How to Determine an Interference Fit?

I am making a plinth (base) for a quick-change tool post. The plinth has a

16.05mm vertical hole in the center. Through this hole, I will press a post (threaded at the top) for locking down the QCTP. There is a lip on the bottom of the post that fits into a recess in the bottom of the plinth to prevent the post from being pulled out of the hole (the post will be pressed through the hole from the bottom of the plinth). But I need an interference fit to prevent the post from rotating as I torque down the nut that clamps the QCTP. The length of engagement (the axial length of the hole) is 22mm.

Question: What should be the diameter of the post for a proper interference fit?

I thought it would be easier to fit if I heated the plinth and then pressed the post in. However, I still needed to know the proper diameter to make the post. So I tried heating the plinth in the oven to 400deg F and then measuring the diameter of the hole. But I could not get an accurate measurement (I suspect that the telescopic gage was expanding during measurement, due to the brief contact with the hot plinth, and then cooling and contracting by the time I got it in the micrometer). So I think I will have to rely on the cold diameter measurement of 16.05mm.

Thanks in advance for any assistance!

Reply to
Loading thread data ...


I have fitted a QCTP recently to a 9x20 lathe with the same situation. The original 10mm bolt went into the carriage through a 10mm hole that was c-bored from the back at a .625 dia where it was held by a splitpin drilled axial into the middle of the outer dia splitline to keep from rotating.

I considered the 8mm thread on top of the original bolt to be to weak for the QCTP I also figured the 14mm bolt that came with the QCTP was grossly oversize for the small upper carriage. I decided a 1/2 bolt would be the best size and fit. I also choose a fine thread there to reduce the amount of torque required to tighten this bolt. I also wanted a easy way to replace this new bolt for wear or to use different length stud bolts.

What I did: Used a standard 1/2-20 x 4 grade 8 hexhead bolt. I machined 2 opposite sides of the hex surface to a width of 0.625

I then machined the hole on the top carriage to 0.500 dia. (the bolt dia. was 0.501 just under the head) I elongated the c-bore on the back to fit the 0.625 width of head with some clearance for the hex area to keep from rotating. I reused the splitpin in a small hole drilled in the splitline of the fit area to aide in holding it in.

I made a brass sleeve to bridge the clearance 1/2 dia. of the bolt to the 14mm of the QCTP. This sleeve needs only be about 3/4 high.

Worked out great. I opened up the through hole of the other original parts also like the 4 way toolholder and the single toolclamp so they can be used with the new bolt also. Good Luck

Reply to

Experience tells me 1 to 2 thousands should be enough to prevent rotation based on the part description and application. Much more inference is likely to damage the parts if you are not careful. Be sure to have a lead (chamfer) on the parts. Just for reference -The Coefficient of Thermal Expansion for 1020 steel is 6.5 x 10-6 "/"/degree F. The expansion is calculated by multiplying the diameter x temp change x coefficient. .632" x (400-70) x (6.5 x10-6)=.00135" change. One thing to keep in mind when using thermal differences for assembly purposes is that as soon as the parts touch they quickly begin to move toward the same temperature and therefore the smaller the mass of any of the parts the quicker the assembly must be. Obviously its best to have the parts just freely drop in but should they hang up you are pretty much screwed and will have to press the rest of the way.

Reply to

"James P Crombie" wrote

Hmmm... Prior to posting, I had looked in the index in my Machinery's Handbook, but only found a reference for "Interference Threads". I guess I needed to look under a different index entry.

Can you point me to the page where you found that info?


Reply to

That's what I was originally going to do. But I didn't have the correct-sized tap on hand for the internal threads (one is on order, but hasn't arrived yet) and I didn't want to wait. I suppose I could have fashioned a custom single-point threading tool and tried to thread the internal bore on the lathe, but I figured that was more trouble that it was worth.

??? Do you mean if it were a slip fit with locktite applied? I figured that could be a fall-back solution if I screwed up the press-fit. :O

The other fall-back option would be to make a slip fit, then drill a hole parallel to the post, and from the bottom, such that the hole passed partially through the lip at the bottom of the post, and partially in the plinth. Then I could drive a pin in there to prevent rotation of hte post. Sort of like a shaft key.

I've been meaning to get some silver solder....

No dovetail on my lathe (it's one of the 9x20 class lathes). As far as teh chips in the threads go, I was planning later on to replace the open nut at the top with a custom threaded cap with a handle.


Reply to

A thou is a thousandth of an inch yes.

Heating helps but isn't required. 2 thou would be a good tight fit for general use but if I understand it you need something that will withstand bolt torque as well. 3 thou should be as tight as you need. If there's a bit of a lead in on the part you could press that in with a fly press.

It depends on the size of the hole.

Dave Baker - Puma Race Engines

formatting link
I'm not at all sure why women like men. We're argumentative, childish, unsociable and extremely unappealing naked. I'm quite grateful they do though.

Reply to
Dave Baker

No, just that the friction will probably be enough to keep it from spinning. If the bottom of the stud is *real* small it might not though. I think this is one of the original CF problems, needing to make a long slender shank with a big fat disc at the bottom. Which means you need to start with a big piece of stock and trim it all away except at the bottom.

Sorry, I meant to say the "T-slot" in the comment above. But this is how my indexable turret in my south bend is set up, the T-slot nut has a 3/8-16 tapped hole in it, and I have a bolt that drops down from the top. The toolpost has only the hex of that bolt visible:

Basically I made the bolt manufacturer do the work of producing a long skinny part with a head on it. :)


================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Reply to
jim rozen

I have done a little heat shrinking and you should have the part going into the hole about .0003" bigger than the hole. +- .0001" Sorry I don't use metric . Make a little taper on the end of the rod and have every thing all set up including a hammer. Have a guide to make sure the rod goes in strait and heat your part up in your oven. The driving in of the rod should take only 10 second to do. The actual driving in of the rod should be only one blow of the hammer. If you put the rod on the hot metal you will expand the rod and it might only go in half way a stick. I mean realy stick. So go through a cold run and make sure you remember the name of my club if you procrastinate. SCRAP Ream your hole and emery your rod to size. Let the rod cool down to measure dia. Don

Reply to
Don Huseman

Reply to
James P Crombie

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.