rubber analysis

[Tmoose]

We are designing a part to replace an unsatisfactory commercial component.
Pretty much a metalastik type bushing with a steel inner and outer sleeve.

https://www.thunderbirdspares.com/e...metalastik-triumph-and-bsa.-97-1527-342-p.jpg

Reportedly the rubber filling is some version of natural rubber.

To specify a good replacement "rubber" we are going to start by confirming the the properties of the original part.
We won't be buying the rubber directly, but will need some information to specify a decent performing replacement material.
For instance if we are considering a urethane replacement we might want similar stiffness, elongation without tearing, etc,

A rubber lab has proposed all of these tests.

• Polymer ID by FTIR – ID of polymers present
• Thermogravimetric Analysis - % breakdown of polymer/plasticizer and fillers
• Extraction and Qualitative Analysis of Extract by GC-MS – ID of plasticizers
• Sulfur Content by Ion Chromatography – overall sulfur content in cure system
• Metal Content by ICP – quantitative analysis of Zinc levels, along with detection of other metals
• Organic Filler Analysis by SEM (Ash to 450C) – analysis of organic fillers present
• Inorganic Filler Analysis by SEM (Ash to 800C) – analysis of inorganic fillers present
• Sample Extraction for Physical Testing – sample prep from provided material
• Specific Gravity - Density
• Hardness (Shore A)
• Tensile Strength – ultimate tensile/elongation
• Drum Abrasion – Abrasion resistance of the material
• Pendulum Rebound – rebound properties
• Outline of suspected formulation based on collected results

I'm thinking some of them are not needed. For instance I'd skip the Abrasion resistance, specific gravity, and ID of plasticizers and fillers.

Any insight is appreciated.

 

[3DDave]

Hardness, maybe. But strength should be based on loading rather than what the failed component is doing.

AFAIK Urethanes are available that are superior in toughness/strength to rubber but at the expense of being stiffer. Which is why natural rubber isn't used on skateboard wheels and urethane is not used on car tires, but the popular car bushing replacement is urethane to make the car feel stiffer.

Since the result is very geometry dependent I think you'll need to set up a small number of trial samples of various compounds to see what works best.

 

[GrahamBennett]

FTIR and GCMS can be run at the same time so those tests are needed to ID the polymer, I would also order a TGA to ID the quantities of filler and softeners present. The only organic filler likely to be present is carbon black so I would ignore that test and I wouldn't bother trying to ID any inorganic fillers either (most likely to be CaCO3 or a clay in any case). I'm not so sure about the total sulfur content as this will include any sulfur present in carbon black, accelerator system, free elemental sulfur, etc. I would skip everything else.

http://www.over50schat.com

 

[TugboatEng]

I don't really see the point in analyzing the failed material in this case. Assume there was no engineering done in the first place. The BOM probably says "rubber".

Once you do all of your engineering you're probably going to settle on EPDM or Chloroprene rubber. %elongation is going to be a big driving factor in selection as in this type of bearing all of the movement is in the rubber. EPDM will be the longest lasting and lowest cost material... until the engine leaks oil on it.

 

[Tmoose]

Hi Tugboat,

I should have been clearer . What was unsatisfactory were the manufacturing tolerances of the recent parts received from an absolutely first tier suppler,

That's not quite true either, What was, and remains wildly unsatisfactory is the first tier supplier's response to this problem.

 

[3DDave]

The envelope rule doesn't apply to average dimensions.

 

[TugboatEng]

Is the rubber distorting the inner sleeve? It sounds like the sleeve needs to be undersized and surface ground after installation to meet the dimension tolerances you require.

 

[Tmoose]

Thanks all.

ASME 14.5 is pretty specific about what average means, and the correct required display of "AVERAGE" when that is the intent.

Section 8.5 is only applied if the abbreviation “AVG” is applied directly to the dimension in question. Section 1.4.5 of the standard indicates that “Notes depicted in this Standard in ALL UPPERCASE letters are intended to reflect actual drawing or model entries…” (emphasis added). In other words, if the note is in all caps in the Standard, it must be applied EXACTLY as shown in the Y14.5 Standard. The only way to apply section 8.5 to the bore ID dimension would be for the dimension to have the letters “AVG” on the actual dimension,

In several conversations the supplier mentioned finishing the ID ( and maybe OD) of the finished assembly .
It seems likely on this batch of parts the inner and outer steel sleeves were finish machined before pressing the parts together.
Quite a time saving.

It also appears to finish the ID the relatively thin inner steel sleeve was gripped tightly on the OD with a 3 jaw chuck.
Thus the sleeve was initially elastically distorted inward over .03", and then the bore was machined round.
Then when the finished part was removed from the chuck it sprung to the 3 lobed shape on one end.

The OD of the part shows a similar 3 lobed shape, but only +/- .003", The OD highs spots are in line with the ID highs spots.
My hunch is the badly lobed inner sleeve is affecting the outer sleeve to a reduced degree because of the rubber compliance.

 

[3DDave]

Remove note 1 from the drawing. Fixed.

So frustrating to the "letter of the law" types that the note is right there, a clear statement. It has only one purpose to exist and that is to over-ride the envelope principle. "are intended to" not ARE REQUIRED TO.

So, fix the drawing and remove the note.

 

[TugboatEng]

I have had a the distortion during fixturing problem before. Part measures good in the machine but distorts as soon as it's removed. The vendor tried to blame residual stress. I had to work with them to come up with a better fixture.

Thin walled sleeves are a bit of a specialty to produce. You can't chuck the sleeve in a lathe. Better to finish it on the end of a tube and then part it off.

 

[Tmoose]

Update - Rubber analysis overcome by events but reanimating nicely. All previous inputs appreciated.

-----
Tuesday August 16 - The 1st tier supplier replied to my heads-up that our SVP is simmering every time he thinks of the parts we sent them for re-work June 1, 2022.
- "I just received work this morning that our supplier has finished the re-work. The parts are en-route to our plant and will be inspected with the go gauge and our scanning CMM."

Response to Tuesday Aug 22 heads-up -

https://www.youtube.com/watch?v=CpGtBnVZLSk

https://c.tenor.com/cT8K-hQhoroAAAAC/terrell-owens-popcorn.gif

https://i.gifer.com/1ySa.mp4