11-23-2021 11:12 AM
I am new to this and might be asking multiple questions here.
I have a voltage from a pressure transducer that needs to be filtered due to noise and I would like to know if the filter VI in labview is enough or would i get a PXIe card? I also will need signal conditioning in my circuit so I was thinking a card with filtering and conditioning would be best, unless that can be handled in the labview software.
My thinking is to get one of the PXIe Analog input module (PXIe-4302, 4303, 4304, 4305, 4309, 4481) to do the signal processing I need then send it to the controller to monitor and display.
Solved! Go to Solution.
11-23-2021 11:21 AM
Software-based signal filtering is possible up to an extent and it depends on the use case.
For example, software-based equalizers work for audio as long as the actual sound has sufficient resolution and sampling rate.
It is best to use analog filters on analog signals than implement digital filters for those sampled signals.
Could you please elaborate on the type of signal conditioning and filtering requirements?
11-23-2021 11:45 AM
I think your answer helped but the exact requirements I am not sure of. I am new to this project. From my understanding they do not want change the sampling rate that they have (which I still don't understand why) and want to filter out noise from a pump that is part of the system but also filter it down to avoid aliasing. I believe this will be a low pass filter, but the cutoff i do not know.
The signal conditioning requirements I am lost on. I assume we will filter the signal then condition it to a readable signal.
I think the limiting factor here is the sampling rate locked in. Sorry for lack of information.
11-23-2021 12:22 PM - edited 11-23-2021 12:28 PM
@Guy4196 wrote:
I am new to this and might be asking multiple questions here.
I have a voltage from a pressure transducer that needs to be filtered due to noise and I would like to know if the filter VI in labview is enough or would i get a PXIe card? I also will need signal conditioning in my circuit so I was thinking a card with filtering and conditioning would be best, unless that can be handled in the labview software.
My thinking is to get one of the PXIe Analog input module (PXIe-4302, 4303, 4304, 4305, 4309, 4481) to do the signal processing I need then send it to the controller to monitor and display.
What is the source of the noise on the voltage from the pressure transducer? It is generally better to try to eliminate the noise from ever entering the signal than to filter it out later.
Edited to add: It looks like our replies crossed. I see that the noise source is a pump. I would still try to avoid as much noise a possible from getting into the system. How do you have the wiring from the PT to the DAQ? Are you using twisted shielded wire? If so do you have the shield grounded properly? Have you routed the wire to try to avoid the pump?
11-23-2021 12:35 PM
@johntrich1971 wrote:
@Guy4196 wrote:
I am new to this and might be asking multiple questions here.
I have a voltage from a pressure transducer that needs to be filtered due to noise and I would like to know if the filter VI in labview is enough or would i get a PXIe card? I also will need signal conditioning in my circuit so I was thinking a card with filtering and conditioning would be best, unless that can be handled in the labview software.
My thinking is to get one of the PXIe Analog input module (PXIe-4302, 4303, 4304, 4305, 4309, 4481) to do the signal processing I need then send it to the controller to monitor and display.
What is the source of the noise on the voltage from the pressure transducer? It is generally better to try to eliminate the noise from ever entering the signal than to filter it out later.
Edited to add: It looks like our replies crossed. I see that the noise source is a pump. I would still try to avoid as much noise a possible from getting into the system. How do you have the wiring from the PT to the DAQ? Are you using twisted shielded wire? If so do you have the shield grounded properly? Have you routed the wire to try to avoid the pump?
Just to elaborate a tiny bit. The reason you want to apply filters as much as practical before it gets into the measurement system is that the unfiltered signal might actually be able to damage your instrument! Doesn't do any good to use a software filter to filter a voltage spike if that voltage spike blows up your detector before it ever gets to the software.
11-23-2021 12:46 PM
That's a good point, didn't think of that.
the wire appears to be twisted shielded pair which leaves the "electronics room" gets wired up into the concrete wall and into the testing room so it's hard to determine what wires are used. It's an old system that is looking to get updated. Changing wire path's most likely isn't an option as it seems to be a set path.
I don't think noise is enough to damage the equipment, it sounds like its a hum from the way it is described. But it is enough to disrupt the signal.
11-23-2021 01:38 PM
The cards that you are considering are 24bit, which means they have digital Anti-aliasing filter built in. (I believe all of NIs 24 bit cards have this capability.)
The pump will inject noise/background/pickup/clutter at a frequency along with harmonics of that frequency. You would need a bandstop filter to try to remove the primary pump frequency. The higher harmonics of that frequency may still pass the bandstop filter.