PicoScope 7 Automotive
Available for Windows, Mac, and Linux, the next evolution of our diagnostic scope software is now available.
Nvh for hydraulic flow
Nvh for hydraulic flow
Hello all. I have wanted to find a way to hear a slight flow of hydraulic oil to find cylinder drift in a valve. Has any one used it for this purpose? Also will it find noise in gear boxes? Thanks
Jason
Jason
-
Steve Smith
- Pico Staff Member
- Posts: 1587
- Joined: Sun Aug 25, 2013 7:22 am
Re: Nvh for hydraulic flow
Hi Jason, thank you for the post
You can use NVH to listen to airborne noises via the Mic or structural borne via the accelerometer
The Mic has the disadvantage of capturing all ambient noise and so your frequency of interest may be lost in the noise floor (ambient noise) if its amplitude is low.
The Mic does have the advantage however of being able to capture frequencies of interest up to 20 kHz (The limit of human hearing)
On the flip side, the accelerometer can be used as a contact mic (to listen to your hydraulic oil flow) with the advantage of no intrusion from ambient noise but with the disadvantage of being limited to approx. 2 kHz
Think of using the accelerometer as you do a stethoscope or placing a screwdriver to a transmission housing.
You only hear and capture the noise/activity at the tip of the stethoscope, screwdriver or in our NVH case, the accelerometer
To listen to your accelerometer you have 2 options:
1. At the set-up screen, configure your accelerometer as a mic (See below)
With the above configuration you will be able to pause your recording and listen to the audio playback within the software. Your amplitude will be displayed in "Decibel" (dB) rather than "milli g" (mg)
2. At the set up screen, configure your accelerometers as accelerometers
With the above configuration you will need to export your data in .wav format to listen to the audio captured by the accelerometers via a 3rd party media player. Your amplitude will be displayed "milli g" (mg)
I know Ben may well have additional information surrounding hydraulic diagnostic techniques when he returns. (This is likely to be next week)
Regarding gearbox noise etc. please could you take a look at this post which discuss Tooth Contact Frequency detection and calculation viewtopic.php?p=103777#p103777
Please feel free to post your captures here too.
I hope this helps, take care......Steve
You can use NVH to listen to airborne noises via the Mic or structural borne via the accelerometer
The Mic has the disadvantage of capturing all ambient noise and so your frequency of interest may be lost in the noise floor (ambient noise) if its amplitude is low.
The Mic does have the advantage however of being able to capture frequencies of interest up to 20 kHz (The limit of human hearing)
On the flip side, the accelerometer can be used as a contact mic (to listen to your hydraulic oil flow) with the advantage of no intrusion from ambient noise but with the disadvantage of being limited to approx. 2 kHz
Think of using the accelerometer as you do a stethoscope or placing a screwdriver to a transmission housing.
You only hear and capture the noise/activity at the tip of the stethoscope, screwdriver or in our NVH case, the accelerometer
To listen to your accelerometer you have 2 options:
1. At the set-up screen, configure your accelerometer as a mic (See below)
- Image 1
2. At the set up screen, configure your accelerometers as accelerometers
With the above configuration you will need to export your data in .wav format to listen to the audio captured by the accelerometers via a 3rd party media player. Your amplitude will be displayed "milli g" (mg)
- Image 2
Regarding gearbox noise etc. please could you take a look at this post which discuss Tooth Contact Frequency detection and calculation viewtopic.php?p=103777#p103777
Please feel free to post your captures here too.
I hope this helps, take care......Steve
Re: Nvh for hydraulic flow
Thanks Steve for the info. I’ll have to give it a try.
-
ben.martins
- Pico Staff Member
- Posts: 543
- Joined: Tue May 16, 2017 1:02 pm
Re: Nvh for hydraulic flow
Hi Jason,
Firstly thank you for posting as this is a topic I'm very keen to keep exploring in and its great to see others out there thinking in the same way.
Listening for flow across a cylinder is quite challanging though as the sound are usually at an ultrasonic level. We do have a parking sensor detector which can be used for detecting higher frequencies but I've yet to have a real issue to test it out.
As Steve mentioned, the accelerometers can be used for structural borne noise which can be very helpful especially when it comes to pumps. The problem with checking across a cylinder though is the structure the accelerometers are connected too is one piece so you'll be detecting a lot more than just the leak.
I'll follow up next week with some extra bits I've done with ways of listening to fluid borne noise and see if I have anything where the Ultrasonic detector was used.
Nothing is easy as Steve will say but if it was easy everyone would do it!
Kind regards
Ben
Firstly thank you for posting as this is a topic I'm very keen to keep exploring in and its great to see others out there thinking in the same way.
Listening for flow across a cylinder is quite challanging though as the sound are usually at an ultrasonic level. We do have a parking sensor detector which can be used for detecting higher frequencies but I've yet to have a real issue to test it out.
As Steve mentioned, the accelerometers can be used for structural borne noise which can be very helpful especially when it comes to pumps. The problem with checking across a cylinder though is the structure the accelerometers are connected too is one piece so you'll be detecting a lot more than just the leak.
I'll follow up next week with some extra bits I've done with ways of listening to fluid borne noise and see if I have anything where the Ultrasonic detector was used.
Nothing is easy as Steve will say but if it was easy everyone would do it!
Kind regards
Ben
Re: Nvh for hydraulic flow
Thanks for the info. What kind of cost is the ultrasonic sensor. That’s one thing I’m not understanding is about different frequencies. I have a steel man wireless head set and I can hear cylinders that are buy passing but no luck on the small leaks inside a valve as it has some static. Really hoping to come up with I good way of looking at leaking valves without opening the system. Thanks again for the info.
-
ben.martins
- Pico Staff Member
- Posts: 543
- Joined: Tue May 16, 2017 1:02 pm
Re: Nvh for hydraulic flow
Hi Jason,
The ultrasonic, Parking Sensor Detector, is inexpensive at just £24 Ex VAT. https://www.picoauto.com/products/test- ... r-detector
I feel like it might need some modifications to be honest like the inclusion of a probe tip as it doesn't really lend itself to this type of testing, nor have I been able to confirm it's ability to do the job!
However, all that aside I have held the Parking Sensor detector on a relief valve whilst a service was held on it's stop. From memory I think we were holding the bucket ram on an excavator. This caused the relief valve to open as expected and allow oil to flow back to tank.
I did this test in both the NVH software and the normal PicoScope software of which there are pros and cons to both. In NVH we playback, harmonic rulers, audible sound during playback, exporting to .WAV files etc as well as having the engine speed captured over the CAN network. It is limited to just 20kHz as this is the audible spectrum for us humans. Have to say, it's been a few years since I could hear anywhere near 20kHz!
In the following capture I have included 2 WPS600C on the output of each pump. As in any hydraulic system there is certain 'ripple' created by the fluid which is heightened when under pressure. This can be used to determine pump health and was something myself and Chris Groff, Science of Diagnostics, were looking at investigating further. In this machine the engine is connected directly to the pumps which means the input shaft to the pump is the same speed as the crankshaft. This was a variable displacement pump which had 9 pistons and so you could say that the pump would have a 9th order of engine speed. This can be seen quite clearly from WPS600 where 9 pulses in the ripple match the engine speed.
The green channel is actually the parking sensor detector where you can see that it is picking up the frequencies associated with the hydraulic fluid ripple and associated harmonics. Looking further up the spectrum though it becomes less obvious what is just noise and what is a frequency of interest.
The file can be seen here -
If we move to PicoScope 7 software we can start to explore this same setup but in time domain but also including a spectrum view in the analysis. If needed, I could have included the crank sensor or an optical pick up to include engine speed.
Channel A Blue - Pump 1
Channel B Red - Pump 2
Channel C Green - Parking sensor at relief valve
The red channel is the pump doing the work where we can see the pressure averaging around 200 bar. This generates a fairly even sine wave as seen in top right view port. Measuring a complete cycle we have a frequency of 295 Hz which is also seen in the spectrum view underneath. I have included a ruler to highlight 20kHz which is a our cut off in NVH. To the right of this you can see a lot more activity with a spike of 42.44kHz. Could this be our fluid bypassing across the relief valve? It's quite possible but the spectrum view is across the whole buffer so taking into account all the data present. Hindsight being a wonderful thing, I should have just taken a buffer where no action on the service was being taken.
There's a lot of unknowns here and it's something I would like to continue looking into. Non destructive testing, NDT, of hydraulic components is especially interesting as to open up a system is often difficult and can introduce contamination. Just thinking back to the Manitou case study where this type of testing could have saved a great deal of time! https://www.picoauto.com/library/case-s ... nch-part-2
Lots to think about and please let me know your thoughts. It's a great topic and one I think could be really helpful in diagnosis. If there is a way we can set the scope up to take this type of measurement easily then I'm here to help and look through any captures you may make.
Kind regards
Ben
The ultrasonic, Parking Sensor Detector, is inexpensive at just £24 Ex VAT. https://www.picoauto.com/products/test- ... r-detector
I feel like it might need some modifications to be honest like the inclusion of a probe tip as it doesn't really lend itself to this type of testing, nor have I been able to confirm it's ability to do the job!
However, all that aside I have held the Parking Sensor detector on a relief valve whilst a service was held on it's stop. From memory I think we were holding the bucket ram on an excavator. This caused the relief valve to open as expected and allow oil to flow back to tank.
I did this test in both the NVH software and the normal PicoScope software of which there are pros and cons to both. In NVH we playback, harmonic rulers, audible sound during playback, exporting to .WAV files etc as well as having the engine speed captured over the CAN network. It is limited to just 20kHz as this is the audible spectrum for us humans. Have to say, it's been a few years since I could hear anywhere near 20kHz!
In the following capture I have included 2 WPS600C on the output of each pump. As in any hydraulic system there is certain 'ripple' created by the fluid which is heightened when under pressure. This can be used to determine pump health and was something myself and Chris Groff, Science of Diagnostics, were looking at investigating further. In this machine the engine is connected directly to the pumps which means the input shaft to the pump is the same speed as the crankshaft. This was a variable displacement pump which had 9 pistons and so you could say that the pump would have a 9th order of engine speed. This can be seen quite clearly from WPS600 where 9 pulses in the ripple match the engine speed.
Channel B Red - Pump 2
Channel C Green - Parking sensor at relief valve
The red channel is the pump doing the work where we can see the pressure averaging around 200 bar. This generates a fairly even sine wave as seen in top right view port. Measuring a complete cycle we have a frequency of 295 Hz which is also seen in the spectrum view underneath. I have included a ruler to highlight 20kHz which is a our cut off in NVH. To the right of this you can see a lot more activity with a spike of 42.44kHz. Could this be our fluid bypassing across the relief valve? It's quite possible but the spectrum view is across the whole buffer so taking into account all the data present. Hindsight being a wonderful thing, I should have just taken a buffer where no action on the service was being taken.
There's a lot of unknowns here and it's something I would like to continue looking into. Non destructive testing, NDT, of hydraulic components is especially interesting as to open up a system is often difficult and can introduce contamination. Just thinking back to the Manitou case study where this type of testing could have saved a great deal of time! https://www.picoauto.com/library/case-s ... nch-part-2
Lots to think about and please let me know your thoughts. It's a great topic and one I think could be really helpful in diagnosis. If there is a way we can set the scope up to take this type of measurement easily then I'm here to help and look through any captures you may make.
Kind regards
Ben
Re: Nvh for hydraulic flow
I remember seeking Chris’s you tube video about that. Was not sure if this Somthing he was still working on.I’m real new to the scope being a heavy equipment guy it’s just something no one uses. Seams like tons of things it can be used for besides just looking at electrical. I’m wondering if there’s would be a good resource out there. I have watched some YouTube videos about vibration testing for gear boxes. I’m guessing this would be similar but I’m far from an expert.