External Circlip groove - effect of deviating from nominal widths/depths 1

[Nereth1]

Hi all,

I have found nominal external circlip groove dimensions at various sites, such as this one:

http://www.circlips.com.au/contentblank.php?sec=products&sec2=circlips&sec3=d1400

, which I assume are valid and come from DIN 471 (which I don't have).

I typically use the above dimensions in machinery designs without issue, but our main product line deviates (or we want to deviate it) significantly from the nominals - and I can't form a good argument either way. There are two scenarios:

1) Circlip depth

Our circlip depths are typically much deeper than the standard. The person who set it said that he did so by experimentally deepening the grove until the gap in the split was back to its pre-stressed size - i.e. there would be very little deformation in the circlip. The best reason I have to not do that, is that the circlip should not have radial clearance for fear of spinning around the shaft and wearing, or going skew from the shaft axis. Neither of these feel like strong arguments. By contrast one could argue that the shallower grooves in the standard are more prone to the circlip being forced out of the groove. What is the established knowledge here? What happens if the groove is too deep or does it not matter?

2) Circlip groove width

Currently we have an opportunity to enhance manufacturability by increasing circlip width, but again the question is what are the detrimental effects. I have noticed that where groove dimensions are listed, typically width is listed as a minimum with no maximum. Intuitively, while under load, the position of the non-load-bearing side of the circlip groove doesn't matter as it is not touching anything. However circlips are not always under load - some searching found stories of Honda NSX transmission failures due to circlips going sideways in an excessively wide groove. I can't quite understand that failure mode though so I am hesitant to assume it applies to us. Again, what happens if the groove is too wide, in an application where it is not always loaded, and in fact the item it is retaining can happily retract a few mm from the circlip groove (thus allowing the circlip to shift)?

I wander if the two items are related - the circlip cannot go skew on the shaft unless both the depth and width are too high?

Thank you in advance for your responses and expertise.

 

[tbuelna]

Slightly excessive width of an external retaining ring groove only matters if axial play in the retained components is an issue. The ID of the groove might be a problem if it is excessively undersized, since the ring could shift off center enough to significantly reduce its shear capacity. The flange OD of the groove for an external retaining ring is critical to prevent overstressing of the ring at installation. It is also important to ensure the groove fillets are sized correctly to prevent excessive contact stress from the sharp ring edge bearing against the fillet surface. Lastly, if the shaft is rotating at any appreciable speed and there is radial play between the ring ID and groove, there will be wear of both parts as the ring rotates about the groove.

 

[Nereth1]

Everything you say makes sense. I also got a copy of DIN 471 to check and it is covered there - for single sided forces where positioning doesn't matter, the groove can go wider.

Indeed it does also make sense that if the groove diameter is small enough that a clearance opens up, that would allow the circlip to go off center and potentially be an issue.

I did not realise that the groove fillets are important - we have tiny chamfers on the edges, but only to deburr them so that another part can slide over without being damaged. I didn't think it was critical for the circlips. Since I presume circlips are a good rockwell 40-50 or so, I would think that against a groove in soft material(we are using low carbon steel, yields of around 300-350MPA), any burr or hard edge would mostly just take care of itself?

 

[Tmoose]

Have you had any issues with wear or clip loss, or components shifting in service? Are service conditions going to change?

============

For years some engines have used some form of circlip or retaining ring to retain the wrist pin ( gudgeon to some of our brothers and sisters) in the groovy piston.

In the 60s and 70s race engines derived from V8 passenger cars went different ways in the search for reliable writs pin retention, which was lacking at higher rpm.

Double truarc type retaining rings installed sharp edges outward in opposite dimensions, spirolox,round wire clips with and without removal tangs, round wire clips with beveled wrist pin ODs.

For a while small block Chevrolet engine builders advocated minimal wrist pin end play ( 0.000 - 0.002" ) with room temperature components. The LR grooves in the piston could be quite wide, and the wrist pins were then adjusted to achieve nominal zero end play. The stated concern was extra clearance permitting a battering ram effect and increasing axial loading.
Here is a link to the Lemans winning Ford GT 427 engine development.

http://www.gt40s.com/images/howto/LynnMiner/Reference/SAE_Papers.pdf

Page 32/34 discusses issues with lock ring loss, and the use of experimentally determined end play to eliminate it. Note the minimal endplay to reduce battering was tried first, unsuccessfully.
The difference in philosophy could easily be attributed to piston design differences.

No mention is made of lock ring groove depth. My presumption is Standard groove depth (OD) was used.

My recollection is The Waldes literature did not specify or even recommend a particular orientation of the LR, regarding sharp edge in or out, but hot rodders were keen on sharp edge out to "dig in."

I can picture excessive bevel on a shaft OD or chamfer on the bor ID causing the ring to bow and causing problems, like trying to squirt the LR inward, then out.
And of course At speed the acceleration at TDC can be well over 4000 gs.

 

[Nereth1]

No issues with wear or clip loss at the moment. Tough to disassemble them when necessary, which is one reason I am proposing we go to the nominal groove depth instead of the 'relaxed circlip' method we are using, which is notably deeper.

I also imagine that the slightly deeper groove we use could put extra stresses on the shoulder, as our shoulders skirt the minimum allowed by the standard.

But we have a range of products with several different nominal circlip sizes ranging from Ø25-Ø50mm, and I am going to suggest that all of them can be cut with the same tool: The one for the 50mm circlip. All of these clips are securing on one side only.