8
\$\begingroup\$

When writing software for instrumentation, it is often required to specify a small gap of time between taking measurements. Say for example that 250 voltage measurements must be taken, and they must be 1ms apart.

Is there a word or phrase in electrical engineering that describes that 1ms interval that takes place between the voltage measurements?

\$\endgroup\$
  • \$\begingroup\$ or measurement interval. But sampling ... is good as well. \$\endgroup\$ – Neil_UK Jul 28 '16 at 13:19
  • \$\begingroup\$ Sampling increment. \$\endgroup\$ – Chu Jul 28 '16 at 14:43
  • \$\begingroup\$ Generally most engineering tools specify the problem as how many samples per second and then how many seconds to sample. \$\endgroup\$ – slebetman Jul 29 '16 at 3:04
29
\$\begingroup\$

The name for the frequency the samples are taken is Sampling Rate (not specific to audio only). It is measured in samples-per-second or in Hz (which is 1/s). The time between two samples is called sampling period, and is given in units of time. So in your case the period would be 1ms and the frequency 1/1ms=1kHz. Or in terms of rate it's 1000 samples per second.

\$\endgroup\$
  • 2
    \$\begingroup\$ ms is a time, not a rate. \$\endgroup\$ – Scott Seidman Jul 29 '16 at 1:24
  • \$\begingroup\$ "how frequently" is a frequency. \$\endgroup\$ – Eugene Sh. Jul 29 '16 at 1:30
  • 1
    \$\begingroup\$ This answer is incorrect. Sampling period is correct. As mentioned a unit of time is not a rate. Rates are time^-1 \$\endgroup\$ – Lamar Latrell Jul 29 '16 at 6:22
  • 1
    \$\begingroup\$ Oh my... The OP is not asking "how 1ms between the samples is called", but "how to describe the fact there are samples 1ms apart". And it can be described either in terms of period or the frequency dually. Especially frequency, given the title. \$\endgroup\$ – Eugene Sh. Jul 29 '16 at 17:18
  • 1
    \$\begingroup\$ I afraid he didn't understand it fully... Anyway. We have the right to have different views on it. \$\endgroup\$ – Eugene Sh. Jul 29 '16 at 17:48
20
\$\begingroup\$

I would call it the sampling period or measurement period. For short periods like 1 ms, specifying the sampling frequency is another option (1 kHz).

\$\endgroup\$
  • \$\begingroup\$ Don't be surprised to hear sampling time in some contexts, which (insofar I encountered it) means exactly the same as sampling period. \$\endgroup\$ – Sanchises Jul 29 '16 at 9:13
  • \$\begingroup\$ @sanchises, for me this is a different quantity. For instance, one may measure the quantity for 10ms, repeated every 1s. In this case the sampling time is 10ms, the sampling period 1s. \$\endgroup\$ – Gremlin Jul 29 '16 at 12:11
  • \$\begingroup\$ @Eoin I agree - but I've seen otherwise quite often indeed. For example, MATLAB's Simulink invariably calls it 'sample time'. Hence my remark 'Don't be surprised', since it may very well be surprising to see such inconsistent naming conventions. \$\endgroup\$ – Sanchises Jul 29 '16 at 12:28
  • \$\begingroup\$ Downvoted: Time between measurements is not the sampling rate or frequency. rate/frequency = 1/period. Period = sampling time + time until next measurement. The term for the latter is the OP's query. \$\endgroup\$ – iheanyi Jul 29 '16 at 17:13
6
\$\begingroup\$

The word is "period". Could be used in the expression "sampling period", in your case.

\$\endgroup\$
  • \$\begingroup\$ Downvoted: Time between measurements is not a sampling period. Period = sampling time + time until next measurement. The term for the latter is the OP's query. \$\endgroup\$ – iheanyi Jul 29 '16 at 17:14
  • \$\begingroup\$ @iheanyi how would you call it then? Note: the time to actually sample the valu wasn't specified by OP and can therefore considered null. And this is really an implementation detail. Not relevant here. \$\endgroup\$ – dim Jul 29 '16 at 18:16
  • \$\begingroup\$ Huh, time to sample can't just be ignored. Anyway, the question mentions writing software for instrumentation. Clearly, the OP already understands sampling terminology, so this question is not asking what you thought. As for what I think it's called, you can read my answer and the OP's comment on it. \$\endgroup\$ – iheanyi Jul 29 '16 at 20:37
  • \$\begingroup\$ Well, it doesn't make sense to me. But I'm not going to argue anyway. And to be frank, I find it quite disappointing that such a trivial thing attracts so many upvotes. So I can easily accept as many downvotes you'd want on that (although not for the reason you give, but let's move on...). \$\endgroup\$ – dim Jul 29 '16 at 21:32
  • \$\begingroup\$ The downvote is pretty much because I don't think this answers the question. Courtesy when downvoting is to say why, so someone has a chance to address the issue in their answer (or defend it and change the downvoter's mind). \$\endgroup\$ – iheanyi Jul 29 '16 at 21:39
0
\$\begingroup\$

It is the sampling interval. Sampling period is the time during which the sample is taken, i.e. the time taken for the actual conversion.

\$\endgroup\$
-2
\$\begingroup\$

There is not an actual term for that time. As a reference, given all the wrong (based on my interpretation of the question) answers, Sampling frequency = rate at which something is sampled = 1 / sampling period.

If the sampling frequency is 100Hz, it means the sampling period is 10ms. So, if the actual act of sampling takes 9ms, and there is 1ms until the next sample, that 1ms is not the sampling period.

What an engineer would term that 1ms time is context dependent. Most won't have a name because there may not be a need to actually talk about that time. For the purposes of filtering or operating upon the sampled values, that datum is useless.

If on the other hand, you need to discuss behavior or properties of your sampler, for example, if you're sampling as fast as possible and the sampling device needs a certain amount of time to "reset" or some other action before it can sample again, it may be called dead time, blank time, or OFF time. There, the term that is chosen conveys information about the sampler's behavior.

So really, the name depends on the purpose of the information you need to convey about the 1ms time.

\$\endgroup\$
  • \$\begingroup\$ You seem pretty sure that the OP was referring to the time between the end of one measurement and the beginning of the next measurement. \$\endgroup\$ – Justin Jul 29 '16 at 17:29
  • \$\begingroup\$ @Justin "that describes that 1ms interval" pretty much... \$\endgroup\$ – Snoop Jul 29 '16 at 17:32
  • \$\begingroup\$ @Justin I'm basing that on the complete text of the OPs question. It references writing instrumentation software suggesting that the OP is familiar with the concept of sampling and thus would know the meaning of terms like "frequency" and "sampling rate". Together with the actual question posed suggests the OP is specifically referring to the time from the end of sample N acquisition to the beginning of sample N + 1 acquisition. \$\endgroup\$ – iheanyi Jul 29 '16 at 17:32
  • \$\begingroup\$ Based on the OP's own comment above and reading his profile, my assumption seems to bear out. \$\endgroup\$ – iheanyi Jul 29 '16 at 17:44
  • \$\begingroup\$ How about "inter-sample interval" or "inter-sample gap"? \$\endgroup\$ – David Aug 8 '16 at 6:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.