Hello and welcome to another installment of “Not All O-Rings Are Created Equal.” I’m going to take a few minutes today to talk about one of the common failure modes of an o-ring and some of the factors that should be considered or be aware of to alleviate any such failures.
For example, did you know that in dynamic reciprocating applications an o-ring is not recommended for strokes of 12 inches or longer? A lot of applications out there, especially reciprocating that tend to stroke longer than the 12 inches. There’s a lot of factors that must be considered and we’ll talk about some of those as we go along.
As a rule, anything over 12 inches is not recommended.
In addition, in dynamic reciprocating applications, the direction of pressure and the seal friction should oppose each other.
Failure will likely occur if pressure and friction both are of the same direction.
Now, these are just two examples of a common failure that is given the term spiral failure. I want to talk a little bit more about spiral failure and an understanding of what that is. According to the Parker o-ring handbook, which is his kind of the Bible to the o-ring industry, this type of failure was given the name because when it occurs, the seal looks like it’s been cut halfway through the o-ring.
The cross-section of the o-ring actually has a corkscrew or spiral look to it. An o-ring usually seals in that condition until you get a complete break, but it does cause fatigue and it doesn’t lend itself to susceptibility to breaking once it has spiraled.
Having said that, a properly used o-ring slides during the reciprocating stroke. There’s actually movement of the o-ring. The hydraulic pressure produces a greater holding force within the groove that produces the sliding effect. If it doesn’t slide, it is going to roll. The smoother finish of the sliding surface in relation to the groove surface finish produces less friction. Stands to reason. Running friction is lower than breakout friction. The torsional resistance of the o-ring also resists twisting. The compound, the material, as we discussed in other sessions, what are the differences in o-ring materials themselves?
I want to talk a little more about the torsional, or spiral failure, and the fact that it is not limited to just an o-ring cross section.
There are many applications where you’ll use what they call a quad ring, a square section or a lathe cut, an x-ring – these types of seals are also susceptible to spiral failure given those same conditions. I’m not going to go into a lot of detail in all of those but there’s actually quite a list of things that need to be factored into the design when applying o-rings in reciprocating applications to reduce or eliminate spiral failure:
– The speed of the stroke itself
– The media that is being used, or the lack of lubrication
– We talked about the pressure differential
– Squeeze – if you have too much
– What’s the shape of the grooves or are the grooves split?
– The length of stroke
– Surface finish
– The type of metal that is being used
– Is there a side load? Is it absolutely concentric all the way around?
– And there are several others
These are all factors that can a tribute to spiral failure which can be a catastrophic failure in o-rings. To learn more about it, go to the ESP International website to find the Parker O-Ring handbook. It’ll give you much greater detail to how to avoid spiral failure in application.
FDA, 3A, and NSF are common standards for the production of o-rings within the food grade and potable water industries. Each standard is derived from a different agency who contains unique sets of rules, requirements, and regulations.
They are very different, and very specific in their place in the market.
Click on the image to open in a New Tab.
Welcome to our next installment of not all o-rings are created equal. Today, I thought we would take a little bit of time to talk about the differences in FDA, 3A, and NSF. These are common standards in the North American marketplace that seemed to have a lot of confusion about certifications, validations, approvals etc. We’re going to try and simplify it a little bit and then in future editions will get into greater detail to each. Again, specific to North America – the China GB4806 the European WRAS – those are standards to be addressed another time as well. But in general, as we talk about food grade and potable water, there are three standards that in the production of o-rings you need to be able to consider and be aware of; and those specifically being FDA, 3A, and NSF.
FDA is the standards that are derived and managed by the Food and Drug Administration. What this is, is simply a list of ingredients that are neither toxic nor carcinogenic that can be included in the formulations of compounds for o-rings.
They go through a series of extraction testing so that as the o-rings are in use in application, what is leaching of those o-rings that may or may not create a problem. Once that is done, the o-rings and the o-ring material is deemed to meet FDA requirements. Not to be confused with being certified.
The FDA does not certify o-ring materials or compounds but rather creates a list of acceptable ingredients.
The next one we want to talk about is 3A. It is not, despite what some people might believe, part of FDA. It is derived from the International Association of Milk, Food, and Environmental Standards. A secondary group is the Milk and Dairy Supplier Association. What 3A does is it determines the rubber that’s capable of contact with antibacterial treatments and the maintaining of the physical properties of the o-ring after those treatments.
As you can imagine in in the milk production world, there’s a lot of chemicals. There’s a lot that goes on in order to continue to service that equipment. What is the o-ring able to survive in that environment? There’s also an E3A Standard – basically similar to the 3A standard, but specific to egg production.
Finally, we’re going to talk about NSF. NSF not to be confused with the Food and Drug Administration or the Dairy Group, but it is specific to the National Sanitation Foundation. NSF has two primary standards or certifications that we need to be aware of:
With the NSF and the various testing that is required in order to meet those approvals, the NSF does the testing and does, in fact, certify to the compounds, to the materials for their standards.
Welcome to our latest edition of “Not All O-rings Are Created Equal.” In previous editions, we’ve talked about some of the physical properties of the o-rings and things that you need to be aware of and account for when selecting an o-ring when selecting material for your application.
Some of the more common physical properties that we’ve talked about so far are your durometer and hardness, your compression set, elongation and the important factors that they are in determining the compound of materials to use. Today we are going to talk a little bit about tensile strength.
Click on the image to open in a New Tab.
Short version according to the Parker Hannifin O-ring Guide:
Tensile strength is simply measured as the PSI or megapascals required to rupture a specimen of a given elastomer when stressed.
Put it under stress and at what point does it burst – at what point does it break?
One thing I want to point out, as I’ve done this in other sections regarding compression set and elongation, it varies by material, but it also varies within the material. Which is why I emphasized that they’re not all created equal. A nitrile isn’t just a nitrile, isn’t just a nitrile.
Specific to tensile strength, it can range anywhere from 6.9 to 27.6 megapascals. Just within the nitrile family alone. FKM can very 3.4 to 20.7 – a pretty good size range. Even EP which tends to be a fairly simple, stable material in most applications – 2.1 to 24.1 megapascals.
Why is that important? In the grand scheme of things, as you increase tensile strength, you improve the modulus of the material. As stated in other sessions, when you make a change to a compound or material, to get something you’ve got to give up something.
What am I giving up when I improve my modulus?
I’m sacrificing elongation. In most applications that’s a non-factor. As we’ve talked to the elongation section, elongation is predominantly about installation. How far can it stretch? If you’re not having to overstretch your o-ring, it’s less of a factor but rather gives you some important properties.
With the modulus – which is the stress at a predetermined elongation usually measured at 100% – the higher modulus is apt to recover from peak overload. That’s important. It helps support and aid to the strength and wear-ability of the o-ring and it usually increases with hardness.
Tensile strength while not talked about a lot can be an important factor and one of those key physical properties that you need to be aware of as you evaluate why and how all o-rings are not created equal.
1 of 5
Today we’re going to talk about durometer of rubber products. Durometer is a measurement of hardness and like other hardness test measures the depth of indentation in the material created by a given Force using a standardized pressure foot. The ASTM D 2240 standard recognizes 12 different durometer scales.
Click on the image to open in a New Tab.
One important aspect to note is the test methods used when testing durometer.
When you’re using any of these Shore OO, Shore A , or D, or the IRHD scales, you need to be using a large thick flat piece of rubber. The piece of rubber needs to be a minimum of six mm thick and large enough that all of your measurements can be taken at least 12 mm from the edge of the material. This can create some obstacles when you’re trying to measure small rubber products such as o-rings. Most O-rings don’t have a 6 mm thickness. So if you’re trying to use a Shore A durometer tester on the typical o-ring that’s going to be an incorrect measurement method. It’s not valid.
In the case that you do need to measure physical parts with small cross sections, you’re going to have to use the shore M durometer scale. Again, you can see that this is a 30 degree cone. It is a little bit smaller diameter than the short D and also would have a different spring force applied to it. But with this scale you can measure samples as little as a 1.25 mm in diameter.
1 of 5
The AS568 o-ring size chart, published by the Society of Automotive Engineers (S.A.E), sets a standard for universal o-ring sizing. The chart specifies the inside diameters, cross-sections, tolerances, and size identification codes for 349 o-rings used in sealing applications.
DJ Rodman spends some time covering these attributes at a high-level.
Click on the image to open in a New Tab.
Today we’re going to talk o-rings. We get frequent calls, request for quotes, emails, regarding o-ring and o-ring sizing. Today we are going to run through and explain how o-rings are sized and talk a little bit about the language of o-ring sizing. O-rings are sized using AS568; which is an aerospace standard that was published back in the seventies. The original AS568 was published in 1971, since then there have been several versions A through D – you can see the years that they were published – 1995, 2001, 2008, and 2014 is the latest.
When we talk o-rings, we are of course are talking about circular seals, but more importantly, when we say o-ring, we mean the o-shaped cross-section – or round cross-section. Because a number of seals are round but the cross sections are not circular. So today we are talking about the circular cross-section seals.
O-rings are sized using three different dimensions. There’s the ID which is the inner diameter. There is the OD which is the outer or outside diameter, and then there’s the cross-section which is also known as the width. So those are the three dimensions that are most important.
As we look at the standards themselves. The first thing that I do want to clarify is that sizing is important when it comes to o-rings. You can get close to dimensions, but you want to be careful because those o-rings are sized using a squeeze and fill percentage, which is critical in sealing and it does affect the overall sealing of the application. So if you have questions about the groove or groove size console an application engineer.
The AS568 dash numbers refer to a specific ID and cross-section. The OD is listed when you look at tables, or in particular when you look at the tables for a manufacturer, but it’s not necessarily needed.
The AS568 numbers are uniform across manufacturers. It does not matter who. You are going to see the same dimensions and you’re also going to see the same tolerances.
So in our example, we’re looking at a -001 o-ring. That o-ring has a 1/32 inch ID and it has a 1/32 inch cross-section. We can look at it from a nominal standpoint and we can also look at it from an actual standpoint. That actual standpoint- it would be an ID of .029 inches with a tolerance of + or – .004. The cross section would be .040 with a tolerance of .003 inches. Again, this is universal across manufacturers. As long as it’s AS568 referenced, all of these dimensions should be the same.
As you look at the dash numbers – as they go up higher you get into thicker cross-sections. So for instance, a -120 o-ring is 1 in by 3/32 of an inch – actually translates to .987 + or – .010 with a cross section of .103 + or – .003.
Again, use the table, use any manufacturer – we have Parker in particular – use the table provided by the manufacturer and then use tools. We don’t necessarily recommend a caliper, although you can get close to those dimensions. Also, you can use an o-ring cone which would get you the actual AS568 reference for that part.
All in total, there are 349 AS568 dash numbers that are available – these are tooled-up. Any o-ring manufacturer should have 349 sizes available.
Now in addition to your standard AS568 dash numbers, when it comes to o-rings, I do also want to reference that there are 3-9XX sizes available – few limited. These are considered boss or the tube-fitting o-rings – sized specifically for threads and the end tube fittings.
There are also nonstandard sizes that are available, and those sizes depend on the manufacturer. They’ll be a list of available sizes that have been tooled up outside of the AS568 references.
One last note as you’re talking about AS568 sizes, do keep in mind that tooling for these sizes at manufacturers is unique to the material or the compound that the o-ring is made up of and this is due to shrinkage rates and shrinkage rates on material varying. So the tooling itself will have to vary with that.
I appreciate the time. If you have any questions, feel free to contact our website at espint.com or industrial seal.com.
2 of 5
The Society of Automotive Engineers (SAE) and the American Society for Testing and Materials (ASTM) established ASTM D2000 to help provide guidance when determining elastomer compounds. By using a method called the “line callout,” engineers have a readily available classification system.
Andrew Rommann breaks down the individual elements that compose this “line callout” and the benefits of using this method.
Click on the image to open in a New Tab.
In many elastomer products, ASTM D2000 is utilized as the standard to communicate the performance requirements of the materials based on the customer’s expectations or the demands of the application.
An ASTM D2000 line callout looks like this and is the entire line that you can see between my arrows. We have a specification that applies to the line call out. We have some basic requirements information and then we have what’s known as the suffix requirements portion of the line call out.
So, within the line call out, within the basic requirements:
So we’ll see that the durometer if that is a specific target you’re going for, you may need to add an additional suffix requirement to explain that.
What you see on the left-hand side of the line call out – this is actually the minimum requirement that you need to specify an ASTM D2000 material. With this requirement, there is a set of basic requirements automatically imposed regardless of the grade of the material and without the existence of any of the suffix requirements. Those basic requirements include tests and performance results for heat aging, oil immersion, and compression set.
The suffix requirements as you can see the line call out is actually the greatest portion. What I have written on the board is the longest standard line call out that you could come up with for a M2BG710 material. This is a nitrile compound. Grade 2 correlates to the performance results of each one of these tests. For a grade 2, a grade 3, grade 4, 5, and 6, each grade will have different applicable suffix requirements. It will also have different levels of minimum performance to qualify as that grade of material.
In the suffix requirements section we see that we have a preceding letter or set of letters for each suffix requirement.
These special requirements are very powerful to help clarify specific items that may be required by a manufacturing process. A typical Z could be in this case for Z1. I wanted to clarify that the seven in the durometer call out is actually applicable to a durometer of 75 plus or minus five. I wanted to make that clear so I added the Z1 call out for that.
Z2, this could be the special processing in the manufacturing that I was referring to so maybe this elastomer component goes on to an assembly that goes through a paint line and ultimately through a paint oven there could be a small degree a small amount of time short duration or we have an elevated temperature and you wanted to evaluate the effects of that Temperature of the paint booth on the elastomer itself. So in this case, I’ve included a Z2 call out to say this ASTM method D 573 and I want to check it one hour at 125 degrees Celsius.
And then Z3 in this case. I wanted to come up with something a little bit out of the ordinary and this one I wrote down is must smell like vanilla birthday cake. It’s very unlikely that you actually need your product to have a certain fragrance, but it is possible to create a Z call out to impose any special requirement of any kind on the material. Keep in mind in doing that, you can prescribe a Z call out that is impossible to meet or could have a major cost impact on the overall material price.
So with these Z callouts, you want to make sure that you’re using what is applicable to your needs and not imposing anything above and beyond your requirements on the material.
Some additional suffix letters are shown here. In addition to the ones that I’ve had this particular call out did not include a C12 call out and the C suffix would indicate an ozone resistance test. You could also have a G call out which is an air resistance test and there’s a small list of additional suffix letters that correspond to different types of tests that can be applied to different types of material. The combinations of grade, type and class could have a different list of suffix letters applied.
So with all of this, based around the ASTM D2000 standard, and included on your drawing the major benefits of using it –
So with those things defined -both the grade, type, and class – along with the ASTM D2000 suffice requirements, we know exactly what tests need to be performed on the material and what the minimum requirements of those tests need to be to qualify for this requirement. It provides very clear information to the design team, to the manufacturer, and also to the quality assurance teams for products.
In most applications, to create an effective seal an o-ring must be slightly smaller than the groove it sits on – stretched during installation. But how much can it stretch and still be effective?
In the second part of Not All O-Rings Are Created Equal, Don Grawe covers elongation – how it’s defined and how it’s impacted by the base polymer, hardness, and curing process.
Click on the image to open in a New Tab.
Welcome to our next installment of how all o-rings are not created equal. Today we are going to talk about elongation and to start us off we’re going to use the trusty Parker handbook to give their definition and then go from there.
Parker defines elongation, as it pertains to o-rings: as an increase in the length expressed numerically as a percent of initial length. It is generally reported as ultimate elongation and further, like tensile strength, elongation is used throughout the industry as a quality assurance measure on production batches of elastomer materials.*
Okay, what’s all that really saying?
What we need to know when looking at how far an o-ring can stretch is:
The primary reason that’s important is especially the smaller the o-ring gets, there are applications where you have to be able to stretch it over a fitting, over some type of a ledge to get to the groove that you ultimately need to be able to get to.
Always consider the elongation when it comes to installation. With that, elongation gets measured by ASTM requirements in our physical properties. We talked about compression set, when we talked about tensile strength, we talked about abrasion resistance – they all fall within the ASTM D2000 requirements.
How that affects the different materials and the ratings for elongation. Let’s look at three of the most common materials that you work within the industrial marketplace. That being nitrile or NBR, FKM or fluorocarbon, and EPDM.
A 70 durometer NBR o-ring, typically by ASTM D2000 requirements, will stretch 250%. In practicality, most quality compounds will exceed that significantly. In the case of a 70 durometer EPDM, you’re looking at a 200% stretch. Again, most compounds, quality compounds, will exceed that by a good measure.
The closer you get down to 100%, the tighter those variances will be. With an FKM or a fluorocarbon material – 75 durometer very common – now you’re looking at only 150% stretch. So from a pure base polymer standpoint standard compounds, these are your norms.
Let’s stay with the two common NBRs and FKMs. If the only thing I change is hardness and go from a 70 durometer nitrile to a 90 durometer nitrile, look what happens to my elongation. Now, my elongation is only 100%. Common applications for fittings and couplings etc. 90 durometer is a common durometer because of the properties that that hardness will give but also understand you can’t stretch it near as much as you can the 70 durometer nitrile. In the case of fkm also a significant hit from 150% for 75 durometer to only 100% for the 90 durometer.
Let’s throw one more variable in there. If compression set is a significant factor in your application, a peroxide cure o-ring is a common go to when it comes to compounds and materials.
Well, a sulfur-cured 70 durometer nitrile is really the baseline that we were comparing over here at 250%. But by going to a peroxide cure material, higher compression set material, one with a little higher tensile strength – look what happens to your elongation at the same 70 durometer hardness. Drops from 250% down to 125%
Base polymer, hardness, curing process all have an impact on elongation.
There’s some give and take as there is with most compounds and most elastomers when you’re trying to reach a particular physical property within that. Elongation just like any of the others.
To get performance in one property sometimes means giving up performance in another physical property.
You want elongation, you got to stretch this part significantly without breaking, but in order to get that based on the other criteria:
Know the basics – elongation- how far can I stretch but what else do I need my material to do?
Again, not all o-rings are created equal!
SOURCES: *Parker Hannifin Corporation, “Parker O-Ring Handbook – ORD 5700”
2 of 5
“Shelf life” is the maximum time (beginning with manufacture date) that an o-ring or elastomeric seal – with proper packaging and storage, becomes unable to meet its original specifications.
Aerospace Recommended Practice (AP 5316) is the most comprehensive basis for establishing shelf life, however, it is not a binding specification.
Let’s look at a few Methods at which shelf life is being calculated.
Click on the image to open in a New Tab.
Some materials are listed as unlimited, however unlimited is not equal to forever. For some companies, they use 25 to 30 years as unlimited shelf life. According to the study EPRI NP-6608, which is one of the references in ARP 5316, which replaced by Aerospace Standard 5316, in the study it talks about five different methods to obtain elastomer shelf-life.
It is acceptable to use field experiences data or lab test data, however, using military standardization handbook or rubber products or warranty, then other methods are recommended.
Data is collected by products being stored for a number of years in an average room temperature. This method is time-consuming and yet it is a case-by-case basis.
Shelf life can be obtained by using this extrapolated graph or EST equation, which requires materials in qualification data or using EDT equation and only materials data is needed.
By using a test chamber at a given temperature, data can be obtained by measuring critical properties periodically until data drop below acceptable values. Then the EST equation can be used to estimate the shelf life.
Appendix B is where roughly 70 generic materials were summarized by using Method A, B, and c and is listed in years.
A premature o-ring failure is often times due to a combination of root causes. Let’s dive into one of those causes – Compression Set – what is it, how to diagnose and how to keep it from happening.
Click on the image to open in a New Tab.
I will start by saying not all o-rings are created equal; different materials, different hardness’s, different curing processes, they’re not just round, they’re not just squishy and just to fill in a hole. Having said that many of you also know a lot of the basics when it comes to o-rings, but one of the things that I want to focus on today is what Parker Hannifin states is the number one cause for o-ring failure and that is compression set.
Compression set is a term that’s used for one of those physical properties that gets measured with every batch of material that gets produced by reputable o-ring manufacturers. Which is maybe clue number one. If you’re buying o-rings from a manufacturer that does not test every batch and every lot and has full traceability- that’s a red flag.
Let’s go on the premise that you’re working with people that do have full batch and lot control of every material that they produce. Tensile strength, elongation, specific gravity, TR10 values; these are all different physical properties that get measured. Compression set is one that, in my experience, [I will validate what Parker’s telling us] is probably the most important of all of them. If you don’t know what compression set is – in simplest form – it’s taking o-ring material squishing it between platens, applying pressure to squeeze at 25%, then baking it in an oven for a prescribed period. Taking it out. Undoing the platen, pull out the material and measure how much did it come back.
Now in a perfect world, it comes back all the way.
If it comes back all the way, it is taking no compression set. Also, factor in that when you design o-rings for seals you are utilizing a squeeze rate that will vary from 10%, 15%, to 20%. So, in the scenario where you have an o-ring that takes a compression set and it doesn’t come back 10%, 20%, 30% – what does that do to that design squeeze that you’ve put in there? So now similar to other variables in your design now, in essence, you have another form of tolerance stack-up.
So, let’s dive a little bit more into compression set. Compression set values vary widely by material. You use different materials for different applications. From NBRs for a lot of hydraulic applications [oil/petroleum uses in the like]. You use HNBR for lot of the same things, but when you get involved in higher temperatures or different types of fluids some compatibility issues. You use EPM, again for different applications often for ozone resistance external uses for brakes in some of those applications. FKM is used often for chemical compatibility and for high-temperature applications. Probably the four most common o-ring materials in the industrial market. Yes, there’s silicone, AFLAS, Neoprene, and others, but these are the four common.
Within these four common materials is a variance in the amount of compression set they just inherently take. Start with NBR. One thing to note about NBR is how it’s processed. There is a difference between a standard sulfur cure – which the world has known for decades – and a peroxide cure material. A peroxide cure material is a step up from the sulfur cure and one of the reasons for that, [you can see in the illustration here] is the compression set. Under standard hundred degrees C testing for 70 hours, you can expect a good sulfur cure to still have 15% to 25% compression set. Peroxide cure material you’re going to reduce that to 10% to 15%. Significant Improvement for a lot of face seals, for both static as well as dynamic applications.
You go to an HNBR which is considered an upgrade in the NBR family, and you’ll see very similar compression set characteristics of the peroxide cure. If compression set is all that we’re dealing with peroxide cure might be an alternative because there’s also a premium on price to go to the HNBR.
Another reason to go with the HNBR would be for higher temps and for other application concerns. Look at EPDM. Again, EPDM, something very similar to the sulfur cured – 15% to 20% compression set on EPDM material. FKM which gets tested at 200 degrees, so you’re not exactly comparing the same thing, but because of the higher temperature material is often tested at 200° C – you’re looking at 15% to 20% compression set. Now I want to point out that these compression set figures are also going to vary on what and how the methods are being used to test it. Slabs, buttons, dog bones – often when you’re just testing material, it’s not done with the actual o-ring. Which brings us then to the next set. When you’re looking at compression set values by o-ring, you’re going to get some wide variables as well.
Taking an o-ring material that has been molded into a shape, gone through all its curing processes and everything like that, you’re going to get variance. If nothing else do to sheer size alone. A very thin small cross-sectioned o-ring to a big fat o-ring. What are those values going to differ? You are going to see significant differences from the o-ring to just the material. At a hundred degrees for 70 hours a standard NBR 1/16th cross-section can take a 100% set. The fatter or thicker cross-sections take a 60% set.
You go to a peroxide cure, the thicker cross-sections not quite as much difference, but when you get to those thin cross sections, there’s a lot of applications out there that use the small cross-section, 90%. That 10 points can make a difference of sealing and not sealing. HNBR again a little bit better improvement. 85 to 60. 85 being the thin cross-section.
So, wide variances when you’re looking at physical properties and when you’re looking at compression set from your suppliers, understand what are you are looking at? Am I looking at raw materials, slabs, buttons, dog bones, or am I looking at actual o-rings? What are those figures and those values that you want to be able to measure to – size and cross-section do matter. Where I’m going with this – beware of low-cost materials. Not all o-rings are created equal. There are differences and low-cost materials often times will impact your compression set values. Know what you’re getting.