PicoScope 7 Software
Available on Windows, Mac and Linux
Is A PicoLog 1012 Suitable -2 ?
-
BilloBillo
- User
- Posts: 4
- Joined: Fri Dec 11, 2020 3:13 am
Is A PicoLog 1012 Suitable -2 ?
I have found the answer to my question in my "Is A PicoLog 1012 Suitable?".
The programmers-guide.pdf for pl1000SetInterval() says "Sampling of multiple channels is sequential." therefore each channel does not have an A-D converter and so for BM_SINGLE mode with no_of_channels = 2 the channels will be logged as 1, 2, 1, 2 with 1 uS between each data point.
Does PicoLog correct the 1 uS offset between channels 1 and 2, or must I save to a file and subtract 1 uS from the channel 2 values?
The programmers-guide.pdf for pl1000SetInterval() says "Sampling of multiple channels is sequential." therefore each channel does not have an A-D converter and so for BM_SINGLE mode with no_of_channels = 2 the channels will be logged as 1, 2, 1, 2 with 1 uS between each data point.
Does PicoLog correct the 1 uS offset between channels 1 and 2, or must I save to a file and subtract 1 uS from the channel 2 values?
Re: Is A PicoLog 1012 Suitable -2 ?
Hi BilloBillo,
I answered your query on your original post.
You are correct about the sequential sampling and the 1us delay between channels. However, I didn't mention the sequential sampling, because I didn't think that a phase displacement of 1us would be a problem for you, as you said
" Channel 2 has a 100 kHz sine wave.
I want to see the phase of it at the position of the notch in the sync pulse."
which would hardly be affected by a 1us delay (i.e. would only marginally change whether or not it would be a negative going crossover, for a 5us interval, assuming that by negative going crossover you mean negative going section of the sine wave). But if that is an issue then it makes the decision of what to use even more watertight.
More importantly, I didn't think that a PicoLog 1012 would be the best choice for your application anyway (as you will see if you read my response).
Regards,
Gerry
I answered your query on your original post.
You are correct about the sequential sampling and the 1us delay between channels. However, I didn't mention the sequential sampling, because I didn't think that a phase displacement of 1us would be a problem for you, as you said
" Channel 2 has a 100 kHz sine wave.
I want to see the phase of it at the position of the notch in the sync pulse."
which would hardly be affected by a 1us delay (i.e. would only marginally change whether or not it would be a negative going crossover, for a 5us interval, assuming that by negative going crossover you mean negative going section of the sine wave). But if that is an issue then it makes the decision of what to use even more watertight.
More importantly, I didn't think that a PicoLog 1012 would be the best choice for your application anyway (as you will see if you read my response).
Regards,
Gerry
Gerry
Technical Specialist
Technical Specialist
-
BilloBillo
- User
- Posts: 4
- Joined: Fri Dec 11, 2020 3:13 am
Re: Is A PicoLog 1012 Suitable -2 ?
Hi Gerry,
Thank you very much for your detailed answers.
The question was for an initial acceptance test by my customer.
I will reply to both my topics in each one so a forum member who only reads one will see the complete situation.
I considered using the 1012 logger because of the number of channels.
Thank you for suggesting the PicoScope 2204A, unfortunately 10 channels are eventually required but I will inform the user that only 8 are possible and I will quote for using a PicoScope 4824.
I will also quote for designing and building a 10 channel logger with A-D and buffers in each channel and all channels driven by a common clock.
The huge cost mainly due to including precision channel amplifiers, etc will make the PicoScope 4824 seem inexpensive.
My reason for wanting to align the channel data is that the user is smart enough to see the misalignment (in a table or charts) and say something like "while one channel is being sampled the state of the other channels are unknown" and will reject that data and the whole logging system.
By cheating and making the data appear to be aligned the data will be accepted and the tiny notional errors will still be there but will not be anywhere near significant.
I know what is needed and it is nothing like what has been specified.
I suppose you have experience with people dreaming up specifications without understanding what they mean and what is required to achieve them.
Thanks again for your help.
Thank you very much for your detailed answers.
The question was for an initial acceptance test by my customer.
I will reply to both my topics in each one so a forum member who only reads one will see the complete situation.
I considered using the 1012 logger because of the number of channels.
Thank you for suggesting the PicoScope 2204A, unfortunately 10 channels are eventually required but I will inform the user that only 8 are possible and I will quote for using a PicoScope 4824.
I will also quote for designing and building a 10 channel logger with A-D and buffers in each channel and all channels driven by a common clock.
The huge cost mainly due to including precision channel amplifiers, etc will make the PicoScope 4824 seem inexpensive.
My reason for wanting to align the channel data is that the user is smart enough to see the misalignment (in a table or charts) and say something like "while one channel is being sampled the state of the other channels are unknown" and will reject that data and the whole logging system.
By cheating and making the data appear to be aligned the data will be accepted and the tiny notional errors will still be there but will not be anywhere near significant.
I know what is needed and it is nothing like what has been specified.
I suppose you have experience with people dreaming up specifications without understanding what they mean and what is required to achieve them.
Thanks again for your help.
Re: Is A PicoLog 1012 Suitable -2 ?
Hi BilloBillo,
Thanks for your explanation. and I'm glad you have what you need to respond to your customer.
Just as an aside, there is never any need to 'cheat' the customer. For data logging, the data needs to be aligned because the purpose of collecting it is for making comparisons. If relative channel timing is important, the misalignment (1us of clock skew between the channels) can be legitimately explained as the timing error between channels. The significance of this, and how it will affect the measurement application can then be debated.
In data logging applications (which is what the PicoLog 1000 series devices were designed for) it is typically the overall sample rate that it is more important. So, for instance, a customer will be typically less concerned that when sampling on all 12 channels of the PicoLog 1012, channel 12 will be skewed 12us from channel 1, rather than the fact that the minimum effective sample interval on any single channel will be 16us (which would be more important in your customers case, as he wouldn't have the 1us effective sample rate per channel that you were referring to originally, if he's using all of the inputs).
Yes, we do get some 'interesting' requirements, however, that's what this Forum is for, as not everyone has the benefit of the knowledge that you've acquired
.
Regards,
Gerry
Thanks for your explanation. and I'm glad you have what you need to respond to your customer.
Just as an aside, there is never any need to 'cheat' the customer. For data logging, the data needs to be aligned because the purpose of collecting it is for making comparisons. If relative channel timing is important, the misalignment (1us of clock skew between the channels) can be legitimately explained as the timing error between channels. The significance of this, and how it will affect the measurement application can then be debated.
In data logging applications (which is what the PicoLog 1000 series devices were designed for) it is typically the overall sample rate that it is more important. So, for instance, a customer will be typically less concerned that when sampling on all 12 channels of the PicoLog 1012, channel 12 will be skewed 12us from channel 1, rather than the fact that the minimum effective sample interval on any single channel will be 16us (which would be more important in your customers case, as he wouldn't have the 1us effective sample rate per channel that you were referring to originally, if he's using all of the inputs).
Yes, we do get some 'interesting' requirements, however, that's what this Forum is for, as not everyone has the benefit of the knowledge that you've acquired
.Regards,
Gerry
Gerry
Technical Specialist
Technical Specialist