| ??? 04/06/07 13:20 Read: times Msg Score: +1 +1 Good Answer/Helpful |
#136736 - But how much do you save... Responding to: ???'s previous message |
When you do as you have described you save one crystal and a couple of capacitors. If the on-board oscillator between the X1 and X2 pins is a indeed a workable scheme with an external crystal one always has to ask "how reliable". Embedded engineers who have watched this market place develop over many years will know of many instances and have experiences with X1/X2 pins and external crystals not working perfectly on some MCUs, on some boards, or with some components. These engineers will question the external crystal with X1/X2 pin sensitivity even more. Now when you hang a long trace on an oscillator trace to take it to another MCU have you compromised reliability of something many are already leary about? And if so is the savings that you realized worth it if even only 5% of your product fails in the field?
Only you can answer this for your application but I can point out two things. Whenever I need a common frequency I use a separate standalone oscillator and buffer it to the various points that it is needed on the board. Secondly if X1/X2 pins with a crystal were the _best_ oscillators I would guess that the number of available catalogue pages and web pages of standalone oscillator components would be far far smaller. Michael Karas |
| Topic | Author | Date |
| single crystal for two 89c52 | 01/01/70 00:00 | |
| I think you might find it doesnt work like that | 01/01/70 00:00 | |
| Schmitt trigger for crystal oscillator?? | 01/01/70 00:00 | |
| yeah possibly | 01/01/70 00:00 | |
| As it happens ... | 01/01/70 00:00 | |
| I yield to the voice of experience | 01/01/70 00:00 | |
| I was quite surprised, too! | 01/01/70 00:00 | |
| One quartz to | 01/01/70 00:00 | |
| not a good idea | 01/01/70 00:00 | |
| But how much do you save... | 01/01/70 00:00 | |
commentary - corollary | 01/01/70 00:00 |



