The first step to constructing the accumulator enclosure is cutting the outline of the thick aluminum base plate. I triple-checked that it would barely fit within the travel of the Bridgeport before beginning!
The sides and lid are assembled using airplane rivets and clecos were used for fit-up. I enjoy this style of fabrication!
(*yes, chronologically out of order with the above photo*) This is what the four sub-modules (missing front insulation shields) look like stacked in the enclosure.
This photo shows the front and rear “Retaining bars”. They are pinned into the front G10 Radsok insulators and into the blocks bolted to the back of the packs. This stabilizes the packs and makes it a more monolithic and rigid assembly. The front bar has the Poke-Yoke feature as I realized that the pitch between the positive and negative sockets in the cell packs was almost the same as the pitch between the packs, so it would be conceivable that if the front bar were installed wrong one could short a single cell pack with the shorting bar meant to put adjacent ones in series– we deemed this severe enough of an outcome to add the pin and hole features to prevent the front bar from being installed in the wrong slot.
This photo more clearly shows the retaining bar purpose in the assembly. It is pinned to the blocks which are bolted to the cell packs, and then is bolted to the inside wall of the pack. In this way the shear strength of the box wall is used to tie the cell packs in. This makes for a very rigid assembly without requiring the space and installation difficulty of tie-rods.
This is the lid with the power electronics assembly attached to its underside. You can see the main positive and negative Radsok pins where they mate with the cell packs adjacent to the front retaining bar. The PVC insulators which hold the Radsok pins are mounted to the box lid using rubber sandwich fittings to prevent them from over-constraining the lid fitment. The lid is bolted to the front and rear retaining bars with four small screws, since it is not responsible for any of the structural retention of the heavy cell packs. Note the sills attached to the sides of the lid– the accumulator assembly is not sealed using gaskets, but the lid, sides and base all have drainage sills to prevent any water from running inside. The water intrusion test at the competition is something like IP52-IP54, and isn’t very hard to pass as long as drainage paths are well defined and prevent pooling or seeping into enclosures. This enclosure worked well in the rain test and no evidence of intrusion was seen in subsequent disassembly.
Before final installation several guards of Formex were made to shield all the exposed busbars.
Here is the view into the lower section of the enclosure with the lid removed, with the BMS wiring harness, note the connector for BMS signals which mates with the pigtail on the lid.
Without the sides of the enclosure, for clarity, this is how the blind-mate main positive and negative of the lid interface with the cell stacks:
The Gigavac GX-11 contactors define the height of this assembly!