The "Noise" in NVH

Ask questions relating specifically to the PicoDiagnostics NVH equipment and software here.

The "Noise" in NVH

Postby Mark Dalton » Fri May 22, 2015 1:28 am

Has anyone had success in diagnosing noises using the NVH Kit, or at least had a go at trying to diagnose a noise that is not closely tied to a vibration? Maybe the sort of noise you would normally use chassis ears for. Have you used the microphone? What kind of noise, what sort of signal did you get from it and what information could you interpret from it.
User avatar
Mark Dalton
Advanced User
Advanced User
 
Posts: 51
Joined: Mon Dec 01, 2014 9:35 pm
Location: Melbourne

Re: The "Noise" in NVH

Postby Steve Smith » Fri Jun 19, 2015 4:59 pm

Hello Mark and thank you for your post. Sorry it has took so long to reply, I have seen your post here and been gagging to reply but found myself side tracked.

NVH using the microphone moves us into the realms of audio frequency analysis (above 200 Hz) and of course all the challenges that come with it.

I have posted an NVH file below from a Toyota Yaris with an whining alternator. If you playback the NVH file you will hear the whine via your PC speakers.

YARIS ALT NOISE.pddata
(3.63 MiB) Downloaded 174 times



The first step to capturing audio frequencies is to ensure we have a large enough frequency spectrum to view the amplitude of the sound ( in this case a high frequency whine)

When looking at vibration our spectrum is around 0-200 Hz ( low frequency) which we experience as vibrations we can feel and so detected using the accelerometer.

Because we have a whining sound ( which we experience as a sound we hear) we switch to the microphone for detection and increase our spectrum to cover up to 20,000 Hz. Above this frequency humans cannot hear as we move into Dolphin territory and ultrasound.

Coming back to our whining alternator, looking at the frequency graph we can see the fundamental frequency identified by the letter "A". The is derived from the pulley dimensions entered into the NVH software during the set-up process.

FREQUENCY CHART.jpg


Crank pulley diameter 164 mm driving an alternator pulley with a diameter of 55 mm at an engine speed of 3514 rpm.

164 / 55 = 2.98 (Crank pulley to alternator pulley drive ratio)
1 x crank pulley rotation equals 2.98 alternator rotations.

Engine speed at 3514 rpm / 60 = 58.56 Hz

Crankshaft frequency is therefore 58.56 Hz at 3514 rpm when the alternator whine is present.

Alternator fundamental frequency with a crankshaft frequency/speed of 58.56 Hz is therefore:
58.56 Hz x 2.98 = 174.5 Hz Identified in the frequency graph by our letter "A"

However our whine is a lot higher than 174.5 Hz otherwise we would feel a vibration rather than hear a whine. Looking at the frequency graph we can see a high peak detected by the microphone around 1250 Hz, how could this be attributed to the alternator?

1250 Hz is certainly audible (94dB) and happens to be visible when the alternator is whining at high speed. Let's look at the frequency of the alternator AC output (before rectification)

Alternator output frequency = Number of alternator poles x RPM of alternator / 120

Our alternator has 14 poles: 14 poles x (174.5 x 60)= 14 x 10,470 = 146,580 / 120 = 1221.5 Hz

Therefore our whine can be linked to our alternator as a result of generator noise from an internal electrical error during generation, not a mechanical failure.

FREQUENCY MARKERS.jpg


The microphone has enabled the capture of the relevant frequency (whine) whilst the operating principle has enabled the application of the theory to confirm the component failure.

When looking for intermittent or sporadic noises, the typical "Chassis Ear" is still a viable option as we rely upon the noise level being present for a period of time before the NVH software can identify and display the frequency/amplitude.

Body creaks/ speaks etc. are evident for minuscule amounts of time, and whilst they can be detected by the sensitive NVH microphone, they would be lost to background noise and delayed transfer to the frequency graph.

In these scenarios we require multiple microphones placed about the suspect area of creak/ squeak with a rapid live update speed to in order to display the amplitudes on screen.

Here the NVH kit is not applicable but watch this space!

A big thank you to Alex (NVH Guru) for his input here at Pico as I was lost looking at this one

I hope this is of some help, take care......Steve
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Re: The "Noise" in NVH

Postby Mark Dalton » Wed Jul 01, 2015 10:21 am

Thanks again for your help with my questions. Would be great to see more examples like this in YouTube videos or in the support section.
User avatar
Mark Dalton
Advanced User
Advanced User
 
Posts: 51
Joined: Mon Dec 01, 2014 9:35 pm
Location: Melbourne

Re: The "Noise" in NVH

Postby Belet » Thu Jul 16, 2015 4:01 am

steve smith wrote:Hello Mark and thank you for your post. Sorry it has took so long to reply, I have seen your post here and been gagging to reply but found myself side tracked.

NVH using the microphone moves us into the realms of audio frequency analysis (above 200 Hz) and of course all the challenges that come with it.

I have posted an NVH file below from a Toyota Yaris with an whining alternator. If you playback the NVH file you will hear the whine via your PC speakers.

YARIS ALT NOISE.pddata


The first step to capturing audio frequencies is to ensure we have a large enough frequency spectrum to view the amplitude of the sound ( in this case a high frequency whine)

When looking at vibration our spectrum is around 0-200 Hz ( low frequency) which we experience as vibrations we can feel and so detected using the accelerometer.

Because we have a whining sound ( which we experience as a sound we hear) we switch to the microphone for detection and increase our spectrum to cover up to 20,000 Hz. Above this frequency humans cannot hear as we move into Dolphin territory and ultrasound.

Coming back to our whining alternator, looking at the frequency graph we can see the fundamental frequency identified by the letter "A". The is derived from the pulley dimensions entered into the NVH software during the set-up process.

FREQUENCY CHART.jpg


Crank pulley diameter 164 mm driving an alternator pulley with a diameter of 55 mm at an engine speed of 3514 rpm.

164 / 55 = 2.98 (Crank pulley to alternator pulley drive ratio)
1 x crank pulley rotation equals 2.98 alternator rotations.

Engine speed at 3514 rpm / 60 = 58.56 Hz

Crankshaft frequency is therefore 58.56 Hz at 3514 rpm when the alternator whine is present.

Alternator fundamental frequency with a crankshaft frequency/speed of 58.56 Hz is therefore:
58.56 Hz x 2.98 = 174.5 Hz Identified in the frequency graph by our letter "A"

However our whine is a lot higher than 174.5 Hz otherwise we would feel a vibration rather than hear a whine. Looking at the frequency graph we can see a high peak detected by the microphone around 1250 Hz, how could this be attributed to the alternator?

1250 Hz is certainly audible (94dB) and happens to be visible when the alternator is whining at high speed. Let's look at the frequency of the alternator AC output (before rectification)

Alternator output frequency = Number of alternator poles x RPM of alternator / 120

Our alternator has 14 poles: 14 poles x (174.5 x 60)= 14 x 10,470 = 146,580 / 120 = 1221.5 Hz

Therefore our whine can be linked to our alternator as a result of generator noise from an internal electrical error during generation, not a mechanical failure.

FREQUENCY MARKERS.jpg


The microphone has enabled the capture of the relevant frequency (whine) whilst the operating principle has enabled the application of the theory to confirm the component failure.

When looking for intermittent or sporadic noises, the typical "Chassis Ear" is still a viable option as we rely upon the noise level being present for a period of time before the NVH software can identify and display the frequency/amplitude.

Body creaks/ speaks etc. are evident for minuscule amounts of time, and whilst they can be detected by the sensitive NVH microphone, they would be lost to background noise and delayed transfer to the frequency graph.

In these scenarios we require multiple microphones placed about the suspect area of creak/ squeak with a rapid live update speed to in order to display the amplitudes on screen.

Here the NVH kit is not applicable but watch this space!

A big thank you to Alex (NVH Guru) for his input here at Pico as I was lost looking at this one

I hope this is of some help, take care......Steve


Steve, the attached NVH file was obsolete, can you re-upload a new one?

And can the NVH kit deal with the noise presented in auto audio? For example, the auto audio of a car caused the noise like a mobilphone beside it with the electromagnetic interference when turning up the volume to a specific value. Can the NVH kit find out the source of the noise and separate it out?

Belet
User avatar
Belet
User
User
 
Posts: 28
Joined: Mon Jun 01, 2015 1:29 am

Re: The "Noise" in NVH

Postby Steve Smith » Fri Jul 17, 2015 11:40 am

Hello Belet and thank you for the feedback

It would appear there is an issue with my post as none of the images or the pddata file will down load. I have reposted them above.

Moving onto your “noise enquiry” Can I clarify the symptom?

Auto audio, do you mean auto sound levelling (ASL)? (Increase in road speed, increase in volume)

With ASL functioning are you experiencing this interference/noise?

Can you confirm there are no mobile phones present?

The NVH microphone is ideal for identifying the frequencies of “noise” assuming they are present for a specific time period 3 to 4 seconds.

Positioning the microphone inside the cabin whilst monitoring the audio will no doubt “pick out” a variety of frequencies which could in itself, present you with a challenging frequency view given the multiple frequencies present during audio playback.

Should your interference be present for the specified time and you have selected a spectrum wide enough to capture the event, then the NVH kit will display the frequency of your noise/interference

You could also use the NVH microphone with PicoScope in Spectrum mode to capture the noise/interference frequency of interest.

I hope this helps, take care……Steve
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Re: The "Noise" in NVH

Postby Mark Dalton » Wed Aug 26, 2015 12:04 am

Hi Steve,
Just a quick question. In you initial response you gave a formula "Alternator output frequency = Number of alternator poles x RPM of alternator / 120". I'm just wondering what the "120" is, how is that number derived?
Regards,
Mark.
User avatar
Mark Dalton
Advanced User
Advanced User
 
Posts: 51
Joined: Mon Dec 01, 2014 9:35 pm
Location: Melbourne

Re: The "Noise" in NVH

Postby eric989 » Wed Aug 26, 2015 6:55 am

I am looking at this.
The 120 shoule be 2*60=120, I guess.

Frequency= (poles/2)*(RPM/60)=poles*RPM/120.

In these scenarios we require multiple microphones placed about the suspect area of creak/ squeak with a rapid live update speed to in order to display the amplitudes on screen.
What do you mean, Steve? We only have one microphone within NVH kit. Sorry, bad English :cry:

Thanks,
Eric
eric989
User
User
 
Posts: 31
Joined: Fri Apr 16, 2010 6:53 am
Location: Hongke China

Re: The "Noise" in NVH

Postby Steve Smith » Tue Sep 01, 2015 3:43 pm

Hello and thank you all for taking the time to read and digest this post.

The 120 value refers to the windings of the three phases of the stator.

The alternator stator has three internal windings wound 120 degrees between each phase.

The alternator output frequency formula is based upon a single phase winding.

Moving onto Multiple Microphones, I was referring to a possible future upgrade to the NVH software along the lines of the Chassis Ear Tool.

You are correct Eric, at present we only have 1 x accelerometer or 1 x microphone to be used individually at any one time.

This may well change in the future (I have said too much already)

No need to apologise Eric your English is fine.

I hope this helps, take care.......Steve
Steve Smith
Pico Staff Member
Pico Staff Member
 
Posts: 426
Joined: Sun Aug 25, 2013 7:22 am

Re: The "Noise" in NVH

Postby Mark Dalton » Wed Sep 02, 2015 12:56 am

Thanks Steve,
That makes much more sense now. It's so cool that it can all be tied together so well.
User avatar
Mark Dalton
Advanced User
Advanced User
 
Posts: 51
Joined: Mon Dec 01, 2014 9:35 pm
Location: Melbourne

Re: The "Noise" in NVH

Postby Mark Dalton » Mon Sep 07, 2015 2:19 am

For anyone wondering about Multiple Microphones, I have experimented with a Chassis Ear/Oscilloscope Hybrid set up. I use'd the Oscilloscope rather than the NVH PlugIn. But I was chasing a squeak in the rear end of a car that was difficult to reproduce in the workshop but was relatively easy to reproduce out on the road. So I rigged up some Piezo Transducers glued to some rare earth magnets so I could attach them to the car body and suspension components. If you look at the attachment you can see how I rigged them up.
Using the Oscilloscope on single trigger I could position the trigger point so that the trace only ran when the squeak produced a voltage signal. I could then change which transducer I was watching using the selector position switch from our Chassis Ear Kit. I then went back through the Signal buffer to see which signal produced the greatest amplitude, then focus on where that Transducer was located on the Car.
Attachments
Chassis Ear.pdf
(94.28 KiB) Downloaded 101 times
User avatar
Mark Dalton
Advanced User
Advanced User
 
Posts: 51
Joined: Mon Dec 01, 2014 9:35 pm
Location: Melbourne

Next

Return to PicoDiagnostics NVH

Who is online

Users browsing this forum: No registered users and 0 guests