Here's one to tax the collective brain...it's certainly doing mine in!
I have a one year old Myford Super 7 and I am having a major problem cutting threads, any threads. I have cut dozens on my old Myford and am fully familiar with the general techniques.
The problem is that each time I offer the tool to the work, the position of the cut has moved about a sixty fourth of an inch towards the headstock and is no longer cutting in the original spiral made by the first cut, set on the topslide. After about three or four passes, the embryo 'thread' is completely destroyed This is the case even when there have been no alterations to depth of cut and happens regardless of the point of engagement of the Thread Dial Indicator, even when re-engaged into the same position as the previous cuts, ie always engaging the TDI at position number I, for example. I have thoroughly checked everything possible for security, there are no loose gibs, the tool, work and topslide are all secure, the TDI has been removed and cleaned, the gearbox settings have been checked (I'm currently trying to cut 3/8 Whit by 16tpi and should, therefore, be able to engage the TDI at any point since 16 is a multiple of the 8tpi Myford leadscrew) The TDI is maintaining engagement with the leadscrew and there are no teeth missing from the small gear on the TDI or on the main gear train at the end of the machine.
I'm at a complete loss as to where to go next.....what have I missed??? Answers on a postcard please........!!!
--
Chris Edwards (in deepest Dorset) ..."There must be an easier way...!"
I'm just starting cutting threads on the ML7. Not having a TDI, I stop the lathe, wind the tool back, reverse the lathe past the start, (I have an electric reverse now, I used to turn the chuck by hand), wind the cut back on, and run the lathe forwards.
If you do this, does it cut right? This would elim>Gentlemen
Might have read the original post wrong, but I got the impression that the indicator was being disengaged and then re-engaged at the start of the new cut.
I always worked on the basis that the indicator remained engaged at all times and the spindle was reversed to bring the whole thing back for the new cut, thus retaining the relative positions of everything?
The spindle has to be reversed back past the start point and then brought forward to eliminate backlash of course.
Like I say, I may have misinterpreted the original posting.
Peter
-- Peter & Rita Forbes snipped-for-privacy@easynet.co.uk Engine pages for preservation info:
Thanks to everybody for their comments. I will certainly dismantle and inspect the leadscrew primary drive gear train later this morning, I will also experiment with leaving the tool engaged and winding the leadscrew back to the start point, as Martin suggested. I was trying to cut
16tpi (3/8 Whit) although the problem effects any threadcutting attempts, regardless of pitch. For John, the machine has a Norton-type gearbox driven through the standard primary gear train, ie without the addition of any supplementary gears.
Sorry, Peter, I didn't make it clear - the TDI remains engaged all the time, of course, and it's the saddle clasp nut which is operated at the appropriate marking well back beyond the start point to accommodate any backlash.
It's so nice to have friends at times like these! --
Chris Edwards (in deepest Dorset) ..."There must be an easier way...!"
Check the number of teeth in the primary gear train, you may have an odd sock one on which will give you the same results. On 16 tpi it can't be anything to do with the TDI as anywhere will work. On 16 tpi you don't need a TDI and can do it blind. The error has got to be odd setup or slippage in the train. Also check end float on the leadscrew.
Has your lath got a metric leadscrew, if so when cutting imperial threads I think the TDI comes out right once every 127 turns due to the back gear ratio.
-- Jonathan
Barnes's theorem; for every foolproof device there is a fool greater than the proof.
As far as I am aware Myfords never made a metric leadscrew. The metric lathes had metric screws on the cross and top slides but relied on the old 8 tpi leadscrew and conversion gears.
Well, after another frustrating day, I'm no further forward than I was yesterday.
The only significant bit of new information is that the second and subsequent cuts produce new tracks towards the tailstock end of the work, rather than the headstock, which I thought yesterday.
I still produce rubbish even if I leave the tool engaged with the work, the clasp nuts engaged, and run the saddle back to the start of the cut by means of the reverse idler gears.
A call to Myford brought forth puzzlement a promise to call me back this afternoon....I'm still waiting.
I note John's comments regarding gear teeth on primary train, I'll check tomorrow. Interestingly I did discover significant end float in leadscrew this morning (engage clasp nut and wiggle saddle handwheel back and forth while watching leadscrew)...I thought I'd cracked it when I discovered that the nut holding the leadscrew handwheel, at right hand end, was slightly loose allowing about ten thou end float in the leadscrew, easily seen where the leadscrew passes through the thrust collar next to the right hand leadscrew bracket. Sadly, adjusting this out made no difference to the problem.
Is there somebody out there with a gearbox-equipped Myford who would be good enough to try the following simple experiment for me? I live in the boonies and know of nobody else with a lathe within thirty miles...sad lot! Set the lathe for screwcutting and the box to cut eight threads per inch, in order to produce an easily seen coarse spiral, and engage the thread dial indicator with the leadscrew. Adjust a tool so that it just scratches the surface of a bit of rod held in the chuck and, using the power traverse, cut a light spiral on the work for about an inch or so noting the position at which the tdi was engaged. Return the tool to the start point and repeat the exercise, taking care to engage the tdi at the same point on its dial as for the first cut and without altering the depth of cut or the setting of either the top or cross slide dials. Do the two spiral tracks coincide? Should they not exactly coincide???
--
Chris Edwards (in deepest Dorset) ..."There must be an easier way...!"
Backlash in the train and endfloat in the leadscrew cannot affect the thread if you don't do ANYTHING to break the engagement of the train. i.e. don't use the tumbler reverse, or touch the clasp nut. Simply wind the chuck backwards with the key by hand, its only 8tpi after all, until you are past the start of the thread, to eliminate backlash, and then try another cut. If it still won't follow the thread there must be slippage somewhere, loose chuck, loose toolpost broken woodruf key etc. An "odd" gear somewhere in the train has to cancel out if you wind the chuck backwards
Sorry I can't help as I don't have a screwcutting Myford any longer but the answer to your question is yes they should coincide. On an 8 or 16 tpi pitch [ or any figure divisible by 8 ] there is no way when engaging the clasp nuts you can go wrong.
According to 'THE BOOK' the gear train for a screwcutting machine is 24 teeth driving the first 57 /19 driving the next 57/19 then on to the 72 T gearbox infeed gear. Ahhaarrggh the light has just dawned. The second 57/19 cluster gears is reversible, one way for fine feed, one way for screwcutting. For fine feed you have 24 / 57 - 19 / 57 / 72. For screwcutting you have 24 / 57 driving the second 19 and it's 57 onto the 72.
I don't suppose it's that simple ?
Next thing not in the book but behind the apron on the bottom 1/2 nut here is a small screw that stops you tightening the nuts too much. If that is too tight then it can stop the nuts closing enough and causing free play. To adjust you need to back off, close the clasp nut lever and tighten the screw up until it just lifts the lever off the screw. There is a spring under the head to stop this vibrating loose.
I have never done any screw cutting so sorry if I am teaching Grandmother to suck eggs, but you do have the cross slide set at 90º to the bed don't you?
I have just tried what you asked on my gearbox-equipped Myford. The spiral tracks coincided perfectly; there was no creeping whatsoever. I had a strange situation once involving my previous lathe, a long bed
9" South Bend. I had to thread a piece of stock, one end of which was gripped in the 3-jaw chuck and the other supported with a steady. I was not aware of it at the time, but the bed had a slight twist in it, which apparently was just enough to induce the work to wind itself back into the chuck, a minute amount with every turn. It drove me nuts until I realized what was going on.
Just food for thought, but it might be interesting to try your set up again with a crisp witness mark on the work, against a jaw, and see if if changes position after a couple of passes of screwcutting. If it does, it might indicate a problem with the lathe being out of level, or slightly twisted. No flames, please. Pat Dworzan
You really must do as Martin suggested... take the first cut, switch the lathe off, and don't disengage anything at all. Then back the tool out of the cut, rewind the chuck by hand until you are in a position to take the n ext pass, reset the tool and switch on the lathe to take the next cut. This way you divide the problem into two: if it now works you know the gearing is wrong, if it still doesn't work then something is slipping. You can then concentrate on just one problem.
With so many helpful suggestions to try and to report back on, please allow me to report progress here, rather than answer each individually.
First, and most encouragingly, Martin's suggestion regarding turning the chuck by hand to run the tool back to the start of the cut, leaving clasp nut and TDI untouched, has worked and a second cut, under power, runs in the spiral formed by the first cut. If the cutter is then withdrawn and the tool run back to the start under power and a third cut is taken, again under power, the problem returns. This seems, therefore, to eliminate slippage of chuck, work, tool, topslide etc from the equation and to leave only gearing in the frame....this tends to be confirmed by the fact that the distance between errant spirals is proportional to the pitch being cut. In other words, if cutting a coarse 8tpi spiral, the second and subsequent cuts are a good eighth of an inch back towards the tailstock on each cut, if cutting 26tpi it's more like a sixty fourth each time.
I have checked the bed for flatness, (Pat Dworzan) and the cross-slide is square to the bed (Cliff Coggin). There is no significant end float in the leadscrew and the clasp nut adjustment is OK (John Stevenson) and I -do- have an imperial 8tpi leadscrew (Jonathan Barnes). The tumbler reverse gears remain engaged throughout ( Mark Rand). Both John S and Pat D have confirmed that each spiral should be identical to the first ..I'm starting to doubt my own sanity- which brings us to back to gears.
The primary gear train between the leadscrew and gearbox and the gearbox internal ratios are as supplied by Myfords. I notice, however, that there is a difference between the primary gears described by John and what I have.
John says (quote) According to 'THE BOOK' the gear train for a screwcutting machine is 24 teeth driving the first 57 /19 driving the next
57/19 then on to the 72 T gearbox infeed gear. Ahhaarrggh the light has just dawned The second 57/19 cluster gears is reversible, one way for fine feed, one way for screwcutting. For fine feed you have 24 / 57 - 19 / 57 /
For screwcutting you have 24 /
57 driving the second 19 and it's 57 onto the 72. I don't suppose it's that simple ? (unquote)
I have the two 57/19 gears on studs one and two of the quadrant and a wide 72 on the gearbox input. The 57/19 on stud 2 - is - large gear outwards for screwcutting. However, the tumbler gear (right name? - the gear which drives the 57 part of the 57/19 on stud one is a 38/20. So, instead of having the 24 tooth driving the 57/19 on stud one, as John suggests, I have a 38 tooth. Can this be it? Hard to imagine that Myford's could fit the wrong part...anybody else with a gearbox machine care to count the teeth on the large gear immediately below the pair of fibre tumbler reverse gears? The Myford handbook calls this the tumbler sleeve gear. Unfortunately, Myford is effectively closed for routine business for the next two weeks due to their participation in the show at Castle Donington. I'm rapidly getting out of my depth now so any knowledgable comment is entirely welcome! Thank you all for your contributions.
I have also accepted Mark Rand's kind offer to look in on his way south this week-end...I need all the help I can get!
Watch this space! --
Chris Edwards (in deepest Dorset) ..."There must be an easier way...!"
Couple of things I have thought about but being at work I don't always get time to keep up to speed on all the replies, so apologies if this has been covered already.
If the spiral thread cut by the gearing is identical to the previous one, but is displaced, then the gearing would appear to be OK, which is what I would expect given the previous details of your checks etc.
If the difference/displacement on each cut is consistent and moves only one way (also consistently) then a mechanical problem would seem to be the cause. If the difference itself is inconsistent then something sounds loose.
Going on from there, a consistent movement one way on each cut sounds almost mechanically impossible given that the gear train is the same for each cut, so is there anything in the train that is either slipping or loose ?
Again, apologies if this has already been covered. Kind regards,
Peter
Peter Forbes Prepair Ltd Luton, UK email: snipped-for-privacy@easynet.co.uk home: snipped-for-privacy@easynet.co.uk
Err this may be rather stating the obvious, but you must never, never, never touch the tumbler reverse when srewcutting, or it will make split threads.
ASAIK with the dti, you get ready to cut the 1st thread, with lathe running and clasp undone, and wait for 0 to line up on the dti, drop the clasp, and cut. after cutting, open clasp, wind saddle back by hand, lathe still running, past the start, then drop the clasp on the same number, ie 0.
Not sure if this helps, but my ML7 has a 25t on the chuck, driving the
20t tumblers, driving the 25t primary gear.
For 16tpi, a 20t goes on the primary, and drives a 70t idler on the
1st stud, nothing on the second srud, as the 70t will mesh with a 40t on the leadscrew.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.