Making a Special Beaver Partsmaster CAT 40 Pull Stud

[awemawson]

Ever since I first got my Beaver Partsmaster CNC Milling Machine I've been short of the special pull studs that retain the CAT 40 tools in the spindle. They are similar, but not interchangeable, with other manufacturers studs.

Some time ago I decided to have a go at making some but two things held me back. Firstly these are a highly stressed item - the originals are made from an unknown steel type that has obviously been heat treated. Secondly I wanted to avoid an extra machining operation milling the spanner flat with which they are tightened, and wanted to find a way of doing it on the Beaver TC-20 CNC lathe that I've been rebuilding recently.

The second of these conundrums was solved by much help from other forum members, with Archie (a new member) being very generous with his time helping me, so I have now a method of cutting flats using an axially mounted endmill in a powered tool holder as featured in my recent plea for help with the mathematics of the motion.

So I have decided to go ahead and try and make some initially from mild steel as a 'proof of concept' but no way will these serve as the final items as they will not be resilient enough for tool changing service.

Some time ago I measured up a pull stud and drew it in Autocad, but for this exercise I have re-drawn it, hopefully somewhat more accurately - it's not the easiest thing to measure, but careful cogitation hopefully got me inside the head of the original designer deducing that his angles were standard at 90 and 60 degrees (included angle) thus forcing some of the dimensions to work out as exact millimetres allowing me to predict other dimensions such as fillet radii.

So here is a picture of what I am aiming for, and my drawing of it in AutoCAD

[awemawson]

Now the intention is to make several blank threaded ends with the flats milled on - so standard M16 by 2 mm pitch with flats of 19 mm AF (yes I know in my maths plea thread I was doing them at 18 mm - some are and some are not!)

Once these threaded blanks have been made, I'll screw them into a female M16 thread in stock in the chuck as a 'receiver' then make the 'pointed end'

So next job, draw up both sequences in FeatureCAM and model the cutting and choose the tooling

[awemawson]

Now FeatureCAM lets you run a simulation, which is very handy to check tool clearances and make sure that tools can get at the faces that you want to turn.

So I ran the two operations and took videos of them. On the threading video you will see that there is a threading tool interference issue with the adjacent shoulder - it may not happen in reality as my threading insert has a different approach, but if it does then I'll have to tweak things a bit more

[awemawson]

So that's where I'm up to at the moment. All tools are in the FeatureCAM crib, but I need to mount the SE facing 55 degree diamond tool in a holder on the actual lathe and check the tool setting is spot on before cutting metal.

Workshop time limited today due to external influences (dentist, worming pigs, wife flying to Houston and preparing for one set of guests leaving and two sets of holiday cottage guests coming tomorrow) So probably no more will happen until the week end 

Here is the code generated by FeatureCAM for both operations and also the flat cutting program (yet to be modified for 19 mm flats) just to whet your appetite

[Muzzerboy]

Hi Andrew

A couple of years ago I bought a dozen or so BT40 toolholders from John Stevenson. As you may or may not know, he had a Beaver turret-type vertical milling machine with CNC control. Although it wasn't a Partmaster, the pullstuds look very similar. I have 7 of them that I removed from the toolholders, which you are welcome to.

The critical dimensions agree with your drawing, give or take, and although your example looks a little different (pointy nose, no register next to the thread etc), they would certainly work the same. Mine look somewhat better made, with ground surfaces etc, so if anything I'd guess they are possibly "more original"? Just guessing.

PM me if you are interested. They are yours if you have a use for them and there is little to lose....

Murray

Edit - make that 8.

[awemawson]

Murray that's extremely kind of you and yes I'd love to try them. PM incoming

I live in fear of a slightly wrong stud getting stuck in the spindle as I understand that the gripper assembly and it's Belleville washer stack are an absolute b****** to get apart and back together.

It's always been a bit of a mystery why all mine have the very pointy top - I have a suspicion that it's something to do with aiding alignment as the stud enters the gripper, and as there wasn't ever a through coolant option there was no need for a central hole.

[awemawson]

Murray's Pull Studs arrived this morning - many thanks Murray 

Sadly they are not going to work. The shoulder that the machine pulls on, on my machine is inclined at 45 degrees and starts 21.16 mm from the reference face and runs to 23.25. On the ones Murray very kindly sent, they have a 90 degree face to pull on, situated 28.25 from the reference face. Some of them have the 45 degree face but still too far from the reference.

Never mind, it was an extremely thoughtful gesture 

[Muzzerboy]

Bugger. Worth a try but annoyingly different. I don't know how much movement your drawbar can accommodate but I think you mentioned Belville springs, which suggests very little movement. I have a stiff die spring on mine which can tolerate a fair degree of tolerance on the pullstud position.

Ah, well. Might come in handy as fishing weights...

[awemawson]

Well none of you spotted the syntax error in the thread code


No - nor did I and it took me two days on and off finding those two extra comas 

Turns out that they are a hang over from a previous version of the FeatureCAM to Siemens post processor that I failed to correct last year so variables R23 (number of none cutting passes) and R31 (finished o/d) are terminated by a coma AND a space instead of just a space.

Controller just stopped reporting 'General Program Error Line N605'

These things are always SO obvious once you've spotted them 

[ddmckee54]

Happens all the time, we see what we are expecting to see - not what's actually there.  If you read a sentence where letters have been intentionally left out, we will automatically fill in the blanks with what we expect to see.

It helps to have someone else look at your work, but not all the time.  I remember one instance about 25 years ago where I had one particular line of code in a PLC program that would cause the program to fault.  It was a simple little thing, read this word of status from the PLC's reserved memory location, but every time we ran that code it faulted the PLC.

I looked at it and couldn't find anything wrong with it, my boss looked at it and couldn't find anything wrong with it, the local factory representative looked at the code and couldn't find anything wrong with it.  We finally bit the bullet and paid a factory service engineer to come out and tell us what was going on.  He looked at the code for about 30 seconds and asked "All of the instructions are supposed to be "reads", correct?"  We all three more or less simultaneously answered yes, to which he replied "This one's a WRITE".  We were faulting the PLC by writing garbage into its' protected memory.

You wee what you expect to see, not necessarily what's actually there.

Don