1000 watt air cooled transformer

[Greg1707]

Please excuse me but I do not have much experience with low voltage. I stopped by a mansion today that had LED strip lights installed in the cove molding in a ballroom. The lights were supplied by two 1000 watt air cooled 24 volt transformers located in the basement. Each transformer was connected via 12-2 NM to an 8 amp controller/dimmer. None of the lights worked and one of the dimmer/controllers had melted.

Does this sound like a typical install? Someone suggested this may have been an older installation that used Halogen lights? I don't understand using 12-2 NM on the load side of a 1000 watt transformer.

It has been suggested to bypass the controllers and go directly from the transformer to the LED lights.
Any thoughts?

 

[ptonsparky]

We need more detail.but a 1000 watt xfmr at 24 volts puts out 41 amps which would be more than twice what the 12 is rated .

 

[GoldDigger]

We need more detail.but a 1000 watt xfmr at 24 volts puts out 41 amps which would be more than twice what the 12 is rated .

If each transformer is driving one 8A dimmer/controller at 24V, then I do not think the main problem is in the wire size. I would definitely see how the connected lamp load compares to the 8A rating of the controller though.

 

[ptonsparky]

If each transformer is driving one 8A dimmer/controller at 24V, then I do not think the problem is in the wire size. I would definitely see how the connected lamp load compares to the 8A rating of the controller though.

You’re right, I’m sure the secondary has been fused properly and no loads have been exceeded.

 

[JFletcher]

A standard dimmer switch is only good for 600 watts. I've seen numerous ones pushed beyond their design capacity and they all have one thing in common: they were melted.

LED lighting I think you'll find that a standard dimmer is only good for 150 watts.

Yes I have seen old lighting that ran halogen bulbs at 24 volts. I consider them Legacy or obsolete systems.

Eta : as mentioned above 1000 watt Transformer can put out 41 amps at 24 volts which is significantly more than either the cable or dimmers are rated for. If they have LED lights chances are they only need 300 watt Transformers.

 

[gar]

180307-1011 EST

The following information is inadequate, and thus, a useful answer is very difficult to provide.

Please excuse me but I do not have much experience with low voltage. I stopped by a mansion today that had LED strip lights installed in the cove molding in a ballroom. The lights were supplied by two 1000 watt air cooled 24 volt transformers located in the basement. Each transformer was connected via 12-2 NM to an 8 amp controller/dimmer. None of the lights worked and one of the dimmer/controllers had melted.

What is an LED strip light? I will assume it is what I might call a DC load or more particularly a load that should normally driven by a DC or modulated DC source. I believe what is call an LED strip light will consist of two power distribution wires run the length of the strip. One labeled + and the other -. Distributed along these bus bars are many parallel light items each consisting of a series string of several LED chips and a series resistor.

An LED chip is a semiconductor component that is a diode that in the forward direction that emits light, and in the reverse direction is a high resistance until a breakdown voltage.

In the forward direction there is a maximum allowed current, the series resistor helps keep this current within an acceptable maximum with varying input voltage.

The resistor is of virtually no use relative to protection in the reverse direction. The voltage source has to be designed to never produce too much reverse voltage.


What is a transformer? I have no idea here.

What is a dimmer/controller? I have no idea here.

Where is the dimmer/controller located in the circuit?

Can a DC light strip be supplied by a sine wave AC source? Possibly. It will depend upon the peak inverse voltage rating of the DC LED light strip. Would you want to? Doubtful.

Two general ways to dim a DC light strip. An adjustable, but otherwise constant DC voltage source, or a constant DC voltage source with a pulse width modulated output (might be preferred method).

A more complete description of this circuit is needed.

.

 

[DrSparks]

I highly doubt the lights are LED up in the cove. You would need an insane amount of LEDs probably miles of strip to draw that much current.

Sent from my A574BL using Tapatalk

 

[gar]

180307-1352 EST

Some data from an actual DC LED light strip. This is a normal brightmess unit 2.5 meters (100", 8' 4") long. I have just one Armcost medium brightness strip. At 12.08 V DC I read 1.45 A, the far end of the strip is 11.36 V.

Armcost claims about 19 W at 12 V. This calculates to 1.58 A. The number of LEDs per meter is 60, or about 12 LEDs per ft. Looks like 3 LEDs per resistor. LEDs are about 0.65" apart which corresponds with 60 per meter.

Their high brightness strip is 120 per meter, and that would double the current.

If the area being covered is large, then required current can add up quite easily.

In my opinion the best way to dim DC LEDs is with pulse width modulation. This provides 100% range control, linear brightness adjustment, and no quirkiness in adjustment.

As with most products today there is inadequate technical information.

On-line I could not find good information aboit the Armacost DC dimmer. On the package is listed 12 and 24 V DC loads. No maximum input voltage. 8 A is mentioned with the 12 V comment, might be the same at 24 V since this is an on-off modulation at several thousand Hz. Would I run it at 8 A? Probably not.



Greg1707:

You probably need to find out more about your circuit before you can make progress in solving the problem.

If the dimmers are on the DC side of the circuit, then find out what your total load current is per dimmer at full brightness. This will be the worst power dissipation in the dimmer for a pulse width modulation dimmer because the full current is passing thru the switching device (likely an FET) 100% of the time.

.

 

[Greg1707]

problem solved

problem solved

OP We removed the controllers and and wired the lights directly to the transformers. I spoke with the manufacturer before doing this. I believe the transformers were for an older lighting system.
In the cove there were 4-16' LED strips. There were two on each transformer.

 

[gar]

189307-1711 EST

Greg1707:

What is a transformer? If you take a Fluke 27 or 87 or similar instrument what voltage do you read at the transformer output using the AC range, and then using the DC range?

.

 

[hbiss]

The OP, having described it as an "air cooled transformer" makes me believe that they are really DC power supplies. Betcha it has a heat sink with fins that suggested the air cooled description.

This really points to a problem with some people in this trade. To some, a transformer is anything that takes a higher voltage and reduces it to a lower one and that leads to all sorts of confusion.

A transformer only works with AC and steps up or steps down the voltage. It is not (in our work) used with DC nor can it, by itself, produce a DC voltage. So if you come across a "box" that takes an AC voltage and produces a DC voltage it is a power supply. Don't call it a transformer. And if you are not sure check it with your meter!

-Hal

 

[gar]

180307-1736 EST

If the transformer output is AC, then does anyone notice flicker from the lights? If AC excitation goes to the LED strip, then the LEDs are on for slightly less than 1/2 cycle at a pulse rate of 60 times per second.

Another question is what is the PIV (peak inverse voltage) rating of the LED strip? How does this compare with the AC peak voltage?

.

 

[gar]

180307-2413 EST

hbiss:

Your post was very good. Greg1707 never provided very useful information, and he probably won't provide any more information.

If we extrapolate from the measurements on my LED strip, 20 W for 8 ft, and assume double brightness (120 LEDs vs the 60 in my strip), then for 32 ft of lights at 24 V current would be about 160 W/24 V = 6.7 A. One might expect moderate life from a DC pulse width modulated dimmer rated at 8 A. But this depends upon real life conditions (heat transfer) and how valid is the manufacturer's specification.

We don't even know if the dimmer was a DC dimmer, or if it was in a DC circuit.

Far too much information was not provided.

.

 

[Greg1707]

Yes, power supply or driver not transformer

Yes, power supply or driver not transformer

180307-2413 EST



hbiss:

Your post was very good. Greg1707 never provided very useful information, and he probably won't provide any more information.

If we extrapolate from the measurements on my LED strip, 20 W for 8 ft, and assume double brightness (120 LEDs vs the 60 in my strip), then for 32 ft of lights at 24 V current would be about 160 W/24 V = 6.7 A. One might expect moderate life from a DC pulse width modulated dimmer rated at 8 A. But this depends upon real life conditions (heat transfer) and how valid is the manufacturer's specification.

We don't even know if the dimmer was a DC dimmer, or if it was in a DC circuit.

Far too much information was not provided.

.

I am not sure why you would think I would disappear and not provide further information?

I showed by ignorance about this topic by referring to the power supply or driver as a transformer. I have no experience in this area and just lumped everything together that changes or reduces voltage or current. I am glad to have this pointed out.
So, yes, the LED lights were powered by a 24 volt DC power supply. All is working well. Thanks for the support.

 

[gar]

180308-1344 EST

Greg1707:

Why? Because many posters want help and then never come back with feedback.

That we now clearly know that the "transformers" are DC power supplies that helps greatly.

We don't know where the failing dimmers were.

If I have a DC power supply consisting of a transformer, rectifier, and filter capacitor, with a very large capacitor so there is little output ripple, then I can change the output DC voltage by varying the magnitude of the input sine wave voltage. Easy smooth control of DC voltage, it is proportional to input voltage. Basically the sine wave peaks determine the DC voltage because of the large output capacitor. If no capacitor, then the output is probably a half wave or a full wave rectified. If the load is a resistance then average DC voltage is the average of a sine curve. Full wave this 0.636 of the sine peak, and half wave is 0.318.

If the input to said power supply is from a Triac type phase shift dimmer, then we only get dimming control with a phase shift to turn on of 90 to 180 degrees, and it is not linear with phase angle. But there is a further problem. The + and - half cycles must have exactly the sane average current. Otherwise there is a net DC current component and this will drive the transformer core into saturation. In turn either the dimmer and/or the transformer burns up.

If the AC to DC power supply has a voltage regulated output, then over a reasonable change in input voltage there will be little output voltage change and thus input voltage adjustment is not a workable dimming approach.

If dimming was done on the DC side of the circuit, then the DC dimmers were almost certainly overloaded, or a bad design.

.

 

[GoldDigger]

180308-1344 EST

...
...

If dimming was done on the DC side of the circuit, then the DC dimmers were almost certainly overloaded, or a bad design.

.

One other remote possibility with low voltage, high current, wiring is that there were bad terminations or screw connections at the input and/or output of the dimmer module. Voltage drop there would both produce local heating and possibly require a higher current to satisfy the needs of the load (not the case with unregulated LED strips). That local heating would combine with the normal heat dissipation in the dimmer to potentially overheat it.

 

[gar]

180308-1553 EST

GoldDigger:

Good point.

On the Armacost LED Dimmer I obtained at Home Depot it has the following characteristics:

1. It is in a small plastic box about 1.5 H x 2 W and 2.3 long.
2. Not designed for heat transfer.
3. Power connections are via a small Phoenix separable terminal block.
4. Nice looking package.

I have not looked inside. Without testing I would probably only loaqd it to 4 A. Half of the rqting.