O-ring material and sizing?

[KerryH]

Hey guys, without getting into the specifics of it all, I've got an oil pump that goes to an engine I'm rebuilding that isn't a common engine. I need to source some o-rings that will work for the oil pump as the oil pump looks like this:


front


back (with o-rings in question)


This pump is internal and chain driven and can be seen here The relevance in this is in the purpose of the o-rings in question)


This is where the back of the oil pump goes and the o-rings seal. It is just machined flat aluminum.

The grooves I measured as:
Smaller
1.050 OD
0.842 ID
0.055-0.060 depth
~0.100 width

Larger
1.125 OD
0.927 ID
0.055-0.060 depth
~0.100 width

Now I'm assuming the oem o-rings were silicone since they're red/orange.

taking a look at this chart: http://www.oringwarehouse.com/files/documents/oring-sizechart.pdf

Does a silicone -118 and -119 make sense here or is that too thick for the groove?

Would a different material and or size work better?

 

[A_Pmech]

The OD dimensions sound about right. For statically sealing pressure from the inside, the OD of the o-ring should contact the outside diameter of the boss by 1% to 3% of circumference.

The squish seems VERY high though! Normally, squish on an axial joint is only about 20% of cord diameter more or less. If squish is too high, you run the risk of cutting the o-ring when it extrudes out of the groove during installation.

 

[Techie1961]

I would do a quick calculation on the area for the O-Ring compared to the groove. Do a measurement of the groove width and depth and make sure that you don't put too much O-Ring in it but still provide enough compression. Like A Pmech says, you don't want to extrude the O-Ring out the sides of the groove.

 

[gorilla]

The design rule of thumb that I've used for years is that the depth of the O ring groove is ~80% of the cross section of the O ring deep and ~ 120% wide. The O ring should be a snug fit to the side you wish to seal. The Parker O ring manual is a good source for O ring material selection and design data.

 

[Hpozzuoli]

I never knew there was a science to O rings.

 

[zkling]

I never knew there was a science to O rings.

There is since behind almost everything, especially mechanical. Parker puts out a good O ring hand book that is a crash course on seal design of all sorts. As gorilla brought up.

 

[duc916]

Those might be metric, depends on the engine. You should be able to use Viton (that is what we use on all our o-rings that come in contact with oil). You can look up the o-ring chart from parker on the net, search for "parker o-ring handbook". In my book on page 4-14 is where they have face seal glands, which is what you have. Mcmaster Carr has a great selection if you do not have a local source.
You are close to a 2-021 and 2-022 size.

 

[dlcwent]

And who says you don't learn something new every day. ????

 

[teamdfl]

The first place I look for o-ring info is the Parker O-Ring Handbook.

http://www.parker.com/literature/ORD 5700 Parker_O-Ring_Handbook.pdf

 

[yaidunno]

-020 and -022 is what i would use in that application. Buna-n is my first choice in oil, vition should work as well. The parker book is a great resource. Pick up a copy if you can. McMaster sells what you'll need.

Looks like somebody took something too aggressive to the sealing surface

 

[Techie1961]

The design rule of thumb that I've used for years is that the depth of the O ring groove is ~80% of the cross section of the O ring deep and ~ 120% wide. The O ring should be a snug fit to the side you wish to seal. The Parker O ring manual is a good source for O ring material selection and design data.

That is probably a good rule of thumb but like you mentioned, you need to look up the dimensions. There are many ways that an O-Ring can be utilized such as on hydraulic/air cylinders, stationary compression on a flat face, moving compression on a flat face, stationary on a shaft, rotating shaft, linear movement attached to the shaft, etc. Each one has a different groove design and advised compression.

I consider the application that the OP is referring to as the least worrisome. No movement and completely enclosed as the two metal surfaces meet in a planer fashion. There is no chance of it being damaged during assembly like you would have when you slide a shaft into a hole.

 

[Fueler]

I consider the application that the OP is referring to as the least worrisome. No movement and completely enclosed as the two metal surfaces meet in a planer fashion. There is no chance of it being damaged during assembly like you would have when you slide a shaft into a hole.

Agreed that it is the least worrisome but too thick of an o-ring can distort or even cause enough resistance to break the item being torqued down. uhh, let's just say "been there".

 

[Aaron10647]

Not a perfect fit but I would slap a 2-020 and 2-021 in those grooves and it should be close enough. Stick with a .070 diameter o-ring with those gland depths. Common material for oil is Nitrile NBR.

 

[KerryH]

Awesome, this is why I asked you guys!

I've replaced o-rings before, but they're never been in a groove like this against a flat surface so that was throwing me off a little.

Thank you so much for the replies, it has helped tremendously!!

 

[bggrnchvy]

Looks like somebody took something too aggressive to the sealing surface

That.

Spec for static liquid o-ring sealing surfaces is 32rms max IIRC.

 

[yaidunno]

That.

Spec for static liquid o-ring sealing surfaces is 32rms max IIRC.

Surface quality is something that a lot of people don't seem to understand when it comes to sealing. I can't begin to count the number of heads that i've seen needing .015-.020 material mowed off because someone went at it with a 90* die grinder and a roloc pad.

In this application, i don't really see why the surface needed any cleaning? I'm assuming that it is just the 2 o rings, an no further gasket?

 

[DenisG]

The application may use this type of o-ring:
http://www.epm.com/encapsulated.htm

Plain silicone o-ring seems to be wrong choice.

 

[driz]

Take this FWIW but I discovered the holy grail of O rings when I had my dozer. Go to your local Cat heavy equipment shop. Not the truck shop though they probably have them too but not as many, the dozer / excavator repair shop. Take the part and ask them if they have any o rings to fit that. I once ventured into the back room of their part room. They had them covering an entire wall laid out hanging on nails on a shadow board. There was over a thousand. I always remembered that for the day I had a weird o ring that needed replacing.

 

[KerryH]

-020 and -022 is what i would use in that application. Buna-n is my first choice in oil, vition should work as well. The parker book is a great resource. Pick up a copy if you can. McMaster sells what you'll need.

Looks like somebody took something too aggressive to the sealing surface

That.

Spec for static liquid o-ring sealing surfaces is 32rms max IIRC.

I suppose that was my local machine shop when I had the block hot tanked =/.

So I went and got a -020 and -021 o-ring from a local industrial supply for a couple nickels. When I got back to the house I test fit them on a spare oil pump and clamped a piece of 1/4" plexi to it so that I could see how it was fitting and I don't think the -020 and -021 are the best choice:





You can click on the pics to see a higher res version, it's hard to see but you should be able to make out how the o-ring contacts the plexi. The plexi has some flex away from the clamp so the best representation is closest to the clamp.

Basically on the larger groove, the o-ring is definitely too thin. It fits well with just a tiny amount of stretch that allows it to stay in place once placed in the groove.

On the smaller groove it appears to be decent, but I'm not sure. It does not stay in place meaning the ID of the o-ring is greater than the ID of the groove.

I'd love to hear further input, I will definitely try to swing by the supply house tomorrow and try another size.

 

[yaidunno]

Your not looking for the grove to hold the o-ring in place. The grove is larger then the profile of the o-ring to allow it to "bulge" as its tightened. The depth being .060, and the profile of the o-ring being .070 means that, in theory, you will have .010 of compression.

Now, the reason i suggested a -022 is, like A_Pmech stated, when sealing static pressure from the inside, the o-ring should make contact with the OD of the grove. This will prevent the o-ring from expanding funny under pressure. It will in essence be "caged" in place.

It does appear that both o-rings are making contact in your photos. You can also tell the plexi is flexing looking at the contact area. Is this motor metric? They do have metric sized o-rings if needed.

Hope this helps...

 

[KerryH]

Your not looking for the grove to hold the o-ring in place. The grove is larger then the profile of the o-ring to allow it to "bulge" as its tightened. The depth being .060, and the profile of the o-ring being .070 means that, in theory, you will have .010 of compression.

Now, the reason i suggested a -022 is, like A_Pmech stated, when sealing static pressure from the inside, the o-ring should make contact with the OD of the grove. This will prevent the o-ring from expanding funny under pressure. It will in essence be "caged" in place.

It does appear that both o-rings are making contact in your photos. You can also tell the plexi is flexing looking at the contact area. Is this motor metric? They do have metric sized o-rings if needed.

Hope this helps...

The smaller o-ring appears to be sitting very well and may be contacting the outer wall of the groove when compressed, the larger o-ring is definitely not (you can slightly see it at the very bottom of the o-ring in the first picture).

The engine is a japanese engine, and most parts of it are metric. However the measurements I have don't appear to convert to metric very well but I'm not familiar w/ metric o-rings.

 

[EdT]

As suggested above, the Parker book is a wonderful source for O-ring info. One thing to keep in mind is that the info in the Parker book is usually for the max pressure situation which can be thousands of PSI. Most likely the two ports on your oil pump will not see anything like that kind of pressure. The one on the inlet/suction side will see about -15 psi suction MAX and the output side will see a couple of hundred pounds MAX and probably more like <100 depending on the relief valve in the oiling system. So, in O-ring world, this is a pretty easy application. Most designers will try to use a standard O-ring size because it makes their life simpler. They don't have to design an O-ring and the part it goes into. Some marketing guys like the special part because they can sell it for big bucks, but I doubt that fixation extends to o-rings. If the engine you are working on came from Japan or Continental Europe or anywhere except the USA and Great Britain or its colonies, the O-rings are probably metric. Later stuff from USA and GB is tending toward metric since that's what the rest of the world uses. Very early Asian (Japanese) stuff may not be metric, but that looks like a fairly new engine design.
Unless the O-rings are special, the cross section is going to be standard and the diameter is going to be standard. You're not going to find a metric o-ring with a 1.732 mm cross section and a 27.471 OD. It's either 1.5 or 2 and 27 or 28. Even if it's a special, it's very likely that the cross section is ordinary. Even though it's tricky, try measuring the O-rings themselves. The red color MAY indicate silicone although I can't think of a reason to use it in your application. I know silicone does not like gasoline or other high end hydrocarbons, but I think it's OK with oil. Hope you find what you need. BTW the squeeze on your samples from the CAT store looks a bit light (subjective opinion).

 

[KerryH]

As suggested above, the Parker book is a wonderful source for O-ring info. One thing to keep in mind is that the info in the Parker book is usually for the max pressure situation which can be thousands of PSI. Most likely the two ports on your oil pump will not see anything like that kind of pressure. The one on the inlet/suction side will see about -15 psi suction MAX and the output side will see a couple of hundred pounds MAX and probably more like <100 depending on the relief valve in the oiling system. So, in O-ring world, this is a pretty easy application. Most designers will try to use a standard O-ring size because it makes their life simpler. They don't have to design an O-ring and the part it goes into. Some marketing guys like the special part because they can sell it for big bucks, but I doubt that fixation extends to o-rings. If the engine you are working on came from Japan or Continental Europe or anywhere except the USA and Great Britain or its colonies, the O-rings are probably metric. Later stuff from USA and GB is tending toward metric since that's what the rest of the world uses. Very early Asian (Japanese) stuff may not be metric, but that looks like a fairly new engine design.
Unless the O-rings are special, the cross section is going to be standard and the diameter is going to be standard. You're not going to find a metric o-ring with a 1.732 mm cross section and a 27.471 OD. It's either 1.5 or 2 and 27 or 28. Even if it's a special, it's very likely that the cross section is ordinary. Even though it's tricky, try measuring the O-rings themselves. The red color MAY indicate silicone although I can't think of a reason to use it in your application. I know silicone does not like gasoline or other high end hydrocarbons, but I think it's OK with oil. Hope you find what you need. BTW the squeeze on your samples from the CAT store looks a bit light (subjective opinion).

I would typically measure the o-rings that were used and find the closest match, but the o-rings that were used were in the engine for 100k++ miles and they are holding their "squished" shape which makes it very flat.

I definitely agree that a thicker o-ring is in order, I will try some others from my local industrial supply (the CAT store was another poster).

 

[gbsmithy08]

Check with Mc Master Carr and tell them what you have. They are a great company to deal with and they have a great source to help you. You can even call them and tell them what you have and they can look it up for you.

http://www.mcmaster.com/#

(562) 692-5911

 

[KerryH]

An update to this thread, I spoke with someone that is more familiar than I with metric O-rings and this is what they had to say:

Often in axial face-seal applications like yours, you want 25% compression of cross-section and 2% stretch on the ID. 2mm cross section o-rings would give you 25% compression at 1.5mm groove depth. Using a 23mm ID o-ring will give 2.3% stretch for your 23.55 ID groove, and a 21mm ID o-ring will give 1.9% stretch for your 21.4 ID groove.