I did some current testing this morning using the new Power board and the last version of the Logic board. The new Power board outperformed the previous version of the Power board and did so without an expensive and difficult to work with copper bus bar. The “secret” to this is two fold. First, I more than doubled the amount of copper foil on the PCB which is now equal in cross section to a 6 AWG wire. Secondly, and more importantly, I’ve increased its thermal mass and ability to radiate heat with the addition of the screw down power terminals. These terminals are massive, are well thermally coupled, and will be exposed to air outside the epoxy and will act as radiating surfaces to push the heat off the Motobrain. Below are photos of the unit pulling 80A and achieving temperature stability at around 62C (better than the previous board which got to around 80C) and the board pulling 95A and leveling off at around 74C. These tests were conducted for about an hour and no significant temperature change was observed after about 15 minutes.
The next steps are to build up some new Logic boards, confirm they work, and mate them to new Power boards. Then I will need to do a little firmware/software update for compatibility with the 4 new analog inputs. Finally, I’ll epoxy them up and do more thermal tests. Assuming these are good, I’ll box them up a ship them off for 3rd party testing.
Save for a single piece of hardware, a large gauge small diameter ring terminal which is on the way, I have all the parts necessary to begin testing this version of the power board. What do you think of the new terminals? Preferable to the Weatherpaks we had before? I will supply this with Philips head screws rather than hex head bolts or hex drive screws as are shown in the photo. Those are just what I had on hand this afternoon.
I am temped to move the main input terminal (the unpopulated one) to the left about a centimeter in order to allow the ground wire to be completely clear of the power input (instead of being kicked out to the left). After I’ve had a chance to install the power cable with its fat ring terminal I will have a better idea about space though. I’m not too concerned either way at the moment. I went for smallest overall size rather than concern myself with tidiness of the wires. I’ll cogitate on it for a while and see how I feel.
The new PCBs arrived a couple days ago. I got 100 of each type (Power & Logic). I noticed right away that the holes drilled and plated for the power terminals were too small. It took a couple of days to come up with a procedure that will allow me to use them, and it remains untested, but I expect it will work. The vendor believes they met the agreed upon specification but I am unconvinced. I am having a calibrated measurement being made now and we should have some clarity as to the solution going forward (we are arguing over about 40 microns). The boards as is will not work for production (because the assembly labor is excessive), but they should be usable for prototypes which is good enough. I’m more concerned about the time cost of running to boards than I am about the dollar cost of producing another run at this point so I’d rather just move forward if at all possible.
You will notice in the photo below that I got rid of the spade connectors in this iteration and instead went with high current power bushes. These screw down terminals are waterproof and very high current capable. Plus they will package up nicely in a retail box and will be comprehensible to the users without being “scary” looking with a mass of wires hanging everywhere.
The logic boards have only modest changes. The only significant update is the 4 addition analog inputs (a total of 8 now) and the way I bring the inputs lines out of the device. I’ve got a standard header which I will use with a sacrificial breakout board which will cheap to replace and easy to change going forward. I went with maximum future flexibility solution.
I’m going to build up prototypes ASAP and get them tested and shipped out to a few prospective customers for testing.