To power the fan and any other electronics, we need to step down the voltage (or electron pressure) from the batteries' natural 24V-ish to around 12V-ish.
After some googling and searching and waiting, we came about this LM2678 chip from National Semiconductor.
It comes with a standard design, and all that is needed to do is to follow the directions as given by the datasheet for picking the right parts.
It can provide up to 5A (would not subject it past 3A in retrospect), has a build in On/Off switch and is very efficient at around 90% or greater.
Below is the design we finalized before printing 2 of them. It was made to be fixed at converting 24V to 12V. It could also be designed to be adjustable and provide other voltages, but different parts need to be purchased to make it so.
The schematic is free for anyone to download and use. It is available through our Redmine page.
We use an open source circuit design program gEDA to draw out the board and print out the layout to be transplanted to the copper board.
Apart from all that work, which is a post in its own, we followed the steps necessary to print out the board based on this fabrication method from PCB Fab in a Box.
First we print out the circuit on a piece of paper in a regular printer to see where it will come out. Our board is 2 sided, so this process is repeated two times, while covering the already completed side with masking tape before subjecting it to the etching step.
Then we cut out a piece of the appropriate transfer paper and tape it to where it will print out, making sure it will be straight. Also, making sure the shiny size of the paper is facing up, to be printed on. This is the side that will stick to the copper board.
Cutting out an appropriately sized piece of copper board.
Then using a laminator, the shiny printed side of the paper is transferred to the board.
Unfortunately we do not have a picture of that transfer, but it will look similar to the pictures below. The copper board with the paper now sticking to it is thrown in water, and the paper will peel off, showing only the traces.
Once that transfer is done, another layer is transferred using the laminator. That is this green paper below. It will stick only to the black traces previously transferred. The previous transfer looks just like this one, except black instead of green.
The black marks are from a regular Sharpie pen covering the traces that were not perfectly solid. The sharpie will withstand the etching. Everywhere around the green traces and below it, there is a thin layer of copper. The green will prevent the copper underneath to a good extent from being removed.
For the etching we used ferric chloride. When dissolved in water, it becomes an acidic corrosive solution and if in contact with copper will have a chemical reaction so that the iron and the chlorine in the solution will stick to the copper. These new Iron+Copper and Chlorine+Copper chemical compounds will stay in the water mixture and in this way the copper is removed from the board.
Doing this for about 5 minutes... The hotter the solution the faster the reaction is. A good mild scrubbing action with the sponge also helps it along.
Rinsing in water.
The green masking comes off with a non polar solvent. We used acetone but any alcohols should work.
Here you can see the clean board with copper existing only where the traces were printed. The other side of the board has different traces. We needed two sides of the board to be able to make all the connections between parts without going over another line.
When doing the other side of the board, it is important to line up the traces, we we drilled a couple holes to line up both sides. This is seen in the video but there are no pictures of it.
Drilling out the holes. We had to re-drill a number of the holes bigger because some of the components could not fit.
To make it all clearer here is a video of the process.
No comments:
Post a Comment