Strength of a shrink fit

[alanganes]

I have some ongoing projects that involve doing a bunch of work in stone, mostly hard New England granite. Having the need to drill a fair number of 5/8" holes, I recently I got ahold of an older Bosch hammer drill for next to nothing that uses spline drive bits. This drive seems to have sort of fallen out of favor with most newer tools using SDS or SDS-MAX bits. I was able to get an adapter that gores from spline to SDS for not much money, so no problem there.

But as it happens I have access to a decent supply of old "B-taper" masonry bits for close to free. The A and B taper style bits go way back and were common on small pneumatic hammer drills used in quarries and such. B-taper is a straight 1:10 taper. I also was given an old B-taper socket to some sort of hex drive used on the old air tools.

So wanting to make a B-taper to spline drive adapter for myself, I cut the spline off of a junk bit I had and also cut the hex portion off of the taper adapter. I figured that welding the two together would probably result in a brittle joint that would break in use. Each of the cut off pieces had a straight round section behind the working end so I turned a sleeve with a matching holes for a shrink fit in each end. After doing a bit of reading about shrink fits, I was aiming for a 0.003" interference. I hit that on one and ended up with something a bit more than a 0.002" fit on the other. I heated the sleeve up to 800F and as expected it slipped right onto the two pieces. I seated the larger diameter piece first then the smaller one to assure that the ends were butted together. This is what I ended up with:




On cooling down everything seems tight as can be. The pen is pointing to the shrunk on sleeve, the taper socket is at the top of the photo.

So the question is: will the shrink fit be sufficient to hold this together in use or would I need to add pins through the sleeve or something? My thinking is that the shrink fit is likely way stronger than the holding strength of any taper, so if something is going to give it would be the connection between the taper socket and the bit. I have never really done anything using a shrink fit and am just wondering what the collective brain here thinks...

[DrMike]

What is the nominal diameter of the shrink-fit portion?
How long is the shrink fit portion?
Is the hole a simple sleeve, or is there a stop or locating part?

[alanganes]

What is the nominal diameter of the shrink-fit portion?
How long is the shrink fit portion?
Is the hole a simple sleeve, or is there a stop or locating part?

In kinda round numbers the nominal diameter of the "inner" parts is 0.780", the smaller of the two is about 0.030" smaller if I recall. The overall length of the sleeve is 1.5" and each part goes on about 0.75" more or less. The hole is just a sleeve, but there is a step that the larger of the two stopped against when it was put in. The smaller diameter part seated such that it was tight against the end of the larger diameter part. I figured that would be important to transmit the impact from the hammer part of the drill.

As mentioned, on each end hole in the sleeve is about 0.003" smaller than the OD of the mating part and was heated to about 800F to assemble them. They went in with a smooth fit that required a bit of light tapping with a small hammer to seat them together.

Thanks for taking the time to reply.

[DrMike]

I agree, the step will take the brunt of the hammering leaving the press fit to just hold the pieces together.
An interference of 0.003-in on a 0.780-in diameter is pretty hefty, I don't think you'll have any problem with it.

[Mcgyver]

its not 100% clear to me all the dimensions, but the strength of an interference fit is subject to the strength of the parts . e.g. a given fit will not be the same if the wall around the bore is 1/8 vs 2". 3 thou is too much for that diameter and wall thickness, rule of thumb is .001 per inch, but even that is limited given what looks like a thin wall.

I think it would be better/faster/easier to loctite it.

Will what you've done be enough? Who knows, but since its done, give it a try.

[Cuttings]

I don't know but is the hammering all vertical or does it have some rotation forces like an impact wrench? If it ever started to move I think it would soon let go. I would be inclined to put a couple of pins in to prevent it from starting to move then you should have no trouble.

[mikey553]

its not 100% clear to me all the dimensions, but the strength of an interference fit is subject to the strength of the parts . e.g. a given fit will not be the same if the wall around the bore is 1/8 vs 2". 3 thou is too much for that diameter and wall thickness, rule of thumb is .001 per inch, but even that is limited given what looks like a thin wall.

I completely agree with Mcgyver. Shrink fit is a very strong connection especially with the amount of interference you provided. It may be limited by the wall thickness of your sleeve and the strength of the sleeve material. I do not think additional pins are needed - shrink fit is much stronger than pins.

Looking at your original post we can see that you need to make a number of 5/8" holes in hard granite. You did not say how deep the holes are, how big the granite pieces are and how accurate the holes need to be. I just wanted to give you one more option - diamond tubular drills. It is a piece of steel tube with the shank on one end and diamond grit bonded to the other end. Such drills are available very inexpensively in different sizes. I have a set of them ranging from about 6 to 50 mm in diameter. They need water for cooling, but produce good quality holes and best suited for a machine work, not being hand held. The only drawback - they are shallow. Mine can only go about 1" deep.

[alanganes]

Thanks for all the input, everyone.

A few more details:
The sleeve was made from what was the unthreaded shank of a 1" diameter 2' long steel allen head bolt. I don't know what grade it was but I do know the application and place it came from so know it was a high quality (and probably high grade) bolt and not a cheap thing of unknown origin. It machined beautifully with a carbide boring bar. The wall ends up a tad under 0.125 on the end with the larger bore. I got the 0.003" fit from some online source but don't recall what it was specifically. It looked like it was easy enough to get the required amount of expansion of the hole at what are reasonable temperatures for me so that's what I did. I'm clearly not an expert...

Cuttings, the hammering is all vertical along the axis of the bit. The rotation is continuous. The general theory of these things is the strike from the hammer pulverizes the stone under the cutting edge and then the bit turns a fraction of a turn to a higher spot and then it gets hit again. But there certainly can be significant torque applied especially when the bit catches. Then it tries to wrap your arms up around the drill if you aren't paying attention.. But yes, the shrink fit is pretty much only to transmit the rotation, the hammering should all be steel-on-steel.

As to your question about what I'm doing Mikey, these holes are for using wedges and shims to split the granite blocks. The holes do not need to be very precise but do need to be at least 4" or 5" deep, and a bit deeper is often better. The blocks range from maybe a 12" cube to a rectangular block 8"x 18" x 7 feet long. I am familiar with the tubular diamond bits but they are not ideal for what I am doing. When it's all said and done I may have ended up drilling several hundred such holes. I do appreciate the suggestion though.

Mcgyver you are correct, I am at the "try it and see what happens" stage and really have nothing to lose beyond the time I spent on it. That was all enjoyable shop time so I have nothing at all to lose really. I figured I'd give something new to me a try (doing a shrink fit) but always like to hear the input from guys that have the experience to actually know what they are talking about.

In the end this is sort of a project done just for it's own sake. SDS bits are fairly inexpensive and readily available and I have a diamond wheel that I can touch them up on. I was just offered some of these B taper bits along with the adapter and thought it would be fun to try to use them. As I often like to say: If I can't overcomplicate a simple project, why have all these machines! Right?


Quick edit to add: A suitable adapter can actually be purchased ready made, but that only takes money.

[nickel-city-fab]

I recently welded up a rock bit for one of the contractors at work, he needed it made 4 feet long (!) with a spline end on it. . I used 309 SS rod and he said it worked fine. In your case, if the shrink fit doesn't hold, I could recommend running a bead around it. The 309 rod can save the day when you have a medium o hi carbon joining to mild steel. It's hard to crack that stuff.

[no704]

If it slips, just keep running it and it will spin weld!

[RB211]

Shrink fits have worked for the railroads for 180+ years

[boslab]

Just use it, if it falls to bits it wasn’t up to it!, I must have destroyed dozens of big drills, buildings are often rock over here, sometimes 4’ thick, I’ve lost a few in the wall, hear a rumble drill stops turning or snaps, I think your shrink will hold, essentially your hammering it together
mark

[alanganes]

Thanks everyone, I'll report back once I have a chance to try this out.

Nickle-City, yeah those stone drills are an entire world in themselves. The guy who gave me this stuff (along with most of the stone for the projects I am working on) owns a granite quarry. They have sectional drills with a screw-on carbide tip and a 7/8" hex shaft that they can screw on as many 8' long rods to go as deep as they need. Sort of like a mini oil rig. Very old tech but really interesting stuff. I have a few pieces of that drill rod, it's a bit tough to machine but makes great stock for making anything you plan to bang on.