PDA

View Full Version : Dual scales on slide dials.



Peter.
01-10-2009, 04:18 AM
My lathe has dual scales on the dials, imperial and metric. The leadscrew seems to be 1/2" x 10TPI, and the dials I have are graduated as .100"/turn (.001" per graduation) and 2.5mm/turn (.02mm per graduation). Obviously the two don't relate perfectly .100" being 2.54mm, so one or the other (I'm pretty sure the metric one) is .016" inaccurate over a turn of the handle. Is it normal for makers to do this or standard practice?

Ian B
01-10-2009, 04:36 AM
Peter,

Are you *sure* that the two dials rotate at the same speed? My lathe also has dual dials, made by CMT. The feedscrews are imperial, the imperial dial is fixed to the feedscrew. The metric ring is driven by internal tooth ring gears at a 125:127 ratio, and rotates at a slightly different speed. This exactly compensates for the "inaccuracy".

Regards,

Ian

Peter.
01-10-2009, 04:55 AM
Hi Ian.

Never thought of that. This isn't a 'high-quality' lathe, having been made for the schoolroom market, so I doubt it. All of the lathe is metric bar the slide screws as far as I have seen. I'll go and check it out.

Peter.
01-10-2009, 05:10 AM
No, the scales are fixed together. By using my vernier and measuring across 20 divisions on the metric scale at various points and either side of 0 all I can conclude is that they must have simply spread the error across all of the 125 divisions on the dial.

JCHannum
01-10-2009, 08:41 AM
The best solution is to rig an indicator on the slide and see what one turn of the screw actually equates to.

Some of the cheaper imports use either metric or imperial thread and fudge the dial graduations of one system to make it work while others graduate the dial correctly in both systems, leaving one with an odd finish count.

There are also a few with metric leadscrews and imperial dials that are fudged and don't relate to anything at all.

Peter.
01-10-2009, 08:53 AM
That's a good idea.

I figured it's an imperial screw because the OD is 0.500". According to the machine plate the leadscrew is 3mm pitch and it has a metric quick-change threading box.

John Stevenson
01-10-2009, 09:03 AM
If it's an import it could be anything.
They often decide on a diameter and then add the pitch needed so you could have a 1/2" x 10 tpi and also 1/2" x 3mm pitch. It means the build up for the family of machines, bearings etc, is still the same.
Often even using the same dials.

Of the cheaper fixed dual dials only one is right and the other fudged. Some times it's easy to spot as they religiously keep to an accurate division but at the end it may read 73 and a bit.
Other times they stretch the reading to get a full dial and then you have to indicate to see what's what.

Better class machines have the internally geared dials so they do read accurately

Peter.
01-10-2009, 09:05 AM
This is no import!

Denford Viceroy made in Brighouse. Don't get much more British than that :)

1st pic is the model I have - 280 synchro: http://www.lathes.co.uk/viceroy/page6.html

John Stevenson
01-10-2009, 09:36 AM
Of course it's an import, it came from Brighouse, that two continents away from Kent :D

You will have to clock it out and then see but if they are equal division one is out. Not a good way to do it, the unequal divisions are better as once you are close, lees than full turn you can use the divisions accurately.

I think my metric TOS has unequal thou markings, not certain as in 8 years I have never looked at them :rolleyes:

Yup

http://www.stevenson-engineers.co.uk/files/dualdial.jpg

This is direct reading so each imperial division is actually 0.001"

Peter.
01-10-2009, 10:28 AM
I think this is going to be 'one of them days'.

Using the dial gauge, four turns of the cross-slide screw moves the slide 9.98mm. Since there are 100 one-thou increments on the imperial dial if it read correctly it should have move .400" or 10.16mm. I guess it must be a 2.5mm pitch screw.

What I can't understand, is why on earth someone would turn a thread with 2.5mm pitch and an OD of .500". More to the point where the heck will I find a tap to make a new cross-slide nut when the time comes?

John Stevenson
01-10-2009, 10:40 AM
Bite the bullet now.
Replace the screw and nut with standard 1/2" x 20 thread and fit one of these.

http://www.stevenson-engineers.co.uk/files/digital%20dial.jpg

Digital dial from Arceurotrade.
reads imperial or metric at the touch of a button, ability to zero at any point and you don't have to count turns.

It's not a DRO it's a digital dial and doesn't take care of backlash but neither does a dial.

[EDIT] it needs 20 tpi no matter what diameter to work the encoder correctly.
.

J Tiers
01-10-2009, 11:37 AM
What do those dials cost?

Appears to be a workable solution so long as the dials plus the required new screws don't cost close to a cheap DRO.

As far as the screws and dials, early 1980s chinese import machines apparently sometimes had metric screws with an english dial just put on... the "nearest" metric was used, with no real thought as to whether the dial would be accurate.

Maybe the makers thought the dial would "convert" the machine by itself. Or maybe they had seen a picture of the dual geared dials, and didn't realize they were geared.

Davidhcnc
01-10-2009, 12:54 PM
I find myself using different lathes on different sites and never try to figure out the scale language used.

I keep a sticky back (magnetic) long travel dial gauge. Stick it on and read it!

Teenage_Machinist
01-10-2009, 01:05 PM
MAAAKKEEEITTTTT!!!!!!!

If it is that large you might single point internal it.

djc
01-10-2009, 01:15 PM
...it needs 20 tpi no matter what diameter to work the encoder correctly.

Just wondering if with 20tpi it works in radius or diameter mode? Hence, might there be the possibility of using it in the other mode with 10tpi (sod's law says it will require 40tpi - a bit fine for a leadscrew)?

P.S. For anyone interested in dual dials, have a look at UK patent 1 192 831.

Peter.
01-10-2009, 02:28 PM
What do those dials cost?

Appears to be a workable solution so long as the dials plus the required new screws don't cost close to a cheap DRO.

As far as the screws and dials, early 1980s chinese import machines apparently sometimes had metric screws with an english dial just put on... the "nearest" metric was used, with no real thought as to whether the dial would be accurate.

Maybe the makers thought the dial would "convert" the machine by itself. Or maybe they had seen a picture of the dual geared dials, and didn't realize they were geared.

Those digital dials are cheap - 75 for the dials, screws,nuts and all to to both cross-slide & compound.

John Stevenson
01-10-2009, 03:03 PM
Peter, that kit is a bolt on conversion for the C3 lathe and unless you have that lathe the screws and nuts are useless.

Nothing to stop you buying the kit and just using the digital dials.

Peter, I may have one spare sample dial, I'll email you later.

At 20 tpi these give true travel so 10 thou takes 20 off diameter.

A 10 tpi screw will work but you will have to halve the reading.

Peter.
01-10-2009, 05:36 PM
Cheers John, I appreciate that.

Trouble is I don't even have a 10 TPI screw. It's a strange arrangement for sure. The screw is certainly 1/2" diameter, but has a 2.5mm pitch.

http://peterrimmer.myby.co.uk/images/csscrew.jpg

http://peterrimmer.myby.co.uk/images/csscrewcheck.jpg

But the nut is made in imperial dimensions. This and the fact that the screw was 1/2" caused me to assume the pitch was imperial too.

http://peterrimmer.myby.co.uk/images/csnut.jpg

They did make an imperial & metric version of the lathe so I'm wondering
if it's been fitted with the 10TPI nut, or whether they just made the stupid metric/imperial hybrid thread in that too. Certainly adds a new dimension to the imperial/metric - which is best argument :D

So no matter what happens I can't use that digital dial without making new screws & nuts at 20TPI, at least my tests show the metric scale to be pretty accurate.

Peter.
01-12-2009, 02:22 PM
Just to follow-up, I've now got the factory drawing for this nut and it is indeed a mixed imperial/metric thread, and a left-hand one at that! Gawd knows where I'll get a replacement when the time comes - and I know of at least one other person who needs one.

http://peterrimmer.myby.co.uk/images/nutdrawing.gif

Peter N
01-12-2009, 03:27 PM
Bastard threads must be flavour of the month, I've just cut a 12mm x 40 TPI on two parts of a current little project :D

As for making a new nut, I think you mentioned your leadscrew has a pitch of 3mm, so I'm assuming you should be able to set the gearbox or geartrain up to screwcut a 2.5mm thread L/H thread? So why not get a lump of bronze and have a go at making one yourself. Easier than an internal R/H thread as you're cutting away from the chuck.
All you have to lose is a little bit of time and a few bob for the bronze.

Peter

Peter.
01-12-2009, 03:35 PM
I might have a go at making one in aluminium. Gaining threadcutting experience is one of the main reasons I bought this machine.

John Stevenson
01-12-2009, 03:57 PM
You probably want a new screw anyway so why not make a new screw and a short piece for a tap at the same time.

Sounds more work but you then have a method of replacing the nut later.

Peter.
01-12-2009, 04:21 PM
Strangely enough, I might have tracked-down a piece of screw with a mind to making a tap from it. My cross-slide screw appears to be in quite good condition, and the nut isn't at all bad either. Just looking ahead here.

What's the most reliable method of determining the wear-condition of a square-threaded screw?

miker
01-12-2009, 05:02 PM
Peter,

At 20 tpi these give true travel so 10 thou takes 20 off diameter.

A 10 tpi screw will work but you will have to halve the reading.

John, does that mean a 10 TPI screw will give you the actual material removed reading? In other words, if it reads 10 thou, that is what will be removed and you wouldn't have to halve the reading.

Rgds

Peter.
01-12-2009, 06:48 PM
I think it would go the other way - 10 thou on the dial would remove 40 off the diameter.

miker
01-12-2009, 08:04 PM
Thanks peter. It's no wonder I ruin so much material. :)

Rgds