Wheel bearing failure

Ask questions relating specifically to the PicoDiagnostics NVH equipment and software here.
Martinjensen
TwoWaves
TwoWaves
Posts: 55
Joined: Wed Feb 13, 2013 6:22 pm
Location: Denmark

Wheel bearing failure

Post by Martinjensen »

Hello all.

I have a Corolla Verso i the workshop, with a failed front wheel bearing making exessive noise.
The customer is on Holiday, so the car is left with me for a week.
I thought it would be a good opportunity to practice with my NVH kit.
So i attached 2 sensors to the left and right strut tower ind the engine compartment, trying to measure if its the left or right bearing that failed.
After that i moved the sensors to the lover left and right control arm.
So here is the problem:
When i measure on the strut towers, the noise leavel is higest on the right tower
But when i measure on the controlarm the noise level is higest in the left controlarm.

Is there a better way to try to narrow down the offending bearing using the NVH kit.

This is not a question of fixing the car, but trying to learn more about my new Pico tool

Note:
Blue channel is left, and yellow channel is right.
Channel markers is placed when steeringwheel is turned right, putting weight on the front left Wheel.

Regards Martin.
Attachments
Lower control arms, Z axis.pddata
(615.66 KiB) Downloaded 777 times
Strut tower Z axis.pddata
(541.23 KiB) Downloaded 723 times

victor2k
TwoWaves
TwoWaves
Posts: 337
Joined: Thu Jan 28, 2016 7:17 am

Re: Wheel bearing failure

Post by victor2k »

Hello,
Your NHV sensors will read until at 200 Hz.You have some peaks recorded at 38Hz and 169 Hz...
What is the noise frequency measured with the microphone?
You told us about a noise,not a vibration...
Regards

Martinjensen
TwoWaves
TwoWaves
Posts: 55
Joined: Wed Feb 13, 2013 6:22 pm
Location: Denmark

Re: Wheel bearing failure

Post by Martinjensen »

Hello Victor.

The reason for using the accelerometers is that i know its a Wheel bearing failure, and i was hoping that when using 2 sensors i would see a greater vibration amplitude from the offending Wheel.

Plus I can feel the bearing failure from the driver seat.

I have made a capture using the microphone, its the first time using the microphone, so i could bee completley wrong, the mic is suspendet from the rear wiew miror.

I am signet up for NVH training later this year, i think its needed :D
Attachments
Microphone Bad wheel bearing.pddata
(2.28 MiB) Downloaded 680 times

victor2k
TwoWaves
TwoWaves
Posts: 337
Joined: Thu Jan 28, 2016 7:17 am

Re: Wheel bearing failure

Post by victor2k »

Hello,
Then you can attach at the driver seat the sensor to record the vibration...
Also use the microphone in the same place.
Regards

User avatar
Mark Dalton
TwoWaves
TwoWaves
Posts: 106
Joined: Mon Dec 01, 2014 9:35 pm
Location: Melbourne

Re: Wheel bearing failure

Post by Mark Dalton »

I think your praticing on the wrong thing here, it's difficult to get any practice time with tool, but I think this is the wrong kind of vibration / noise for the NVH tool. Because...
-You could probably diagnose it much faster and just as effectively with a stethiscope and the NVH tool is the tool you turn to when that sort of diagnosis won't work.
-A wheel bearing is unlikely to have an issue where the vibration it causes is stuck at the same frquency, or an order, of the frquency of the drive shaft, so the NVH tool won't be able to tell you where the vibration is coming from. I think the fact that you can hear the noise tells us that the vibration is a much higher frquency than the drive shaft rotation.
-A relative test would seem the best option, ie testing side v's side, but then you have to take into account the transfer path, responder etc. and is that why you get different results from different test points. You just end up confused about what's going on.
-The microphone in the car, seems to me to be no good at all because the mic can't determine direction, only frquency and amplitude. For the same reasons as above the NVH software probably won't be able to tie the wheel bearing vibration to a known frquency.

I think a better way to get to grips with the tool, since you have this car for a week, would be to clamp a 40 gram wheel weight to drive shaft or stick a piece of wood or a socket under a engine mount. You'll clearly see the frquency tied to a known frequency, so you'll know, ok it's a P1 or E1, but you still have to work out LH or RH, front mount, rear mount.
Hope that makes some kind sense. You'll get a lot out of the training.

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: Wheel bearing failure

Post by Steve Smith »

Many thanks for the posts and especially raising the wheel bearing issue.

We are working with a wheel bearing failure at the moment and creating a test plan with regards to defining the offending bearing using NVH.

First of all "It’s challenging". Whilst you may have an offending "stand out" frequency at a high amplitude how can you attribute what you have found to the wheel bearing?

First of all we have no idea just how worn the wheel bearing is in terms of damage or the number of rollers/balls within the bearing.

Having said that, I did find this website useful which can display an image of your bearing where you can just see and so count the rollers http://www.skfpartsinfo.com/

E.g. 20 rollers (known) with 3 "pits" (unknown) with a bearing race at 10 Hz (known).
Theoretically 20 rollers x 3 pits x 10 Hz = 600 Hz but this is often not the case.

600 Hz is certainly a frequency you will hear rather than feel and so the mic used in conjunction with the accelerometer would be best placed here. I have found it best at present to keep the mic inside the cabin as outside the cabin introduces all forms of noise that can mask the bearing noise.

With regards to wheel bearing diagnosis, the basic inspection prior to measurement is paramount.
The MOT test sequence suggests "feeling" the lower suspension arm for roughness whilst spinning the road wheel in order to determine bearing condition.

We all know this may or may not reveal anything given no bearing load. (If you can hear the bearing with no load then no need to measure with NVH)
The lower suspension arm however may be favourable mounting point for an accelerometer on road test as opposed to the shocker tower or chassis rail as you have highlighted in your initial post Martin.

I have learned that different metals of different mass/structure will effect amplitude and so when placing accelerometers in different locations to differing components we introduce variables where amplitude is concerned

Better still attach the accelerometers to both front hub assemblies (containing the wheel bearing) and compare the amplitudes of suspicious vibrations that are present when the noise occurs with one another. (Use the mic in the cabin to correlate the appearance of suspicious frequencies during playback/analysis)
Accelerometer mount re wheel bearing
Accelerometer mount re wheel bearing
During the road test, try to maintain the peak noise level from the wheel bearing for a minimum of 2.5 seconds on a smooth road surface at fixed engine and road speeds. (Avoid periods or acceleration, deceleration and potholes)
Use this same technique for repeat road tests when using only one accelerometer for comparison measurements. (Basically keep all possible variables to a minimum)

I will post the findings of our test plan next week once the vehicle has been rectified and I can review the damage to the wheel bearing.

I hope this helps in the interim, take care…….Steve

Technician
TwoWaves
TwoWaves
Posts: 705
Joined: Thu Apr 24, 2014 8:32 pm

Re: Wheel bearing failure

Post by Technician »

I'm impressed with the NVH kit as I see the results here, very interesting. I used to call the wheel bearings when rough in operation 'Rarefraction', then one day a customer pointed out he did not know what I meant, so I had to stop writing my reports that way, shame really.

Martinjensen
TwoWaves
TwoWaves
Posts: 55
Joined: Wed Feb 13, 2013 6:22 pm
Location: Denmark

Re: Wheel bearing failure

Post by Martinjensen »

Thanks for answering.

First of all, the car is now repaired with a new left front Wheel bearing.
This was diagnosed on the hoist with a stethoscope, prior to posting here.

The goal was not to diagnose this car, but to learn more about my new NVH Tool.


So what did i learn... :-)

As Mark pointet out, i should be more avare of transponder path. and when i moved the sensors to the lower control arms, insted of the strut Towers, the NVH kit measured the highest amplitude from the left side, as i would expect.

Afterwards i moved the sensors to the knuckle as Steve sugested. (why dident i do this in the first place.......)

After installing the new bearing the vibration at ~174hz is gone, confirming the repair.

The old bearing has 14 rollers, and 1 "pit" on the outer bearing race.

This illustration demonatrates how the frequency will change, if its the inner or outer race, or its a bearing roller that failed: http://www1.udel.edu/present/Becky/bearing/

Regards Martin
Attachments
NVH
NVH
Sensor Placement
Sensor Placement
Failed bearing
Failed bearing

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: Wheel bearing failure

Post by Steve Smith »

Hi Martin, thank you for the feedback and in particular the animation.

This is a real eye opener for me and will go a long way to explain the complexities predicting the offending frequencies from a worn bearing.

I hope to recover the bearing from our test vehicle today and will post ASAP.

Take care.......Steve

User avatar
Mark Dalton
TwoWaves
TwoWaves
Posts: 106
Joined: Mon Dec 01, 2014 9:35 pm
Location: Melbourne

Re: Wheel bearing failure

Post by Mark Dalton »

It makes my head spin watching that animation and thinking about the different frequencies that can arise from the wheel bearing and when those different frequencies are going to interact with each other.
Is there any advantage in playing with the slider in the FFT Tab of the advanced options?
I only have a very basic understanding of the concept of FFT anyway, but for noises like a wheel bearing are you better with a 50 second capture or a 5 second capture and the slider all the way one end or the other?

Post Reply