Material Properties 1

[GrumpyPlasticBastard]

We keep track of our machine settings, for a given viscosity of the material.

Say our viscosity is 28MU, I have over a dozen sheets showing that we ran at a historical high of 365 for the mold temp and a historical low of 340. No notes from the operator no numbers in regards to scrap.

Go to start up with a fresh batch of 25MU, same as last time. Started at 355 degrees mold 195 water... Nothing but bubbles and cracks.

Over the course of an hour I dropped the temps nearly 15 on the mold and 10 on the water. 18 scrap pieces. That's too many on this machine.

Then it hit me. I went back and looked at the property sheet from the supplier for this batch, as opposed to last batch.

There is MUCH more to the material than just viscosity.

Turns out that the section marked "modulus 300%" and the one labeled "Tensile" way different than the previous batches. The Mooney Viscosity, the three T-test columns and the two "G" columns were within 2.

I went back through our historical file and matched up historical setup sheets with the material data sheets for a given lot of rubber.

New policy is to match the material as close as possible using this data, not just match the setup sheet to the viscosity number.

Looking over the data, I now understand why we had such wild swings in machine settings for a given viscosity range.



No doubt some of you are shaking your head and laughing, because this is obvious to you... others, I hope I have passed on a valuable tip for you.

----
BFL Dreamworks
"If we don't have an answer we'll get one"

 

[btrueblood]

A star for that post, GPB. Rubber is funny stuff, our molder likens it to baking a cake more than any other process. Seemingly minor changes in materials can wreak havoc, as we've found from long-term cycle testing for fatigue properties on our parts.

 

[GrumpyPlasticBastard]

When I did injection molding of thermoplastics, a material variation of this magnitude was UNHEARD OF.

I am told batch to batch variations of this magnitude are normal, in fact it is SOMEWHAT UNUSUAL to get batches that were nearly identical.

----
BFL Dreamworks
"If we don't have an answer we'll get one"

 

[GrahamBennett]

That depends on many things GPB e.g. SBR from two suppliers each using the same technology have given and continue to give wild variations in processing from mixing through extrusion to curing and field performance. The same applies to carbon blacks.

 

[GrumpyPlasticBastard]

What we have is a standard material from a standard supplier, we always order this material from them.

For over two decades.

Same plant. Same mixer.


Not complaining really. I have a tool now to help overcome the swings in material consistency.

----
BFL Dreamworks
"If we don't have an answer we'll get one"

 

[edine]

Right now the rubber industry is being plagued by raw material discontinuations and processing location changes. It may be that your mixer or his supplier is having supply issues or has made changes. If this is a new job, consider that you may not have seen all the ups and downs of normal variation in the compound.

I track custom mixed material statistically on all the QC tests I require and watch for outliers which will tend to cause problems. Viscosity alone is not a good enough test because so much of what happens in rubber is shear dependent. Data from an RPA or PPA may give you more insight into what could be going on.

I would be curious to know what tool you are using to compensate for swings in material consistency.

 

[GrumpyPlasticBastard]

Edline:

Right now my "tool" consists of process changes. Now that I know that variable A leads to condition X so I need to change Z in the process I'm much better off.

As it was before we were flying blind. We have a captive product line, we've made the exact same product out of the same two types of rubber since 1995. They were tracking viscosity and that was it.

As we go forward and my sample size increases I'll be able to look at the numbers, devise a solution and correct the problem on the fly... rather than the "shot in the dark" method we used previously.

----
BFL Dreamworks
"If we don't have an answer we'll get one"