Acoustic Emission DAQ need

[Jason93]

Simple question. Someone must do this. I want to get an Acoistic Emission sensor. I don’t want the various proprietary company daq hw and sw. I want to use NI hw and labview. I’d like a usb solution not pxi. Anyone have experience on this? NI-9223? Just FYI this is not microphone measurements. Typically 1mb/s. I know I could just browse for something but I’d like to get user AE input.

also signal express ok?

thanks

Jason

[ClintT]

Hi Jason, 

There are a lot of specifics that would help us give you a recommendation here. I'm not going to list them all, but here are a few!

What is the application? 

What are the sampling specifics that you need?

What's your budget?

What sensor are you getting? Do you have a datasheet to reference. 

We like to make sure we have a full picture before recommending hardware. This inquiry may be better addressed on a call to our sales engineers who are trained to develop whole systems. (877) 387-0015

Best, 

[Jason93]

Clint thanks for the reply.

The main idea is that I don't have an exact application. I'm looking for something fairly general purpose. I'm looking at wideband AE sensor probably from Kistler or Physical Acoustics and probably in the 100k-800kHz range.

The main application will probably be maching process monitoring. Mills, CNCs, etc for tool wear. But could expand to rubbers & plastics.  Budget is as reasonable as possible. hopefully less than $4k. But that will include the sensor, conditioner, and DAQ HW. All my NI-DAQ is less than 100kS/s so I need to get something faster..1MHz probably. hence my question. Plus ask if anyone uses AE sensors and NI DAQ.?  Places like Physical Acoustics want to sell you their own software and DAQ equipment and I want to use NI.

Jason

[ClintT]

Jason, 

I'm glad to hear you want to stick with NI products! As I mentioned before, calling one of our sales engineers would be the most valuable thing to do for getting some ideas of what solutions we could offer. That number is in my first post.

As another aside, I would typically recommend a PXI solution for this kind of application due to the amount of bandwidth and sampling frequency you are looking for. 

To somewhat answer your question, the 9223 would be a good solution. I can also see the NI 9775 being a valid option. Now depending on your application and how you want to transfer/process the data, there could be specific chassis recommendations to keep in mind.

Cheers!  

[Jason93]

Thanks Clint.

Going down the road of PXI might be daunting. I have a few cDAQ chassis around that I like very much plus a few individual chassis for "1 of"  testing. The two modules you suggest look good. The one is 20MS/s! that is a lot. The other one is cheaper at 1MBs/s. Seems like spending the extra $s for the 20MS would be worth it. The sensor I was looking at was up to 900kHz so I would normally be looking to sample at 2MB minimum.  I will look into a local rep.

Still would love to hear some NI stories using AE sensors and NIDAQ hardware and the success they had.

[ClintT]

Jason, 

Glad to help. If you will be sampling that high I would surely recommend using the 9775. There is no way to verify the integrity of your data if you are sampling below Nyquist frequency! 

This doesn't have a ton of info, but here is a link I found on the forums regarding AE sensors. Out of curiosity, what is the need to sample this quickly? Is that the frequency of interest for this application?

Cheers

[Jason93]

Honestly I don't know why to sample so fast. Again this is just exploratory work? I do condition monitoring mainly with vibration and sound (like everyone else) I came across AE sensors and thought it would be a good tool "in the bag" Most AE sensors are narrowband (i.e. focused on a frequency) or wideband that are often 100kHz - 900khz or so.  I figured wideband would be best to start with since it would give me some flexibility to change the application, since at the moment I don't have an application. So if I bought an AE sensor up to 900kHz bandwidth I'd need to sample at 2kHz to use it properly. (that is a ton of data collected) I want to focus on machining processes and it seems like a good sensor. So I wanted to see the cost point. The sensor, the corresponding HW, the DAQ and the software. I have a research licence for NI and have a bunch of cDAQ modules that I've had some good success with (other than the time I was trying to sync data from a normal AI module and a accelerator module...not as easy as they make it sound to be)

 But I''m comfortable with the software and the sensing is going to run be about $2k with the NI module about $2k. Not terrible, but on a prof budget that is not on a major funded grant that is still sizable, but could lead to a good future investment.

Jason

[Henrik_Volkers]

Hi Jason,

I follow this threat because I work with accelerometers (and Laser Vibrometers as primary standards 😉 ) ..

I 'only' look at frequencies up to 130kHz but I sample with 10MSPS (PXI 5922, nice card.. heavy price) .. in sync with a 200MSPS for the 40MHz LDV raw signal.

I would go with the 20MSPS unit... just because I assume you measure a lot of noise and you will love every extra sample for signal processing 😄   and you always can lower the samplerate or decimate... in high res mode it's 5MSPS ....

Since the 9775 has only one range you will need a flexible Amplifier with 1MHz bandwidth ... (or DIY ...modern video amps (CFB OPs)) .. the transfer functions of the AE sensors datasheets I peeked are so ... mh  mummocky?    so even if the amp isn't flat .. it doesn't seem to harm.

[Jason93]

Thanks Henrik.

I was really on the fence with the 9775 vs the 9223. I knew the 9223 could not use the AE sensor full range but it seemed like a more common module in the NI universe. I could find very little on the 9775 and it applications in the field. I'm sure there are some but I was hoping for more success stories. (not that the 9223 had a ton) However hearing from someone that takes high sampling rate measurements is a major plus. The cost increase for an academic is still noteable at about 33% more but I think it is worth it.

As far as amplification both companies I've contacted Kistler and Physical Acoustics who I believe are leaders in the AE field have amplifier/signal conditioning units with it. Frankly they cost more than the sensor. Kistler was around $700+$750 and PA was less but still was $300 + $500 or so.

thanks for the note.

Jason

[Henrik_Volkers]

josepaul, this thread is about acoustic emission in the 100 kHz and higher range..

But since your both post only include a link to another (not relating)  website ... spam?