o-ring width 2

[kakalee1]

Hi,

I used the Parker o-ring handbook for most of my o-ring groove design. I just have one question regarding different o-ring width.
What is the advantage of using a smaller or bigger o-ring? Imagine if the ID of the o-ring groove is 1 inches, is there a different if I use a 1/16, 1/8, 3/32, or 3/16 width o-ring?

All feedback are greatly appreciated.

 

[Compositepro]

Thicker o-rings can accommodate more movement or variation in the gap they are sealing. The important factor from the o-ring's perspective is percent compression in the rubber.

 

[KENAT]

Yep, Compositepro has what I'd think is the main point - the compression ratio.

There can be secondary factors like available space, strength of o-ring if it needs to be stretched during installation...

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[kakalee1]

Thanks guys,
Since you brought up compression ratio, I assume it is the same as Squeeze percentage on Parker handbook. In term of stretching the o-ring during installation, is there any guideline on how much should the o-ring be stretched to achieve certain compression ratio? I am having trouble find this info.

What I usually do is use the nominal ID of the o-ring as my ID for the groove, and get the groove width and depth based on Parker handbooks. I noticed that the o-ring only stretches very little. I haven't run into any problem by doing so, but I want to make sure that I am doing it right, not by dumb luck.
Example: take O-ring # 131 as example, it has an actual ID of 1.674, so I make the ID of my groove to be 1-11/16 = 1.688, slightly bigger than the ID of the o-ring. Would this provide enough stretch for the o-ring?

 

[KENAT]

I thought if you looked at the appropriate section of the Parker handbook it's pretty explicit on the ID of grooves etc. - but it's been a year or two since I looked at it and my application was atypical so I had to take everything in there with a pinch of salt.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[CoryPad]

Isn't there a third part that completes your system? The dimensions of that part would determine the compression in the ring.

 

[Compositepro]

In service, the rubber should always be in compression so that any flaw or nick will be squeezed closed rather than pulled open. This is why o-rings can be cut from stock cord and the ends do not even have to be bonded together to create a seal. Rubber under tension will always eventually fail by cracking.

 

[kakalee1]

I guess I am not using the Parker handbook correctly then. Most of my application using the face seal gland. According the the book, for external pressure, I should dimensions the groove by its inside diamenter H = Mean I.D of Oring ( see table 4-1).
However, I can't find mean I.D of o-ring from table 4-1 at all.
Can you give me some guidance on this?

 

[kakalee1]

Here is table 4-1

 

[dineshakabari]

In case of dynamic seal (sliding) bigger dia O-ring is better to use to have longer service life. As per my understanding smaller dia O-ring need much accurately manufactured metal parts. So in this case if you use larger dia O-ring then there is more compression is possible (in terms of microns) since the percentage compression range in both the cases are same (approx. 10-15% for rubber elastomers). If you do not have such criteria then better to select middle of range (not smaller or bigger). In this case you may optimize the situation.

 

[Timelord]

kakalee1,

Oring squeeze is always a percentage of the x-section, therefore the bigger the x-section, the larger the actual squeeze. The larger the squeeze the more forgiving the seal is to flatness variations in the sealing face, and to scratches somewhat, due to the wider contact width. Another consideration is bigger orings are simply easier to handle, the groove is easier to wipe clean and things like that.

Timelord

 

[3DDave]

The big ones take more space but the little ones are easier to damage and more difficult to work with.

One project I worked with used a rectangular groove (face shape) for a box. I put in a large groove for a large section o-ring but this was changed by later workers to a very small groove**. Pretty much ever single gasket got pinched or cut because, as the lid was closed, the difference between almost closed and closed-enough to trap the gasket in the groove was tiny. For those not cut on install, whenever the lid came off, continued curing of the gasket would bond it a little to the lid and box. Being little it always broke. Not only was it uncertain that the seal actually sealed, it was so often cut on install that the production and spares items went into the prototypes.

A thicker gasket has n^2 times area for strength and n times area for sticking, so it is n times less likely to break and it's a lot easier to guide it into the groove in the first place. My gasket was 3/16; the substitute was 1/16th. They kept the flange the same size; just made the groove smaller, so no savings on size.


**Why, you ask? Because it was the same groove size used on a small connector and -that- worked, so why change the size when the groove length is maybe 30 times longer?

 

[tbuelna]

3DDave makes a good point about the damage tolerance of small section o-rings. I work in aerospace, and I have often seen design requirements that discourage the use of small section o-rings (like 1/16") due to the greater potential for damage during installation.

 

[KENAT]

kakalee1 Been a while and I don't have time to download the manual. However, I thought that either ID was given or you had to calculate it based on other dims given in a big table that gave all the standard sizes - I forget sorry.

Maybe someone that uses it more often can point you to it.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[dgallup]

Very small o-rings are a genuine PITA. The tolerances of the o-ring do not decrease below the 1.78 mm cross section and by the time you get to 1 mm there is almost no tolerance left for the mating parts. I have seen some companies state minimum cross sections, one automotive company we supplied would not allow anything below 2.5 mm. Bigger cross sections are much more forgiving as they can bridge over small imperfections that would cause a leak with a small o-ring. Larger o-rings don't have as high a % squeeze requirement either so they are less likely to take a compression set. Mind you, I have never used anything bigger than a 3.6 mm cross section so there may be issues with big o-rings I've not experienced.

Note that while it is always a very good idea to use standard sizes when possible, every o-ring manufacture makes hundreds of additional sizes and if you are using large quantities will want you to pay for your own tooling which makes if easy to justify any size your heart desires. Every o-ring we buy comes off tooling we purchased. O-ring tooling is cheap and quickly made compared to injection molding or stamping tooling.

I never use the design guide tables because I made a spread sheet that calculates the stretch and corrected squeeze for the full range of component tolerances over the full range of operating temperature.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.