Making Cams

[Chris CrewParticipant @chriscrew66644]

I am making the Parkes’ gear hob relieving (backing-off) device which requires the making of a 3-lobed cam with a 3mm lift, or at least the instructions seem to indicate this although I would have thought that a greater movement of the cam follower would be more effective. I have followed the instructions given in the construction notes, MEW 57, with some success, or at least not a total failure but there is a bit of ‘guestimation’ involved.

Not being the most skilled or knowledgeable of back-shed workers, I wondered if anyone can suggest a better approach for producing such a cam or point me in the direction as to where instructions might be found?

Has anyone else built this device? Could they provide any advice?

T.I.A.

[ChrisLHParticipant @chrislh]

Not attempted to make this device I’m afraid but out of curiosity I’ve tried laying out the cam according to Giles Parkes drawing dimensions. It looks nothing like his drawing ! Using his “throw” of 0.030 inches (you quote 3 mm, inside info.?) it’s hardly possible to tell that it’s not a circle. Scaling GP’s drawing a throw of about 0.070 inches looks to have been used. Or am I missing something here ?

As to making such a cam, I have used the E.T.Westbusy method for IC engine cams with success. Basically, the work is mounted on a shaft which is provided with, in your case, 4 centres at each end. One on centre and 3 equally spaced on an offset circle. The work, in principle, is then mounted between centres on the offset centres in turn to produce the cam flanks, etc.

[JasonBModerator @jasonb]

I’d make an arbor to fit the central hole with three equi spaced keyway slots and set it to run eccentrically in the 4 jaw. Mount the blank with a key in one slot and turn one face, reposition and do the next, then the third.

[Chris Crew]

On 21 November 2023 at 12:56 Chris Crew Said:

Could they provide any advice?

Post a picture/drawing of the part you are trying to make and you will likely receive plenty of advice.

My copy of MEW 57 is in the loft and there is a ghost up there, so I won’t be going to look for it just now.

If you are unsure that the dimensions are correct, someone might make you one or more 3D printed prototypes which would allow you to explore many options very quickly without having to worry about workholding.

[bernard towersParticipant @bernardtowers37738]

I have looked in the archive and can’t see it

[JasonBModerator @jasonb]

It is in the archive as I looked at it yesterday otherwise my answere would have been a blind guess

[Chris CrewParticipant @chriscrew66644]

“Not attempted to make this device I’m afraid but out of curiosity I’ve tried laying out the cam according to Giles Parkes drawing dimensions. It looks nothing like his drawing ! Using his “throw” of 0.030 inches (you quote 3 mm, inside info.?) it’s hardly possible to tell that it’s not a circle. Scaling GP’s drawing a throw of about 0.070 inches looks to have been used. Or am I missing something here ?”

In the construction notes it states that one jaw of a 3-jaw chuck should be packed out by 3mm (this is the only metric dimension given in all the drawings and notes) to produce the off-set and a cut put on of no more than 0.030″, the cam blank being fixed to a hexagonal mandrel. I have used both a cam blank of 1.4″ as called for on the drawing and, assuming this may have been a typo, a blank of 1.250″, i.e. one inch and a quarter as all the other dimensions on the drawings are given in this ‘imperial’ format. The smaller blank produced the more defined cam with the larger hardly appearing to deviate from a circle, but in either event the throw seems very small. Unfortunately Dr. Parkes is no longer with us to clarify his intentions, hence my asking if anyone else has built this device, I cannot believe that I am the only one.

I know that this device works as I had a conversation with the late Dr. Parkes when he was demonstrating it at one of the ‘Doncaster’ exhibitions several years ago.

[Chris Crew]

On Chris Crew Said:

In the construction notes it states that one jaw of a 3-jaw chuck should be packed out …

Oh that’s good … I hadn’t bothered to comment, because more experienced cam-profilers had already made their points [no pun intended], but that would have been my approach.

MichaelG.

[ChrisLHParticipant @chrislh]

Point taken re. 1.4 typo as the magazines’ proof readers appear to be frequently on holiday. Just to give you some comfort regarding the small cam throw I have in the past had passable results with a home made hob with no backing off at all.

[Chris Crew]

On Chris Crew Said:

In the construction notes it states that one jaw of a 3-jaw chuck should be packed out by 3mm

Just for completeness, please note that packing out a three jaw chuck jaw by 3mm will not produce a lift of 3mm.

It might be in Tubal Cain’s book: there is a formula relating eccentricity produced by this method to the amount of packing used.

If it is not in his book, it is included in the programs Marv Klotz provides on his website.

https://www.myvirtualnetwork.com/mklotz/files/eccent.zip

Even if you do not use the program, do at least read the notes that accompany it as the packing needed to achieve any particular eccentricity is dependent on the geometry of the chuck jaws (specifically, the width of the part of the jaw that contacts stcok).

In practice, that means unless you have an identical chuck to Giles Parkes, the result you produce will differ from his.

[DC31kParticipant @dc31k]

I think I am too late to edit my previous post: the information on eccentric turning is Section 13, page 12 in the 3rd Edition of Model Engineer’s Handbook.

[DC31k]

On DC31k Said:

… the information on eccentric turning is Section 13, page 12 in the 3rd Edition of Model Engineer’s Handbook.

Thanks for locating that

Happy to remove it if the ‘the management’ objects, but I have put a good copy of that snippet in my Gallery.

CAD-jockeys and Spreadsheet aficionados might enjoy demonstrating the formula

[or they might not]

MichaelG.

https://www.model-engineer.co.uk/members/michaelgilligan61133/mediapress/michaelg/

[JasonBModerator @jasonb]

well this CAD jocky would go straight to CAM and then the CNC.

But for your enlightenment on the left I opted for a 25mm AF hexagon, added 3mm packing to the right hand side  and plotted where the ctr comes. Nice round figure of 2mm offset.

On the right is the resulting part set to a 2mm offset and a 0.030″ deep cut taken at three positions base on 1.25″ round stock to start with.

This is from the article

Worth noting that the various formulas are based on holding round stock but using hex removes the need to allow for the width of the jaw end as the surface always stays parallel to the end of the jaw.

[Michael GilliganParticipant @michaelgilligan61133]

Looks good to me, Jason

MichaelG.

[JasonBModerator @jasonb]

Having now plotted it over the original the only thing is that the original blank looks like it was larger than 1.25″ and a much larger offset would be needed to get the radius of the flank shown.

 

[Michael GilliganParticipant @michaelgilligan61133]

So we’re back to the opening question really … but with your tidy demonstration for reference.

… is there an Excel expert in the house ?

I’ve never got to grips with ‘Goal Seek’ but I presume it could find the answer.

MichaelG.

.

Edit: __ this page apparently has an offset calculator program available

https://www.cncwrapper.com/freeware/

… for MS Windows, I presume

Another Edit: __ looks like Marv Koltz did it in DOS

https://www.myvirtualnetwork.com/mklotz/

[JasonB]

On JasonB Said:

Worth noting that the various formulas are based on holding round stock but using hex removes the need to allow for the width of the jaw end

That is a good observation because it means that so long as the (round) stock is small enough, you could hold it in a hexagon collet block that is held in a three jaw chuck. That would simplify the mathematics.

Would it be correct to say that as the jaw width tends to zero (i.e. line contact), its influence on the calculation reduces? If so, a six jaw chuck with three jaws removed could help as six jaw chuck jaws are naturally a lot ‘pointier’ than three jaw. Pointed soft jaws in a three jaw chuck could also assist.

 

[Michael Gilligan]

On Michael Gilligan Said:

 

Another Edit: __ looks like Marv Koltz did it in DOS

His program is written in C and the source code is provided.

It would be interseting to put his formula up against that of Tubal Cain, to see if they are the same.

If you are fluent in Rosetta Stone and love the challenges of obscure, arcane languages, a translation might be possible.

[Michael GilliganParticipant @michaelgilligan61133]

Regrettably … I would be way out of my depth

MichaelG.

.

[ The nearest I ever get to “in C” is Terry Riley ]

[JasonBModerator @jasonb]

For completeness this is the cam from my first image overlaid onto the drawing using a 1.25″ blank. Either it is a good case for “do not scale from drawing” or somewhere the sizes and text are wrong.

1.4″ blank would be too big as it looks like 1.312″ is about right but offsets and DOC would not match what it says.

Probably wants a 1.4″ blank, a deeper DOC and then a bit of file work to round/blend the corners where the three faces meet will reduce that overall 1.4″ to closer to the 1/312″

[JasonB]

On JasonB Said:

For completeness this is the cam from my first image overlaid onto the drawing using a 1.25″ blank. Either it is a good case for “do not scale from drawing” or somewhere the sizes and text are wrong. …

Thanks again, Jason

I have an uneasy suspicion that the published image is “for illustrative purposes only” … almost certain that it could be drawn with a geometry compass at one simple re-setting.

MichaelG.

[Michael GilliganParticipant @michaelgilligan61133]

It is worth downloading ECCENT.ZIP referenced earlier

… because the included notes are useful in themselves.

MichaelG.

[Martin ConnellyParticipant @martinconnelly55370]

If you want to you can mark the centres of the 4 circles that make this part, the outer Ø1.4″ and the three centres of the cam surfaces, on a bar of say 1.5″ diameter. Then you can set it up in the 4 jaw chuck to do the Ø1.4.

Move to each of the other three centres and do the 0.033″ cut.

Follow this with returning to the centre of the Ø1.4″ to face, drill and ream the bore before parting off.

Martin C

[Michael Gilligan]

On Michael Gilligan Said:

I have an uneasy suspicion that the published image is “for illustrative purposes only” … almost certain that it could be drawn with a geometry compass at one simple re-setting.

… and to demonstrate that: Here is one that I made [considerably] earlier

.

.

as a school woodwork project in 1966

MichaelG.

[Author]

Posts