Order of Operations Question

Topics include, Machine Tools & Tooling, Precision Measuring, Materials and their Properties, Electrical discussions related to machine tools, setups, fixtures and jigs and other general discussion related to amateur machining.

Moderators: GlennW, Harold_V

User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Order of Operations Question

Post by rmac »

Here's a kind of general machining question that came up today as I was making the piston for my little Stirling engine. It's a simple steel part that looks like this:

piston.png
piston.png (7.2 KiB) Viewed 4057 times

The outside surface has to be smooth and just the right diameter so it slides freely in the cylinder but doesn't leak. As far as I can tell, there's nothing critical at all about the inside surfaces. I'll do all the machining on the end of a long piece of bar stock, and then part it off at the very end.

So the question: Is it best to do the fussy outside surface first while everything is nice and rigid, and then hope that hollowing out the inside doesn't somehow distort the part? Or would it be better to do the non-critical inside first and live with the resulting less rigid tube when working on the outside?

I'm actually more interested in learning how to think about this sort of thing than I am in this particular part.

-- Russell Mac
User avatar
NP317
Posts: 4557
Joined: Tue Jun 24, 2014 2:57 pm
Location: Northern Oregon, USA

Re: Order of Operations Question

Post by NP317 »

Bore the inside first, then turn the outside cylinder to almost (<.001") the correct diameter.
Test fit with the completed power cylinder and finish the turning and polish until they JUST fit together, with some drag.
Then cut off the piston and deburr.

At this point I told my students to take the piston and cylinder home and use toothpaste and water to hand hone them together while watching their favorite TV show.
They always brought back the properly honed pair with smiles.
And their engines ran too.

There are other ways to fit them together, but this worked well. Low tech. Clean teeth...
RussN
User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Order of Operations Question

Post by rmac »

NP317 wrote: Fri Apr 23, 2021 10:58 pm Bore the inside first, then turn the outside cylinder to almost (<.001") the correct diameter.
Thanks, Russ. I don't disagree with your suggestion, but I'm still curious why it's better to do the inside first rather than the other way around.

For the record, my own instinct was to do the outside first. At the moment I have what I think is a really nice fit between one piston and its cylinder, but I haven't bored it out yet or started on the second piston.

-- Russell Mac
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: Order of Operations Question

Post by Harold_V »

If you don't mind my intervening, what Russ said is spot on.

Rule of thumb, one that will never let you down.

In almost all cases, roughing parts before taking any finish cuts is the best approach. That means that you'd have to open the hole before taking any finish cuts. I'd also recommend you take a cut on the exterior surface as well. Do not rely on stock finish, not even when you're using drill rod, for it is often not round (even though it may measure like it is).

When you have removed as much metal as you can, but your part is still oversized on all features (I like .030" for most instances), you should then finish the bore before finishing the OD.

The reason is simple.

When you remove material, the remaining portion is prone to changing (due to internal stresses being relieved by the privious cut(s)). While on round pieces the change may be equal, there's no guarantee that it will be (there's always the risk of going out-of-round). By roughing, you'll have relieved stresses that might alter the critical diameter if it is finished first. With all features rough machined, you'll have made the part as stable as you can hope to have it, so when you remove the last amount of metal, it shouldn't change much, if any.

By finishing the bore first, you will have eliminated any further risk of the OD changing from alterations, as it is the last feature to be machined, so there are no further changes made that might alter its state.

When you take your finish cut on the OD, I recommend three passes, each of equal depth of cut. My preferred approach is this:
Measure the roughed part, pick up the roughed surface and remove 1/3 of the remaining material.
Measure the part after the cut, then take ½ of the remaining material.
Measure again, and take the amount you deem necessary to establish the target diameter, keeping in mind that your finish diameter will require polishing.

If your cut isn't torn (mild steel loves to tear), a half thou will usually be enough material to allow polishing to bring up the finish. I do NOT recommend you leave any more than is necessary, for once you've removed tooling marks, the remaining material is slow to remove.

If you find chatter is a problem, you can usually alter the condition by changing speed (go slower) and feed rate (increase feed). As finish is critical in this case, you won't have the option of increasing feed, however. Something as simple as placing a finger on the cylinder as it is cutting will usually stop chatter. Armed with that thought, and you have trouble with chatter, you might consider installing a rubber plug. A rubber stopper from a chemical lab might be a good example, or you can simply push in a small square of rubber, with the corners tight enough to keep it in place.

Use a sharp tool, and I highly recommend positive rake, so cutting pressure is low. A radius on the tool tip encourages chatter, so keep the tip radius as small as you can, while still creating a smooth cut, one without sharp peaks.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
Bill Shields
Posts: 10460
Joined: Fri Dec 21, 2007 4:57 am
Location: 39.367, -75.765
Contact:

Re: Order of Operations Question

Post by Bill Shields »

Bore the ID. Rough the od per Harold's suggestion. Put in oven @400 degrees and true up id.

Then Superglue it on a mandrel and finish the od.

Back in the oven to kill the glue.
Too many things going on to bother listing them.
User avatar
NP317
Posts: 4557
Joined: Tue Jun 24, 2014 2:57 pm
Location: Northern Oregon, USA

Re: Order of Operations Question

Post by NP317 »

Harold:
Thank you for the detailed lesson describing WHY and HOW the suggested order of machining should be followed.
A generous offering from you. Again.
RussN
User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Order of Operations Question

Post by rmac »

Harold_V wrote: Sat Apr 24, 2021 1:42 am If you don't mind my intervening, what Russ said is spot on.
My goodness, Harold ... you are always welcome to jump in as far as I'm concerned. I look forward your thoughtful advice. Always.
Rule of thumb, one that will never let you down.

In almost all cases, roughing parts before taking any finish cuts is the best approach.

[big snip ... see quoted post]
Thank you. That is exactly the kind of information I was after. The whole idea of internal stress is not something I would have thought of on my own.
When you take your finish cut on the OD, I recommend three passes, each of equal depth of cut. My preferred approach is this: Measure the roughed part, pick up the roughed surface and remove 1/3 of the remaining material. Measure the part after the cut, then take ? of the remaining material. Measure again, and take the amount you deem necessary to establish the target diameter, keeping in mind that your finish diameter will require polishing.
You taught me this almost ten years ago. It worked then and (big surprise) it still works today!
I do NOT recommend you leave any more than is necessary, for once you've removed tooling marks, the remaining material is slow to remove.
Very slow! I found this out yesterday trying to get the first piston to fit.
Use a sharp tool, and I highly recommend positive rake, so cutting pressure is low. A radius on the tool tip encourages chatter, so keep the tip radius as small as you can, while still creating a smooth cut, one without sharp peaks.
I've been experimenting with a homemade version of the so-called tangential or "diamond" tool holder (see the picture) with encouraging results. It's not the answer to all the world's problems as its proponents would have you believe, but it seems to work well for general facing and turning. As you suggest, the tool has lots of positive rake, and a very small tip radius. It leaves a so-so finish when turning from left to right, but if you then feed in a tad and kind of drag the tool backwards for a final pass, it will shave off just a bit more and leave a very nice finish most of the time.

-- Russell Mac

th_done.jpg
User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Order of Operations Question

Post by rmac »

Bill Shields wrote: Sat Apr 24, 2021 6:15 am Bore the ID. Rough the od per Harold's suggestion. Put in oven @400 degrees and true up id.

Then Superglue it on a mandrel and finish the od.

Back in the oven to kill the glue.
Bill: What's the thinking behind the trip to the oven? And how long does it need to cook?

-- Russell Mac
User avatar
Bill Shields
Posts: 10460
Joined: Fri Dec 21, 2007 4:57 am
Location: 39.367, -75.765
Contact:

Re: Order of Operations Question

Post by Bill Shields »

First oven is to get it to relax so it will be a bit more stable when machining. Exact temps will depend on msterial.

Loctite falls apart at 400 f in about 1/2 hour.

Or you can soak it in acetone and wait
Too many things going on to bother listing them.
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: Order of Operations Question

Post by Harold_V »

Bill's suggestion of placing the finish bored piston on a mandrel is quite good, as it addresses the possibility of chatter, common when machining thin cross sections.

If using glue isn't in the cards, you can achieve the same results by simply making the piston a light slip fit on the mandrel (machined in situ, longer than the piston bore depth, and faced), then drive the piston with a live center. As the finish cut(s) would be exceedingly light, it takes very little effort to keep it from slipping under cutting pressure. A small faced plug with a center hole drilled is used between the piston face and the live center.

If this method is chosen, a generous chamfer on the end of the mandrel assures there is no contact of the corner radius inside the piston. The piston should be faced internally so it is flat and parallel with the external face.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
Bill Shields
Posts: 10460
Joined: Fri Dec 21, 2007 4:57 am
Location: 39.367, -75.765
Contact:

Re: Order of Operations Question

Post by Bill Shields »

You could put it over a rubber mandrel and cause it to expand by squishing from the end...maybe a push with the tailstock or a central deadbolt of some sort.

Be creative...maybe blow it full of expanding foam?

Anything to stiffen it up and keep it from ringing
Too many things going on to bother listing them.
Lew Hartswick
Posts: 775
Joined: Sat Jan 04, 2003 10:45 am
Location: Albuquerque NM

Re: Order of Operations Question

Post by Lew Hartswick »

ie. leave the most critical dimension for the last operation. :-)
...lew...
Post Reply