• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 84 subscribers
  • Views 828 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Poor Fitment between BoostXL-DRV8301 and the LaunchXl-F28027F

Mark Hunsberger
Prodigy 130 points
Other Parts Discussed in Thread: BOOSTXL-DRV8301, DRV8301, CONTROLSUITE, MOTORWARE

Sorry I did not attach have a picture, and I probably will later, but I have a problem with the fitment of the BoostXL-Drv8301 and the LaunchpadXl_F28027F board.  The 2 devices mate together, but the inserted length of the mating contacts do not allow a reliable fit.  It looks like, the connector tail length of the LaunchpadXl-F28027F board is not long enough.  When I connect the boards together, if I move the BoostXl-Drv8301 board, there is a 50%  chance the boards will separate.  The jumper switch height causes the booster pack to bottom out without making a physically reliable connection.

If I plug the BoostXL-Drv8301 into my old LaunchpadXl-F28027 board that is 2 -3 years old, and only has male headers top and bottom, the board has a very nice fitment, but then I do not have the FAST processor for InstaSpin

Is anyone else having this problem?   I received the parts from the TI e-store this week.

I really like the time TI has put into this product, and I have a very soft spot in my heart for TI, but not being able to install CCSv6 Linux_64 and not having a reliable way to develop this system could be a deal breaker.  I dusted off my old laptop (MSWIN JUNK) to install CCSv6, but I still need a reliable way to prototype; my environment is not that ideal 

  • 0
    TI__Mastermind 22025 points
    Mark,

    Sorry to hear you're having trouble. While yes, the stacking of the boards isn't ideal, I've never encountered problems with unreliable connections between the two boards. The LP will still function without the switch if you jumper across the third set of contacts, so this may be a possible solution for you.

    You mention not being able to install CCS linux. Is this an architecture limitation you're facing or is something going wrong with the installer?

    Best,
  • 0 in reply to Trey German289
    Prodigy 130 points
    Thank you for the response. I never would have guessed that you would have time to personally reply. I really just wanted to confirm that the poor fitment is, or was rather, an actual design choice made by the team and not just some fluke quality control issue I am having. I really do not want to "bypass" both sets of dip switches to test functionality in my design, but I have to do something. The connection just is not reliable enough. I only get about 40% contact before the shield bottoms out on the 2 dip switch assemblies because the BoostXl-Drv8301 uses what looks like a half height smd female header. I may try replacing the headers on the launchpad with Arduino parts. I think the solder tails are longer, but I do like the design choice of having female headers on the bottom. Since there is no mounting provisions, placing the old design LaunchpadXl-f28027 on my metal desk was a little tricky with the exposed male headers, which is why I never really tried using it in a design.

    As for the CCSv6, I think that is a question for another thread. I understand it is very hard for a manufacturer to implement software in Linux because there is no real standard way of doing things, but "ia32" has gone from current, to legacy, and almost totally to legend. The dependency could also be from Eclipse, but I use Centos v7, which is actually the open source version of RHEL (v7) and there is no 32b compatibility anymore. The workaround is dust off the 10yr old Win7 32b laptop and update from CCSv4, which works, but I, like many other intelligent people who perform technical work with their computers, see Windows as a problem more than a solution.

    I probably will continue troubleshooting CCSv6 for my os, via the log files, and probably through the Eclipse forums because it does not start at all, but I have more pressing issues right now, like finals.

    Thank you very much for taking the time to genuinely address my concerns. I really appreciate it and trust me it goes a long way. It is people like you that make TI an exceptional company.


    Have a wonderful holiday,

    Mark R. Hunsberger
  • 0 in reply to Mark Hunsberger
    TI__Guru** 119590 points
    I agree that the new headers on the Launchpad are the issue. The old ones had little stand-offs, the new version has solder balls. It's not a very clean solution.
  • 0 in reply to Mark Hunsberger
    TI__Mastermind 22025 points
    LaunchPad has been a learning experience for all of us. The F28027 LaunchPad was the second in the ecosystem before we had any idea what it would blossom into. This should explain some of the issues you've noted. As an alternative, you can also use the F28069M LaunchPad which should allow the BP to mate more reliably.

    I would highly recommend posting your CCS issue to their forum. They want to fix issues like you're having but because of all of the nuances of these larger operating systems sometimes its difficult for us to catch software bugs like this in house.

    Totally agree with your comment about winblows... ;-)

    Happy Holidays and good luck with Finals!
  • 0 in reply to Trey German289
    Intellectual 700 points

    Hi Mark,

    Fitment issue:

    I’ve experienced the same issue. The connector solution between the Launchpad 28027F/28069M boards and the DRV8301 board could use some improvement. The solder on the pins of of the Launchpad board sometimes wick upwards and result in an inconsistent fitment. It also wouldn't be unreasonable to expect the pins to ablate the soldermask on the underside of the DRV8301 board and cause a short to its ground plane.

    To get around this, I just bent all of the jumper pins on the Launchpad board by 90 degrees. It works well enough.

    My wish list for future Launchpad/boostxl board revs:

    • Add mounting holes for optional stand-offs between boards

    • Add mounting holes on the corners of the Launchpad board so it can be attached to things easily (sometimes it's useful to use a launchpad boards in one-off projects rather than spinning custom copper.)

    • Replace connector on Launchpad with shrouded and keyed 0.100” headers

    • Replace receptacle on boostxl-drv8301 with keyed 0.100” header (it’s easy to insert the thing backwards or pin-offset. The getting started guide isn't crystal clear on orientation.)

    Linux toolchain issue:

    I was able to get CCS6 to install pretty easily on Xubuntu 14.04.3 LTS. The problem is that controlsuite and motorware only have windows installers. You will have to install those with wine. The ControlSuite.exe and MotorWare.exe apps are really just indexes for the application examples, and are not needed for development.

    Be sure to check the installation instructions at the link below to grab all of the package dependencies for your distro.

    Best Regards,

    Grant

  • 0 in reply to Grant Freese
    TI__Mastermind 22025 points
    Thanks for the feedback Grant!

    Future C2000 LaunchPads all have mounting holes and will be ditching the boot switches to simplify operation. Connectors on the LaunchPad will remain the same as they are standardized across the ecosystem to enable stacking of the boards.

    Best,