Stuart Twin Victoria (Princess Royal) Mill Engine

[HopperParticipant @hopper]

Posted by Roderick Jenkins on 12/12/2017 15:32:32:

This one was made by the previous owner of my lathe. I believe it is the Mason design as mentioned above. It works very well.

HTH,

Rod

From another thread. Illustrates my garbled description above. I must get around to making on. Handy for graduating dials etc too with a couple of depth stops.

[Dr_GMJNParticipant @dr_gmjn]

Continued with the housings, a repeat of the previous process with a few minor tweaks for the additional radii:

I made a spigot to do the O/D, which worked well enough:

I also left the 1mm oil hole until after boring, because last time it caused a flaw when boring the hole:

Milled a pocket for the oiler bosses:

JB welded in place:

And a Milliput “rolling ball” fillet – far more satisfying than in CAD:

Currently awaiting primer:

Next, the crankshaft I guess.

[Dr_GMJNParticipant @dr_gmjn]

So regarding the crankshaft. Following Ramon's advice, I had a go at grinding an HSS tool. It's not pretty:

But somehow it did give a very good finish, at least until the 'step', when it chattered like mad. I tried increasing the relief at the sides, but to no avail. Also tried slowing the speed a bit, but again no luck:

Something I've never been totally comfortable with was turning to dimension in the Z-axis. Diameters aren't much problem, but specific lengths I've always struggled with in the absence of a DRO.

This is the basic geometry of the shaft, not checked the dimensions, but they're about right:

So, what's the procedure for getting accurate lengths, and the undercut bits?

Do I somehow angle the tool so that the left edge is perpendicular to the Z-axis, turn to diameter, move along to a scribed mark, the feed inwards and hope for the best?

Also, it was mentioned previously to work at the tailstock end, and turn the shaft to enable both ends to be finished. Presumably this is after the central portion is initially turned to size?

When swapping the ends, is it OK to put the carrier on finish machined diameters (maybe with a thin aluminium packing piece), or should the bar be significantly over-long so that the carrier tightens onto a bit that's cut off later, when finishing to final length? I want to face off the centre drillings to leave clean faces for this one anyway.

Thanks.

[Ramon WilsonParticipant @ramonwilson3]

Doc,

The edge faces of the tool need a reasonable back rake, the top rake is about 2-3 degrees, 5 at most, and there's no side rake on the top surface but you could put a little on if you are only cutting on the left hand side of the tool – mine has none so it cuts in both directions. Going right into a corner with all in contact puts a lot of cutting pressure on which quickly leads to chatter if the tool is not really sharp and the head stock a tad worn etc. Try coming up to the corner in several cuts inwards gradually creating the full form – that will improve things considerably. You do not need to undercut the shaft at all but set the leading edge of the cutter square to the shaft and let the tool create a radius.

If you haven't got a dead stop for the saddle then create one by clamping a piece of bar on the bed and coming up to it for depth. Make a simple gauge for the amount of movement required to set the stop bar position.

A method of stopping the saddle at a fixed position is a facility well worth having – once you have it you will quickly realise that. I use mine all the time using slip gauges to set the movement but as said a simple gauge can be quickly made.

Personally I would not machine the centres off in case you need to remount it between centres. EG to polish a tight spot.  Use the smallest centre drill and by all means face the ends to leave a small diameter cone if desired – once you remove them you've lost that ability to set it up again – your choice, but if you do face them off and you do require to remount it you will soon see exactly why I'm saying that.

Make the initial shaft to finished length plus a tiny amount for facing and use packing as you say to prevent the carrier from doing any damage.

 

Best – R

Edited By Ramon Wilson on 24/08/2022 23:54:11

[Paul KempParticipant @paulkemp46892]

If I were making that I would rough out the steps with the raw bar in the chuck at the same time as putting the centres in, hold on the largest diameter in chuck, face, centre, rough out steps and reverse to do the other end leaving diameters say 0.5mm oversize and lengths maybe 1mm depending on the width of the undercuts desired. Set up between centres, finish turn centre and stepped diameters. Grind a tool like a parting tool but with a radiused nose for the undercuts, working from the end kiss the step with the tool, measure the length with the vernier from the end, use the top slide (set parallel with bed) to advance the tool the required amount and put the undercut in, repeat for second and third measuring always the total length from the end (saves tolerance build up) Reverse the shaft, set your dog on shim as you suggest and repeat for other end. However I would be leaving the centres in so you will need to add an allowance to machine them off after.

Paul.

[JasonBModerator @jasonb]

If using that HSS tool I would also rotate the toolpost so that you are presenting the left cutting edge of the tool almost at right angles to the lathe axis but with a bit of clearance. Infact just like your carbide tooling does. That way you can also use it for facing a shoulder square and if fed in a little deeper than the diameter cut it will also do your undercut.

As you have the tool at the moment you get a lot of contact along teh cutting edges and that is going to cause chatter, angled only the radius of the end will be cutting. I'm sure I have seem picture sof Ramon's tool mounted like this, he may be able to post one if not I'll dust off the HSS and take one.

Rather than turn the smallest diameter first which leaves a big step get some ctr support and do the long length upto the central section this will mean that when you get to the step the edge of the tool won't be taking such a deep cut which will reduce the chance if chatter.

Then do the next smallest dia and so on all will end with shallow steps.

I don't have a DRO on my lathe and unlike the myfords the carriage handwheel is insensative and marked in 0.020" increments. I tend to set my topslide parallel to the lathe axis and it's handwheel dial to set a final cut which is then applied with the cross slide

 

Edited By JasonB on 25/08/2022 07:14:26

Edited By JasonB on 25/08/2022 07:17:56

[Dr_GMJNParticipant @dr_gmjn]

Thanks all.

What about using an HSS parting tool, ground with a radius, then do the undercuts to final depth (after roughing to the largest o/d?

Then work on the diameters knowing that the tool will run into the groove and the lengths are already correct?

Ramon – I made a saddle stop a while ago, but probably don’t use it as much as I should. My saddle hand wheel is a bit sloppy, and if using power feed it’s easy to strain the drive when it hits the stop if I do t release the feed at precisely the right time.

Jason – so for the long central length, how do you get an accurate length? Not sure my top slide has the range (will have to check). Do you check for top slide x-alignment with a dti? Mine can turn and just had an angle scale on it with two clamp bolts. So potentially all the careful setup of the tailstock to turn parallel could be negated by using the top slide unless it’s set perfectly parallel.

Edited By Dr_GMJN on 25/08/2022 07:38:45

[JasonBModerator @jasonb]

Measure from the end. You can use the "third option" on a digital calliper against the shoulder to get a good measurement of where you are and just put on the final amount with the topslide

[JasonBModerator @jasonb]

Topslide is set with a dti

BUT it is only used to advance the tool towards the the headstock to put on a cut or two which faces the shoulder to final distance from the end. Not used to turn the diameter that is done with carriage and power feed.

[Ramon WilsonParticipant @ramonwilson3]

Doc,

I don't use the top slide for turning unless I'm turning a taper and certainly would not use it to turn parallel – just too much effort to get it set right only to lose it as soon as it needs moving. You can soon get bogged down in how to set the TS dead parallel which for this operation simply isn't required but, advice is what it is, you do the choosing.

I can't recall my ML7 but setting the topslide parallel interferes with the tailstock on the Super 7 when turning between centres giving restricted movement on the cross slide. My top slide is permanently set at 15 degrees for that very reason and only gets moved to change the angle if required

There really is no need to undercut the shaft, there's no justification for doing so besides it's not full size practice.

The tool you have ground is fine – you may just need to tweak it a bit to improve the cutting – part of the learning process in grinding tools by hand. Here's a short version of the one I posted previously roughing out the crankshaft for the Corliss engine – virtually the same as your shaft just larger

It was finished with the other, longer, tool shown previously. Note the radii – an undercut would be considered to reduce a crankshaft to a thinner diameter for absolutely no practical reason

If you have a dead stop then use it in this case. Carefully face a small length(s) of steel to the dimension(s) required and set the stop. With one hand on the hand wheel lift the self act lever a little before the stop is reached and finish off the last few thou by hand. A dead stop is not designed for running into under power! As Jason I have no DRO on the lathe – doing the above provides an accurate method by simple means.

As always though the advice is based bearing in mind on what kit you have and how I would tackle it as such. You do the choosing

Best R

[JasonBModerator @jasonb]

Yes does come down to what you have, the big extended handwheel that some fit to a Myford will give a better feel as can having a 0.001" graduated scale. Having neither I turn most with power feed and then face the shoulder back using the topslide scale.

I do have one of the extended holders for the Dickson post which allows the topslide to be parallel when using HSS and the tailstock or a DC** holder will reach in a standard holder. This pick also shows how I would angle the tool so it can square up any shoulders as well as turn diameters at one tool position.

As Ramon says it is not usual to undercut a shaft like this but I can see why the article suggested it and would likely do it myself if using the tool you have ground.

The step down in diameters is only 1mm per step, that is going to leave a shoulder of only 0.5mm, now given that the HSS tool looks like it may have a 0.75mm tip radius you are not going to get a flat face to the steps. Add to that any deburring of the mating parts and they are not going to locate along the shaft with any exactness Only way is to do the undercut the depth of which will be governed by the dia of the tool.

Easy to see that the "shoulder" on the left would not give very positive location but the one on the right gives a flat shoulder and provided you only deburr the mating part lightly will give a positive lengthways location to the part

Edited By JasonB on 25/08/2022 10:15:50

[Ramon WilsonParticipant @ramonwilson3]

I would agree to a degree with Jason here Doc as I didn't take on board that the steps were so shallow.

Even so there is no need to undercut unless something is going to sit firmly against the shoulder in question – money and choice are the keywords but an exposed undercut in a shaft for no practical reason is not going to look right – well not in my book

On the other hand if the shoulder is to locate something then yes an undercut would be needed if the step is minimal.

Though I have quite a few Dickson toolholders I don't have an extended version – just using a longer piece of HSS is enough if the occasion warrants it.

To my mind the tailstock clearance on the top slide is one area where Myford could have made a big improvement to the lathe had they chose to do so. Doesn't the Warco have that Jason?

[Ramon WilsonParticipant @ramonwilson3]

As an addition to the last post if you take a look at this image again you will see that where there is need for location – Crank web and bearing seats – the radius is reduced to virtually nothing and the mating parts lightly chamfered to accommodate any slight radius in the very corner – no undercuts required.

 

All other shoulders have a radius – why would there be a need to undercut such and besides it will probably lead to comments such as "What are those undercuts for ?" if you do – horses for course etc

Best – R

Edited By Ramon Wilson on 25/08/2022 11:32:33

[JasonBModerator @jasonb]

Doc's earlier sections through the various parts on the shaft do show them locating against the shoulders though probably only needed for his eccentrics and not the bearings as shown, even then it's debateable if the eccentrics really need it.

EDIT the steps on your crankshaft shown above while I was typing are a lot more than 0.5mm so there is still good location even with internal radius and chamfers of the mating parts. I do a lot of my work with a 0.2mm tip radius insert so apart from threads seldom add an undercut but Doc's tool has a bigger radius

The warco tailstock is not symmetrical when you look from the end unlike the Myford as the vertical part is set towards the rear of the machine so you do get better clearance for the top slide which goes almost under the barrel. I find it's more of a case of clearing the revolving ctr than the tailstock . It could do with a bit more overhang of the casting at the nose as there is a lot further to reach when working on shorter items due to the much wider topslide.

 

Edited By JasonB on 25/08/2022 11:53:43

[Ramon WilsonParticipant @ramonwilson3]

Yes I thought that was how the Warco tailstock was relieved, just like on a Colchester 2000 (and many others no doubt)

As you say though the live centre major diameter can be an issue at times. I have a very nice but larger diameter Jones and Shipman that's rarely used due to that situation so my 'turn to' one most of the time is one of those 'Skoda' centres (remember that constant advert in ME ?) bought years ago that still does the business but I recently purchased one of those small diameter ones from Arc – a real asset for sure as it's been well used since.

The only point that actually requires a step for location are the two faces between bearings and that for the crank webs. As you infer it's much better to be able to position the eccentrics from a lining up point of view so allowing that ability is much the better option

[JasonBModerator @jasonb]

Interesting thoughts about bearing location there.

As this is a twin with a crank either end just outside the bearings I would have thought that the large bearing surface between crank arm and bearing flange would be best to locate the shaft from sideways movement.

If the inner step on the shaft were to be used on Doc's model there is very little area for the side of the bearing to run against and could possible wear quicker than if using the whole side of the bearing to locate things.

[Ramon WilsonParticipant @ramonwilson3]

I suppose that depends on how the bearings are finished off – I don't see this as an issue as you can see the bearing face on the Corliss is just that circle on both sides so the bearing face area is the same either on crank or shaft.

The important thing is that there is clearance on one side or the other so it doesn't matter which side as long as the crankshaft is captive and not allowed to float. Trying to bring both faces on each bearing to a smooth running fit risks the possibility of a tight fit but then the bearing could be tweaked on the faces if so.

This really is a choice thing I guess but I prefer to use the inner faces to keep the shaft in place axially – after all if you don't have a second crank but a pedestal then that's all there is.

As you say Jason, all interesting thoughts throughout a build such as this especially from Doc's perspective

Regards – R

[Dr_GMJNParticipant @dr_gmjn]

Understood about the need for a bit of end float on the bearings. I've modified the cylindrical lengths to give about 0.25mm float, which can be adjusted to zero by snugging the eccentric against the bearing face, the eccentric subsequently being secured with grub screws (the fitting of which has been previously covered by one of Ramon's posts):

(upper geometry new, lower original).

I think we covered securing the crank webs previously, but my intention is to make them a snug fit on the shaft ends and loctite them on during final assembly. I've seen an axial pin put into the interface of the boss/shaft in some applications, maybe this would be a good mechanical solution to prevent rotation, or maybe there's a better method?

Also increases the bearing annulus area by moving the bearing face from the shaft shoulder to the eccentric:

I'm assuming this is what you meant?

[JasonBModerator @jasonb]

Good Loctite joint should be strong enough, there is plenty of area.

Check your eccentric drawing, there is a small raised boss on the opposite side to the main one

Edited By JasonB on 25/08/2022 16:47:19

[Dr_GMJNParticipant @dr_gmjn]

Yes, bosses on both sides of the eccentric with blend radii to the disc. Drawing just represents the overall width of the part to get the shaft dimensions about right.

[Ramon WilsonParticipant @ramonwilson3]

Doc,

I can't think of a single full size crankshaft with an axial pin to prevent the crank moving but there are the odd few that have a tight fitting axial key. As far as I'm aware all full size engines of the size you are replicating would have had the crank shrunk on so a key is unnecessary. Loctite 638 or similar replicates this with ease. The crankpin(s) is just Loctited but I make the last mil or so on the spigot a tight fit in the crank web hole

Apart from one, all the cranks I have made so far then have been Loctited and pinned but with a radial pin going in from the outer diameter of the crank through the shaft and just into the crank opposite. This gives the strongest shear but with only one face to disguise the pin on and a curved one at that – much easier to do that than a flat one if using the likely file and emery paper. The odd one was an attempt at shrinking on but it was well and truly stuck before reaching it's location – never attempted it again.

Crankpins are never pinned but I do make the last mil or so a tight fit as the crankpin reaches its location

That said the shaft and pins are both pinned on this latest one, mainly to assist in reinforcing the shaft as a whole, a 2.5mm pin if my memory serves right. Getting the pin to be invisible is a bit of an art if the crank is a finished item but easy enough if gone about the right way – that can be covered when you get to it. You can just get a hint of it on the upper surface on the right

There's nothing wrong with using the eccentric face as a wear surface just that the eccentric position is then committed.

 

Just realised I have to go out tonight, I need to get a grip – back later perhaps

R

Edited By Ramon Wilson on 25/08/2022 17:45:20

[Dr_GMJNParticipant @dr_gmjn]

Re-worked the shaft drawing slightly, to revise the step lengths to give about 0.25mm float for bearing adjustment. I've also removed the undercut from the first steps adjacent to the eccentrics, since they're no longer needed (nothing abutts them). *ETA I guess I could also eliminate the undercuts at the inner ends of the bearing steps, since they no longer abutt the bearing faces. Would just need a decent chamfer or radius on the inside of the bearing bore?

Just thinking, the exposed shaft each side of the flywheel boss – will it look a bit crappy just seeing a thin gap next to the keys on the left? On the right side my intention is to move the pulley over to contact the eccentric, eliminating the gap at that side. Difficult to visualise how it will look in reality. The left end of the keyway extending into the step also looks a bit bodged, but will ultimately be invisible. Just seems it might look like something wasn't machined quite right (see lower image, basically the area between the left end of the key and the eccentric)…any comments on this area?

Ramon, I pinned the fabricated crankshaft on the 10V without much issue. Pins ended up barely visible:

Edited By Dr_GMJN on 26/08/2022 10:02:24

[JasonBModerator @jasonb]

I don't think anyone else looking at the engine will see that slight gap so not worth worrying about

[Ramon WilsonParticipant @ramonwilson3]

Posted by Dr_GMJN on 26/08/2022 10:00:42:

Ramon, I pinned the fabricated crankshaft on the 10V without much issue. Pins ended up barely visible:

Edited By Dr_GMJN on 26/08/2022 10:02:24

Looks like you've mastered that then Doc

[Dr_GMJNParticipant @dr_gmjn]

Didn't have much workshop time today, so prepared material for the split main bearings.

Hacksawed the square section brass bar into radius sized pieces (the only size I could get was too small to cut in half and re-solder, so I had to offset the cuts:

Milled them flat, with identical thicknesses:

Tinned and sweated them together:

Them milled the assembly to clean-up the sides:

Centre drilled ready for fitting in the 4-jaw chuck. I did this knowing that the offset from each side to the join was identical, and after checking everything was correct with the edge finder. I thought it would serve as a method of double-checking my centralising of the piece in the 4-jaw; I can centre using the telescopic pointed shaft thing, and also using the dti on opposite faces. Pretty confident it's spot-on though:

One question regarding this: I assume that if this single bar is for both bearings, then I bore all the way through (in order to be able to get the journal lengths fully engaged beyond the crank spigot), periodically checking fit on the shaft? Once parted-off for the first bearing, I'd double-check for fit on the other journal? I assume that the bore will taper slightly, so, assuming both journals are identical diameter, the second bore might need a spring pass or perhaps a few?

[Author]

Posts