Stuart Twin Victoria (Princess Royal) Mill Engine

[Ramon WilsonParticipant @ramonwilson3]

Doc, I'm not quite sure I follow you but once you are happy with the first end and have a reference point from the cross slide dial then I would reverse the shaft and work towards the same reference point as before but measuring as you go to be certain the cut is the same. Technically finishing at the same point on the dial should be the same but there are very small variables that can add up to it not always being so – by measuring you can see if there are any.

Don't rely totally on the dial is what I'm saying – a thou under is under size, a thou up can be rectified and it doesn't take much to be a thou out even if you are boring the parts to suit but you already know my opinions on going about it that way.

When machining to a fine tolerance once within a few thou take another pass without altering the dial and see how much more that takes off – a spring cut will usually take a very small shaving cut but you need to know how much it will remove if you do.

Best – R

[Dr_GMJNParticipant @dr_gmjn]

Thanks Ramon, I understand about not relying on the dial, but in this case where absolute accuracy isn’t necessary, it’s not a problem.

Indeed, if, once I was happy with the final size at one end I left the dial alone and did the opposite end, I’m not relying on the dial at all – in terms of getting the ends identical?

This is assuming the incremental cuts have also been done at each end with un-touched dial settings.

Thanks.

[Ramon WilsonParticipant @ramonwilson3]

If you do the first end in a series of cuts to get to size then do the second at the same finish setting on the dial the extra material you will be taking off in one pass will in all probability lead to a differing size. You hold the key to accuracy in your hand – your Mic – and your ability to work to the dial

As a matter of interest do you have the old fixed Myford ML7 dials or have you got the replaceable moveable type?

The former does make things a tad awkward (but a felt tip mark helps) – the latter is something that once fitted you would never change back.

You most definitely would not be changing end for end to ensure each path was identical to each cut – do one end complete then repeat the process – if you've done it once you need to be confident that you can repeat it. Doing what you suggest is not the way to go about I'm afraid.

Best – R

[JasonBModerator @jasonb]

Posted by Ramon Wilson on 08/09/2022 11:39:40:

If you do the first end in a series of cuts to get to size then do the second at the same finish setting on the dial the extra material you will be taking off in one pass will in all probability lead to a differing size. You hold the key to accuracy in your hand – your Mic – and your ability to work to the dial

………………….

Doing what you suggest is not the way to go about I'm afraid.

Best – R

That is not what Doc is suggesting

He is 10thou over dia. He will take 6thou off one end then leaving the cross slide as is flip the work and take the same amount off the other end.

he will then advance the tool 2 thou and cut one end and then the other again without touching the cross slide

And so on until he is where he wants the size.

It's not the way it would be done in industry* but I can't see anything technically wrong with it if he wants to go that way.

*Really is is not different to swapping parts in a repetition lathe, they all get cut at the same stop setting of the tool.

As I said this morning and you have said sneaking up on one end and then taking a big cut in teh other could well give different results.

[Dr_GMJNParticipant @dr_gmjn]

Posted by Ramon Wilson on 08/09/2022 11:39:40:

If you do the first end in a series of cuts to get to size then do the second at the same finish setting on the dial the extra material you will be taking off in one pass will in all probability lead to a differing size.

 

Thanks Ramon, sorry but you're answering a question I've not asked. As I keep saying, that's not what I'm suggesting:

"This is assuming the incremental cuts have also been done at each end with un-touched dial settings."

So…for all cuts at each end, exactly the same amount of material is being removed. There's no sudden jump from 0.002" to size at one end, and 0.010" to size at the other.

I'm asking where the error would come in if I do it as I've outlined?

Any error in the dial is removed, because the dial isn't involved in the cuts, so to me it's more consistent in getting identical diameters on each step, because I've removed a significant source of error.

Yes, there's an error in length due to the depths of the centre holes being different – we've covered that. But for diameters, the centres aren't moving radially, and nor are the centre holes, so…where's the error going to come from?

To me it's little different from machining to a depth stop on the mill or lathe – you're relying on a repeatable datum – in this case the centres – to get multiple parts the same (as I did on the mill with the bearing blocks).

 

I removed the original conical cast dials on the lathe, and fitted the adjustable dials to the cross and top slides instead. Helped a lot.

Cheers.

 

ETA sorry Jason – crossed posts saying the same thing.

Edited By Dr_GMJN on 08/09/2022 13:37:07

[Ramon WilsonParticipant @ramonwilson3]

Jason, Doc,

Yes I am aware that's what is being suggested but it's not the normal way to go about things when machining between centres.

That the part is being machined between centres, yes, it can be swapped end for end between cuts without changing the settings and yes it might get you very close but if that's the only way one is able to rely on to get a repetitive dimension then I'd say maybe it's time to rethink things.

Whilst he object of turning between centres is to be able to remove and replace the component without loss of concentricity it is not to ensure radial dimension is equal both ends

Had a turner gone about it the manner you describe in the machine shops I ran I would have been seriously under impressed

The point I'm making is that is not the correct way to approach things Jason. Doc has already proved he is capable of machining to a limit so why the sudden reluctance to do the same.

If anyone using a lathe has to require doing what has been suggested then I would say they have no confidence in their ability – how would they get on if more than one part is required to be identically machined in a chuck all in one hanging.

Believe it or not many, many, of us did that on a daily basis with nothing more than a Mic and confidence in our own ability.

Adjustable dials are a huge improvement over the fixed ones so I really don't see or understand what you are so concerned about Doc

R

Edited By Ramon Wilson on 08/09/2022 14:33:47

[JasonBModerator @jasonb]

As I said I can't see any reason why it should not work if Doc wants to do it his way and feels happier with that method.

Should soon be easy to assess if it is working as measurements can be taken after the 6thou and again after the first 2thou cuts and compared, if they are significantly off then think again if not then a tenth or two can be taken care of as has been discussed previously

[Dr_GMJNParticipant @dr_gmjn]

Guys – I still don't understand why it's the case that it's the wrong way to do it. Ramon, you said:

"Whilst he object of turning between centres is to be able to remove and replace the component without loss of concentricity…"

If that's the case, then how could different radial dimensions result?

"…it is not to ensure radial dimension is equal both ends."

I'm aware that 'my' method isn't favoured by some machinists, but..why not? As Jason pointed out, it's no different to using hard stops on any other repetitive machining process where the tool feed is fixed.

If anything I'd think 'my' method is more accurate because you're not trying to machine to undefined increments between dial divisions, guided by measurements that are also have inherent errors.

Just asking!

[JasonBModerator @jasonb]

The main reason would think of as to why it's not the done thing in industry for small runs is that a lot of time will be spent swapping the drive dog from end to end as well as the shorter time of actually swapping the part. But what you get up to in your shed is up to you.

[Dr_GMJNParticipant @dr_gmjn]

Jason, fair enough I’ll give it a go on the two roughing cuts for the crank webs and see what happens. If they’re identical to the limit of my micrometer I’ll crack on. If not I’ll measure and cut each end at a time.

[Ramon WilsonParticipant @ramonwilson3]

Doc

Why not indeed – by all means do it as you want to I'm just pointing out to you that you are over complicating things unnecessarily.

You're an academic – I've spent years on the tools – that's the difference.

It may not have registered but I'm trying to show that you really don't need to go down this route in order for accuracy but if that floats your boat then jump in.

I began my model engineering journey just like you – completely without machining knowledge and experience. The only difference was there was no Internet to guide me on a daily basis – just the monthly magazine and club meeting and the odd person – well only one actually – who really helped me. I think I may have mentioned somewhere being passed from person to person at a club meeting when I asked someone what is the best way to cut a thread on silver steel – a difficult material to do that on at the best of times. Each person just passed me on to someone else until I gave up. It would be years later that I realised that in all probability none of them actually knew!

I made an awful lot of mistakes before I gave up a lucrative career in order to machine metal for a living such was its attraction to me. Correct training was a beacon – doing it for a living was a passion. Three of the four jobs I held until retirement were jobbing workshops where time was money and mistakes were an anathema but eventually I ran two of the four machine shops.

I'm just passing on that experience – not trying to vie with Jason or force you down a route you don't want but if something's fundamentally wrong, it's wrong – I think you might agree I would not be doing you any favours by not pointing something out.

Just do it how you think best – what can go wrong, will, and if it doesn't, then it doesn't matter

Best – R

[JasonBModerator @jasonb]

With the bits you are working on at the moment you have a good chance to try it both ways. Do the shaft as you have proposed and then when you soon come to do the crank pins which are most likely to be turned one at a time on the end of a bit of bar you can see how you fare with feeding the cross slide to handwheel readings. You can then decide which way worked best for you and apply that to future jobs.

The crank pins will probably need to be the more accurate of the two as you will probably be turning them to fit a previously reamed hole in the bush so won't be able to bore to size if you end up with an undersize part though the could be bored if you really wanted.

[Ramon WilsonParticipant @ramonwilson3]

So what you are recommending then Jason – if I follow correctly – is

Set roughed out shaft between centres and tighten carrier in place at one end

Take first pass of 6 thou. Stop, remove component

Swap carrier to other end and, without altering the dial, repeat

Take second pass of 2 thou cut. Stop, remove component

Swap carrier to other end and, without altering the dial repeat

Take third pass of 2 thou. Stop, remove component

Swap carrier to other end and, without altering the dial repeat

Really ???

Well each to his own but that's a strange way of going about things to me. Like I said if one hasn't got the confidence to work to a thou or less off a Mic then maybe it's time to reconsider the options

Sorry – not onboard with this one at all

[Dr_GMJNParticipant @dr_gmjn]

So this evening I machined the shaft using ‘my’ method. I was careful to evaluate after the first pair of diameters. I made up the three slip gauge stacks ready for the three length setups:

To compensate for the centre drill error, I used a piece of doubled-up Rizla paper. I wasn’t bothered about the tiny error anyway, but thought it couldn’t do any harm to try to (approximately) eliminate it.

Apologies in advance for mixing units again, but: Got the metric step diameters spot-on (at least to the 0.01mm accuracy of the calipers), and as discussed, left the final cut tool setting well alone before swapping ends round.

Removing the workpiece and swapping the carrier and re-fitting took less than 60 seconds, so it’s really a non-issue.

DIAMETERS:

The result was 3 out of 3 diameter pairs were identical to within the 0.0005” accuracy of my digital calipers. I then switched to my micrometer (0.0001” accuracy) which showed 1 out of the 3 diameter pairs to be identical, the remaining two were 0.0002” different.

There is no way I could have got that consistency by turning individual ends. Clearly I’m relatively inexperienced and not saying for a second that others couldn’t do it, but for me, this method seemed completely legit, and the micrometer doesn’t lie.

As an aside I asked one of our machinists at work about it (time served, very experienced) and he said he’d turn that particular shaft using exactly the same method; there was no reason he could think of why, if the centre setup was correct, it could ever yield different diameters (assuming no significant tool wear).

I had to make at least one error, and so it proved. The middle continuous turned centre piece was supposed to be 14mm diameter, but ended up 13.9mm due to me forgetting to halve the difference to apply to the tool. Thankfully I caught it at 13.95mm due to incrementing down (I took the 0.05mm off to make final cut a fine one).

LENGTHS

Lengths seem OK to within a few thou, but are difficult to measure. I’ll set up in the mill at some point and use the edge finder to bet a better idea of how accurate and consistent they are.

This is after finishing with fine abrasive. I’ve left 0.25mm on the ends (as per instructions), so the centre drillings will end up a bit smaller:

Also added 0.005” undercuts, just because they seemed to look OK, and would remove any doubt as to binding in assembly.

So whatever the wrongs or rights of the method, I’ve finally got the beginnings of a crankshaft!

Edited By Dr_GMJN on 08/09/2022 23:19:18

[Paul LousickParticipant @paullousick59116]

Ref Cutting keyways

The Dec.2011 edition of MEW has an article for making an adjustable broach for cutting internal keyways.

[JasonBModerator @jasonb]

Posted by Ramon Wilson on 08/09/2022 20:51:15:

So what you are recommending then Jason – if I follow correctly – is

 

…………………………………….

Sorry – not onboard with this one at all

I'm not recommending any particular method, but I can't see any reason why the Docs method won't work and it's upto him to choose. As you often say there is more than one way to skin a cat

I have said he can then try the cross slide dial method on say the crank pins as there are two of them to get "identical" and he can see if he gets as good a result.

I'd more than likely do it one end at a time with the dials as I have found that way works best, with time hopefully the Doc will too.

2 tenths shows the method works for him, will be interesting to see if the pins can be made the same.

Edited By JasonB on 09/09/2022 07:06:14

[Ramon WilsonParticipant @ramonwilson3]

Doc

As I said if you don't have confidence in your own ability to turn to a given diameter it's going to be a testing time at times ahead

As for the remark about being time served you've touched a raw point. I've had the pleasure to work alongside some fine time served men in my days from whom I gained much knowledge but I've also witnessed more than enough poorly skilled ones to know that a comparison cut's no ice with me.

The finest precision form grinder I knew who worked in a machine shop I ran was a press operator before he was trained in house.

Yep I've heard all the arguments about time served versus non time served but its what you put on the table at the end of the shift that counts

Jason – I would have thought it obvious by now that I've not been saying it won't work merely that its a fandangled way of going about the process of machining between centres just to get two ends identical. Besides it's a bit disingenuous to say now 'that way works best' for you after the event

Horses and water spring to mind however.

Nice result Doc but now try it the conventional way – you might surprise yourself but ditch the digi calipers for accurate measuring of parts on the lathe – use the mic at all times.

Now that has opened a real can of worms no doubt, nothing wrong with digi calipers of course but they are not the tool for the job for measuring accurately in situ even thou they do measure down to 2 tenths.

I'm off to ponder what I missed by spending my first years of 'working life' in the Army when I could have remained a low paid welder on the shipyard – not even a single bit of protective gear supplied – summer or winter and all the rust you could eat.

[Dr_GMJNParticipant @dr_gmjn]

Ramon, if you read the summary, I mentioned I’d turned the three diameters on one end to the drawing, within the limit of my calipers, so it’s not like I’m totally inept. Using ‘my’ method just ensured the opposite ends were identical. Plus I’ve proved it works with no issue. What’s not to like apart from that some machinists wouldn’t do it that way in industry?

[Ramon WilsonParticipant @ramonwilson3]

Nothing Doc, it's always down to individual choice – just not the way I would advocate doing it based on personal experience. – and no one I should think, certainly least of all myself, considers you 'totally inept'

There always many ways to skin the proverbial cat – just some are recognised and others are not – doesn't matter which you choose if the end result justifies it as you have proved – doesn't make it right, or wrong, for that matter.

FWIW On lathe or mill I use digi calipers a lot to give me an idea of 'where I am' but always use a mic to tell me exactly.

Good luck with the rest of the shaft parts and make up

Ramon

[Ian HewsonParticipant @ianhewson99641]

Ramon, you have more patience than me.

Ian

[Dr_GMJNParticipant @dr_gmjn]

Did some length measurements:

Best case end-to-end is 0.01mm difference, worst 0.11mm.

Difference from nominal lengths best 0.05mm, worst again 0.11mm.

(I think I may have twitched when doing an undercut, hence the odd high readings).

Central length is 0.02mm off nominal, overall length, 0.01mm off.

Central diameter is consistent all the way along.

Surface finish looks ok too, so all’s well that ends well I suppose. Now for the bearings.

[ZanParticipant @zan]

Impressive stuff doc, and it’s interesting reading your quest for high accuracy and you are a really skilled worker oozing quality, but it’s really time you purchased a decent micrometer Cheap digital callipers are great to get you started and for basic use. You are very much a cut above that level of action and with one you will find things a lot easier. I only use mine for quick’ reference and on non critical turning, otherwise out comes the quality micrometer

[JasonBModerator @jasonb]

Look in the photos he has a Mitutoyo mic for final measuring. Same make as his callipers

Edited By JasonB on 10/09/2022 06:51:54

[Michael GilliganParticipant @michaelgilligan61133]

… the process of ‘moderation’ nicely demonstrated.

MichaelG.

[ZanParticipant @zan]

Agreed, my apologies I missed that photo but why then did doc say

“Got the metric step diameters spot-on (at least to the 0.01mm accuracy of the calipers)” when there’s a micrometer I wouldn’t trust my mitutoyo calliper for measuring here but would use my digital Mitutoyo……. Still impressed though!

[Author]

Posts