It's getting towards the time of year when all the blinky lights are hooked up and turned on. All the sequencing is done, all the decorations are prepared and now everthing just needs powering. that's where the following info might save you from a costly mistake or 2. Judging by forum posts and chat in the chatroom it seems that the go is to put the power supplies and controllers into the smallest watertight box that can be purchased for the least amount of money. That's all good until you turn things on and everything starts radiating heat into that lovely plastic box that is a lovely thermal insulator. If you don't want your own mini Chernobyl to be part of your display this year you're going to have to think about all that heat inside the box and how to get rid of it.
Some stats to start with. The switchmode power supplies that are being squeezed into nearly every plastic box it seems are typically around 90% efficient. This means that the fairly common 350W power supply that is being used this year will lose 35W in heat when at full load.
Transformer based plugpacks are terribly inefficient. A 30W AC output will typically be less than 70% efficient so you could expect heat losses of greater than 9W per plugpack. A DC output transformer based plugpack is even more inefficient due to the internal recifier. You could safely say that they are going to be 60% efficient and a 12W loss would be expected.
If you're putting a dimming controller into a box then you've got heat losses there as well. An AC controller will lose about 3W for each amp of output current that you are switching. If it's 240V lights and you are switching 2400W or 10A then they will be losing about 30W. If it's 24V lights and you are switching 240W or 10A then there again be a loss of around 30W.
DC controllers get a bit more complicated. There are a number of contributing factors to the losses with them. The transistor/mosfet used as the switching device, how fast the transistor is switching, the length of the output cabling, the type of output cabling, what portion of the maximum current per channel is being used etc, etc. As a super rough guesstimate you could expect to get better than 90% efficiency but it wouldn't be a bad thing to work on 90%. So if you're driving 48 channels from 24V with 100mA per channel you can expect around 10W of heat loss.
All these numbers are great but what are they going to mean. Assuming for instance that you have a switchmode supply and a 48 channel controller in the same box and that you are running at 100mA with all 48 channels. The controller will be losing about 10W of heat and so will the power supply. What will happen is that everything keep getting hotter until the heat being generated inside the box will equal that being lost through the box. If you've got a lovely sealed watertight box then it is only when the plastic has heated up enough to conduct the heat to the outside air that the temperature inside the box will stabilise. If you have the box ventilated then the air can at least circulate to some extent and the temperature will be lower. If you've got a lot of heat that is going to be lost inside the box then you will probably need to think about using a fan for forced circulation. In an ideal world you will get better results if you have the fan on the inlet and pressurise the box and blow the air out. If that isn't feasible then you could have the fan at the output and suck the air through the box. This means that the air pressure inside the box is a slightly lower pressure (depending on fan size, inlet size etc) and things will cool marginally less. You will also need to remember that because you are sucking air up into the box you need to make sure the the air inlet is not going to pull water up and into the box.
Before things go bang in the night you're going to have to check your controller/power supply boxes under full load remembering that if you're blinking your lights you're unlikely to be running your lights at significantly less than 100% load 100% of the time. As a rough guide to see if things are going to go bang you should run your controller for maybe 15 minutes. Turn the power off as there's likely to be potentially lethal volts floating around. If the temperature inside the enclosure is too hot to leave your hand on ANYTHING then you're looking at a reasonable likelihood of premature failure. At some times during Xmas season there are going to be controllers that get turned on when the ambient air temperature is 40 Celcius. If the controller is in the sun then the temp inside the controller enclosure could be 50C. This doesn't leave much headroom until the typical maximum operating temperature of commercial grade components of 85C.
I had intended to do some testing and measuring to show the difference between sealed, unsealed and fan forced but it got to blinky season and lights took precedence.
Some stats to start with. The switchmode power supplies that are being squeezed into nearly every plastic box it seems are typically around 90% efficient. This means that the fairly common 350W power supply that is being used this year will lose 35W in heat when at full load.
Transformer based plugpacks are terribly inefficient. A 30W AC output will typically be less than 70% efficient so you could expect heat losses of greater than 9W per plugpack. A DC output transformer based plugpack is even more inefficient due to the internal recifier. You could safely say that they are going to be 60% efficient and a 12W loss would be expected.
If you're putting a dimming controller into a box then you've got heat losses there as well. An AC controller will lose about 3W for each amp of output current that you are switching. If it's 240V lights and you are switching 2400W or 10A then they will be losing about 30W. If it's 24V lights and you are switching 240W or 10A then there again be a loss of around 30W.
DC controllers get a bit more complicated. There are a number of contributing factors to the losses with them. The transistor/mosfet used as the switching device, how fast the transistor is switching, the length of the output cabling, the type of output cabling, what portion of the maximum current per channel is being used etc, etc. As a super rough guesstimate you could expect to get better than 90% efficiency but it wouldn't be a bad thing to work on 90%. So if you're driving 48 channels from 24V with 100mA per channel you can expect around 10W of heat loss.
All these numbers are great but what are they going to mean. Assuming for instance that you have a switchmode supply and a 48 channel controller in the same box and that you are running at 100mA with all 48 channels. The controller will be losing about 10W of heat and so will the power supply. What will happen is that everything keep getting hotter until the heat being generated inside the box will equal that being lost through the box. If you've got a lovely sealed watertight box then it is only when the plastic has heated up enough to conduct the heat to the outside air that the temperature inside the box will stabilise. If you have the box ventilated then the air can at least circulate to some extent and the temperature will be lower. If you've got a lot of heat that is going to be lost inside the box then you will probably need to think about using a fan for forced circulation. In an ideal world you will get better results if you have the fan on the inlet and pressurise the box and blow the air out. If that isn't feasible then you could have the fan at the output and suck the air through the box. This means that the air pressure inside the box is a slightly lower pressure (depending on fan size, inlet size etc) and things will cool marginally less. You will also need to remember that because you are sucking air up into the box you need to make sure the the air inlet is not going to pull water up and into the box.
Before things go bang in the night you're going to have to check your controller/power supply boxes under full load remembering that if you're blinking your lights you're unlikely to be running your lights at significantly less than 100% load 100% of the time. As a rough guide to see if things are going to go bang you should run your controller for maybe 15 minutes. Turn the power off as there's likely to be potentially lethal volts floating around. If the temperature inside the enclosure is too hot to leave your hand on ANYTHING then you're looking at a reasonable likelihood of premature failure. At some times during Xmas season there are going to be controllers that get turned on when the ambient air temperature is 40 Celcius. If the controller is in the sun then the temp inside the controller enclosure could be 50C. This doesn't leave much headroom until the typical maximum operating temperature of commercial grade components of 85C.
I had intended to do some testing and measuring to show the difference between sealed, unsealed and fan forced but it got to blinky season and lights took precedence.
