Vibration on Acceleration & General NVH Questions

wmarlexingtonsc · Post by **wmarlexingtonsc** » Sat Mar 02, 2024 5:56 pm

Hello everyone,

I am currently working on a 2011 BMW X3 28i 3.0l AWD. The vehicle has a resonation/vibration that is felt especially badly in the 50-60 mph range. It seems to get progressively worse the more you drive it. It is definitely related to load. Whenever I put the vehicle in neutral in these same mph areas and let it coast, the vibration is not felt. It is felt under light throttle/med. throttle/heavy throttle. I seem to be picking up T related vibrations the worst. T1 and T3 stick out to me. The tires were all out of balance by about one ounce, but that has been corrected. Nothing noteworthy was seen with regards to runout when spinning the wheels and tires on the spin balancer. CV axles look OK visually, but it is hard to run the vehicle up to speed on the rack, since it's an AWD.

I did perform an initial test at the seat frame with a 3 axis sensor. I unfortunately did not save that capture, but the main offender there was Z axis (lateral) T vibration. That was clearly the worst. So, I hooked up a sensor on each wheel at the brake caliper mounting bolts. I hooked them all into the Z axis, since that was so much worse than the rest in the initial capture at the seat frame. I did three runs. One is before balancing the tires. One is after balancing the tires. Finally, the one labeled "bmwbest" was after balancing the tires when I was feeling the vibration very good.

This is another situation where I feel like we can figure this out by traditional means, but I do want to learn how to use my PICO NVH better, and this seems like a good candidate before the vehicle is fixed. My questions, which are both specifically about this diagnosis and PICO NVH in general, are as follows:

1) There was a slight drizzle this morning when I was testing the vehicle. Should I be concerned about the electrical health of the accelerometers mounted at the wheels with regards to rain? Is it OK to test a vehicle in the rain for NVH? Light rain? Heavy rain? I would think Heavy rain is a big no-no, but is any water allowable on the road when doing NVH with a computer?

2) How often do you have to replace the batteries in the accelerometers? Or does it take power from the scope and computer when it's plugged in?

3) Intensity discrepancies. Why is it that when I am analyzing, e.g., T3 data in "BMWbest" file, and cycling back and forth between a paused capture on the bar graphs view and a paused capture on the frequency view, the T3 vibrations do not look like they are the same intensity. For example, on the frequency view, I have a paused moment where it looks like the T3 vibration for the RF wheel is 74 mgs, but when I cycle over to the bar graph view, it looks to be less than 20 mgs. Is there a time delay between the frequency view and the bar graph view? Is it because the "T3" vibration is slightly off center from what the mathematical calculation for where T3 actually is? Why is there this discrepancy between the bar graph view and frequency view in terms of intensity (mgs)?

4) Has anyone been successful diagnosing which CV axle is at fault with cameras mounted underneath to watch for vibrations at the joints? Please share your experience, if so. Was it a slam dunk or not very helpful to use the camera to find with certainty which CV axle is bad?

5) I notice that on the capture BMWbest, when I go to advanced options and set the harmonic marker count way up, many of the vibrations under the frequency view line up with the T vibrations. Anybody have a good feel for why this is so? T8, T4, and T5, all seem to be notable offenders, just to name a few. What is the logic of how things reverberate down into other orders? I am happy to read some engineering document about this, if anyone has the information available. Is it worth programming in all the transmission ratios, front diff, rear diff, and transfer ratios? Would this help me diagnose this issue in any way? Would it help me identify all those outlying frequency offenders, which I am identifying as somehow echoes of the T vibrations?

Thanks in advance to anyone who might help me with the questions above.

On Monday, based on the captures attached, we intend to remove the RF CV axle for inspection of the joints and possible replacement. T-3, channel B (RF), seems to me to be the worst offender. Additionally, I notice that aggressive lane changes either exacerbate or lessen the vibration while it's happening. This is something we have often practiced with wheel bearing diagnosis (sound), but not something I typically think to do with CV related vibration issues. When lane changing right at 50 mph (loading left side and unloading right side), the vibration lessens. When lane changing left at 50 mph (loading the right side and unloading the left side), the vibration increases. I suppose one final question:

7) has anyone successfully diagnosed WHICH cv axle is causing a vibration based on aggressive lane change while the vehicle is in its vibrating state? This is the first time I have tried it, and I am not sure how to decipher the data. The results however were very interesting. (I read about trying this turning motion for CV diagnosis in John Kelly's NVH app.)

Josh

wmarlexingtonsc · Post by **wmarlexingtonsc** » Wed Mar 06, 2024 2:56 am

Update: the right front cv axle inner joint was clearly defective when we pulled the cv axle out. The joint was basically falling apart, even though it looked OK visually before we removed it for inspection. A new axle fixed the issue 100%. I have attached a capture of the NVH data after the cv axle replacement. For some reason, the MPH is not reading correctly in this attached capture. I was travelling 50 MPH, but the NVH tool/J2534 device captured me as travelling at 50 KPH. Perhaps this had something to do with loading an OLD capture first (so that I could skip the setup step), then capturing from there. I don't know why that speed thing happened, but you can clearly see a large reduction of T3 and T1 vibrations in the final capture. I have also attached pictures of the blown apart CV axle inner joint.

Post by **Steve Smith** » Tue Mar 12, 2024 2:01 pm

Hello Josh, thank you for taking the time out to post and I will get to your questions ASAP rest assured

Thank you again ,take care.....Steve

wmarlexingtonsc · Post by **wmarlexingtonsc** » Tue Mar 12, 2024 3:57 pm

There's no rush at all. Take however long you need, Steve. Thank you.

Post by **Steve Smith** » Tue Mar 12, 2024 9:18 pm

Hello again Josh, thank you for posting, I would like to include this case study in our Newsletter if this is OK with you? https://www.picoauto.com/library/newsletter (Especially given the questions raised)

Starting with your initial diagnosis and file “bmwbest”, knowing what we know now (driveshaft tripod joint failure) we are looking at an excessive T3 vibration under load conditions

T3 = 3 disturbance for one revolution of the road wheel

Below I have chosen an area of the signal history where I believe the load has increased due to a clear increase in road speed accompanied with a slight increase in engine speed

: 1

The above conclusively highlights a T3 vibration above all others at the RF (almost twice the height of the LF) this I would imagine was the point at which the vibration was clearly “felt” in the cabin?

Your capture has given me an idea and a “New Feature Request” to enable the graphing of the OBD PID “Calculated load” (along with RPM/MPH) as this would be very useful in the signal history

On that very subject, we can graph OBD PID’s for ICE and Hybrid’s (post49491.html#p49491) within our NVH software, but can only display them in the Time Domain view. Either way, this is still useful information to correlate Time Domain data to the Frequency view

Now, it is very easy to look back at data and make it fit the offending component, but at this stage let us forget about the driveshaft tripod joint and click on the T3 vibration order marker for Help on what can cause this level of vibration at 3 times the frequency of the road wheel (T1)

: 2

Above we have the “Pop-up Help” with a link to “More information”, clicking on the link takes us to “Wheel related noise and vibration”

I must apologise here, as reading through this Help page, we are missing a T3 description, but our Help does ask the user to “see the Drivetrain vibration topic” (I will add another New Feature request)

: 3

As we move onto the “Drivetrain related noise and vibration” topic we find that piece of GOLD

: 4

We should add the wording “load” to the above text which I will also suggest to the publications team

I think this would be a good place to add a few tips in terms of vibration orders and typical faults

(Please note: these are based on our experience in the field and are not the sole cause of these vibrations)

: 5

I will add some links to the above “notes” that hopefully reinforce the sweeping statements above!

Moving onto Question 1

There was a slight drizzle this morning when I was testing the vehicle. Should I be concerned about the electrical health of the accelerometers mounted at the wheels with regards to rain? Is it OK to test a vehicle in the rain for NVH? Light rain? Heavy rain? I would think Heavy rain is a big no-no, but is any water allowable on the road when doing NVH with a computer?

I have to say, come rain or shine, I have used the accelerometers without any impact on their performance. My bigger concern is mounting to the front calipers and trying to ensure the cables do not wrap themselves around the driveshaft along with confirming there is enough “purchase” to allow the wheel to move from lock-to-lock. Following the route of the brake flexi-hose normally resolves this issue

The accelerometers I use have been mounted to the underside of vehicles for months (during a variety of tests) and I still use them today. I am by no means recommending this practice, we are of course requested to “stress test” everything. I can therefore say, you have nothing to worry about

Question 2

How often do you have to replace the batteries in the accelerometers? Or does it take power from the scope and computer when it's plugged in?

The accelerometer and mic (sensors) take their power from the internal battery inside the TA259 (NVH 3-output interface) The internal battery is only drained when these sensors are connected, so best practice is to disconnect the sensors from the TA259 when not in use. If the TA259 LED is flashing when the sensors are connected, the internal battery is sufficiently charged.

Should you wish to replace the battery inside the TA259, you will require a Lithium CR123A cell

The image below highlights a test carried out (using PicoLog) to determine the internal battery performance when a sensor is left connected to a TA259 interface

: 6

Note above the battery voltage falls to 2.2 V after 66 days. (I have never had to replace a battery to date)

Question 3

Intensity discrepancies. Why is it that when I am analyzing, e.g., T3 data in "BMWbest" file, and cycling back and forth between a paused capture on the bar graphs view and a paused capture on the frequency view, the T3 vibrations do not look like they are the same intensity. For example, on the frequency view, I have a paused moment where it looks like the T3 vibration for the RF wheel is 74 mgs, but when I cycle over to the bar graph view, it looks to be less than 20 mgs. Is there a time delay between the frequency view and the bar graph view? Is it because the "T3" vibration is slightly off center from what the mathematical calculation for where T3 actually is? Why is there this discrepancy between the bar graph view and frequency view in terms of intensity (mgs)?

This behaviour we can attribute to the theoretical frequency calculation of T3 (denoted by the vibration order marker) and the actual frequency of T3

When obtaining road speed via OBD, the theoretical T3 frequency is first calculated using T1 x 3 using the tyre circumference method

Given NVH knows the tyre diameter and road speed it calculates the theoretical frequency as follows:

Wheel Frequency calculated from OBD road speed and tyre circumference.
Road speed 47.8 mph = 76.92 Km/h
Tyre size 205/55/16
(Vehicle speed (km/h) * 1000 (m)) / (2 π * Tire Radius x 3600 (sec))
Wheel Frequency (T1) = 11.34 Hz
T3 = T1 x 3 = 34.02 Hz

Our NVH software will place the vibration order markers (T1 & T3) at these locations in the frequency view, regardless of whether the peak vibration from the tyre or tripod joint aligns with these markers (often they don’t, but are instead, half way up the peak)

The bar graph view indicates the peak value of T3 according to the position of the vibration order marker (T3). The frequency view gives you the advantage of seeing the vibration order marker relative to the position of the peak vibration in close proximity to that maker

: 7

Below, the bar graph view indicates the theoretical position of T3 (and all other vibration orders markers) not the actual

: 8

There are several reasons why the vibration order markers may not sit directly on-top of the peaks and they are covered here using “Tyre correction factor” viewtopic.php?p=36541#p36541

To summarize, given we are relying on OBD road speed & tyre circumference to calculate wheel frequency we need to factor in the variables

For example:
• Tyre pressure
• Tyre wear
• Tyre deformation
• OBD road speed v actual road speed

Referring to the link above, we can modify tyre correction factor (post capture) to enable our vibration order markers to sit at the top of our offending peaks.

Note, this practice is OK for “fine tuning” the position of the marker, but not for large adjustments to relocate a marker to a peak frequency that is clearly not a T1 or T3 etc.

Below I have amended the tyre correction factor from the default 1.030 to 1.017

: 9

Note the effects on the position of the T3 marker in the frequency view below after amending the tyre correction factor above

: 10

Note the effect on the bar graph view

: 11

Question 4

Has anyone been successful diagnosing which CV axle is at fault with cameras mounted underneath to watch for vibrations at the joints? Please share your experience, if so. Was it a slam dunk or not very helpful to use the camera to find with certainty which CV axle is bad?

The following case study has one such example which I think we have discussed before https://www.picoauto.com/library/case-s ... celeration

In the above case study, I have attached a webcam under the vehicle where you can see the lateral movement of the transmission due to a tripod joint failure.(See video below)

I have not tried measuring the transmission oscillations in slow motion replay to obtain the frequency (oscillations per second of playback) but it should equal the offending frequency displayed in the pddata file

Question 5

I notice that on the capture BMWbest, when I go to advanced options and set the harmonic marker count way up, many of the vibrations under the frequency view line up with the T vibrations. Anybody have a good feel for why this is so? T8, T4, and T5, all seem to be notable offenders, just to name a few. What is the logic of how things reverberate down into other orders? I am happy to read some engineering document about this, if anyone has the information available.

Before we move onto harmonics, below is the "BMWbest" capture with additional harmonics indicated by multiple harmonic rulers

: 15

There is a lot to study when it comes to harmonics/vibration orders as we require a theoretical understanding of how and why they develop.

A golden nugget of information to fall back on; everything naturally generates harmonics and, (other than artificially generated) nothing generates a pure sine wave at a fixed frequency without harmonics. Harmonics are what give a sound or vibration its unique warmth, depth, or harshness, technically referred to as its “timbre”

For example, a note played on a musical instrument will sound like a pure fixed frequency, however, when you analyse the musical note using a spectrum analyser, you will discover not only the fundamental frequency (we perceive as the pure note/ tone) but the accompaniment of harmonics that give the note its unique “timbre”.

This is how we recognise different musical instruments playing the same note, as each will have bespoke harmonics of varying amplitudes and count

Harmonics are distinguished by their relationship to the fundamental frequency, given they are multiples of this fundamental. i.e. a low frequency “note” at 60 Hz will be accompanied by harmonics at 120, 180 and 240 Hz all of which give the note its unique sound (timbre)

So how does this relate to a vehicle vibration?

Remember, vibration and sound are one and the same; a vibration is sound you feel, a sound is a vibration you hear. With this in mind, can you imagine for 1 second a pure, single frequency (sine wave) generated by a faulty component on a motor vehicle that would be unaccompanied by harmonics?

Below I have tried to replicate an artificial, pure fixed frequency sine wave at 20 Hz using a signal generator in order to demonstrate the absence of harmonics

: 12

Note: Even using the signal generator above, there is a potential harmonic evident around 60 Hz!

Below I have tried to replicate a real-world scenario where our accelerometer is subjected to multiple vibrations (especially when connected to a caliper bolt) at differing frequencies and amplitudes. One of these frequencies is fixed around 27 Hz the other is fluttering between 19 & 22 Hz

: 13

Note the complexity of the sine wave in the time domain!
Without the spectrum view, it would be impossible to determine the frequency content of this signal and display the associated harmonics that have developed as a result.

Whilst I have mentioned this above, I believe it is worth repeating as we can now relate the theory to our auto application:

All components that rotate and generate noise or vibration will consist of the fundamental frequency (the frequency at which the components rotate) plus harmonics (orders) that are multiples of the fundamental.
Other components that do not rotate may still vibrate, resonate or generate noise at their natural frequency as a result of the vibrating forces imparted by an imbalanced component. The image below will help with resonance

: 14

These non-rotating parts will naturally include harmonics, the amplitude of which depends on the severity of the fundamental noise or vibration.

As you can imagine, the spectrum (NVH frequency view) is going to fill up very quickly indeed given the variety of components scattered about a vehicle, the majority of which are rotating at different fundamental frequencies.

Let’s take a look at the phenomena’s we need to consider in auto applications and the harmonics/orders to expect

• Imbalance: First order vibrations but will produce multiple harmonics if severe
• Misalignment: Second order on a prop-shat, Third order on a Tripod joint
• Tooth contact: Any number of orders of the fundamental depending on tooth count
• Resonant frequencies: Excited and linked to orders of other vibrating components
• Combustion: Dependent on No. of cylinders: 4-cylinder engine has a dominant second order
• Engine misfire: 0.5 order due to one disturbance every 2 revolutions of the crankshaft
• Tyre orders: First, second and third orders. (Think deformation)

Hopefully the above goes someway to explaining harmonics/orders, why they are formed and what to expect into auto applications utilising the internal combustion engine

To summarise
• Think of harmonics/orders as the number of disturbances per revolution of a component
• The speed/frequency of a rotating component is referred to as the “fundamental”
• A dominant offending vibration on a vehicle, will not be the only vibration
• A vibration or sound measured on a vehicle will be accompanied with harmonics
• The more severe the vibration, the greater the potential for multiple harmonics

For some awesome viewing and dedicated study to NVH, please watch and listen to Jason Tranters video from the Mobius Institute here https://www.youtube.com/watch?v=Vj1xmze3GlE

Question 6

Is it worth programming in all the transmission ratios, front diff, rear diff, and transfer ratios? Would this help me diagnose this issue in any way? Would it help me identify all those outlying frequency offenders, which I am identifying as somehow echoes of the T vibrations?

It is worth populating the Advanced Vehicle settings found in the Vehicle information dialog during NVH set-up as this will provide you with additional information surrounding gear position and vibration orders if pulley measurements /ratio fields have been populated

: 16

As we can see above, during acceleration, the gear position is plotted in the signal history (your laptop will also announce the gear position if not muted)

For example, if you are chasing a transmission whine, the offending frequencies may only appear in a particular gear position (let’s assume 7th gear)

With the gear ratios entered, this will be immediately apparent as the offending peaks in the frequency view will align with the period in the signal history where 7th gear was displayed and hence selected

With pulley measurements /ratio fields populated, pulley vibration orders can be easily identified with Custom vibration order markers (See below where we have “A” for the Alternator fundamental frequency)

: 17

To answer your question “Would it help me identify all those outlying frequency offenders, which I am identifying as somehow echoes of the T vibrations?

It is possible that an offending vibration that was first thought to be a harmonic of T1 (e.g. T7) could also be the fundamental frequency of the alternator. Without a vibration order marker for the Alternator, this would not be apparent

In your BMW scenario, you entered the minimum amount of data required (Engine cylinder count, Differential ratio & tyre size) and managed to zone in and identify the offending component / source of your vibration.

From these minimal entries and an OBD connection, we can identify the fundamental frequencies of “The Big Three” (Engine, Propshaft & Tyre)

The following forum post will help with the various options surrounding the gear ratio & engine speed entry fields and the expected results

viewtopic.php?p=99593#p99593

Question 7

Has anyone successfully diagnosed WHICH cv axle is causing a vibration based on aggressive lane change while the vehicle is in its vibrating state? This is the first time I have tried it, and I am not sure how to decipher the data. The results however were very interesting. (I read about trying this turning motion for CV diagnosis in John Kelly's NVH app.)

There can be no doubt that “load” is one of the fundamental properties to consider when it comes to NVH diagnosis. For example, if we consider a worn wheel bearing, loading the vehicle by cornering or providing subtle inputs to the steering wheel can influence the bearing noise dramatically, simply because we are applying “load”

The additional weight and forces applied to the wheel and tyre during cornering are transmitted through to the hub and ultimately to the supporting bearings

Likewise, subtle steering inputs (e.g. a progressive lane change) will provide a differing load due to the twisting action of the tyre resulting in alignment alterations within worn bearing

I used to use a rule of thumb (prior to using NVH) whereby if a wheel bearing noise from the front of a vehicle increased when cornering left, then the “loaded” front right had bearing was at fault.

The caveat to the above was, if the vehicle in question was FWD then you reversed the theory, i.e. when cornering left (whilst the loaded bearing was the front right) you replaced the front left!

This rule of thumb worked more often than not but some came along chassis ears, then NVH software to provide objective data upon which to accurately diagnose the cause

If we then extend the above to our Rzeppa CV joint which again consists of bearings and all-encompassing roller bearing guides, not only is it handling rotational forces but also angular forces presented by cornering and vehicle/suspension articulation (See image below from https://en.wikipedia.org/wiki/Constant-velocity_joint#)

: 18

Whilst I have not used the suggested technique of “aggressive lane change while the vehicle is in its vibrating state” I can see how a worn CV may well respond to changes in steering or suspension angles based on the above

Where NVH can help is by identifying these components by their inherent frequencies
E.g. A Tripod joint = T1 x 3 or A Rzeppa joint housing 6 bearings = T1 x 6

Adapting our road-test may help to pinpoint the area of concern as hard acceleration (when climbing) will often reveal issues from drive shafts due to torsional/rotational forces being applied to the joint

Equally, cornering/angular forces can be applied to the joint/shaft by manoeuvring the vehicle over various terrain that forces not only a change in joint angle but also vehicle height

Note: Exceptionally work joints will introduce momentary shaft alignment issues

On the subject of vehicle height, we can include this in our road test if the vehicle has the ability to change its ride height. In conjunction with adapting our road-test, knowledge of the vehicle usage pattern and service history is invaluable

For example, it would come as no surprise to suspect a worn CV joint on a vehicle operating Off-Highway in a quarry dragging heavy machinery out of the mud

I hope this helps, take care……..Steve

wmarlexingtonsc · Post by **wmarlexingtonsc** » Thu Mar 14, 2024 2:10 am

Steve,
Thank you as always for your thorough and well-reasoned responses. You are the only one I know who not only gives an answer about battery life expectancy, but also a scope capture to back it up! That's awesome that you just happened to have that in your back pocket. Feel free to use this case for the newsletter and thanks for asking. I particularly like the chart that you posted with different axes of vibration and their corresponding vibration sources. I look forward to your continued responses, but really take your time. I am in no hurry at all.
Have a good night!
Josh

wmarlexingtonsc · Post by **wmarlexingtonsc** » Fri Mar 15, 2024 10:22 am

Steve,
the explanation about tire correction and the discrepancy between the bar graph and frequency views makes perfect sense. As a general rule of thumb, since I do not fully understand what number to set the tire correction factor to in each situation, I suppose I should make it my habit from now on to not fully trust the bar graph display, but rather the frequency display, and to consider that view as the closest to the raw results from the road test.

wmarlexingtonsc · Post by **wmarlexingtonsc** » Sat Mar 16, 2024 10:39 am

Steve,
what a response you are giving here! I am so thankful to you for taking the time to explain all of this to me--and in such incredible detail! I am currently in the process of studying all this information that you have given me, which might take a little while, but suffice it to say that it is all very helpful, indeed, for wrapping my brain around these concepts.

Two thoughts occur to me here: one an idea, the other a question. The idea first: perhaps it would be nice to have a sticky post at the top of the NVH section with all the best NVH theory posts of the forum or elsewhere, including links to PDF files, videos, etc., where people can learn more about the theory of NVH. At the top, of course, could be all the videos and case studies put out by PICO, which are very helpful, but it could also include other things, outside of the scope of the original NVH video series. To me, the most helpful information that I have found about NVH theory, besides the PICO multi part video series, were a VW SSP pdf on NVH, which is easily obtainable via a google search, and an old Toyota NVH manual from 1989 (the number is not immediately in front of me), which I was able to purchase off EBAY, and which I saw referenced in an old forum post of yours about NVH. I could not find it online. I wonder if there would be a problem with me scanning that and posting it on the forum? Or would there be copyright issues?

Next, the question that bothers me: it has to do with moving the accelerometers from the seat bolt to the caliper bolt, particularly with regards to the orientation of the sensor. Say, for the sake of example, that I put the accelerometer on the seat bolt with the bolt hole facing forward. We then have a X (forward to back), a Y (up and down), and a Z (side to side) axis. Say then that a test drive reveals the Z axis to be the main offender, T1. The vibration is seen in the steering wheel, so we move the two accelerometers to the front two caliper bolts. By repositioning the accelerometers, and screwing the magnet into the "front" bolt hole, and only having one channel on the Z axis, how does this work exactly? Hasn't the X axis, which was facing forward on the seat frame, become then the Z axis (side to side) and the Z axis, which was facing side to side on the seat frame, become the front to back axis? Why shouldn't I then, if I see the main vibration on the Z axis on the seat frame, when I am moving the accelerometer to the one channel mode and placing it on the caliper bolt, then plug into the X axis, since it then seems to have become the Z axis by moving around the orientation of the sensor? Does that question make sense? I feel like if I understand the answer to this question, I may be more confident and free in the areas and orientations in which I place the accelerometers.
Thank you!
Josh

Post by **Steve Smith** » Wed Apr 24, 2024 1:39 pm

Hello again Josh and sorry for the delayed reply

Some nice ideas once again, thank you

Re NVH topics in one location, this is not quite what you had in mind but if you click on this link topic15231.html and use “CTRL + F” keys and type “NVH” there are a list of topics that will cover several NVH features and questions

We are always in the process of updating the website and library here https://www.picoauto.com/library so please keep those new requests coming in here https://www.picoauto.com/tech-support and click on “Get Support” to share your feedback which will be triaged to the relevant team within Pico HQ

Thank you also for the offer to scan and share the Toyota NVH manual but I would advise against it just to protect yourself and others due to copyright. You can however share a link if you have found this manual on the open web

Moving onto your question surrounding accelerometer orientation, you are correct.

To run through the scenario you describe, we have our 3-axis accelerometer mounted on the seat bolt, in the vertical position with the screw thread facing forward. (see below)

: 1

You then identify a high T1 lateral vibration in the Z axis (Channel C, Green)

We now wish to add an accelerometer to channel D to help zone in on the offending corner of the vehicle (You can use multiple accelerometers as you describe but using 3-axis + Single channel mode will help with interpretation)

Placing an accelerometer on a caliper mounting bolt (see below) results in the rotation of the accelerometer through 90°, meaning or Z axis is now measuring the fore and aft vibration and our X axis is measuring our lateral vibration (Note, the Y axis [vertical] remains unchanged)

: 2

Based on the above, we know our 3-axis measurement confirmed our offending vibration to be “lateral” we therefore connect the axis measuring the lateral vibration based on the accelerometer orientation. In the scenario above, this would be the X axis connected to channel D of our scope (See below)

: 3

The following video may help, which discusses the connection of additional accelerometers to measure the vertical axis

If you are using one accelerometer, you can use the vector sum value obtained at the seat bolt, relocate the accelerometer to the caliper bolt and compare with the vector sum value obtained in this new location without having to worry about orientation.

I hope this helps, take care……..Steve

PicoScopes
& kits

For vehicle manufacturers

EV kits and
accessories

Noise, Vibration and Balancing

Milliohm tester

Pressure sensors

Current clamps

Breakout
leads/boxes

Accessories

Automotive Forum