prevent damaging mosfets

[dollar4bill]

In testing out the 2 new CMB-24D controllers, I’ve done something where I heard a pop and now channel 10 doesn’t work on one controller and channels 10 and 14 no longer work on the other controller. In both instances, I hooked up the cmb-24d controller to a 12v, 350 watts power supply and mounted inside a cg-1500 cableguard.  I then wired 4-conductor waterproof extensions to each channel and the + on the controller (http://www.holidaycoro.com/4-Conductor-Waterproof-Extension-5ft-p/725.htm) and wired the other end of the extension to one string of 12v 50 count dumb rgb led string of lights.  I water proofed this end of the connection on each individual wire with1/8 inch diameter heat shrink wrap tubing and also overall with 3/8 inch diameter shrink wrap tubing that wraps around all 4 wires at once.  I was using the light-o-rama hardware software to test everything and as I made my way to turning on the red, green, blue channel to make the dumb rgb light string light up white, the first couple of light strings worked fine.  Then, when I got to channels 10, 11, 12 on one controller and channels 10, 11, 12, 13, 14, 15 on the other, I heard some pops and channel 10 on one controller and channels 10 and 14 on the other controller no longer work.  The controllers were still under warranty so lor fixed them for free.  I have the controllers back and before i wire everything up again, i'm wondering what could've caused the mosfets to blow?  old mosfets?  A short circuit?  Is there a way to test if there’s a short circuit occurring without blowing the mosfet?  Thank you for any insight you can provide.  

[k6ccc]

The most likely cause is a short circuit in your wiring. A volt ohm meter (DVM or VOM) will tell the tale. You can also try using a power supply that is current limited to a couple amps and clip lead connect the suspect channel to it while watching the current with the DVM. If you don't have a current limited power supply, I can explain how to create the effect.

[dgrant]

Lets be clear on those connections...on the controller, you'll see four wire ports, times 8 sets. Each port has its own +(Positive 12V) and each also has an "R", "G" and "B" connection which are each -12V from the controller. You connect those to the dumb strings only, +12V to the proper wire on the string, and the R, G and B wires to their string connections. Be sure to look closely and connect them correctly. Are you sure they are dumb strings you bought? As k6ccc suggested, you might want to check those extensions and insure they are wired correctly too with a DVM or VOM. I've run into some from another source that were not wired correctly inside.

[mpageler]

When I tested out my new cmb24d earlier, I had a sloppy testing protocal and caused some shorting.  I heard some popes on the board.  In my case, the problemed channels just stayed on all the time.  I opened a help desk ticket with LOR and got authorization to return to LOR for repair.  My screwup was coovered under 1 year warrenty with shipping cost payed by me.

It took a few weeks earlier this summer but as you get closer to the holiday season, LOR warns that it could take longer.

[dollar4bill]

this is the power supply i have http://www.holidaycoro.com/350w-Dual-Output-Power-Supply-p/49.htm.  I don't believe it's a current limited power supply.  How can i create the effect?  

 

also, if there's a short circuit, it has to be in pig-tail, light string connection.  I have a volt ohm meter.  How do i use that to test for short circuits in the pig-tail, light string cable runs?  

 

My controllers were under a year old and LOR fixed them with me only paying for shipping to them.  

[k6ccc]

No, that is not a current limited supply.  Current limited supplies are generally electronics technician bench supplies used for testing.  A current limited supply will only supply a set amount of current (often controlable).  You can create a current limit on a supply by putting a resistor in line with the current.  One of the more common (low tech) current limits is an incandecent light bulb.  Unfortunately it generally takes a little trial and error to determine the correct light bulb because the resistance of light bulbs changes A LOT when they are only slightly on (the part of the curve desired).  For example, if you are using a 12V supply, and run the current through a 40W 120V light bulb you may find that with a dead short, you get 2 amps.  A 60W bulb may pass 3 amps.  IMPORTANT NOTE:  I just pulled those examples out of thin air.  Without testing I have no idea what current you would get under those conditions.  To test the current limiting, wire it up like this:

 

Power supply + >>> current meter >>> light bulb >>>  power supply -

 

BTW, if you are using a 12V power supply and a 120V bulb, don't expect the bulb to light.

 

Try different bulbs until you get a little more current than the desired test current.  Once that is determined, hook up the test device between the light bulb (right side in the example above) and power supply minus.

 

To use a VOM or DVM to find a short, use the Ohms scale of the meter with the leads between the various light connections.  You should NOT be seeing zero (or near zero) ohms.

[Ron Boyd]

One other thing to mention that bit me a couple of years ago. Do a continuity check on the pigtails. It's a simple multimeter test to make sure the pigtails are wired correctly, as mentioned above. It took me a week to find the culprit. Thankfully, mine were ground and data swapped. It was on 2811 pixels, but it can happen on dumb strings too.

I'm not sure what VOM and DVM are, it may be the continuity test under a different name

Edited September 9, 2015 by Ron Boyd

[LOR Staff]

The Cheap Transformer and why You Shouldn't Use It.

 

[dollar4bill]

ok so i'm assuming transformer=power supply and i feel like my issue is what Mike linked to...a cheap power supply.  Does anyone have recommendations of 12v 20+amp regulated power supplies?

[LOR Staff]

In general, I reserve the words 'Power Supply' for those devices that are regulated with built-in regulation.  'Transformers' in my eye are unregulated.  A 'Regulated Transformer' is a transformer that is connected to an external regulator.

[EmmienLightFan]

HolidayCoro mention nowhere that is is regulated, which is bad considering that many people on here will only buy regulated ones.

 

It still could be regulated though.

 

 

Also, I thought there was no such thing as a transformer that puts out DC. I know that some cheap power supplies do just have transformers, then a rectifier, and possibly a regulator, but proper switch mode power supplies don't and are normally regulated.

 

I could be wrong - this is all stuff I read on here - but I thought transformers are only for AC.

Edited September 11, 2015 by EmmienLightFan

[dollar4bill]

So now i believe i'm wrong in thinking the issue is a cheap transformer.  The transformer i'm using is this http://www.holidaycoro.com/350w-Dual-Output-Power-Supply-p/49.htm.  On that page, scrolling down to the "product features" section, there is a line that says "switching power supply with screw terminal for input and output."  Looking at Mike's link above about cheap transformers, he mentions "t he switched supply is going to be even more expensive, but has additional benefits."  Does that mean this power supply is switched and therefore the best choice?  Or is that meaning something else?  Should I ask Holidaycoro?  

[Rayburn]

I don't think you have a power supply problem.  It is odd how channel 10 has faulted on both controllers. You probably have a cable or clearance issue.

 

Are you using the same cables or lights and swapping them between the controllers?

 

How have you mounted the board in the CG-1500 box? Are you using a metal back-plane?

[dollar4bill]

yes, i was using the same lights and swapping them between controllers.  Not 1 dumb rgb light string that was moved through each rgb channel but 8 dumb rgb light strings connected to their sister channels when swapping the controllers.  

 

this is the mount i'm using http://www.holidaycoro.com/Light-O-Rama-LOR-CMB24D-Mounting-Kit-p/641-kit1.htm inside a cg-1500 enclosure.

[Rayburn]

If you used the same light for channel 10 on both boards, I would highly suspect you have a short in that light or its cabling.  As others have said, use an ohm meter to check between each of the wires of the light. Compare the readings with readings on known good lights.

 

You say you heat-shrunk each wire and then covered those connections with more heat shrink.  You may find you damaged the underlying heat-shrink when heating the outer layer.