Author Topic: Wizard  (Read 393 times)

0 Members and 1 Guest are viewing this topic.

Online cabbage

Wizard
« on: Jul 31 2020 13:32 »
In a book called "Locomotives that never were" is a design consideration by the Crewe works of the then LMS. This is an electric locomotive designed to pull 500t trains between Crewe and Carlisle. The design is based on a Swiss AE4/7 (2-A-C0-1) but configured as 2-D0-2. It is shown in the book in full LMS Crimson Lake pulling the LNWR Royal train.
Both Robert (Doddy) and I have a desire to build it.





The main problem, (for me), is the Buchelli drive system. My soul insists that it use this(!) I think I have a valid G3 scale working drive even though another part of my mind insists - would anybody know?

The nice people at HPC realised that it was not a prank call from another gear cutter, (my Son had to translate BSL down the phone), they were quite helpful. The response was "Yes the bull gear could be made from hostaform or delrin - but the transfer gears which have to be 75% cut away will have to be steel". They too were worried anout the stresses on a G3 scale transfer arm. However I live near a source of aerospace grade Titanium alloy...

The wheel size is amazingly enough 5" 3.5". Off the shelf at Walsall Model Engineering though the castings list at N2.5GA might be suitable(?)

It exists only on paper - which it always did!

Regards

Ralph

Offline Doddy

Re: Wizard
« Reply #1 on: Jul 31 2020 15:59 »
I'm taking a different design approach to Ralph in that I am making the assumption that the LNWR 4-8-4 as originally derived from the Swiss Ae 4/7 that was also (in reality) lengthened to produce the SNCFs 2D2 variants.

For that purpose an Editions Atlas HO scale model of SNCFs 2D2 was acquired to get some perspective of the running gear/chassis of the proposed 'Wizard' model in three dimensions.



The 2D2 5500 class was produced in several configurations - particularly with changes to the cabs noses.
  • Nez de cochon (Pig nose)
  • Femme enciente (Pregnant lady)
  • Waterman (after the rectangular shape of a Waterman ink bottle) - as above
I should add that the LNWRs Wizard has a Pig nose similar to the SNCFs 'Nez de cochon' seen here at the Cite du Train museum in Mulhouse




The 2D2 9100 class was derived from the earlier 5500 class and was the last of the SNCFs rigid framed locomotives. SNCF designed Co-Co locomotives from thereon - the most notable being the CC 7100 class built immediately after the last of the 2D2s.
I have decided to forego the miniature Buchli Drive (prototype as below) and use a tried and trusted pair of internal bogie bolsters, which whilst being completely unprototypical, have excellent ride characteristics and lends themselves to tracking through the tight curvatures of garden railways.




TAFN

Robert
"You don't know what you don't know"

Online cabbage

Re: Wizard
« Reply #2 on: Aug 01 2020 11:30 »
Somewhere amoungst our membership I recall seeing the casting manager for N2.5GA (?) The potential wheel xasting for the Buchelli drive would appear to be from a GWR Hall. The only information I have are the nice jpgs of the two wheel castings. It "might" be possible to drill and tap the drive pins into the counterweight but how big is it and is there enough "meat" on the casting to get a decent length of pin into it. My wheels are old spec 8mm thick.

I have never tapped cast iron - machining it is bad enough!!! I had hoped for 2mm pins but 3mm pins would seem to be a wiser option. I can get 3mm rose joints off
the shelf at  eBay. 

The next stage is to write the GCode for the machine to cut the right shapes out of the delrin gear. I am going to run a few tirals before I clamp the gear to the xy deck...

Regards

Ralph

Offline IanT

Re: Wizard
« Reply #3 on: Aug 01 2020 11:43 »
Tapping cast iron is not really a problem Ralph, provided that care is taken to keep the tap upright and steady, as any 'rocking' motion will tend to damage the initial thread form more so than in steel or non-ferrous - as the CI is brittle. So use a tap guide where ever possible and tap dry (e.g. no tapping paste or compounds). In blind holes, empty or blow the dust out frequently if you can.

Regards,

IanT
Nothing's ever Easy - At least the first time around.

Online cabbage

Re: Wizard
« Reply #4 on: Aug 02 2020 09:57 »
Tim has offered to check my maths... The MOD1 teeth of the delrin bull gear are rated at 0.7Nm of force and the type ZPG gear is 9mm thick. Since the double ended rose joint transfer bars are 6mm thick they can lie within the width of the bull gear as per original. This could give me a potential side play of +/- 1.5mm before things clash.

Now the bad news...

Inorder to get the transfer rods to fit I am going to have to machine two 10mm wide slots 30mm long in a 9mm thick MOD1-72T delrin gear.... It is going to be like an explosion in a pan scourer factory!!!

Hostaform is unpleasant to machine delrin is simply vile...

Regards

Ralph

Online cabbage

Re: Wizard
« Reply #5 on: Aug 02 2020 15:42 »
This is my working drawing! As normal I use 200% scale.





The wobble gear are going to have to be steel... The otjer gears can be delrin (ugh).

The RE540/1 supplies 160g/cm of torque.
The mod1 20t reduces this to 80g/cm.
The torque on the centre of the bull gear is 280g/cm.
At the pivot points for the wobble gears ( at radius 20mm) it is 560g/cm.
Thus (I think!) The gear ends of the wobble gears are locked together by plus / minus 560g/cm.

Regards

Ralph

Online cabbage

Re: Wizard
« Reply #6 on: Aug 05 2020 17:02 »
After two days of applied "gear evisceration" I have decided how top proceed. Really all I need from a gear is the boss and the teeth around the edge! What I propose to do is to produce a "sandwich" of 2mm steel plate, gear wheel, and 2mm steel plate. The plates I nut together and begin cutting away. Very little of the rotary torque is going to be taken by the hostaform centre section, (as most of it will not be there!). The steel plates will provide the structural strength.

regards

ralph

Online cabbage

Re: Wizard
« Reply #7 on: Aug 07 2020 12:06 »
There was a "Pardon"? When my Son rang Technobots  to place my order. But, four hostaform and eight steel gear wheels. All of these are MOD 1 (I keep telling people that I am a lazy person). The GWR "Hall" casting has arrived from N2.5GA and after a suitable couple of cheese doorsteps and mug of tea will be measured to within an atom of its life!!!

I think the counterweight is big enough to tap into bit the calipers have the last say...





Regards

Ralph

Offline Doddy

Re: Wizard - Modellers Licence
« Reply #8 on: Yesterday at 07:17 »
Presumably you will use the GWR Halls counterweight to locate the ends of the driving links?




Buchli drive components described:-


"You don't know what you don't know"

Online cabbage

Re: Wizard
« Reply #9 on: Yesterday at 09:43 »
That is the idea Robert! I will need to do a number of drawing at 400% to make sure that everything will fit to the gear and driven wheel. I had intended to thread the Titanium bar but have been advised against it as my dies are only High Carbon Steel. So, it looks like "rough it over with a file" and epoxy it into the driven wheel... The rose joint ends are those used in 1/10th scale model car racing and can take quite high speed impacts(!)

I might have to appeal for help drafting the steel plates for MEL. I haven't done any real heavy design work in ten years!!!!

Regards

Ralph


Online cabbage

Re: Wizard
« Reply #10 on: Today at 10:50 »
This is my solution using first order levers... I don't like it.





What I think I need to do is alter it to use third order levers. This WILL increase the rotational mass - but make the final design a lot simpler, (I hope!).

Regards

Ralph

Online cabbage

Re: Wizard
« Reply #11 on: Today at 12:26 »
Right, mug of tea and doorstep later... This is the design using 40T gears and third order levers. This I like! The main problem may be the Boss diameter on the 40T gears if I slice into them I can spread the distance between the transfer arms at the expense of increasing the movement of the meshed ends of the gears....

Spreading the distance increases the mechanical advantage. What I am wondering with 560g/cm do I really need it?






Regards

Ralph