Measuring a double Vee lathe bed Vee position

[Kim GarnettParticipant @kimgarnett94824]

Hi I am trying to measure the center position of the 2 Vees that the saddle sits on a Boxford lathe bed as I am thinking of replacing my existing bed with a second hand one. Now I have a good saddle which I have measured using a surface plate, angle plate and height gauge and had some surprising results in regards to the saddle Vees. Now my problem is measuring the vees on the bed to find the center distance between the vees without using the same method as I used on the saddle as My surface plate and angle plate is to small for the length of the bed. when machining/grinding the bed they must be able to check it while machining to get the center distance correct. This measurement would be needed if I have to recut the Vees in the saddle to reduce the amount of scrapping afterwards, as I don’t suppose that for one minute that they are exactly the same.

Thanks in advance for the help

Kim

[Pete RimmerParticipant @peterimmer30576]

Put some cling film over the bed and mix up two large blobs of body filler. Dump the filler on the cling film over the two vees,  slap a bit of wood on top and let it go off. You might need some pin heads king out of the timber to grip the filler. Once set you have a buck to take measurements from using dowels.

[peak4Participant @peak4]

Just thinking out loud;

Take 4 round bars, all the same diameter.
Lay a pair on a flat surface, or even across the bed front to back, and clamp together, possibly with a matched spacer between them, depending on diameters and bed wear.
Repeat for second pair.
Balance the parallel pairs on the Vs of the bed, front and back, along the lathe’s axis.
Lay another bar perpendicular across them and clamp that down too, thus ensuring all 4 bars are in the same level plane.
Distance between the V centres will be the same as between the fronts of the bar pairs.
Alternatively, lay an extra pair of round bars on top, parallel to the clamped ones, and use those for the measurements.

Bill

[DC31kParticipant @dc31k]

Two inverted vee blocks, measure either the internal or external dimension of their sides and add or deduct two half blocks. It would not even need to be a matched pair of vee blocks.

[DC31k]

On DC31k Said:

Two inverted vee blocks, measure either the internal or external dimension of their sides and add or deduct two half blocks. It would not even need to be a matched pair of vee blocks.

I did think about that, but it might depend on the included angles of the V blocks compared to the lathe’s prismatic bed, hence using round bars.
This is obviously just a quick lash-up using a pair of magnets, to act both as spacers and clamps; better bars would obviously give more accurate results.

Bill

[peak4]

On peak4 Said: 

I did think about that, but it might depend on the included angles of the V blocks compared to the lathe’s prismatic bed

It is a good point. Are there many vee beds around that are not 90 degrees?

I guess that so long as the vee block is the same as or more acute than the bed, it would work but if the vee block were more obtuse than the bed, it would not work.

A lot of the old metrology books show set ups similar to your own, and include a three dimensional case using balls rather than cylinders, so it is a good general method to know about.

[Kim GarnettParticipant @kimgarnett94824]

Hi thanks for the response that is not a bad solution better than I have come up with and it does give a solution if all else fails the only problem I can see is shrinkage of the filler and not knowing the tolerance that these beds were made to and not sure how much of an issue it would be. But there must be away of checking using some sort of measuring device or method that could be done quickly as when these were machined or re machined/reground can’t see them putting filler on the bed and measuring the imprint and there would also not been any cmm probes at the time.

The only thing I could up with is a checking fixture with say 4 dti set at certain heights say the midpoint of the Vees one on each Vee face but this don’t tell you a lot either the other method was to use Vee blocks and a Parallel and slips which again is not ideal

Thanks  again

 

Kim

[Kim GarnettParticipant @kimgarnett94824]

Hi I like the 4 bars I don’t think you need the top bars as long as the 2 pairs were equally spaced you could measure directly the outside dimension minus the thickness of one pair had thought about doing that but didn’t think of  the bars on the top keeping them parallel across the bed. I will knock something up in the morning and let you know how I get on I will also have a go at the filler method as I can work out center that way too and check the angle and compare results using the same method as i used  on the saddle

Thanks again its been a help

[Pete RimmerParticipant @peterimmer30576]

Thinking on this further – if you take two dowels say 10-12mm diameter and fix them (glue, solder, braze etc) onto a piece of cold rolled flat bar say 25mm x 5mm, with a shim in between say 5mm. Now make another set identical to the first. You can sit these on the two prismatic ways and connect them by clamping a longer bar across. Now measure the inside of the gaps and add 5mm and you have your vee way centres.

[Kim Garnett]

On Kim Garnett Said:

Hi I like the 4 bars I don’t think you need the top bars as long as the 2 pairs were equally spaced you could measure directly the outside dimension minus the thickness of one pair had thought about doing that but didn’t think of  the bars on the top keeping them parallel across the bed. I will knock something up in the morning and let you know how I get on I will also have a go at the filler method as I can work out center that way too and check the angle and compare results using the same method as i used  on the saddle

Thanks again its been a help

The reason I took a photo of that setup, was that it was easier to balance; also I wasn’t sure how big a micrometer you had available.

Bill

[Clive FosterParticipant @clivefoster55965]

I suspect you are overthinking things.

The beds and saddles on Boxford (and the original SouthBend) simply aren’t made to the level of accuracy needed for contact on all four Vee surfaces. Maybe once the lathe is well worn all four will have proper contact (in the worn bits) but generally you will be getting proper contact on only two or three surfaces.

Which is the whole point of such technically over-constrained designs. I’ve swopped beds and saddles on both SouthBend and Boxford machines with perfectly acceptable results. Even between makes. Is that South-Box or Box-Bend? with the nearest thing to measuring being a quick eyeball inspection and run of the finger to verify no objectionable wear or surface damage.

In principle double Vee requires only that the Vees are parallel and of nominally the same height and width for sufficient contact to be achieved on enough surfaces for stable guidance. Vee-flat is significantly more sensitive to the relative heights of Vee and flat.

SouthBend were always the masters of (relatively) inexpensive but effective design. As witness their use of oversize holes and poured in place babbit to establish alignment instead of high precision parts and/or finicky adjustment. No way would they have chosen the double Vee with its for four aligned surfaces over the Vee flat which has only three if there weren’t significant manufacturing cost benefits.

Clive

[DiogenesParticipant @diogenes]

^👍^

[Kim Garnett]

On 16 July 2025 at 15:15 Kim Garnett Said:

Hi I am trying to measure the center position of the 2 Vees that the saddle sits on a Boxford lathe bed as I am thinking of replacing my existing bed with a second hand one

Kim

I’m not a boxford man but surely the saddle does not sit on the two vees, it sits on the front V and the flat surface at the rear so not sure why you need to measure the ctr distance. Tailstock and head make use of the other vee and the front flat. Or do the older boxfords have 3 vees?

before over thinking the method of measurement think about what you need to be measuring

[OldironParticipant @oldiron]

I have swapped 3 Boxford lathe beds without any of that kerfuffle. Always been in alignment after a bit of careful pulling down.  BTW Jason is correct in saying that only one V is used for alignment.

[cogdobblerParticipant @cogdobbler]

Some Boxfords have double V saddle. AUD:

 

[Ex contributorParticipant @mgnbuk]

BTW Jason is correct in saying that only one V is used for alignment.

I have largely given up commenting on here, but I don’t belive that is correct for the earlier Southbend derived lathes with the 3 inverted vee bed – the saddle runs on the outermost of the two inverted vees, with the headstock and tailstock located on the inner vee and flat. Even a cursory look at photos of the bed and saddle arrangement for any model of machine of this type will confirm this. The later gear head machines with hardened & ground beds had a different formation & used a front vee and rear flat arrangement.

I would not expect there to be any substantial difference in the top formation geometry between beds. I spent the last year of my apprenticeship at Boxfords when these machines were the main production type & at the time the later geared head machines started production. I worked on (fixing) the Russian Duplex horizontal milling machine that was used to gang mill the top of the beds in one setup. The gang milling cutter was built up on the arbour from mutliple single cutters and this was setup after regrinding to a master bed gauge. At the time Boxford offered a remachining service for worn beds & the same cutter was used to remachine worn used beds as well as to machine new castings. The assembled cutter was a substantial item that had its own wheeled transport trolley to get it between toolroom & machine and required craning into position.

The underside of the saddle was also gang milled & there was not much time allowed for the fitters to scrape in a saddle to a bed on the assembly line, so the bed & saddle machining was closely monitored so that the parts required minimal fitting.

Nigel B.

(who will now return to “silent” mode)

[cogdobbler]

On cogdobbler Said:

 

Some Boxfords have double V saddle. AUD:

 

Thanks, Only briefly used a newer one.

[Kim GarnettParticipant @kimgarnett94824]

Just to update everybody Had a play with trying to measure with 4 bars without much success due a couple problems the tailstock Vee being in the way and the sides of the Vees are different lengths the inner sides are shorter than the outer faces and the unused surfaces where the inner vees merge into are at different heights not by much but there not same which means that I going to have to support the bars from the top and have to measure the outside of the bars as the tailstock Vee is in the way. I will have to make some sort of fixture on the lines that peter has mentioned. It does not surprise me in what Nigel has spoken about in regards to milling the beds I would be interested to know if that milling machine had always been used ? My gut feeling is that it came latter. In regards about contact on the Vees It appears that its the front faces of the vees that wears lifting the rear faces as the saddle drops down at the front this I seen this by bluing the original saddle and had good contact on the front faces of the Vees but not a lot on the rear faces if any and the new saddle did not touch any where.

I didn’t try placing saddle where the head stock sits which would have been worth doing in regards to the new saddle to see what contact there was. In regards to the Vee angle on the bed I very much doubt there 90degrees now as the wear on the front faces is quit a bit. the saddle has dropped quit a bit and higher on the rear than the front so the cross slide tapers up to center rather than dropping down to center which is what suppressed me when measuring the new saddle as the Vee angle is not 135 degrees from the top face its about 134.7 degrees

which makes the cross slide at its maximum diameter above center allowing for deflection of the tool on its largest cut and I have checked this on other lathes and they all taper down when traveling towards the center by about 0.1 mm mine is traveling up by point 0.1mm at the moment.

Kim

 

 

[old martParticipant @oldmart]

Using the far right hand end of the bed where there is the least ammount of wear, use a pair of vee blocks small enough to not reach the flat part of the bed. Put a parallel on top of the vee blocks and measure either between the blocks or outside of them. Add or subtract the width of one block and that should be the spacing. Of course, I have an unfair advantage, at the museum we have at least a dozen pairs of vee blocks from 3/4″ wide to 6″.

When I made the first of two carriage locks for the Smart & Brown model A, I measured the vee angle at the left hand end and it was at least 1 degree from 90 (i can’t rmember if it was 89 or 91) and the lock didn’t work well until I measured the unworn vee at the far end which was exactly 90 degrees. The saddle dips at least 0.025″ due to wear and the rack has been tilted down by that ammount at the left end, and the leadscrew and feedscresw also with eccentric bushes. We have to make do with what we have got, and still get good results knowing the lathe.

[Kim GarnettParticipant @kimgarnett94824]

I know what saying this all started when i put part on the surface plate and it spun like a top it was not facing concaved it was facing convex at which point I checked and lined the headstock and lined up the tailstock both are good for the age of the lathe It will cut parallel no problem but i did have to shim it up as well as the tailstock. However it still cut convex on the face at this point I took the saddle off and re machined the cross slide dovetails which was a bit scary as if I had got wrong the lathe would be no good lucky it went very well . The lathe now works well so in no hurry to change the bed and I have thought about not doing it but I do use it a lot and its not going to get better. I the idea was sort the bed and saddle out so when I do get round to changing it its already.

The problem people sell the beds but without the saddle so I wanted to make sure that if I can get a good bed I can get a saddle to fit well as the cost of the bed then the cost of picking it up would an expensive mistake if didn’t work. I have learnt a lot and pretty sure that  i could recut the vee ways in the saddle if I needed to getting to the current problem measuring the bed as I was taught many years ago you can’t machine something if you can’t measure it

kim

[Author]

Posts