Skip to content

2026‐05

Jump to bottom
Vector edited this page May 27, 2026 · 3 revisions

May 5, 2026

Caribou Call — Recording

Attendees: far1no, alperenag, errrks.eth, thomasg, kbmollysuh

  • Project Status / Funding

    • Julius published the April Caribou progress report in the forum.
    • May funding has been received and contributor streams have been updated.

  • Battery Connector / 90° Adapter PCB

    • KBM closed most battery connector issues and created the 3D model, KiCad footprint, and KiCad symbol.
    • Because the selected battery connector is not right-angle, KBM is working on a 90° adapter PCB.
    • Julius created quick mockups showing the adapter/battery connector PCB arrangement.
    • A first version of the battery connector/adapter PCB is expected this week, with screw-terminal adjustments still needed.

  • CMAIN / Flight Controller / ESC Routing

    • Julius is working on the Caribou main PCB but has not made major progress yet.
    • First prototype flight controller plan remains Quiver-style: FC PCB setup adapted for a hexacopter.
    • Alperen raised future redundancy/certification concerns for aircraft over ~150 kg. Julius noted that future redundancy options have been discussed, but prototype 1 is focused on getting the aircraft flying.
    • Longer-term options discussed include higher-end redundant flight controllers, Pixhawk 6X/Cube family options, and boards supporting dual Cube Orange-style redundancy.
    • ESC cables may route outside the tube for prototype 1 because the current frame lacks an internal routing hole. Erick agreed this is simpler and may also help cooling.

  • Caribou Hub

    • Julius mentioned early Caribou Hub design mockups/renders, including motor RPM views.
    • Julius is not sure how far along Alex is. Erick noted it looks similar to earlier Feather/companion-computer concepts and may reuse some prior work.

  • Thomas Visit

    • Caribou will not be ready to fly during Thomas’s current visit, which Thomas expected.
    • Thomas said he is eager to help and can return for more time when the aircraft is closer to flying.

  • Longshot Battery Pack Update

    • Julius adjusted the top and bottom printed battery-pack parts so they fit well with the original casing.
    • He plans to define more concrete tasks soon so more people can contribute.
    • New battery interconnect plan uses small rigid edge PCBs instead of Feather-style flex PCBs.
    • Small sheet-metal parts will solder to those PCBs, with four connectors back to the main BMS PCB.
    • Side screws are mainly an assembly/manufacturing aid; glue will still be used in final assembly.
    • Sidewalls will likely remain printed. Top and bottom may use carbon fiber or another material because CF plates are expensive.
    • Weight target is to stay under or close to current Tattu battery weight. Some casing clearances are tight, with printed sections around 1.75 mm.
    • Julius wants to order copper soon because lead time is roughly 1–2 weeks.

  • Battery Cost Reference

    • Thomas reported recent Tattu battery cost at about $745 per battery including tax/customs.
    • Thomas is excited for an alternative because Tattu batteries are expensive.

  • Open Items

    • KBM/Julius: finish first C90D/battery connector adapter PCB version and adjust for screw terminals.
    • Julius: continue CMAIN_PCB work.
    • Erick: review Caribou electronics/frame items and post thoughts in Discord.
    • Julius/Alex: clarify Caribou Hub status and reuse opportunities from prior companion-computer work.
    • Julius: break Longshot battery pack into concrete contributor tasks.
    • Julius: order copper for Longshot battery pack when design is ready enough.

May 26, 2026

Caribou Call — Recording

Attendees: far1no, errrks.eth, kbmollysuh, thomasg

  • CBC PCB / Enclosure Fit

    • KBM had not started the Fusion enclosure yet, but had completed the 90-degree PCB modification.
    • KBM shifted the compression lock outward by about 7 mm to improve clearance on the battery side.
    • Julius said the updated PCB is now ready enough for KBM to use as the basis for enclosure design.
    • Julius asked KBM to review both schematic and PCB routing, not just the schematic.
    • KBM flagged possible clearance concerns around mounting holes, board edge, TVS/gate-protection diode area, and top/middle connector clearance.
    • KBM posted a drawing in the channel; Julius accepted the middle clearance change for connector-pin spacing / short prevention.

  • CBC Power Architecture Update

    • Julius summarized the current board layout: high-current connection pads and current-distribution vias; two 5 V DC-DC converters; two 3.3 V converters fed from 5 V for ESP32/control ICs; isolated 12 V converter expected around 5 A; precharge circuit; redundant gate supply for MOSFETs.
    • ESP32 Wi-Fi/Bluetooth is not required for flight function but may be useful later for a small local battery/PCB status app.

  • High-Current PCB Approach

    • KBM raised 10 oz copper as a possible high-current PCB recommendation for ~300 A designs.
    • Julius and Erick leaned toward local copper busbars instead of increasing whole-board copper weight.
    • Rationale: even 2 oz copper makes boards much more expensive; busbars put copper only where needed and should be much cheaper.
    • Tradeoff: busbars may need to be sourced separately and shipped to the assembler because they are not standard JLC-stock parts.
    • Julius prefers manufacturer assembly/soldering of busbars where possible because manual soldering risks overheating components or exceeding temperature limits.

  • MOSFET / GaN Discussion

    • Current MOSFETs are roughly $2–$3 each and the design uses seven in parallel under the same heatsink size already used on Quiver/Klipper-style hardware.
    • Julius explained that datasheet headline current ratings assume strong cooling; practical continuous current without heatsinking can be much lower.
    • KBM found a GaN option around 200 V / 133 A at about $6 each in quantity, but it is more expensive, has awkward footprint/routing constraints, and may expose drain potential on the top surface, complicating conductive heatsinking.
    • Direction: stay with the current parallel MOSFET approach for this revision; GaN is not compelling here.

  • Manufacturing Timing

    • Julius is aiming to order the CBC PCB around the end of the first week of June, but not before enclosure fit is ready.
    • Erick agreed to review the PR once Julius marks it ready.
    • Julius will continue work on the other PCB in parallel while CBC review/enclosure work proceeds.

  • Aluminum Bracket / Enclosure Manufacturing

    • KBM expects to add or extend existing features for new holes/mounting surfaces on the aluminum bracket.
    • Julius said the aluminum parts have not been ordered yet, so these changes can still be included once the CSMD/enclosure design is satisfactory.
    • Julius expects bending/manufacturing cost impact to be acceptable if the part remains reasonable for sheet usage.

  • Vector Follow-up

    • Julius said Vector finished a PR and he will merge it into main.
    • Julius also noted Vector needs to adjust some sanity checks after the merge.

  • Open Items

    • Julius: update/publish the model space so KBM can start enclosure design.
    • KBM: start enclosure design from Julius’s updated PCB/model space.
    • KBM: document PCB hole/edge/component clearance concerns with screenshots/drawings in the channel.
    • Julius: incorporate the middle/top clearance adjustment KBM drew for connector-pin spacing / short prevention.
    • Erick: review the CBC PCB PR once Julius says it is ready; target within the next two weeks.
    • Julius: continue the other Caribou PCB in parallel.
    • KBM/Julius: add/adjust aluminum bracket holes and mounting features before parts are ordered.
    • Vector: handle sanity-check adjustments Julius mentioned after the PR merge.

Clone this wiki locally