| ??? 03/02/07 23:40 Modified: 03/02/07 23:50 Read: times |
#134170 - Uncertainty Responding to: ???'s previous message |
Ertan Balaban said:
This is an output of sensor. I want to measure the duty cycle with in 16 bits resolution. But no captute input of micros can measure the duty cycle in this precision. Do you mean precision, or resolution. You will have to take an average under any circumstances - your analogue level is an average of a few hundred cycles of the PWM. Why not just measure a few hundred cycles using the CCU ? Flogging the processor at 24 Meg, you'd get 2450 counts of resolution at 50/50 duty cycle. What depth of modulation are you expecting to see ? Averaging 10 cycles will give you sqrt(10) times more resolution, if I am not mistaken, so averaging 700 cycles would give you close to your 16 bit resolution, and would take 70 milliseconds - roughly 14 Hz timeconstant Steve |
| Topic | Author | Date |
| precision duty cycle to analog signal conversion | 01/01/70 00:00 | |
| WHY? and how | 01/01/70 00:00 | |
| 16 bits resolution | 01/01/70 00:00 | |
| Uncertainty | 01/01/70 00:00 | |
| uncertanity | 01/01/70 00:00 | |
| PWM | 01/01/70 00:00 | |
| it's an AWFUL solution | 01/01/70 00:00 | |
| Or stick with the F352... | 01/01/70 00:00 | |
| Is "F352" a 4-letter word... | 01/01/70 00:00 | |
| well, if you want to look like me ... | 01/01/70 00:00 | |
| Really? | 01/01/70 00:00 | |
What sensor? | 01/01/70 00:00 | |
| Resolution does not help you here. | 01/01/70 00:00 | |
| Analog signal is proportional | 01/01/70 00:00 | |
| you are right | 01/01/70 00:00 | |
| Simple answer | 01/01/70 00:00 | |
| true rms convertor | 01/01/70 00:00 | |
| Why RMS | 01/01/70 00:00 | |
| why rms | 01/01/70 00:00 | |
| RMS detectors suffer from high crest factors,... | 01/01/70 00:00 | |
| Why, oh why | 01/01/70 00:00 |