Stuart Twin Victoria (Princess Royal) Mill Engine

[Dr_GMJNParticipant @dr_gmjn]

Thanks Jason. I understand that a tapered key in a flat keyway would wedge the flywheel out at one end, but only if you apply any axial force to it.

Im assuming the key is tapped into the hub until the taper just touches the edge of the hub (the line contact I mentioned) rather than having any load on it. The grub screw holds everything tight.

Even the parallel key option mentioned has a grub screw for retention, which when tightened might displace the hub slightly, so I can’t see any advantage there?

Understood about the angled grub screw hole.

I don’t know why a parallel key with a ‘head’ wasn’t specified on the plans, since it seems purely a ‘dummy’ feature, added for a bit of realism? The taper is never seen and appears to serve no function?

 

Edited By Dr_GMJN on 10/08/2022 08:19:35

[Ramon WilsonParticipant @ramonwilson3]

 

A wedge type key that you show would be used to apply radial force to hold the flywheel in place hence the gib or head to enable extraction. It would likely be used on the kind of flywheel you wish to represent in three's or four's in order, as Jason has already pointed out, to apply equal radial force and to eliminate tilting as already said – 'staking a flywheel' I believe, is the correct terminology but this is usually when the flywheel has clearance to allow the tapers to trim any slight misalignment the wheel may have as well as retain it

Unless the slot is accessible at an open end, ie the end face of a stepped shaft, this is not the type of key to use in a slot cut in the shaft. Unless it has a matching taper as Paul has pointed out, it's short diminishing surface area is the only driving force.

Such keys would usually bear directly on the shaft surface – usually on flat faces and usually from both sides of the wheel.

A sunken key on the other hand can be across the full width of the wheel and is captive by the slot – it gives full drive and only the smallest grub screw is require to prevent lateral movement of the wheel. If it's made smaller than the width of the flywheel then you could put dummy keys in to represent gib keys

I'm assuming you didn't see my previous post re the cutter?

 

 

 

Edited By Ramon Wilson on 10/08/2022 09:28:14

[JasonBModerator @jasonb]

Hold On, hold on!

I just read a bit more of TC article, the grab screws are on the HEAD side of the hub so won't do any jacking.

If we assume tapered keyways in the flywheel are out for the Doc then what about fitting shorter tapered keys in from BOTH sides. If sized so that the taper just engages with about 1/4 to go then a final drive will see them go in a further 1/8" leaving a suitable gap for removal.

As the wedging action will be at both sides of the flywheel it should be equal and only induce an minute eccentricity which the eye can't see.

[Ramon WilsonParticipant @ramonwilson3]

Well isn't that what I said in the first instance Jason

I'm not familiar with TC article and not sure what you mean by the HEAD side of the wheel but lets just get back to basics. We need to fit a flywheel to a shaft on a representation of a large stationary engine.

Now we can do it to absolute scale on one hand or reduce it to the most simplistic grub screw drive on the other. Somewhere we need to meet in the middle with a fair workable compromise.

We need to drive the flywheel round by the crankshaft so need a basic means to attach the two to do so and a simple straight key is the easiest option (setting aside the need for two at 120 degrees for the moment)

If – the keys are made as shown then the ability to slide them in from the end has to be there – that means slots in the shaft where they shouldn't be – for this representation.

If Doc really wants to show keys of this nature then it would be best to have them without slotting the shaft – ie as full size, on flats – or live with extended keyways in order to do so. Obviously if not on flats these would be dummies and would soon lose their grip so a means to drive the wheel has to be there – one slotted sunken key out of sight and two or more dummy keys to hide it – compromise complete. There would not be the need to do the second slot at an exact 120 degrees to match anything if so.

The other alternative would be a simple discreet but more adequate grub screw that provides the drive but as said that defeats the object

Doc picks up on both our points about the single point contact – just touching so not influencing the wheel to any degree. That's not a satisfactory drive to me – if the grub screw is doing that then the keys a dummy anyway

I guess I'm repeating myself so I'll shut up now – flogging and dead horses spring to mind

Edited By Ramon Wilson on 10/08/2022 10:49:57

[JasonBModerator @jasonb]

Ramon this should show what I meant by "head end" the grub screw is towards the head of the key so will only add further tilting force, had it been at the narrow end of the key it may have counteracted and tilt induced by the key.

I think our thoughts are basically the same, For what the Doc can do with his kit:

Single keyway in the shaft which I would have with stopped ends not extending beyond largest dia.

Single parallel keyway in flywheel as it's easy to cut

Parallel key with a head for looks which will provide the drive.

Small M3 grub screw at each end of this slot to take up any slack in the fit of the flywheel on the shaft and also stop the flywheel moving along the shaft. Screws to bear on key so you don't chew up the shaft

Optional additional keys and keyways for looks only is desired

Edited By JasonB on 10/08/2022 11:25:11

[Dr_GMJNParticipant @dr_gmjn]

Posted by Ramon Wilson on 10/08/2022 09:27:59:

I'm assuming you didn't see my previous post re the cutter?

Actually no I didn't see it – as sometimes seems to happen, It appeared after I'd posted this morning – so it may have appeared I've ignored something, but in fact I never saw it until just now. Probably you posted it as I was writing something.

[HopperParticipant @hopper]

When I made my similarly sized model mill engine I just used a grub screw to hold the flywheel on the shaft and no apparent runout. I did make the reamed hole in the middle a good tight fit on the shaft so it could not cant. I can't see a decorative key not driven in hard would make any more difference than the jacking effect of just a grub screw?

[Dr_GMJNParticipant @dr_gmjn]

Ha ha I'm totally confused by all this.

Why is there any jacking using a tapered key, when the tapered part of the key never touches the hub?

Ramon – why is a tapered key secured with a grub screw a dummy? The driving torque is being transferred entirely by the resistance to shear of the key, albeit over a reduced area due to the taper. The grub screw isn't taking any shear or driving torque whatsoever, it's only there to stop the key from moving relative to the hub.

I totally get the wedging action of the tapered key in a tapered slot being used to avoid putting the flywheel out of plane, but that principle is (I think) totally redundant in how this model was designed. The tapered key actually doesn't need to be tapered at all – as far as I can see, it's a completely un-necessary red herring in terms of function for this model, although I could imagine it may be a bit easier to fit than a parallel key.

Jason – I can't put wedges in from both ends of the hub, because there's no gap on the governor pulley side. I can see that having two tapered keys would eliminate tilt, but only if they were tapped into the hub enough that they would wedge the hub out in the first place. But that's not the case, so….I'm lost.

[Dr_GMJNParticipant @dr_gmjn]

Posted by Hopper on 10/08/2022 12:47:21:

I can't see a decorative key not driven in hard would make any more difference than the jacking effect of just a grub screw?

Same here. I can't see the issue.

[Ramon WilsonParticipant @ramonwilson3]

I really must be getting old! I just checked the Waller, Double Diagonal (which is as far as I am aware a true scale model of the original) and the Mc'Onie.

All have headed keys as shown in your sketch Doc but I know I made them parallel, slightly over thick then filed them with a very slow taper so they bind tight – they are all of course driven from the ends of the respective shafts

(This shows the drive gear but the flywheel is done in the same fashion).

There is nothing wrong with this method but you were saying that the key you have will only stop where it touches and then be held by a grub screw. The small area of the key will soon wear from the inertia of starting and you will have a rocking movement develop quite quickly – only my opinion of course but that's how I see it and it's there for what it's worth.

As Hopper suggests there is absolutely nothing wrong from a practical point of view with using a simple grub screw either – it's just that I was under the impression you were going to make this engine as similar to an original one – in which case grub screw drive (Oh we do need the facility to underline at times) was not in my thinking

Not trying to confuse you Doc , certainly not intentionally for sure

R

[JasonBModerator @jasonb]

Posted by Dr_GMJN on 10/08/2022 12:57:08:

Jason's answers in bold

Ha ha I'm totally confused by all this.

Why is there any jacking using a tapered key, when the tapered part of the key never touches the hub?

The key touches at one edge of the hub, see the ringed area on my sketch in previous post as said earlier this can tilt the flywheel particularly if the hole is a bit loose. A grubscrew in the same situation will also tilt a poor fitting flywheel

Ramon – why is a tapered key secured with a grub screw a dummy? The driving torque is being transferred entirely by the resistance to shear of the key, albeit over a reduced area due to the taper. The grub screw isn't taking any shear or driving torque whatsoever, it's only there to stop the key from moving relative to the hub.

A properly fitted tapered key into a tapered keyway DOES NOT need a grubscrew. The friction of the two tapered faces hold the key in place and neither that or the flywheel can move along the shaft. Key that does not make full contact won;t hold the flywheel tightly and it will soon come loose

I totally get the wedging action of the tapered key in a tapered slot being used to avoid putting the flywheel out of plane, but that principle is (I think) totally redundant in how this model was designed. The tapered key actually doesn't need to be tapered at all – as far as I can see, it's a completely un-necessary red herring in terms of function for this model, although I could imagine it may be a bit easier to fit than a parallel key.

Hence my last image of a parallel key, the taper is not needed if a screw is used

Jason – I can't put wedges in from both ends of the hub, because there's no gap on the governor pulley side. I can see that having two tapered keys would eliminate tilt, but only if they were tapped into the hub enough that they would wedge the hub out in the first place. But that's not the case, so….I'm lost.

It can be done and I have done it but not worth going there as it will just complicate things

I could post photos of 20+ models I've made with proper keys and keyways but again not much point

 

 

Edited By JasonB on 10/08/2022 14:05:28

[Ramon WilsonParticipant @ramonwilson3]

I really do think we are all confusing each other on this one

I thought we had already established that a tapered key in a matching tapered slot is totally different kettle of fish and not relevant to this model – and no, its should not require a grub screw to hold it in position

From the outset I said the object is to DRIVE the flywheel by a key – I agree you will be doing this with Docs drawn key but only by a marginal area. That marginal area, as previously stated, will soon wear in my opinion with the starting torque inertia and lead to rocking. It is not the correct way to fit such a key.

I did not say the key IS a dummy but that it could be – with the drive taken up with a hidden, sunken, key within the width of the flywheel – a far more substantial means to drive the wheel. The headed keys could then be dummies

I haven't built twenty models but those I have I have tried to replicate full size practice where possible. That adherence is not needed, it's certainly not a necessity and I'm not advocating it should be a consideration but the means of drive should be sufficient for the task. I don't consider this so

I used the word compromise and this is what I attempt to do – not always possible of course but I do try. An example is the Marine Compound crankshaft – the coupling and flywheel mount are keyed with sunken keys – full size there would be no flywheel and the coupling would, in all probability, be forged in situ with the crankshaft. Eccentrics would likely be split to enable removal. On this model eccentrics are solid and both coupling and flywheel mount will have a very small grub screw to retain them longitudinally but will not play a part as such in the drive

Docs question at the outset was about tapered keys in a parallel slot (though at first I did think he meant a tapered slot) – I'd like to think that I've tried to answer and show that as best I can

Best – R

[JasonBModerator @jasonb]

Posted by Ramon Wilson on 10/08/2022 14:53:08:

From the outset I said the object is to DRIVE the flywheel by a key – I agree you will be doing this with Docs drawn key but only by a marginal area. That marginal area, as previously stated, will soon wear in my opinion with the starting torque inertia and lead to rocking. It is not the correct way to fit such a key.

Not sure I follow that Ramon. if by drive you mean th erotational load from the shaft into the flywheel then the tapered key into the parallel keyways will transmit maybe 99% of the load that a plain parallel key would as there is almost full contact with the side of the key and the keyway in the flywheel.

The only thing it doe snot do is stop the flywheel coming loose and sliding ALONG the shaft

The whole top half of the key in the image below is driving against the side of the flywheel keyway.

[Ramon WilsonParticipant @ramonwilson3]

Yes I agree Jason but I said Doc's key as drawn – marginal at best.

What you show in your drawing here is a much longer shallow taper key with a considerable length of 'drive' and much improved This is just as I described in my three examples above but the taper filed on top is even shallower and the fit over a longer length very tight (radially) as well as a good fit in both slots. No matter how slow a taper it will only bear on one point of course – if over tightened be ready for that slight but possible tilt though a wide flywheel such as this that should be minimal and yes, as I said too, a discreet grub screw is all that's required to prevent lateral movement along the shaft – something despite the eight keys I did on the Corliss flywheel

 

The two keys at 120 degrees as recommended will of course give a three point contact – personally if I were to be fitting multiple keys I would be milling three or four flats on the shaft

I found this pic of the 'staked' flywheel on the Corliss in the Forncett museum – a bit dark but you can see the annular gap and the four keys if you click on it

 

BTW how do you insert or extract the gib key from the sunken slot and  flywheel in your first image (previous post) if such a long taper is machined on

 

Edited By Ramon Wilson on 10/08/2022 15:34:19

[JasonBModerator @jasonb]

Posted by Ramon Wilson on 10/08/2022 15:31:23:

Yes I agree Jason but I said Doc's key as drawn – marginal at best.

What you show in your drawing here is a much longer shallow taper key with a considerable length of 'drive' and much improved

Doc reproduced the key from the drawings, no actual length given but it does say "length to suit" I would take that as just short of full width of the double hub so it does not hit the pulley.

Scaling in CAD it comes out about 1.75" long overall. The pair of hubs total 1.75" wide so about right

This is it drawn with both at 1.75" long, does not look bad engagement to me, extra 1/8" length on the key would be all it could take.

[JasonBModerator @jasonb]

For flywheels I have done with a tapered key in a stopped keyway mid way along a shaft there are two options.

The small amount of extra length in the shaft keyway is usually enough to prize the key out enough to break the taper. Folding steel wedges (I use one side of my drill chuck removal wedges) will go between head and side of flywheel or as mentioned earlier tow flat bladed screwdrivers have a similar effect.

The other way is to hold the flywheel in your left hand and hit the end of the shaft with a nylon faced hammer in your right which will usually break the grip of the key. This method is also the one to use with a non headed key that is the full length of the shaft keyway. The big Easton and Anderson I made was of the later type

[Dr_GMJNParticipant @dr_gmjn]

I drew it to scale as well, and I wouldn't call it marginal. In fact with a 1:100 taper (nominal 1/8" thick), length my interpretation as "to suit" (as per the instructions), you get all the shear area of a parallel key – represented by the dashed line):

I've not shown the grub screw, but that's pretty irrelevant.

The other key is stated as being "rather shorter".

[Ramon WilsonParticipant @ramonwilson3]

No Doc nor would I, but your original drawing was.

However short of knocking on the flywheel (and off) with the key already in the slot which of course one can do so easily when it's in your hand that is very unlikely full size – which is what I thought you wanted to try to replicate. My apologies for this misunderstanding.

I have found some old photographs this afternoon taken at the Science and Technology museum in Manchester. I can't scan them but I can take an image so will do so later. there are several details that may be of help.

[Paul KempParticipant @paulkemp46892]

Posted by JasonB on 10/08/2022 11:19:05:

Ramon this should show what I meant by "head end" the grub screw is towards the head of the key so will only add further tilting force, had it been at the narrow end of the key it may have counteracted and tilt induced by the key.

I think our thoughts are basically the same, For what the Doc can do with his kit:

Single keyway in the shaft which I would have with stopped ends not extending beyond largest dia.

Single parallel keyway in flywheel as it's easy to cut

Parallel key with a head for looks which will provide the drive.

Small M3 grub screw at each end of this slot to take up any slack in the fit of the flywheel on the shaft and also stop the flywheel moving along the shaft. Screws to bear on key so you don't chew up the shaft

Optional additional keys and keyways for looks only is desired

Edited By JasonB on 10/08/2022 11:25:11

The key to this conundrum is as Jason’s last drawing above if you are fixed on a parallel keyway in the flywheel. Do as Jason indicates with a parallel key and keyways and two small grub screws, that removes any risk of tilt and the angle of the screws oppose each other so it should never move.

The second option is do it as per full size with a tapered key and keyway using a broach and suitable tapered bush. Again if the tapers match and a bit of careful filing will ensure that, whack it in and again it’s not moving.

On some of the other comments; I must be missing something as I went back to the first drawing and it didn’t suggest to me the key would be short. A key drives on the flanks, not the top and I was always taught a parallel key should have top clearance (I have fitted a few on propellers and couplings etc over the years and no one has ever complained it broke or fell off). As stated by Jason a tapered key in a parallel keyway will always come loose.

Finally no one mentioned the modern engineer’s friend – loctite!!!! No grub screws required. The key or keys can be any shape you like. For me the second option would be the way to go (broaching will let your lathe breathe a sigh of relief) but it’s not my engine and not my lathe.

All the best,

Paul.

[Ramon WilsonParticipant @ramonwilson3]

No one in the right mind would use Loctite to fix a flywheel in place. Loctite has its right place in engineering but this is not one of them – That would make the flywheel and crankshaft a one piece item to my mind. Whilst not an actual issue it's just not a direction someone making a representation of a full size engine should consider to my mind. Once it's on it's on and IF it needs to be tweaked there's no chance.

I may not have made many engines by some counts but I have made enough to know that the option of taking things apart is one definitely required on any build.

By comparison to most the the Twin Vic flywheel is wide – without seeing the mating part in relation ship there is no indication of length of the key as shown in the original drawing

I was merely trying to point out that short tapered keys may not provide the drive in the long term if fitted in the intended way – from one end of an open slot.

I cannot think of a single example where a part is assembled to a headed key fitted within a closed slot. No doubt there will be someone with evidence to the contrary but it just doesn't make sense to me based on what I have seen over the years.

As always it's down to choice – I know what I would do and fitting a headed key into a closed slot is not one of them

If the intent is to make something as close to full size is there then why not do it and fit the keys in the proper way as already said grub screws are a compromise.

Edited By Ramon Wilson on 11/08/2022 08:24:00

[Ramon WilsonParticipant @ramonwilson3]

Paul, If my last post came across as a bit short it wasn't intended to be – my apologies.

Having reread your post I went back over my own – this is from my first reply to the original one raising 'keys' three pages back

If you cut the keyway parallel and you can slightly taper the key by careful filing until it tightens in the right spot but it will only be on the high point of course. A parallel key with a tiny grub screw is much the better option – the key does the driving, the screw just enough to keep it in place laterally.

On that note I shall say no more – on this matter

 

Best – Ramon

Edited By Ramon Wilson on 11/08/2022 08:54:06

[Martin ConnellyParticipant @martinconnelly55370]

I have been following this section on the use of tapered keys. There is some discussion regarding a tapered keyway in the flywheel but no one has put forward the possibility of putting a tapered keyway into the shaft. I would think that a tapered keyway in the shaft is a relatively simple option compared to cutting a tapered keyway with a tapered bushing down a hole. Having had discussions in the past (not on this forum) with a designer regarding keyways and keys on tapered joints I see no problem with a tapered slot in the shaft. The keyway slots I ended up doing in the tapered hole and shaft were both parallel to the axis of rotation but were not parallel to the line of contact of the two parts. As Jason or Dr_GMJN pointed out earlier the relevant shear section of the key is not changed by its orientation, parallelism or taper. I would use my 5" sine bar with a 0.05" gauge block to set up to cut both the key and the shaft keyway to a 1:100 taper.

Martin C

Edited By Martin Connelly on 11/08/2022 10:47:36

[Paul KempParticipant @paulkemp46892]

Ramon,

lol, I am afraid it was my rather obtuse sense of humour. I was just surprised given the frequency loctite and similar products get suggested on here that no one had chipped in with the familiar cry “loctite” so I thought I would cast the line! For the record I wouldn’t seriously consider it for this application either but do bear in mind that there are various grades of the stuff and it’s not necessarily for ever as some can be separated by application of relatively mild heat.

Quoting you “No one in the right mind would use Loctite to fix a flywheel in place. Loctite has its right place in engineering but this is not one of them – That would make the flywheel and crankshaft a one piece item to my mind. Whilst not an actual issue it's just not a direction someone making a representation of a full size engine should consider to my mind. Once it's on it's on and IF it needs to be tweaked there's no chance.” If someone is making a representation of a full size engine though why not use the techniques used on the full size? Is a grub screw or two more heretic than a blob of loctite? Neither mirror the prototype

On the closed keyway / gib head key issue I have seen that done on line shaft pulleys but by and large I would agree with you this is not a very common way to fit that type of key.

Martin,

I don’t see any issue at all in reversing the taper and applying it to the shaft, should be exactly the same effect.

Paul.

[Ramon WilsonParticipant @ramonwilson3]

I was a bit too quick to condemn there Paul as I'm a confirmed Loctite user on the right application so yes, on a twin crank web shaft such as this once the cranks are set in place the flywheel (unless it's two halves) isn't going very far but, tongue in cheek aside, I still wouldn't advocate it And of course, the grub screw is heretic but it does have it's place and it does offer a much easier method of disassembly/movement if required.

I've studied a fair number of stationary engines over the years not to mention images but cannot recall ever seeing a sunken gib headed key on a flywheel – unless the slot is open ended.

Whatever, Doc G has more than enough to make a decision on but here are a few images that may help, not only on flywheels but other areas too

You can only just see the staked keys on this gorgeous McNaught Corliss valved engine in the Science & Industry (not Technology as previously said) museum in Manchester

This is the flywheel on the Galloways piston drop valve cross compound engine at the same place

A main bearing on another cross compound in the Kew Bridge museum. Built by Simpson it has an identical twin which was situated in the Strumpshaw St. Eng. Museum in Norfolk. Note the centre in the shaft, no key in the crank web and thin nuts on top plus other details.

And another main out rig bearing, more in keeping with Doc's intentions on the Hick and Hargreaves Corliss valved engine in Forncett. Note the typical variation in nuts on a renovated museum exhibit

With regard to the actual machining a slot or keyway. Like most matters there's a huge variation on the manner in which to achieve either. Regrettably however, most are not conducive with doing it on an ML7 with limited kit – to my mind that is the major factor to be taken into consideration before offering advice that the recipient would find difficult if not extremely unlikely to be able to put into practice.

Best for now – Ramon

[HopperParticipant @hopper]

However, if one did want to cut a tapered keyway in a flywheel bore in a Myford, various of the old books by Sparey and Mason and others show various simple set ups to use the topslide as a slotter by taking out the feedscrew and using a long handlever that is pivoted on a column bolted further back on the cross slide and attached to the topslide. It works like a hand shaper by yanking the end of the lever back and forth. By setting the topslide at an angle, a tapered keyway could be cut. Lot of mucking around though.

[Author]

Posts