|Brian Pritchard||05/08/2019 17:14:37|
|17 forum posts|
I have hunted everywhere for information on the size of the groove required for fitting O Rings in pistons. The information on the internet is comprehensive about sizes etc but I can't find a straightforward answer.
I have 1.25" OD , 1" ID o rings. I propose using a groove .189" wide and a depth of .112" (.125 less 10%)
Does this sound about right?
|Bob Stevenson||05/08/2019 17:32:00|
|295 forum posts|
it depends on what this is for, but when I was making piston heads for air rifles using white engineering plastic to basically copy the BSA items I kept the 'O' rings so that there was one third above the side surface of the piston...and that the 'O' ring was a nice fit in the bore with a little compression. They worked very well n the event although there was, as I remember, some R&D in getting the ring compression just right for maximum speed with minimal air loss.
|8 forum posts|
according to my James Walker design notes for O rings
for 0.125" cross section dia o ring
groove width = 0.159" - 0.164"
max piston groove dia = min cylinder dia minus 2 x (0.110" - 0.114" )
min piston groove dia = max cylinder dia minus 2 x (0.110" - 0.114" )
The design notes are available on the net and are well worth hunting down.
Edited By AStroud on 05/08/2019 17:38:05
|duncan webster||05/08/2019 17:39:26|
2197 forum posts
You need to measure the actual cross section, some nominal 1/8" section rings are actually 0.139" designed to go into 1/8" deep grooves.
this link goes into the required squeeze in detail parker seals I suggest looking at the bit on pneumatic cylinders
|Jeff Dayman||05/08/2019 18:11:02|
|1591 forum posts|
Seal and Design here in North America have a good ref page set for groove dims. Link below.
15988 forum posts
For model steam engines and IC engines you don't want to squash the ring as much as most commercial suppliers show
For a nominal 1/8" ring of 0.139" a width of 0.160" and depth of 0.132" will seal without excess friction
taken from Model Engineers handbook and also the same sizes can be found in Reeves paper catalogue.
|438 forum posts|
There is not a straight forward answer. I have worked with o rings over the years in all types of design, for both static and moving designs. On some problem designs I have worked with the manufacturer who comes up with groove dimensions for the specific operation. While the easiest solution is to work to the standard groove dimension these can be modified, particularly stretching an o ring to fit in a size larger piston groove. The main dimension to work with is the extrusion gap between id and od of bore and piston according to the exerted pressure. The manufacturer catalogues contain all the basic information so start from there, o rings are most forgiving of errors except for surface finish if longevity is a requirement.
|Brian Pritchard||05/08/2019 19:09:47|
|17 forum posts|
Thanks for the replies. I have checked and the diameter is 0.139'. i will try the values you suggest, jason.
|John Reese||05/08/2019 22:16:09|
|772 forum posts|
If you are using the O ring for a model engine I don't think you want to follow Parker's groove dimensions. Too much friction. Parker specs are pretty much the bible for industrial O ring use.
|duncan webster||05/08/2019 23:41:41|
2197 forum posts
Jason's figures give a squeeze of 5% which is what Parkers say as a minimum for pneumatic psiton
|Neil Wyatt||06/08/2019 10:13:47|
16415 forum posts
I recall Tubal Cain explained the key issue is making sure the groove is wide enough to allow the ring to expand sideways as it is compressed, but not wide enough to encourage rolling.
|Brian Pritchard||06/08/2019 13:22:12|
|17 forum posts|
I was always told that you must have the groove wide enough for the O Ring to roll.
|Graham Williams 12||09/08/2019 22:32:50|
|25 forum posts|
I have done lots of research over the years for both work and for air rifle design - as you have found information does vary.
My understanding is that both the groove width and the groove depth are important. Obviously you need some compression on the ring to seal. I also always use the James Walker guide - downloadable as a .pdf document.
This gives you a tolerance for groove depth. I would stay within this tolerance to achieve adequate squeeze - just err towards one end of tolerance or the other, depending on wether low friction or a good high pressure seal is more important to you. Also if the groove is within a piston you should take the tolerance of the cylinder it is running in into account, again to make sure you achieve the necessary squeeze. (Andrew has provided the necessary numbers above).
The groove width is important. As Neil says the groove should be wide enough to allow the ring to expand sideways, but most importantly, there must be clearance, after the ring has been squashed and expanded sideways. The air must act on the entire cross sectional area of the ring to make a seal - it seals by virtue of the fact that (because of the initial squeeze) there is always more force acting radially. If the groove is not wide enough (so the ring is also being squeezed against the groove sidewalls) you can imagine that as pressure builds it can't act on the sidewall face of the o ring and therefore will tend to compress it radially inwards, thereby letting pressure slip past the ring, hope that makes sense?
You will note in the James Walker information that only certain sizes of o ring are recommended for dynamic applications (marked with an Asterix in their tables). I am pretty sure this is because you don't want the ring to roll in the groove....
Edited By Graham Williams 12 on 09/08/2019 22:35:57
Edited By Graham Williams 12 on 09/08/2019 22:48:42
|827 forum posts|
Re size of O ring grooves, check out this website :
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.