Activity for TimHâ
| Type | On... | Excerpt | Status | Date |
|---|---|---|---|---|
| Comment | Post #287646 |
@#36396 Thanks for the great answer. We did some testing and found that around 33pF seemed to be ideal for this particular board layout, which reduced the bit error rate by over 99%. Since this is an acceptable amount for our purposes, we'll stop there until we get a chance to completely revise the... (more) |
— | almost 2 years ago |
| Comment | Post #287645 |
I'm not sure what the raise time of the signals are, but they are being driven by an ATmega128 running at 8MHz. The physical length of the SPI lines wouldn't be more than 10 inches without the expansion board connected. (more) |
— | almost 2 years ago |
| Comment | Post #287646 |
Correct, the SPI bus works when the old expansion is connected (lots of extra stray capacitance), and doesn't work when the expansion is removed. This board has a terrible layout, as the original engineer decided he could squeeze lots of ICs onto a 2-layer PCB and also autoroute everything. As a re... (more) |
— | almost 2 years ago |
| Edit | Post #287645 | Initial revision | — | almost 2 years ago |
| Question | — |
Unterminated SPI bus needs expansion board to work properly I'm supporting a piece of legacy hardware that routes the SPI bus through a set of headers to an expansion board. Since the expansion board doesn't actually use the SPI bus, we simply ignored the incoming SPI lines when we revised the expansion board. However, we are now experiencing seemingly ra... (more) |
— | almost 2 years ago |