Sunday, 30 March 2014

PKP (trials and tribulations of building a homebuilt motor controller, Pt. 2)

When I last posted about PK's motor controller, I had just let the magic smoke out of a MOSFET from a cheap cordless power drill, and ordered 8 of these 60V 30A MOSFETs from sparkfun. Well, a little while later I had a pretty little red box show up, and it was time to get started:
 Also on hand for this project, I had the heatsink from an old computer (already cut in half in this picture)
And the ever faithful relay part of my motor controller
I set to work on the heatsink, planning on drilling and taping 6 holes into it, and cutting it in half so each channel would have three MOSFETs wired in parallel. Well, the heatsink had other plans, and I ended up snapping not one, but two drillbits in seperate holes on the heatsink. At this point I decided that rather than try my luck with the other half of the heatsink and probably end up with the same fate, I would cut off the portions of each heatsink that were ruined, tap the holes, and hope that two MOSFETs in parallel would be enough. Here is a picture of the heatsinks that I ended up with:
I then soldered up a board with 4 MOSFETs plus all of the necessary connecting wires:
And put it all together:
Copious quantities of heat sink compound was used, and I put a cut up chunk of circuit board between the heatsinks to prevent them from shorting out on one another (MOSFETs are different than voltage regulators and the like in that the tab on them is not connected to ground, it is connected to the output, so it is important to isolate the channels from each other) The whole assembly is held together by a couple of zip ties (I love my zip ties :) )
I taped an unused computer fan to the top of the assembly, connected it to the relay block and...
Found out that the MOSFET driver board that I made for the last batch of MOSFETs was dead. So this is the second driver board I have built and then fried, and I am getting a little iritated, but I am this far in already, so I gotta keep going, so I broke out the multimeters, battery and electronics, and prototyped another one,
cut the disfunctional one off of the relay block driver,
and soldered up yet another board:
I (kinda, maybe) learned some lessons from the other two that I fried, and made this one completly removable using 0.1" male and female headers so that if it fries, I can swap in a new one without cutting any wires. I also doubled up on the transistors for each channel, hopefully giving a little bit of redundancy and added a header for the computer fan. I stuck it back on the motor controller and now have a fully functioning motor controller:
The male spade connector is for ground, power is supplied to one of the four relay terminals in the centre of the brick, the heat shrink tub covered two pin connector is both ground pins to be connected to the logic circuit, the four pins are for direction control, and the two left are speed control.
Here is a picture of it mounted on PK:
 The interesting thing about large robots such as PK is that, in contrast to smaller ones, it is not as important where things such as a motor controller or MCU is mounted, and it might change around a lot, so that is probably just a temporary resting place. That said, here are a couple more views of how PK is right now:
Due to not having a lot of time and the fact that my front yard is one big mud puddle while my back yard is still under a bunch of snow, I have not yet tried the motor controller on PK's motors yet, I just hope that the two MOSFETs per channel will stand up to the motors, cause if I let the magic smoke out of these, I might not have the motivation to rebuild the speed control part AGAIN, and I am a little on the broke side to buy a commercial speed controller. The paintball gun isn't quite operational yet, I still need to buy a right angle fitting for the remote line so that it will fit between the end of the gun and the main battery. That's all for now, I hope to try that motor controller out soon.

No comments:

Post a Comment