??? 10/17/05 21:14 Read: times |
#102515 - required sampling rate Responding to: ???'s previous message |
The machine will ramp the pressure through the full span of the transducer in one minute. This will result, at ten readings per second, in readings one sixth of a percent of span apart. Since we specify the accuracy of our machine at one quarter percent of span the quantization error will be within our spec.
What we are measuring here is the pressure where the specimen fails. This pressure can vary 10 percent for specimens taken from the same part. I know that some of you are asking "Why require sixteen bits when the need for accuracy is much less?". I asked my customer the same thing and he said it was to have a superior sales feature. This way, if he could get his customer to specify sixteen bits the competition would be locked out. I may have technical reservations about the wisdom of what I am asked to design, but my customer knows what he wants and what he will pay me for. I think I should design the machine to read at a faster rate. That way, when my customer comes back to me for more speed I'll be able to give it to him without major redesign. Don |
Topic | Author | Date |
ADUC848 project getting started | 01/01/70 00:00 | |
Some hints... | 01/01/70 00:00 | |
More! | 01/01/70 00:00 | |
Bible? | 01/01/70 00:00 | |
The "book" is 3 general chapters that re | 01/01/70 00:00 | |
ADUC848 Project: details for hints | 01/01/70 00:00 | |
ground loops, oversampling | 01/01/70 00:00 | |
You have a head start, most have proble | 01/01/70 00:00 | |
Is 10 readings per second sufficient? | 01/01/70 00:00 | |
required sampling rate | 01/01/70 00:00 | |
Thanks, but... | 01/01/70 00:00 | |
books? | 01/01/70 00:00 | |
books? | 01/01/70 00:00 | |
Avoid That | 01/01/70 00:00 | |
C on a PC? | 01/01/70 00:00 | |
OK | 01/01/70 00:00 | |
Starting C on the '51 with no prior C kn | 01/01/70 00:00 | |
Point taken![]() | 01/01/70 00:00 |