Let there be light!

Today I hit the Home Depot again and picked up a solid 3/4″ drill bit to help with conduit holes. It made quick work of the aluminum ribs versus the hole saw I was using, but it had a tendency to get stuck. So it winds up taking just as much time in the end.More importantly, I finally pulled the trigger on insulation. I brought home three 4’x8′ sheets of R-MAX in the 2″ thickness and three sheets of the 1″ thickness. They had to hang out the back of my SUV up 99 West from Sherwood.

Since my ceiling and wall cavities are all 3″ deep, these thicknesses give me the option of Velcro’ing a thick layer to the top of each roof cavity, and a thinner layer below. I’ll custom cut each layer to dodge obstructions like vents and cables.

This insulation is a special type of dense foam (not styrofoam) and it has a layer of aluminum on both sides to form a heat shield. (Remember how hot the aluminum roof gets in the summer sun?) The 2″ thickness yields an R-value of 13.1, and the 1″ yields 3.85. So that’s a combined R-value of almost 17. For comparison sake, typical fiberglass home insulation is R-15.

Next, I ran the first of two dedicated 12V circuits for lights. Here’s how complex the dimmer switch wiring is. The red wire that’s twisted like a pretzel is an inline fuse holder. The little black box holds a 10-amp fuse in this case, on the positive side between the DC panel and the dimmer switch. This protects the switch’s electronics from any surge that escapes’s the panel’s 15-amp breaker (which will serve two banks). All the black ground wires connect as a Y, running back to the DC panel’s negative bus. I realized I had not yet added a DC ground wire from the DC panel to the RETROvan’s steel frame (because it’s considered optional), so I added that too as an additional safety precaution.

Once I got everything connected, I threw the circuit breaker labeled “LIGHTS” and… nothing! Just like Clark Griswold in Christmas Vacation.

So I wasted three hours trying to figure out what happened. There are numerous fail points. It could be that I mixed up some wire colors. I could have misinterpreted three sets of unrelated instructions. Any one of my crimps could be bad. Any one of the components could be bad. The DC panel itself could be bad. Maybe the circuit breaker failed. Or maybe I had mis-designed the whole bus bar scheme.

But it turns out the only defective component was my Craftsman multimeter, which I was counting on to help diagnose the failure. It had died, and when I opened the battery compartment the 9V battery connection was corroded and had a broken wire.

So like in software debugging, I started using brute force to isolate the bug. I swapped out the cable with a new one. I made a little jumper to bypass the fuse holder. I removed the switch from the circuit. Nothing worked. I finally decided to plug one of the LED lights directly into my test cable, but I accidentally did it backwards. And it lit up! What the hell?

It turns out these particular LED lights are stupidly wired for use in a house (even though no one uses 12V lights in a house), so the wire colors were black for load and white for neutral — not white for positive and black for ground. All I had to do was unplug each light and switch the wires.

And WA-LA! There was light, and it was good. The dimmer works brilliantly too. These LEDs are a warm white light, so not too harsh over the bed.

My DC load through the panel has been pretty light until today — just 0.5 amps. So now with six lights turned on high, the draw is about 3.5 amps. Once the forward bank of six lights goes in, that draw should jump to about 6.5 amps. I’ll also be moving the two ventilation fans onto one of these circuits, so that needle will be good for monitoring the overall DC loads. Right now those vents are direct-connected to the Rover solar charger/controller, which isn’t ideal. The wiring inside the helm is getting busier, so I’ll probably have to add a few bus bars in there for all the DC circuits. Either that or test how they handle being branched outside the helm.

But yeah, it was a good day. I even had a stare-down with the doe who has been raiding our garden.

Leave a Reply

Your email address will not be published. Required fields are marked *