BOXFORD IMPERIAL DIAL

[GRIFFIN]

When I recently took delivery of my imperial long bed Boxford AUD MKII, I was very slightly disappointed by the fact that it had metric dials fited. I was told by the previous owner that it had originally been a school lathe and they had probably fitted them  post metrication. I couldn't help thinking that this was a little silly as the machine had an imperial thread cutting gearbox but decided to change them for imperial ones anyway.

I found someone selling them on ebay for £50 each but they were out of stock, so I decided to have a go at making my own. After all, how difficult can it be to scribe 100 equally spaced marks round the circumference of a piece of metal when all you have is a rotary table on a milling machine?
 

I decided to use aluminium, mainly because it would be easier to knurl, scribe and number punch than steel and it doesn't need to be particularly hard wearing. After going to my local Metal Supermarket and picking up a suitably sized off-cut of 50mm round bar for £12, I cut a lump off in the horizontal band saw and popped it in the 3 jaw chuck to start work.

I was pleased with my new Boxford and it was a good job to learn on. I roughed it out first, leaving  some excess metal to allow for errors and finishing to remove any marks made during production. I started the centre hole off with a drill then fitted my boring head into the tailstock to make a precision fit. The next thing was the knurl for which I used a single straight one with the leading edge slightly away from the work piece in order to minimise pressure. At this point I discovered why I shouldn't have lubed the back gears, Etc. quite so much, as the belt started to slip. After liberal wiping of the T lock drive belt with isopropyl alcohol, the slipping stopped and I carried on. Next  turned down the main diameter to raise the knurl above the main body, again allowing an extra few thou for errors in marking the graduations.

Next step was to scribe the increments, which I did using a scriber in the drill chuck of my mill drill and an er32 collet mounted in the 2 morse taper on my rotary table. The ten thou markings were easy as they were 36 degrees apart, I then moved round another 18 degrees and added the five thou marks every 36 degrees. At this point I was glad I left the diameter bigger than needed, as I messed up and got the spacing wrong, so had to remove metal and start again. Second try was perfect as was the addition of the smaller one thou marks which were 3.6 degrees apart. I followed this up by stamping the 2mm numbers round the dial which were not very good, so I was almost happy when I realised I'd done them upside down I then machined a recess in which to add the numbers the right way round and this was the result.

[awemawson]

  It came out well in the end 

[John Rudd]

Nice job Griff, shows how a little effort and perseverance can make a bad job into a good job and less scrap pieces......

Of late, I've had three attempts at getting something right.... Maybe I should take up knitting....?

[DavidA]

Griffin,

Have you checked, with a dial guage, exactly how far each item moves for one complete turn of the dial ?

It may be a metric imperial gearbox, but is it an imperial lead screw ?

same goes for the cross slide.

Dave.

[GRIFFIN]

It's an imperial machine with an imperial gearbox David. the only metric things are the dials. One revolution of the imperial dial is 100 thou or a tenth of an inch that ads up to 2.54 mm, the metric dial says it moves 2.5 mm per revolution, which is 4 hundredths of a mm out per revolution. This was probably quite acceptable in a school environment. Having said that, I will check all screws and gears before I trust it enough to cut any threads.

Cheers, Griff.

[RussellT]

I understand from lathes.co.uk that Boxford metric machines were often fitted with imperial leadscrews.  I've never heard that the top and cross slide screws were actually imperial too.

When I have a minute I must try measuring mine.

Russell

[GRIFFIN]

Just to confirm, it's not a metric lathe, it's an imperial lathe with an imperial gearbox. the only metric things were the dials.

Cheers, Griff.

[SwarfnStuff]

         And if you had not told us the full story we would have simply thought it was a job well done. Still is, Actually. This hobby is full of learning how to make our mistakes look like they are an original feature. (At least that's what I try to do.)
    Great thinking to allow some material for rework if needed Griff.

John B

[RussellT]

Hi Griff

I'm still curious.

My Boxford is a metric lathe.  It has metric dials, imperial leadscrew and imperial gearbox.  All the bolts etc are imperial.

I am wondering how you can tell it's an imperial lathe with metric dials.  Were the dials retrofitted?  Do you have documentation from the manufacturer?  Have you measured the feedcrews?  It seems odd to me to supply an imperial lathe with metric dials.

Russell

[micktoon]

I had a similar thing on my Harrison L5 , the lathe was an imperial lathe, that is imperial gearbox and lead screw but with metric dials. I was told that schools changed over certain parts of lathes to make them 'metric'
  I think what has been mentioned might be right, that is they will have had a fully imperial lathe and to make it sort of 'metric' they will have retro fitted metric dials but also metric feedscrews so the dials read correctly, you could also then cut metric threads with the imperial lead screw with 127 change gears.
   From what I understand if you just swap the dial itself to metric but the feedscrew/nut is imperial it will not read correctly but if you replaced both the imperial feedscrew/nut and dial for a metric feedscrew/nut and dial all will be correct. In the day when they were wanting kids to get into metric thinking the main elements that move would be the crossslide and topslide so swapping these would make them work in metric.
  As mentioned I think you will have to set up a DTI clock gauge somehow on the lathe and then with your old or new dial fitted actually clock how much they move for say a full turn and check it is what either of the dials says , if the imperial dial checks out the feedscrew plus nut will be imperial and if the metric checks out it will be metric feedscrew and nut.
  I am sure I checked mine and it was correct to the metric readings and had metric feedscrews/nuts fitted but can not remember now. I was planning a DRO so did not go any further with altering things.

  Cheers Mick

[Fergus OMore]

Morning Mick et al!

I've just glanced at the post and seem to recall the lathes at Gateshead Tech had  combined metric/imperial dials on imperial feed screws. I'll leave it at that

Happy Easter

Norman

[GRIFFIN]

Russell, Sounds to me like your lathe is the same as mine, imperial gearbox, 8tpi imperial lead screw, this makes it an imperial lathe! Of course you will need the standard imperial end gears in order to get the correct screw feeds.

Of course with the use of different end gears, as Mick said, you can cut metric threads but that doesn't make it a metric lathe. The thing that would make it metric or imperial by definition would be the Boxford name plate, if it is a metric lathe it will have a metric name plate, if it is an imperial lathe it will have an imperial name plate. Also the metric gearbox is kind of a mirror image of it's imperial counterpart, with the tumblers on opposite sides. By all accounts even some of the lathes sold and labelled as metric, with metric name plates and gearboxes, were fitted with imperial feed screws, this was done to use up the stock of feed screws already on the shelf and again they were fitted with different end gears to enable metric thread cutting.

I think I already explained this bit but I'll repeat it in greater detail. As we all know, Boxford lathes have always been used in schools,colleges and with the advent of metrication they couldn't afford to scrap all the imperial parts they already had. Also schools that already had imperial lathes couldn't afford to replace them with new metric ones or fully upgrade them by changing lead screws, gearboxes etc. The simplest thing to do was to change the dials on the cross slides and compound slides. It was a stroke of genius how they accomplished this and a case of using very simple mathematics.

On an imperial Boxford as we all know, one turn of the dial moves the tool 100 thou, or a tenth of an inch, so the dial simply has 100 increments of 1 thousandth of an inch. I'm sure we are all aware that there are 25.4 mm in an inch (approximately), therefore a tenth of an inch, or one turn of the dial, is 2.54 mm, they simply discounted the 4 hundredths of a mm per turn and rounded it down to 2.5mm, after all they were meant to be used by students not in industry. Of course they couldn't put 250 increments on the dial as they would be too close together and hard to read, so they simply made 125 increments, each increment 2 hundredths of a mm. There was no need to change any feed screws or anything else, as this was unnecessary and would have been prohibitively expensive. As Mick said all they did was add a 127 change gear and with the aid of a conversion chart (which is widely available on the internet and in the more recent versions of Know your lathe) and hey presto you could cut metric threads on your imperial lathe.

Hope this isn't too basic an explaination but it really is that simple. Rest assured, as soon as I get a tool holder adapted to hold a dial gauge I will test the theory out with hard facts but I've tried it with a magnetic mounted dial gauge and I'm getting 2.52 mm per turn. I know this isn't a true reading as the magnetic base isn't perfect but it is pointing towards a tenth of an inch per turn!!

Cheers, Griff.

[DavidA]

Griffin,

..Hope this isn't too basic an explaination but it really is that simple..

The whole point bis that the cross slide does not move the tool in as far as the dial says it is doing.

'Rounding up' seems a strange way to go if you are trying to teach precision.

It is short a couple of thou for every turn.  Turn the wheel ten times and you are 20 thou short of where you expected to be.

This isn't too much of a problem as at least you are ending your cut short; you can always take off a bit more.

And, of course, we are assuming that anyone using these machines knows about the shortfall.

I had a similar problem with a mini mill that turned out to be down to the vertical dial having an odd number of lines scribed on it. It took days to work out what was going on.

I have an old lathe that does the same thing. But it doesn't have indicator dials so it doesn't really matter.

Dave.

[Will_D]

It makes sense to me!

After all you are teaching secondary school kids the basics of lathe work.

After the initial guide and 'elf'n'safety lecture (took about 2 mins in the 70s I assume)

The kids were shown how to reduce a bar to a set diameter. I doubt that they were taught to trust the dials and take off say 6 mm to end up at the finished diameter.

They would be taught to rough cut, measure, cut some more and then do a final finishing cut. Done that way the .04 mm error per full turn doesn't matter much!

Are modern day schools allowed to let 15 year olds use lathes and mills etc?

[RussellT]

Hi Griff

Thanks for the extra explanation.

When I have a minute I will have to measure my lathe cross slide and feed screw travel.

Russell

[micktoon]

Hi Griffin and all, very well explained Griffin and I agree with the Boxford set up it makes sense. I also think it makes perfect sense that the schools and colleges of the time just had to do what they could to get metric without the money to go scrapping dozens of lathes as mills, so overall the plan was ok.
  It was bugging me what my dials had been reading on the Harrison but I remembering it was not as simple as mentioned above so I have checked. The dials are metric and I am sure I checked the pitch of the feed screw and it was metric but not 100% about that now time has passed. One full turn of the cross slide dial (graduated and marked 4mm ) is 4mm which is 157.48  thou and the top slide is 2mm which is 78.7 thou.
  So am I right in thinking rather than what has been mentioned above my cross slide will have 4x the error per rev and the top slide 2x the error per full turn. Which would mean if I wanted to fit just imperial dials but correct this error I would have to make a cross slide dial with 157 or 158 divisions and a top slide with say 79 divisions ? I think this is why I thought you could not just swap the dial itself for a stock dial designed for the opposite screw pitch, e.g metric to imperial or vise versa ?
  I think this is why I thought a DRO seemed like a good idea lol
  What do people think , am I loosing the plot or would my options be to make the one off dials as mentioned or get imperial dials and feed screws / nuts.
  Cheers Mick

[philf]

Mick,

Why on earth would you consider a 157 division dial? You'd still have an error and if you wanted to move the slide say 200 divisions you've got to start subtracting 157.

You have metric dials, metric screws and, I'm sure, a calculator. Stick with metric - it's the future

Is 4mm the actual pitch of the cross slide screw or the reduction in diameter for 1 turn? My Boxford Industrial had 0 to 5 mm on the cross slide dial in 0.02mm divisions with a 2.5mm screw. I prefer the actual movement so I made new dial with 0 - 2.5 with 0.01 divisions. My DRO is set to show diameter.

Cheers.

Phil.

[GRIFFIN]

Yeah interesting Mick. I wonder what the graduations are on an imperial Harrison ie how much per full turn?

I'd guess a Harrison would come from a college rather than a school or maybe from industrial use. In either case it would require a more exact solution to the conversion problem and money would have to be spent on a more complete conversion. I'll have to do some more homework and see if there are any Harrison metric conversion kits, etc. mentioned on lathes uk. Certainly the Boxford side of the business was pretty good at using up old stock of imperial parts on metric machines in order to keep costs down.

Just a side note here, the main reason I'm making imperial dials for my Boxford is that I already have a slightly bigger metric gear head Chinese lathe which has a metric thread cutting gearbox and obviously metric dials. Now I have the best of both worlds I can produce metric work on one lathe and imperial on the other without having to use conversion gears every time I go from one to the other. I'm selling my little imperial Warco wm180 as it doesn't have a thread cutting gearbox and it was too much bother changing end gears to do simple threading work. Tee hee, what a lazy b*****d Eh?!!!
 
Cheers, Griff.

[Manxmodder]

Mick/Griff,
The imperial cross slide screw on my Harrison L6 is 10 tpi twin start Acme thread and the vernier wheel has 200 divisions per rev.

Due to the screw being twin start 10 tpi this means the cross slide actually moves 200 thou radial per rev and thus removes 400 thou material cut diametrically.(To take 10 thou off a diameter I have to index in by 5 thou)

This type of screw is commonly fitted to and much more suited to a milling table traverse than a lathe cross slide.

If this screw were replaced with a 10 tpi single start it would then convert the indexing to a more preferable direct diametrical relationship.(10 thou on the dial relates to 10 thou turned off diameter)

My top slide screw is 10 TPI single start and the index wheel has 100 divisions.(100 thou per rev)

I also have a complete set of metric cross slide and top slide screws and index wheels. These are trapezoidal 2mm pitch,though again the cross slide screw is twin lead.

The metric index dials read 4mm per rev for the cross slide and 2mm for the top slide.

.....OZ.

[GRIFFIN]

Thanks Oz,

Thought so, no cheap fix on that one. I wonder how much such a conversion kit would cost in comparison to a couple of dials on the Boxford?

Yeh I remember when I first used my second Warco Chinese Lathe, (sorry for the profanity) the previous one had direct diametrical readings so I assumed the second one also did. needless to say that job went in the scrap bin!!!

Cheers, Griff.

[fishy-steve]

Hi Griffin,
I've just read this thread and must admit I'm  a little confused. From what I understand you think that by changing the cross feed dials from metric to imperial will give you  satisfactory and accurate movement in imperial units?

I've reread Tonys write up on lathes.co.uk concerning the lathes that were partially converted from imperial to metric for educational use during the early  metrication years. From what I understand, the leadscrew remained  8tpi but a metric conversion set of gears were added to the headstock. The imperial gearbox remained. This meant that metric threads could be cut if needed if reference  was made to the correct chart.
You believe that the dials on the cross feed and compound slide were changed to metric but the screws and nuts were left imperial.
That sound like a real fudge.
I made the assumption that the screws and nuts in the cross and compound were changed to 2.5mm pitch instead of 10tpi.
My Boxford AUD is full metric so I have no idea if that is the case.
Could you just put some scrap up and take a light cleaning cut. Zero the dial. Take a measurement with a metric mic. Take a few roughing cuts that amount to just under 3 turns of the cross feed dial (3 turns will amplify the potential  error) then take a light finishing cut. Measure what your left with.
I would be really interested in the result. It would solve my confusion  if nothing else.

All the best.
Steve.

[John Stevenson]

I'll bet Griffin wishes he hadn't bothered now.
Half of these repiles could have been cut out IF they had read the OP correctly 

[RobWilson]

I'll bet Griffin wishes he hadn't bothered now.
Half of these repiles could have been cut out IF they had read the OP correctly 

Did you John 

When I recently took delivery of my imperial long bed Boxford AUD MKII, I was very slightly disappointed by the fact that it had metric dials fited. I was told by the previous owner that it had originally been a school lathe and they had probably fitted them  post metrication. I couldn't help thinking that this was a little silly as the machine had an imperial thread cutting gearbox but decided to change them for imperial ones anyway.

the Op has not even checked the pitch of the leadscrew   

Having said that, I will check all screws and gears before I trust it enough to cut any threads.

OK !  I have a Metric Boxford AUD  , yes the gear box is imperial , this dose not make it an imperial lathe .




So just to prove things a wee test that the dials and LEADSCREWS are metric 


2.5mm per rev dial



ON ZERO


DTI ON ZERO !



One full turn ,,,,,,,,,,,,,,,hey WTF!  it reads exactly , not plus or minus a fannys hair  ,but bang on 2.5mm



mmmmmmmmmmm lets press the imp /metric button 




2.5/25.4 = 0.0984" ,thats a 0.003" error on a part OD/ID  with just ONE full rotation of the leadscrew if it were fit with an imperial 100 thou per rev dial 


Boxford did not fit metric dials to an imperial leadscrew     


Rob

[RobWilson]

PS , my lathe is also from a school



Rob

[GRIFFIN]

Ok the moment we've all waited for, I've fitted a dial gauge into the tool post and as instructed dialled in three full turns on the cross slide !!! Hands in the air' it was 5 thou under three hundred thou, so I was wrong, it has got metric cross slide and compound screws. Looks like it's humble pie for dinner this evening for me!!!

Please don't be too hard on me though, everything else is imperial, I have checked, 8tpi lead screw, imperial end gear set up, imperial gearbox, imperial name plate, HONEST.

The only question now: does anyone have a set of imperial screws and nuts for a Boxford AUD to convert it back to a full imperial machine?