Step 6, Installation
Depending on the order in which things arrive you can start building as soon as it makes sense to do so. No point in assembling a trackerframe when there is no foundation to place it on, but as soon as all items that make up a particular subcomponent are all available you can put it together.
We split up our installation in a number of phases because we all have day jobs and had to take out time during evenings and weekends to complete the job.
The Batteries
After the structures that we put up in step 4 were solid enough to work on we first placed the batteries (to get them out of the way). A solid box made out of plywood was constructed and placed on a footing of concrete blocks to keep the wooden base of the box off the floor, allow for some air to circulate and to put the batteries at a more comfortable height to work on them.
While placing the battery box I found that the blocks were a bit wider than I thought they would be so I shifted them around to keep as much space in the battery room as possible.
Partition Wall
So, with the batteries in their intended position (but not yet connected) I could start work on the partition wall. The partition wall is a really simple affair, with one twist, it has to be close to gas tight to avoid gases venting from the batteries during the charge cycle from escaping from the battery room other than through the vent. So an extra layer of vapour barrier was added to between the drywall and the wood frame, and any space was filled up with polyurethane foam from a bottle. Also the wall was insulated using glass fiber blankets to keep the battery room at it's optimal temperature. Notice how all the wood that touches the concrete is wrapped in thick black plastic film to stop it from drawing moisture up through the concrete which would lead to the wood rotting really quickly.
Solar Panels / Trackers
While the weather was good we decided to work on the solar arrays. First the main posts were placed with the seasonal adjust brace already attached. It helps to have a few people handy during this job, the posts are heavy and it is easy to overshoot the 'upright' position. Especially if - like us - you are working on a hill. Next the main tracker frame was hoisted up and bolted on to the seasonal adjust. Now a few remarks about this picture, this is ABSOLUTELY THE WRONG WAY TO DO THIS... working from a tractor bucket that is extended up like that is asking for trouble, if the hydraulics fail while you're up there (and you know that they'll fail just when you really can't have it) you are going to be in a serious accident. That said I don't really know how else we would have done it, so next time I'll probably do the same thing but I'll lock the bucket in place somehow once I have it at the right height. It sure beats standing on a ladder. If anyone has a better idea on how to go about this then I sure would like to know. Also you'll notice the hill was quite steep. Now that tractor in the picture is a four wheel drive, and we had blocks and stones in front of the wheels every step of the way to stop it from sliding down. BE VERY CAREFUL IF YOU DO SOMETHING LIKE THIS, better to take twice as long than to take one shortcut to regret it...
The next step is to bolt on the crossbars that will support the panels, and when that is done one by one the panels. Of course when you have the first panels on and the whole thing can still flop around you get a flying storm ! Boy I was scared, up there in that bloody bucket holding on to 4x2 ft solar panels fighting to become kites until bolted down. And then when you think you are out of the woods the wind starts to play with the whole assembly. I got one scratch in the backing of one of the panels because of all this, but it didn't go trough and it shouldn't be a problem. Still next time I'll keep a better eye on the weather forecast when attempting a job like this. Notice how all the tracker components were given a double coating of rust proof paint to protect it against the elements. Grease was added everywhere the trackers hinge, and the actuator arm (a device originally designed to control satellite dishes) is connected to the top.
With all 16 panels in place the result looks like this. Isn't it gorgeous ?
Switchboard
Back to the bunker !
With the partition wall in place work could start on the switchboard. First item was to cut the plywood sheet to size and lay out all the components to make sure they fit and to place the studs so that the lag bolts that will hold the inverter later have something to bite into.
Next we fit the board on the wall with anchors and lag bolts all the way through the plywood and the studs. I messed up here because I forgot to put a piece of vapour barrier behind it against the wall and by the time I found out about it half the equipment was already wired up so I left it like that. Time will tell if that was smart or stupid (probably stupid :).
One by one the components were bolted to the board and wired up. Notice the cables coming in from the battery room at the lower right of the picture, and passing to DC disconnect and shunt before going on to the inverter. These cables are extremely expensive affairs, according to the wiring code you can not use welding cable or something like that (even though I'm sure it would work just fine...). The AC disconnect has not yet been placed in this picture, it will go on the top left of the switchboard.
Batteries II
After wiring everything up and double checking (have someone else do this !) I connected all the batteries. The two thin yellow cables run to temperature sensors attached to two of the batteries, these allow the charge controller(s) to compensate the voltages it uses to charge the batteries, the colder it gets the higher the voltage will be. I chose to have all the batteries sitting with their + pole in the same orientation, even if that caused the wiring to be less than elegant in some places. I figured the chances of avoiding an accident weighed up against the disadvantage of the harder wiring, it is *very* easy to make a mistake while wiring up a bank like this and you *REALLY* don't want to have that happen to you.
Cables
One of my favourite jobs is digging trenches... No really, I just can't get enough of them. Fortunately I got to do lots of that while putting in this system, a trench between the two solar arrays, another one from the second solar array to the bunker and a third between the bunker and the house. The trench between the house and the bunker carries one very fat Teck power cable (3x#6, armoured and coated), signal cables (ethernet, instrumentation) for some more advanced plans that I have, the two trenches to the solar arrays carry the power from the panels to the bunker. Trenches are about 2 ft deep, and boards are placed on top of the first few inches of backfill to make sure that when I (or someone else) in the future put a spade into the ground I'll hit the board first. Then, a few inches above that is a marker ribbon, to alert a future backhoe operator that he's about to dig up a piece of cable. Even if this is not in your electrical code this is probably a very good idea ! Stripping this kind of cable is hard work, make sure you don't damage the inner insulation when cutting through the outer layers.
Final Check and power on
After verifying everything once more, both visual and with the DVM we switched on the main breaker and voila! The inverter came up and we had power.
To Do:
High on my wishlist:
* A windmill (working on that)
* Instrumentation on everything to keep some real life statistics on using the system
* Solar hot water system