Automotive Designing a Poka-Yoke for Flat Marks on Rocker Arm Bearing

  • Thread starter Thread starter billy_boy
  • Start date Start date
  • Tags Tags
    Designing
AI Thread Summary
The discussion centers on addressing customer complaints regarding flat marks on the outer diameter of rocker arm bearings, which are crucial in automotive applications. Suggestions include using air gauges and machine vision systems to detect these defects, although the high cost of vision systems is a concern given the large production volume. Participants emphasize the importance of characterizing the flat marks to determine the best detection method, with recommendations for displacement sensors, particularly optical types, to monitor the grinding process. There is also a suggestion that improper assembly lube may contribute to the issue, causing the bearings to function incorrectly. Overall, the conversation highlights the need for effective detection and root cause analysis to resolve the flat mark problem.
billy_boy
Messages
13
Reaction score
0
I am a quality control engineer in an automotive parts manufacturing company. One of our products is a rocker arm bearing which is essentially a cam follower. The OD and ID of the bearing is ground using CBN grinding wheels. The ID contains cylindrical rolling needles. Recently we have received a lot of customer complaints for flat marks on OD of the bearing. Can an air gauge be used to measure or detect these flat marks and also can a simple poka yoke be designed using the same.
 
Engineering news on Phys.org
billy_boy said:
can a simple poka yoke be designed
Interesting, I hadn't seen that term before. Learn something new here at the PF every day...

https://en.wikipedia.org/wiki/Poka-yoke

Have you considered using machine vision to look at the finished parts to catch these problems?
 
What is a poka yoke? How is a bearing to function as a cam follower? is it acting as a roller follower? A picture might help explain your situation.
 
Dr.D said:
What is a poka yoke?
(See the Wikipedia link that I posted...) :smile:
 
upload_2017-8-4_22-3-7.png
 
that cam roller is basically a needle roller bearing
 
berkeman said:
Interesting, I hadn't seen that term before. Learn something new here at the PF every day...

https://en.wikipedia.org/wiki/Poka-yoke

Have you considered using machine vision to look at the finished parts to catch these problems?
we manufacture 1.2 mil. of these in a month. we use a vision system for checking missing needles. it is difficult handling such large volumes. also vision systems cost a lot of money
 
Another vision system is also my guess for detecting the flats.

But you are the engineer in there. Characterise the symptom.
How wide? How deep? Can you see it by eye? Can you see it by camera? Is it one part randomly? Is it a bunch of parts in a row?

You can't ask yourself, let alone us, how to detect it until you know exactly what it is you want to detect
 
  • Like
Likes berkeman
Hmm, first time I've heard the word/phrase/expression "poka-yoke" as well, either way...
So if I understand correctly, you need to ensure that the bearings are circular without any flat spots? It is difficult to get into any specifics without understanding the application in detail, however a displacement sensor monitoring the surface edge of the bearing while it is being ground may suit your needs.
 
  • Like
Likes billy_boy
  • #10
Tolklein said:
however a displacement sensor monitoring the surface edge of the bearing while it is being ground may suit your needs.
I would vote for that approach. The sensor could be mechanical (potential problem due to speed & wear), optical (LASER displacement sensor or similar as supplied by industrial Manufacturers ( two that come to mind are Banner and Keyence)), electronic (capacitance distance sensor). Of the three, probably Optical is the best bet.

As for Root Cause, sounds like a bad bearing in one of your grinders.

Let us know what you find.
 
  • Like
Likes Asymptotic
  • #11
So the bearing is not rolling?

If this is say...a crower rocker arm, maybe the customers are using improper assembly lube which may cause the roller to not roll during that important first few minutes of operation.

Since oil is super slippery, it doesn't take much resistance to rolling to stop the rolling action without causing any damage to the needles...so the outside of the bearing glides on the oil film.

Once there is even a slight flat spot on the outside of the roller where it contacts the cam, it may stop rolling and now acts as a "pad type cam follower" instead of a bearing because oil is such a good lubricant that it allows the cam to turn against a stopped roller causing a flat spot to get worn eventually instead of causing damage that snowballs into galling and or catastrophic heating.

If a hydraulic "lifter" is employed, the bearing could stop turning every time there is oil pressure to keep it against the cam with any amount of pressure while a mechanical follower may allow turning when it's not under pressure of opening the valve.

In stock factory usage, you have less chance of the wrong lubricant being used but in the aftermarket horsepower world, every guy has his own recipe and many are just wrong.

If a customer can send back the rocker arm and the camshaft, you could install it in an engine and just watch the roller to see if it's stopping and you should easily hear the "thumping" if it is.

Maybe a customer can make the observation himself and let you know what he's seeing.
 
Last edited:
  • #12
Can you post a picture of the product you make and the defects you encounter?
 
  • #13
Tom.G said:
optical (LASER displacement sensor or similar as supplied by industrial Manufacturers ( two that come to mind are Banner and Keyence))
can these detect displacements in microns??
 
  • #14
salmansl said:
Can you post a picture of the product you make and the defects you encounter?
upload_2017-8-9_23-2-21.png
 

Similar threads

How do I properly mount a needle bearing for a trike swing arm?
Replies
11
Views
3K

Hot Threads

Recent Insights

Back
Top