Other Parts Discussed in Thread: TMDSHVMTRINSPIN
The eval kits and lab code are great for early design. Since transition to a product is a step that is desired by TI and its customers, a few hints might help. Here are a couple questions and hints in a transition from the high voltage kit TMDSHVMTRINSPIN.
1. Designing product hardware:
A. Downsize cost of micro. In our case, the 28069M is desired and only a dollar or so may be saved with the 80 pin version. However, comments on this process are invited as they may help others (or our low cost product). It looks like heavy hardware mods are needed to test some of the lower performance options.
B. Remove the USB emulator. The XDS100v2 has been great. To connect to the product, it looks like we may need a JTAG pattern, a cable emulator and isolated USB? There are choices on the JTAG pinout, but as far as I can tell, the simple one on the eval kit looks like it would work for an external XDS100v2 emulator, or not?
C. Redesign the balance of the hardware (no specific questions, but open for comments).
2. Making production software (other than the obvious):
A. Chris Clearman and Adam Reynolds pointed out that there are gMotorVars calls in the lab code that eat bandwidth and add to the learning project, but not to the product performance. References:
B. Is there a list of “unnecessary” gMotorVars? I realize they vary by lab, perhaps there is a way other than searching on each name to determine usefulness, or other hints in this regard.
C. Any other comments, or examples, on streamlining or otherwise tailoring lab code to product code?