Mathmatition needed (lathe change wheels)

[ford34/2]

Anyway, if I understand correctly, what John really needs is a 127T as neither the 63 nor the 64T will be close enough...

So why on Earth did Harrison in their wisdom supply the gears as standard ( Inc 63T) with the later metric machines and even include the gearbox changewheel and 9 tumbler selector information plate and fasten that plate to the inside of the changewheel guard..This gave 36 TPI ranges from 4 up to 60TPI

Normally you need the Harrison recommended gers set 50-63/40-120 which is an overall reduction of 0.264550 (50/63)X(40/120)/
The perfect reduction to correct the 6mm pitch leadscrew would be 0.264583 so the error here is -0.012%

Stan, (with the same 140 Harrison lathe as John)

 

[graffian]

Whats this to do, metric? I haven't read all the thread.
I use 55/65 gears. The closest pitch I could get to 1.0mm is 1.0006mm. I needed a second compound
set to get 1.75mm and even then it's the pitch that has most error, I can't remember what it is without
looking to see what gears I have modified and doing some sums;-)

If you work it all out you should have a set of fractions and it's then pretty simple to work any possible pitch and gears. I can never remember how I worked it all out and have to work it out from scratch every time I want to use metric.
I keep meaning to write it all out, laminate it and stick it on the lathe.

 

[Pete.]

EDITED: brain fart - shouldn't post technical stuff so early in the morning.

 

[Wedg1e]

So why on Earth did Harrison in their wisdom supply the gears as standard ( Inc 63T) with the later metric machines and even include the gearbox changewheel and 9 tumbler selector information plate and fasten that plate to the inside of the changewheel guard..This gave 36 TPI ranges from 4 up to 60TPI



Stan, (with the same 140 Harrison lathe as John)

Errr... I dunno, I think I may have misunderstood your post number 6 I guess...? I think I'd got it into my head that what was needed was a 127 but the only one available was a 63 and John wanted to make a 64 as it might be nearer...? Just trying to be helpful

 

[ford34/2]

Hello Mr.Wedg1e,
No problem, As i understand it John bought a dividing head and has just received the cutters from China, but sadly the head has no 63 division, the nearest on it is a 64. He wondered just how near a 64 would be to my 63. I`m finding it very difficult to work out as after the changewheel combinations we have another set of 9 cogs built into a gearbox, with 3 selector levers mooving more cogs,giving an option of 36 permutations. My original workings out gave us the 36 imperial TPI threads from 4Tpi up to 60Tpi. Since that time maybe 2 or 3 years ago we have come accross an original screwcutting plate attached to a machine that was for sale quite recently on E bay. Sadly the seller was not as friendly as us chaps and didnt respond to either of our requests for a good picture. Both our Lathes are Metric versions, all dials leadscrew etc all metric. So as we all know we need a translation gear(s) to convert the metric setup to an imperial one. Now as good as Harrison Lathes are, their documented information and manuals are pretty basic to non exisistant-- compared to say Colchester. In Johns latest Sunday night cap he may well touch on this subject but I havent watched it yet.


I have every confidence John will have this sussed in a very short time, when succsessful he may well produce a few for sale.

Regards
Stan

PLEASE NOTE IN THE SUNDAY NIGHT NITE CAP, VERSION 88, JOHN HAS HAD A PROBLEM UPLOADING THE CORRECT EDITED VERSION. THE LINK CONTAINED SWEARING , SO I HAVE REMOVED THE LINK IN CASE ANY ONE IS OFFENDED
I uploaded the wrong video (the unedited version)
I don't swear on my vids as a rule
The video had 2500 hits before I realised my mistake
I will upload the clean version later
JOHN

 

[Pete.]

For 63 divisions you need a 63 hole plate and index every 40 holes. For 64 you only need a 24 hole plate and index every 15 holes - assuming a 40:1 ratio. In the home shop I'd print & paste a paper scale onto some sheet and spot-punch/drill that. It's how I make all my plates. Could also use a DRO with hole circle function if you had more time.

 

[rikrobson]

I have every confidence John will have this sussed in a very short time, when succsessful he may well produce a few for sale.

There was a lot of swearing getting the brake disk true in the 4 jaw chuck.

 

[doubleboost]

There was a lot of swearing getting the brake disk true in the 4 jaw chuck.

I uploaded the wrong video (the unedited version)
I don't swear on my vids as a rule
The video had 2500 hits before I realised my mistake
I will upload the clean version later

 

[Maker]

Did you edit out all the attempts of getting it true? I was thinking it looks like a good method but you didn't make it look easy, I'm guessing it would be much easier if you were indicating on the outside of a part?

 

[doubleboost]

Did you edit out all the attempts of getting it true? I was thinking it looks like a good method but you didn't make it look easy, I'm guessing it would be much easier if you were indicating on the outside of a part?

That was the video at filmed no editing at all lol

 

[ford34/2]

Had alook in the loft for the heastock spindle off the Milnes lathe with the dividing bit thing on it.
Sadly the graduations dont apear to be of any use, marked as 180,120,112,96.
Dont think thats any use to a 63 requirement.
Stan

 

[AndyStobbs]

Get a 127T timing belt, mount on a plywood form as an outer band. Index from that. That was my plan anyway

 

[Henrik]

You could print a template out here and use that on your rotary table:

http://bilar.co.uk/engineering/

Look under "index plate generator". My plan is to use this for a cheap approximation to real indexing plates to use with my rotary table at some point.


Then you can have a play here to find the required plate and steps:

http://bilar.co.uk/cgi-bin/division-calculator.pl

Unfortunately, as 127 is a prime, you cannot divide it into two whole numbers, so to get an exact match you will unfortunately need a 127 hole plate.

However, the good news is that you don't have to be super accurate when you're making it, as the worm ratio will help in making any error in the indexing plate SMALLER on the actual divisions produced.

And further further to this, you can have a play here to see how close to a perfect pitch you can get with the gears you have or are proposing to make:

http://bilar.co.uk/cgi-bin/change-gear-calculator.pl

Phew

 

[Henrik]

Had alook in the loft for the heastock spindle off the Milnes lathe with the dividing bit thing on it.
Sadly the graduations dont apear to be of any use, marked as 180,120,112,96.
Dont think thats any use to a 63 requirement.
Stan

If you can drive the spindle through a set of gears they might be. Have a play with that division calculator I posted, and set the worm ratio to whatever great set you want to test, and set the available plates to 180,120,112,96

63 is not prime (21 * 3), so you may be in luck

 

[Henrik]

sorry, I've just realised my mistake - your index plates are attached to the headstock..

 

[rikrobson]

63 is not prime (21 * 3),

3*3*7

 

[Gareth J]

Since that time maybe 2 or 3 years ago we have come accross an original screwcutting plate attached to a machine that was for sale quite recently on E bay. Sadly the seller was not as friendly as us chaps and didnt respond to either of our requests for a good picture.

Do you have the "many knobs" Norton gearbox? Or the ABC "peasant" selection? My 140 has the three choice 'box but does have a metric translation plate if it's of any help.

I can't remember if there was a copy of one in the literature you sent me or not...?

 

[ford34/2]

Our 36 speed Norton gear boxes have 3 selector knobs, The machines are both factory produced metric versions,--- dials,cross slide, compound slide, tailstock quil, feed shafts.All factory supplied screwcutting, is metric.
The current objective is for John to be able to replicate the 63T changewheel he has, he has the cutters, but sadly his rotary table doesnt include any 63 combinations on the dividing plate(s).

Stan

 

[Gareth J]

Our 36 speed Norton gear boxes have 3 selector knobs, The machines are both factory produced metric versions,--- dials,cross slide, compound slide, tailstock quil, feed shafts.All factory supplied screwcutting, is metric.
The current objective is for John to be able to replicate the 63T changewheel he has, he has the cutters, but sadly his rotary table doesnt include any 63 combinations on the dividing plate(s).

Stan

As I thought; my single knob, 3 position 'box cutting chart won't help.

 

[doubleboost]

Any 140 owners well worth a look
http://www.ebay.co.uk/itm/HARRISON-...424948?hash=item5b1b8fadf4:g:SioAAOSw~bFWL0-h