Cylinders Take a Tumble

In your April column, you strongly recommended against cylinder media. Yet, in the same and other publications, many other articles on media selection refer to the popularity of cylinders, and they often recommend them for hole deburring and general finishing. Can you provide specific reasons for your objection to this popular media?


Facebook Share Icon LinkedIn Share Icon Twitter Share Icon Share by EMail icon Print Icon

Q. I’ve been involved with vibratory finishing for many years, in three different companies, each of which had deburring applications using vibratory finishers. Two of the three companies always used cylinder shaped media, and one used flat triangles and cylinders. In your April column, you strongly recommended against cylinder media. Yet, in the same and other publications, many other articles on media selection refer to the popularity of cylinders, and they often recommend them for hole deburring and general finishing. My experience with cylinders goes back 20 years and the finishing results have always been favorable. Can you provide specific reasons for your objection to this popular media? P.W.



A. Cylinders do impart a nice finish, but I would argue that there are other shapes to do this without the many negatives associated with cylinders. Often, cylinders are recommended for deburring the ID of drilled holes in castings, or inside diameters of rings, and similar parts. Let’s take a closer look at this. One example might be a part with a through hole of 3/4-inch diameter. A 5/8 × 1-inch angle-cut cylinder can travel through this hole and debur the ID without getting lodged—that is until the media is well worn and reduced in size. That is apparently a typical media recommendation because I see a lot of similar applications in the field.

If you have a similar part and some media samples, you can follow what I’m going to describe. You can readily see that the cylinder needs to be long enough that it cannot turn sideways and get lodged. It is also obvious that only one cylinder can be in the hole at one time. Now, note how many simultaneous contacts the cylinder can make with the ID. The answer is that it has only one contact point at any given moment. And, with the exception of some deburring effect from part-on-part contact, the only deburring that takes place is when the media touches the burr area. The ID will not be deburred until enough time has passed for the total of all these single contact points to have occurred several times all around the circumference of the hole.

Now, consider if the media had been an angle-cut triangle, say 5/8 × 5/8 × 7/8, cut at a 45-degree angle. Again, only one triangle can be in that hole at a time. Note, however, how many simultaneous contacts the media makes with the ID. The answer is two. It stands to reason, and it is easily demonstrated, that deburring of the ID will be 200% faster with the angle-cut triangles.
Next, consider if the media had been a three-pointed star, say 11/8 × 5/16. When a tip of the star enters the hole, there are three or four simultaneous contacts. Furthermore, there will often be two stars working on the ID at the same time, making six to eight simultaneous contacts with the ID. So, you can expect ID deburring to be 300–800% faster with stars compared with cylinders.

Another application in which cylinders are often used is to debur the inside of larger diameter rings, such as transmission rings. Picture a large ring gear or spacer, with a 6-inch ID to be deburred. You will often see larger cylinder media, perhaps 7/8-inch diameter in these, or similar, applications. Let’s visualize what occurs to debur these parts in a vibratory finisher. Note that the maximum number of contacts with the ID will occur when media is nicely positioned, side-by-side, all around the ID. Of course, it doesn’t often get that good, but when it does, the maximum number of simultaneous media contacts with the ID will be 18. Now, consider that the same size angle-cut triangle, 7/8 × 7/8, will contact the ID two times at once. And, if lined up all around the ID, you could have 21 pieces making 42 contacts with the ID burrs. So, with cylinders you have a maximum of 18 contacts in a given moment, and with the angle-cut triangles you have 42; you can expect deburring to be 233% faster with the triangles.

Similar comments can be made regarding deburring straight edges. Figure out how many contacts can be made with different media shapes. A difference that figures here is that when the flat side of a triangle contacts a straight edge, the contact area changes from a single point to a straight line, thereby reducing effective pressure applied to the burr. Suppose that along a short straight edge, three pieces of angle-cut cylinders can touch the burrs. That makes three very small contacts, albeit with pretty good pressure. The same edge may be contacted almost without interruption by triangles, although the pressure from the media is distributed over a much larger area. In both cases, the pressure applied to the contact point is greatly increased by the weight of media backing (on top of) the actual contact media. It may be that the total area contacted is more valuable than the pressure per surface area, or it may be that the geometry of a triangle is more effective at direct transfer of energy from the backing media. I don’t know for sure, but my experience is that triangles will usually come out best on straight edge deburring when compared with cylinders.

There is also the factor known as “media tracking,” the area of relatively unfinished surface in fillet and corner areas that doesn’t get worked well by media. This is particularly bothersome if the purpose of mass finishing is to clean or polish the parts. In the earlier example of a large ring, there are often tabs on the inside diameter, and the corner burr has to be reached by media. Cylinders are less effective in these areas than many alternate shapes, such as the angle-cut triangle. And it may seem a minor point, but cylindrical media tends to align itself and run parallel to part surfaces. This causes some shapes of parts to fall into a smooth, rhythmic flow with the media, with less action against hard-to-reach burrs.

One of the prime considerations in media selection is the potential for lodging. Any media will lodge somewhere. This problem is compounded as media wears down. It is a problem we all face in mass finishing applications. It has been my experience that cylinders are more talented at lodging than most other shapes.

It has been my privilege to see mass finishing installations and applications in most of the U.S., and in four other countries. I recall only one instance in which cylindrical media seemed like the right choice. That was on tumble barrel finishing of steel pieces 0.030 inch diameter × 1.25 inches. It was necessary to debur and radius both ends, remove an extruded burr around a cross-hole on one end, and provide a smooth, shiny finish on the length of the part. The cylinders did a perfect job.

Media selection remains more art than science. There are many things to consider in addition to the shape—propensity for lodging, media density, media size, abrasive content, resistance to fracture, wear rate, cut rate, physical and optical quality of the finish, ability to separate from parts, and cost effectiveness. In this response, I have only addressed media shape. If you would like more comprehensive articles on the subject, let me know.
If your company is using cylindrical media, it’s a safe bet that it’s been losing considerable profit because of inefficiencies. The good news is that you can stop those losses almost immediately. 


  • Impact of REACH Regulation on the Global Finishing Market

    This paper is a peer-reviewed and edited version of a presentation delivered at NASF SUR/FIN 2012 in Las Vegas, Nev., on June 12, 2012.

  • In Mechanical Finishing, All That Glistens Is … Or Is It?

    Surface finish types for commercially supplied stainless steel sheet are detailed in various standards. ASTM A480-12 and EN10088-2 are two; BS 1449-2 (1983) is still available, although no longer active. These standards are very similar in that they define eight grades of surface finish for stainless steel. Grade 7 is “buff polished,” while the highest polish—the so-called mirror polish—is designated Grade 8

  • Taking the (Oxide) Edge Off

    Metal fabricators that laser-cut with oxygen take steps to prepare parts better for powder coating.