Fun with Internal Threading

[WobblyHand]

Is your crank on the spindle or the end of the leadscrew?

I use full form flat inserts so a R/H internal insert used on the O/D becomes a L/H external insert.

Crank is on the spindle. Simple handwheel with an expanding arbor to grab onto the interior of the spindle bore. The expander is the same design as what is typically used to hold handlebars to a bicycle head tube.

 

[mickri]

Threading is a learned motor skill. I practiced on pvc pipe until I had it down pat. Over and over again. I spent hours doing this. I think, right or wrong, the reason why most who have stress cutting threads is that they are learning on a part they have spent a lot of time making and don't want to have to make the part again if they screw up the thread.

As for a relief cut at the end of an internal thread, I don't bother with one anymore. I had trouble one time cutting a relief cut for an internal thread in a blind hole. I finally gave up and just cut the thread. Because of the method I described in a prior post the threading tool bit always stops in the same place. You can let it stay there for as long as you like. The threading tool makes its own relief.

I think another problem is trying to do stuff as fast as we can. That is important in a production shop where time is money. On the other hand in a hobby shop time is meaningless. It doesn't matter how long it takes us to make something. In fact the longer it takes the more time we get to be in our shop using our machines. Case in point. I recently made two bearings out of delrin for a friend. The bearings were 2.375" OD by 1.886" ID by 1.3125" long. It took me around 10 hours to make these two simple bearings. They needed to be a tight fit on both the OD and the ID. So there were lots of test fits as I snuck up on the final fit. I didn't care how long it took. I was in my shop making something instead of toiling away on the never ending remodel of my house.

 

[WobblyHand]

the reason why most who have stress cutting threads is that they are learning on a part they have spent a lot of time making and don't want to have to make the part again if they screw up the thread.

You nailed it. Beginners sweated over getting the part to that point and don't want to blow it! Don't learn on a part you are making!

Practice, practice, practice. The more you do it, the easier it gets. Practicing on pvc pipe is a great way to develop the timing and muscle memory. If you mess up on pvc the only penalty is your pride. Then you just try it again and again until it becomes more natural.

 

[Just for fun]

Good reading guys,

I couldn't get my head wrapped around the cutting threads away from the chuck at first. It finely came clear, I need a different tool.

I did do as some have said and marked a couple of locations of when to disengage the half nut. I also ran the cutter in to the end and took note of the location on the DRO. I gave myself 250 thousand from the point where I disengaged the half nut. I also employed the the same idea that @MrWhoopee designed into his retracter and turned the cutter away from the cut at the same time as disengaging the half nut.

My first try at it I was just going to fast (dumb as mistake). Once I slowed everthing down, it all worked as planned. No stress really it's all in the timing. Now if I was working on a piece that I had been working for hours like @mickri said, it may have been different.

 

[Just for fun]

@Just for fun . When boring , the normal process is to take 3 or so cuts the same depth to establish bar spring etc . Just say .030 on diameter , 3 passes , take a reading , then you can rough it out within .100 and repeat . Taking cuts of all different depths will give you different readings . And we do thread out as Jim suggested , you just use the opposite hand threading bar .

Thanks for the pro tip, that make a lot of since. I know what my line of thought was, but it just wasn't working out. I'll try and get a chance later to try it again.

Tim

 

[Just for fun]

When I tried internal threading for the first time, I used a bushing cut from hex bar (thereby making a nut) so that the cutting tools would travel in to open space after doing their cutting.

How long was your safety (i.e. relief) groove?

I would be interested to see your setup with using the compound. I just used the cross slide with a lot of spring passes after each adjustment on the DOC. Using the compound is better as this reduces the tool pressure by 50% and makes better threads.

Did your setup look like this?

View attachment 416371

Here is another drawing:

View attachment 416372

Well, I would like to say yes.... But it never, I had not changed it from doing external threads. Next time I will do better.

What I did was when I reached my mark, I would move the cutter away from the cut with the cross slide as I disengaged the half nut. I had previously set the cross-slide dial and DRO to zero. Once I retracted the tool, I would set the cross slide back to zero and then dial in a few thousands on the compound and make another cut. To me that part seemed to work really well.

 

[mickri]

You don't need to move the cutter at the same time that you release the half nuts. My sequence is to disengage the half nuts. Then back off the cutter. Then move it out of the hole. Then bring the cross slide back to zero and finally increase the depth of cut with the compound to make another pass on cutting the thread.

When boring I don't worry about spring until I get to around ,100 from my final ID. Also when boring I always take the same depth of cut. My usual depth of cut is .020 which equates to a .040 increase in the ID. This makes it easier for me to keep track where I am at. When I get to around .100 from the target ID I will make several cuts with no increase in the depth of cut. Spring cuts. Then I take several measurements to confirm how much still has to be removed to get to my target ID. The old measure twice cut once mentality. Take more consistent cuts until I get to around .010 from the target ID. Measuring after each cut. If I am say .008 from the target ID I divide .008 by two and then take two more cuts of .004. I find that this typically puts me within .001 of my target ID. Often right on my target ID and very rarely over. I have yet to machine anything that requires accuracy of better than .001.

This is what works for me.

 

[Just for fun]

Thanks Chuck,

I will do as you say on the next one. I was successful today in boring and doing an internal thread with no crashes. Being consistent really does pay off. I took three cuts at .030, three cuts at .050 and one cut at .052 and that put me at .0046 smaller than my target but still right in the middle of a 3B internal thread for a 1"x20 thread. I have another one to make so I will do as you say and take smaller cuts as I get closer.

I didn't have any way of checking the thread that I know of, so I made a 1x20 slug out of 12L14 (That stuff cuts nice) for checking my internal thread.

While my lathe and mill were still enroute to me from Twain I bought a piece of mystery metal 20'x2". That stuff is crap to deal with I have tried all kinds of things and can't get a good finish. It seems to have hard spots throughout the bar. Anyway, that is the stuff I'm working with, so the finishes are not the great.

And another thing that doesn't work very well on larger stock is knurlier that came with my lathe. It is the type that you have to push into the work. It seems to work ok on small stuff but not very well larger stuff.

Here's a couple photos for everyone's viewing pleasure.

Here is one for @erikmannie my setup.







It looks like I should have blowen it off or at least wiped it down before taking the photos, jeeze!

Tim

 

[homebrewed]

Going back to the manually-driven spindle suggestion, there are times when it can be easier to manually drive the lead screw instead of the spindle. That would be the case when cutting threads where the threads you are cutting are coarser than the LS TPI. In that case the change gears from the spindle side are working at a mechanical disadvantage.

On my little machine that would be the case for anything coarser than 16TPI.

 

[Parlo]

Going back to the manually-driven spindle suggestion, there are times when it can be easier to manually drive the lead screw instead of the spindle. That would be the case when cutting threads where the threads you are cutting are coarser than the LS TPI. In that case the change gears from the spindle side are working at a mechanical disadvantage.

On my little machine that would be the case for anything coarser than 16TPI.

Yes that is why I asked the question. Way back in the day I remember driving the leadscrew for coarse helices.