What a difference a few microns make - Numpty Award

[deema]

I made a start on a spindle extension for a replacement motor I burnt out. The original has a custom shaft, so I need to make an adapter. I wanted a real piston fit to reduce out of balance vibration. I was feeling rather smug having cut the receiving bore for the motor spindle to just 5 micron over size. With that tight tolerance I thought I’d better test the fit with the motor whilst it’s in the Chuck. Protective wooden board over the slides and the shaft slid in about half way and them became well a truly stuck. It was a lovely piston fit up to that point.

No prizes for everyone who worked out (it’s my theory at least) that the differential temperatures equalised and it became an interference fit. Took a lot of ‘oh dear me, silly me’ statements and a bit of sweat to get it off. Had to heat up the extension piece and be quick to get it off before the shaft of the motor caught up. Anyway, not so smug now as the bore is actually also tapered, I believe due to the heat build up when cutting, it’s narrower at the bottom of the hole than at the top by a few microns☹

 

[Blister]

A little word of advice , Do not turn the lathe on , If you do, RUN LIKE HELL

 

[RobinBHM]

I had no idea a metal lathe had an accuracy of 5 microns tbh

 

[Dalboy]

I had no idea a metal lathe had an accuracy of 5 microns tbh

It is surprising how accurate you can turn a piece on a lathe especially if you use the cross slide at a small angle and adjust that. also it makes no difference how big something is turned you still can measure to that degree.

Deema did you allow the part to cool down and take a spring pass

 

[deema]

@Dalboy the answer is yes, I did let it cool down, but at these levels of tolerance I think the answer is not enough

Angling the cross slide is a brilliant idea, never thought of that; thank you.
I put a Mahr 1 micron gauge on the slide and adjust it that way after pushing very hard on the tool post to remove any last bit of backlash before being satisfied it’s set. I took about five spring passes just to be sure. It takes me absolute ages to get down to this level of accuracy…….and it’s usually a fluke!! I find getting to around 10micron or 1/2 thou fairly easy. You need the headstock bob on aligned.

 

[Sachakins]

How about using a reamer ?

 

[deema]

@Sachakins That’s would have been a really good solution, but I didn’t have one the right size. Good quality reamers are expensive!

 

[okeydokey]

Other than a lesson or 2 at school many years ago I have never used a metalwork lathe so please excuse the following questions, yes out of ignorance but a desire to learn as well.
I understand the mega accuracy and so the extension will fit without causing a wobble on the shaft extension but thats a very tight fit, how do you secure it into place when you have completed making the extension piece?
Assuming a pulley (the original) will fit onto the shaft extension piece I guess its outside diameter is the same as the original so the pulley will fit, is the pulley manufactured to the same "wobble tolerance" as the shaft or doesn't it matter.
Would be interested to know, thanks
regards

 

[deema]

The fit doesn’t really need to be as close as I was aiming for……but I have time as I’m retired and it was an interesting exercise. I actually need the shaper to make the part!! It will have a keyway and grub screw to secure it onto the shaft. The grub screw will press against the key. Initially the grub screw will hold the shaft extension whilst I make a new permanent shaft extension. The shaper is needed to cut the keyway. I could make the keyway on the lathe as an alternative, using the compound slide.
I’m actually wondering about dropping an end mill that can drill down to clear out the majority of the keyway. I have a Knee Mill. I’ve not made a blind keyway before so it’s a yearning curve. If anyone has any suggestions please shout out.

 

[Dalboy]

@Dalboy the answer is yes, I did let it cool down, but at these levels of tolerance I think the answer is not enough

Angling the cross slide is a brilliant idea, never thought of that; thank you.
I put a Mahr 1 micron gauge on the slide and adjust it that way after pushing very hard on the tool post to remove any last bit of backlash before being satisfied it’s set. I took about five spring passes just to be sure. It takes me absolute ages to get down to this level of accuracy…….and it’s usually a fluke!! I find getting to around 10micron or 1/2 thou fairly easy. You need the headstock bob on aligned.

Did you note the mistake I made when I said cross slide it should have been topslide or compound slide. But I think you got what I was saying

 

[deema]

I’ve got man flu at the mo. I didn’t notice but I did fully understand, it’s a brilliant idea, which I will be using when I resume the spindle machining!!

 

[Chief Sawdust Maker]

Could you remove it from the chuck, apply heat carefully and it should come off?
To refit it i would suggest cleaning up the inside with emery and then warming the sleeve up in an oil bath (we used to use a deep fat fryer dedicated for the job) and then it should push on. It was our standard method for bushes and also very expensive tapered bearing races that had to be fitted on shafts that reached approx. 10,000 rpm
I would also polish up the motor shaft
Well done on your turning tolerances I spent quite a few years back in the 80's boring cast iron and stainless steel from 1" to 15" to those sort of tolerances which before they were removed from the machine were visited by the inspection department with an optical bore micrometre, good set up cleanliness and knowing your machine are the key points
I now run a small wood work shop doing a lot of work on a cnc router but still strive for and usually achieve really good tolerances, I have spent many many hours using my engineering skills to set up all of my wood working machines and get a lot of pleasure from being able to achieve what I am aiming for.

 

[Sandyn]

and get a lot of pleasure from being able to achieve what I am aiming for.

As I get older, that gives me pleasure as well

 

[sploo]

Later today I need to try to achieve an interference/push fit between two pieces of steel, so based on deema's experience here I expect to make something so sloppy it'll be like throwing a hotdog down a hallway

 

[Yorkieguy]

Accurate to 1 micron:

High precision air bearing CNC lathe and grinder Home built high precision air bearing CNC lathe and grinder having 1um (0.00004\") accuracy. www.youtube.com

Dan Gelbart is a most amazing guy. 100 patents to his name. You might wonder who he is. He founded a tech company in Canada is 1983, which grew to 4,000 employees, and sold it to Kodak is 2005 for one billion dollars. He has more than 100 patents to his name:

Dr. Daniel Gelbart | Graduation at UBC

Take a tour of his 'workshop' which is in four sections:

1) High precision room.
2) General Machine Shop. Furnaces, water jet cutter, powder coating, fluidised bed coating, induction heater etc.
3) Room for 'dirty work' (The only sort we have!) for woodworking, sanding, grinding, welding.
4) Laboratory and testing area. (Includes an array of RF gear).

All the air is filtered and each room is at differential air pressures.
Not a speck of dust anywhere, and he knows where everything is!

ttps://www.youtube.com/watch?v=HWPYoE1SNnA

Dan Gelbart Workshop Tour HD Dream workshop for high precision www.youtube.com

 

[Fergie 307]

It is surprising how accurate you can turn a piece on a lathe especially if you use the cross slide at a small angle and adjust that. also it makes no difference how big something is turned you still can measure to that degree.

Deema did you allow the part to cool down and take a spring pass

I would agree with the last comment. Always a good idea if you want this sort of precision. Get it within a gnats of the finished size, then have a cuppa while it cools, the take a final light pass to get to the exact size. If you have coolant facilities then that will also help.