed for the inrush of switching power supplies. Maybe somebody here knows how those affect panel size v. a regular breaker?

I would say that it's generally a poor practice to size a branch circuit (or calculate the number of fixtures on a branch circuit) by assuming reduced current consumption based on control settings ("white at 33%, secondary colors at 50%, primary colors at 100%"). That's because there are too many easy ways to get to "all control channels at full" by mistake. In addition, the current will not be linear in relation to the control setting.

Re: special curve breakers. Hard to source, expensive, and then you'll still have to test them with your specific fixtures to get to the maximum count.

Re: load sequencing. Great idea if you can afford it. Probably the single best rationale for being able to switch individual branch circuits from the control system.

ST
 
I'm curious as to the duration of the inrush current of LED loading and whether it's similar to motors. Circuit breakers have for decades, been designed to handle start up current of motors, thus wonder if LED's have a similar characteristic.

As example, I know that the typical user of a Color Source Par is looking at the recommended loading when the fixture is past it's in-rush phase, to determine how many fixtures on a particular circuit. The ETC literature states (I'm only using ETC as their literature is readily available).

"Up to nine luminaires (15A max) may be linked via power thru connector (10 luminaires total per circuit) when used with R20 relay module or Unison® Echo Relay Panel. Consult breaker trip curves when used with other equipment. Requires power from a non-dimmable source • Inrush - 120V: 35A (First half-cycle) - 240V: 49A (First half-cycle)"

So what am I using as a power source ?, a 35 amp breaker ?, or a R20 relay ?. I KNOW I've used R20's on a number of Aura's without blowing the breakers.
 
I would say that it's generally a poor practice to size a branch circuit (or calculate the number of fixtures on a branch circuit) by assuming reduced current consumption based on control settings ("white at 33%, secondary colors at 50%, primary colors at 100%"). That's because there are too many easy ways to get to "all control channels at full" by mistake. In addition, the current will not be linear in relation to the control setting.
ST

It's not as bad as it sounds. :) With a CV + CC power supply, it has a maximum power output, so if you program white at 100%, it would be current-limited and probably show an orangish white because the voltage would drop. By "33%" and "50%", I meant the linear values, which can be obtained by looking them up in the dimming curve or approximated through guessing. For example, with the HLG-240H-24A power supply, hooking up 600 Watts of RGB LED's and programming all colors to run at 200W. In the event of a programming oops, the power supply would only output a max of 240W and input about 269 VA due to 90% efficiency and 0.99 PFC.

I don't know much about ETC products, but I hear the word "relay" and it sounds like they might be turning them on one-at-a-time to limit inrush current.
 
Last edited:
I don't know much about ETC products, but I hear the word "relay" and it sounds like they might be turning them on one-at-a-time to limit inrush current.

Commonly used are relays built into modules that replace dimmers. The ETC R20 design is similar to its D20 module in that there are 2 dimmers or relays on a module, thus when you change from dimmer to relay you do so in paired circuits. They respond to DMX, thus are useful to power devices not wanting to be dimmed when you want control of closed/open state remotely. The control unit sending DMX can be programmed to have banks of relays sequentially close, but the modules nor control electronics do that that typically. It’s thus possible to have a bunch of relays all close at the same time.

Another power device as sold by ETC is the Thru-Power module, which is 2 dimmers and 2 relays on a module. The state - dimmed or relay, can be controlled by a console as well, useful when you only desire one circuit to be relay.

A different solution now being used is motorized circuit breakers. The control typically allows sequential close, thus you can avoid huge inrush currents. These are also available with DMX control, can be had in different current ratings as well as 1-2 and 3 pole designs.
 
If your concern is the number of circuit breakers buy one big circuit breaker (40A, 50A, whatever you need...) use appropriately sized cable for distribution, and individually fuse each power supply at 5A. There are a bunch of different ways to solve any given design problem it's just a matter of what set of trade-offs are acceptable for the given application.
 
@DIYLED
As I've previously stated, in my experience, it ALWAYS comes down to your AHJ. (And her / his bosses and superiors)
@TimMc Care to comment on AHJ's on your side of Trump's wall?
Toodleoo!
Ron Hebbard

I've not had the kinds of experiences you've had, Ron, mostly because I'm not involved in planning or design of facilities of that scale. Hell, nobody asks me about audio and that's been my trade and craft for nearly 40 years....

My experience with conflicting AHJs came from flipping houses in my off time (back before there were TV shows about that stuff and before Holmes on Homes). I had a house on "the other side of town" where the electrical inspector didn't know me or my partners and was pretty much a jerk, not allowing things that were plainly acceptable under NEC and I know damn good and well that our City does not adopt more stringent codes than the NEC. I called up the inspector who worked with us on properties on the west side and he told me the east side inspector was a Major Dick but was scheduled for vacation soon. If we called in for inspections while M.D. was gone someone more rational would probably sign off.

I can't speak to how other States work but here in Kansas the state exercises little direct control over Codes, interpretation or enforcement leaving those to municipal governments (cities, counties). When our municipal PAC replaced all the dimmers in the building (4 of 5 theaters anyway) nobody outside the City (division of arts and the Code dept) was involved in determining the safety or adequacy of the planned changes (some of which were extensive, including removing the existing load transfer panels, relocation of dimmers & feeder and new on stage distribution of non-dim power for LED use).

As for other aspects of Trumplandia, I leave those to the news media....
 
During the previous recession, a county inspector told me that they were told to write enough citations to justify their salary. The official wording was probably more circumspect, but the message got through.
 
A couple thoughts...

If you are building something for a permanent install, I would adhere to the manufacturer's labeling for the power requirements. This especially if you are say, building a cabinet that needs to be UL approved. You may never draw the current they say, but, even if the "actual" current draw is approved by the AHJ, if something goes sideways it would be a real nightmare.

If you are building a product for retail, I would also say that it should adhere to the manufacturer's labeling of of the electrical requirements of any components you use.

If you are just building a bunch of LED drivers for temporary use I would say you can go by either what you know the max draw is (assuming you measured it) or by the most conservative of your calculated current draws. Just make sure that you have documentation to back up your reasoning for the time the AHJ actually wants to do an inspection.

At the end of the day, the cost of an extra circuit or two is far less than the potential cost of an insurance claim.
 

Users who are viewing this thread

Back