Member postings for Ajohnw

I always suggest APT carbide - as for instance their own threading insert holder is £24 plus vat. I've bought other holders off them and would have no hesitation at all about buying one. All so far have come with a key and if a screw gets messed up they sell those too. Most inserts come in packs of 10 but threading types singly and some useful hobby turning tips that are actually industrial but they sell them in pairs. Good way of trying them.

Which reminds me I must get another 16mm boring bar as the one I have doesn't hold the tip at a sensible angle.

Think I suggested apt to Trevor anyway.

One thing to add. I tend to use a chaser when I want an accurate form. Usually the hand type held in the tool post. The main problem is that they have several teeth on them, as do the machine types, so a bit more "fun" to use. That doesn't cause any problem at all on Boxford back plates as they can be run all of the way through. In other cases they have to be wound out at the end of the cut. They can still be used into a recess though just like single point tools. I've also cut imperial threads close to some metric ones and then corrected by hand with a chaser to save messing with change wheels.

John

-

Edited By John W1 on 05/11/2015 15:46:09

It's improved since last time I looked Michael. I have several microscopes by Vickers and some CTS bits.

This scope is the only one I have managed to resolve Amphipleura Pellucida using oblique illumination and a vickers 100x na 1.25 achromat. An M15C with the separate lamp for if the link doesn't work.

**LINK**

The first time I did it something looked a bit odd so I did a stria count and that was wrong. Several times since that it comes out correct and I can see mounting defects. I have some fine stuff by Zeiss and Leitz and they don't match up in a number of respects. I wonder if this is down to a comment I read in an old august journal - this use of the light cone emitted from an objective to measure NA has to stop. Leitz optics are poor in some respect but it has to be said that they made every conceivable microscope bit imaginable. Zeiss - hard to find good objectives that aren't delaminated and some are awful even when ok. I also have some Nikon gear bought to obtain DIC but to be honest that doesn't really measure up either but then it's more modern in some respects.

John

-

John

-

True but on larger lathes that might well be comparing chalk and cheese. Actually I have seen "used" as new larger chinese lathes sold as seen. I wonder how that happened. Others with a guarantee.

John

-

It's possible to clean up tailstock morse socket alignment with a morse reamer providing it's not much but if there is some variation according to how it's clamped that might not be a good idea and it's barrel alignment needs checking too really - usual answer is to do it half way extended. There also needs to be some spare morse taper sticking out when one is fitted in the socket. The technique is only intended to correct a few thou alignment error.

The usual technique is to hold the morse reamer in the 3 jaw. Half extend the tail stock and then push the whole thing onto the reamer. Relies on heavy tailstocks. I didn't like the sound of this but having done it all I can say is that it does work. I was told specifically not to clamp the tailstock down. That does make a certain amount of sense with some ifs and buts.

The saddle fit problems could be just casting distortion. That will relate to how long they were allowed to cool in the mould. Machining 2 side releasing stress etc. I understand real lathe manufactures leave them lying around outside for a very long time to destress them. Rust softens the skin reducing machining cost and what stress comes out after machining is finally removed by grinding. Opti do offer one small lathe where they say castings are at least 6 months old. As rolls royce used several years on engine blocks I'd suspect that ain't long enough. A unique comment though.

That hole for the leadscrew ????????? Maybe they changed one end and didn't bother with the other or got it wrong on the basic casting and couldn't be bothered to sort it out.

Just add I am not defending them by the way - just pointing out the complications with expecting better = cost in more ways than many would appreciate.

John

-

Edited By John W1 on 05/11/2015 12:49:40

Edited By John W1 on 05/11/2015 12:55:51

This is around on some web sites but can be hard to find

The "original" bearings were by Timken

Front bearing cone 2788, cup 2720

Back bearing cone 19138X, cup 19283X (35.00mm=1.378" 72.00mm=2.8346" 17.02mm=0.6700".

The front makes up a standard imperial bearing so can be bought complete or like that for more or less the same cost.

The rear one is an odd metric bearing that SimplyBearings list as being imperial and costs £300 plus. It can be replaced with a standard metric bearing but it's slightly thicker. This doesn't cause any problems for me as I did replace it but on some of the lathes the dust cap might need relieving by about 1/2 mm. I think Boxford switched at some point. There is something about this on lathes co uk.

So front 30207-TIM 30207-TIM - Single Row Taper Bearing

Rear 2788-2720-TIM 2788-2720-TIM

Some people have ordered the front bearing and had a UBC delivered instead (China)

Or

Front Timken 2788-2720-TIM - 1.5" x 3" x 0.936" (SKF788/2720)

Rear 35X72X17 (Nachi 30207 Bearing - 30207 SKF) metric

This timlken bearing and the other isn't actually 0.936 wide, more like 1.010 so assume that figure is just the shell. I'll check the sizes of what comes off.

There is a need to turn up discs to use a draw bar to press the new shells in and get the spindle out. That ideally needs a spacer but I use a large tool makers vice as it was handy. Discs were aluminium and circa 5/8" thick. It sometimes takes a lot of force to get the spindle out by driving it through the rear bearing. It would be worth making a sleeve to go into the spindle to make sure that the draw bar is central when this is done. Sometimes it's fairly easy which probably means it's been done before and the spindle dia lapped a little. I did that on mine before fitting a new rear. My front bearing cone was so tight I didn't even try to get off. I'm going to take it to some one with a bearing press. I suspect Boxford used bearing heaters.

There are provisions for using a drift to get the old shells out of the headstock. Tap on either side to keep them square and removing the other parts from the headstock before doing anything else is pretty straight forwards. There are parts breakdowns for all of the lathes on the boxford Yahoo group or I could post one some where else if any one needs it.

There preload setting method is shown in the parts break downs. be interesting to see if it works with new bearings. I've noticed that they allow a little axial movement. This may be when there is no grease in them and might be a better method. However I suspect after the lathe has been run for a while they will need adjusting again anyway. Too much grease in new bearings will mess the settings up. I'm going to try a generous smear / 1cc

John

-

I sounds to me like the most serious and troublesome feature was the lead screw location. The rest in terms of finish on hidden surfaces is a pretty standard feature from comments I have seen possibly on all of them but most definitely on the smaller ones. Grinding costs money.

Finding scribed lines marking hole positions is really fascinating. A problem when labour is cheap but then machines to do that automatically cost and that cost would have to be included in the final product cost. The machines don't last for ever and can be very expensive to buy and run. People funnily enough are cheap and above all else are flexible but also more inclined to make mistakes. I'd guess from what I have seen that marginal changes in the position of these holes don't really matter.

Hopper finding he could work induction hardened cast iron doesn't surprise me. It doesn't mean much unless there is a hardness number associated with it. I managed to work a Myford hardened bed but it was hard. Couldn't touch a Raglan one with anything I had at the time. I could probably mark the bed on my Boxford with a file without too much trouble. It's an ME10. If I tied it on a VSL I might well find it a lot harder.

The fact that the lathe has a single V and a flat for the saddle guide probably allows them to get away with just milling that part of the saddle. Top slides can always be milled simply as the gib strip takes out any angular error. The amount of dish the lathe faces is more important. I'd wonder if they can mill saddle guides for twin V beds but these cost more. 3 V's are better anyway, 2 for saddle and one for tailstock and head stock especially as that allows longer saddle guide as they can pass both the head stock and tailstock. None on the market as far as I know these days not even going up to £14k plus "UK" lathes. This might be why most beds are hardened now as cutting loads are lop sided as far as the bed and saddle are concerned and wear would be rapid.

Take away the lead screw and what Hopper seems to be finding is that the milling under the saddle could be better and if I remember correctly bed edge thickness could be more consistent but other factors don't seem to be too bad. The spindle axis is more or less parallel to the bed but any taper will only be seen when he turns something. So looks like the bed is fairly flat too, People should note how he tested that - I'd suggest making a between centre test bar. Nice exercise and useful for checking tail stock alignment later. For some reason we can't buy them. Only the ones with a morse taper on the end which in some ways are less useful.

From what I have seen if some one want well finished ground accurate hidden surfaces it can be bought - try wabeco. Circa £3000 with no extras for what is essentially a mini lathe with some bad design features. Do people really think that China could get that down to circa £500 by the time it was actually at the point where it was sold? Living on planet zod if they do. Sure it could be well under £3000 especially if they sold lots but it would still be at a price where they probably wouldn't so who knows what price it would finish up at.

People should be glad about cheap smaller chinese lathes in some respects because they can afford them and how often has the term casting kits and sorting out been mentioned. What people should demand is a flat bed and decent spindle alignment and no stupid things like a lead screw being madly out. Some better instructions on setting bearings would be a good idea as well because over tighten and they wont last very long at all.

John

-

If you search treatise on milling on the internet archive 2 ebooks will be found. One by Cincinati and the other by Brown and Sharp.

Old but gears are gears.

John

-

Speed depends on the dia of the cutter and if it's hss or home made from silver steel. As far as hss goes similar to what you would use turning bar of that dia with a hss cutter. Lubricant, cutting oil if you have it and that can be used on any metal and will give an improvement but would be a bit messy if it was caste iron / some brass due to bits rather than swarf.

I suspect this is an optical dividing head. They cost a fortune. CTS later Vickers did have a number of military contracts and were bought out by BAS who then promptly bulldozed the place making sure to bury all the bits. Some reckon to make sure nothing was ever made there again. Rather similar to certain other things that "disappeared" during this era.

Ebay 151873513112

John

-

I would 2nd the elbow height or maybe a touch higher but not lower. At one time people used to fit blocks under vices to get them to the correct height for filing flat. Too high is probably better than too low 'cause if you push your elbow out it also goes up. A good thing to do when doing high precision fitting while seated.

When I started my apprenticeship it began with a long period with a file. On the first day one instructor said shall we get the rising blocks out to the other who said no, Never did get on with that bloke. The other tall one who asked was ok, If some one is trying to file a surface flat to mic levels the height is important and when it's low there is a tendency to rock the file.

John

-

People might be interested in any taper when it turns over say 3" Hopper. maybe a ring at each end. Me - more curiosity than anything else. It's a figure seldom mentioned.

John

-

I must have used a cheap skate version as well OB that needed the flux and etc. To be honest I can't remember anyone bothering to anneal after butt welding but it was brief and a long time ago. The flux and etc is more recent.

When I asked where it was on a much smaller affair I was naive and expected what I had seen and used in industry. The question caused some amusement, the company employed a lot of older ex tool makers. This was my 1st real life intro to home machining. Then I bought a Peatol. I was impressed with that for what can be done at that size. It lived up to my expectations which are pretty extreme really - till something, probably the head bent and wrecked what was brilliant alignment. Then a Hobbymat pretty good too, no complaints other than having to deburr the dog clutch periodically. At that point due to job change it was more difficult to use larger machines so went Raglan. Pretty good too but some wear in the bed that was corrected and I had to remake one of the sides of var speed pulley. Then rust started kicking in so moved into the house. I was aware of the reality of used lathes. Then moved indoors so bought chinese, Disaster compared with what I expected and not a suitable design to sort out Then came a Myford so went for an ML7. Spent ages sorting it out and had missed that the seller set the belts super tight to get round rather loose bearings. Another dose of reality. Then came a super 7 head to fit to it - yet more sorting out and an intended trip to myford for a bed regrind. Then I noticed my current Boxford and bought it more or less blind off ebay. I did ask a couple of questions and got the right answers. Luckily not much needed doing, just a few adjustments. It just leaves me with a couple of niggles that I periodically try to sort out.

That's life for a home machinist, in my case anyway. One interesting aspect is my Dad pointed out that Boxfords were not bad machines when I was at school. Not bad because in his youth he had spent time on lathes when they were intended to offer excellent finish and accuracy and they got near to offering that and aren't too big and heavy. His idea of a small general purpose lathe was a CVA tool room lathe. More complicated bearings than most lathes and extremely expensive. He always reckoned that plain bearings in excellent order are the best. I'll bottom that out one day. He also mentioned that the first job on a new one is to make the next set while you can.

John

-

It would be interesting to see how you get on turning steel on it Hopper. It should be capable of taking some sort of cut.

Saddle lifting as it's moved along the bed by the lead screw is a new one for me. One to tuck away and remember.

John

-

The problem with the GTH one will be the quality of the worm wheels. Industrial units may well have been ground however in real terms it works out and error using 0.001 degree intervals for a vernier scale which he probably intended that for wouldn't be significant in use. On the other hand people have made 127 and other prime tooth gears on them. I don't need one now - came with my Boxford. From memory it was possible to get down to circa 0.0001" or so tooth flank error by adjusting the advance periodically.

I think he scaled for 0.01 degrees actually or something like that. Can't remember. Some one offered me a well made one once. I refused it as I wanted to make something myself. I regret that at times and along came a DW one with my miller.

For profile milling there is a simple answer about that is a little like ball turning. A rotary table with a bar that fits in the side to rotate it and an angle scale around the rim plus tapped holes for stops. The stops have slots in them to allow precise setting to the scale. No back play at all and apart from the graduations easy to make.

If you do buy and fit a stepper Michael best make it one of the accurate ones.

John

-

That makes sense. One thing I didn't use at work was a rotary table.

The GHT head adds a 40T wheel plus worm to drive the 60T worm. It's possible to get just about any division ratio any one might realistically want using that by averaging the errors - the 40T worm is driven by a graduated handle rather than a division plate. 0.15 degrees per turn so suppose it could be graduated to 0.001 degrees. More intended for vernier scales I suspect.

John

-

Thanks Jason. I had wondered. My dividing head is the Dore Westbury design with 60T wheel but he used his own idea of hole counts on the 2 plates to go with it. 14 circles of holes and unusual primes so it misses a fair few tooth counts so I had a look at what happens if 60T is used with B&S plates which have been pretty popular and it comes out like this

That's just up to 100 'cause I got fed up. Coverage is more important the lower the count anyway. If I do one I think I will stick to 60T and fit a bigger handle.

I was rather surprised by the hole counts on the vertex plates. As some primes aren't there they couldn't offer the same coverage.

I'd guess the dividing heads I used at work were 40T so that the worm form was strong to allow them to be geared from the table on a miller. They had us make a 1" twist drill and mill some square threads.

John

-

Edited By John W1 on 03/11/2015 23:27:37

If you see one with a green or blue fibreglass PCB I suspect you will be getting a more reliable part. The one you have and all of the ones I can see on ebay appear to be using a rather cheap and nasty type that is well known for producing dry joints. It's looks like hand soldering too - from the appearance of the parts that can be seen.

John

-

Does anyone know what the advantage of 90:1 is?

I wonder if there is one really. Coming from industry I had used 40:1 for dividing plus compound at times. I went to buy a worm and wheel and they only had the 60 as that is what people want so bought that. Turns out that with the usual standard plates 60 is ok. On the other hand I saw some new Vertex bits available and do want to make a bigger dividing head or maybe / rotary table at some point I wondered about buying them to save making a few bits and pieces. When I checked the hole counts in the plates I found that they wouldn't cut 2 primes that I will probably want. Leaves me wondering if 90:1 is just bigger number must be better. 90 also needs to be a bigger wheel to achieve equally strong teeth,

John

-

I've used both braised and butt welded built into the bandsaw itself Mark. The butt welding could probably be done with a capacitor bank but pass on volts and farads.

I have never ever seen a horizontal bandsaw with this facility but a number of conventional ones with it. I happen to have joined up blades on 2 bandsaws.

I didn't mention that there will also be a means of squaring up the ends of the blade.

I'd better add that what I used on one may well have been welding rod. I just used what was there.

John

-

Edited By John W1 on 03/11/2015 19:25:37

Edited By John W1 on 03/11/2015 19:47:39

I looked at one of your links too quickly Michael and then mixed the numbers up. ISO 4 million. 150 times better than cameras that top out at 25,600 , around 7.5 stops more sensitive. I'm busy.

I posted a shot taken at ISO 256,000 on a well known forum taken with an EM5 and some one thought I had cheated. The EM1 is just as capable. Can't expect a huge dynamic range though at that setting and it needs PP but actually the camera jpg's are not bad and can even be pp'd. At more usual iso levels I have had more problems with a Nikon APS.

I'm a bit disappointed by Olympus's weighty pro lenses, all aimed at faster lenses but will probably go for the zoom and converter but the 300mm is too heavy and limited in use for my tastes. I haven't looked at what they are producing or prices for a while now. I'll probably look again early next year.

John

-

His scientific calculator kit was pretty special too, It used rather poor approximations for trig functions. Miles out.

I did manage to make a decent guitar amplifier out of one of his high powered modules but had incredible problems getting round transformer regulation under low loads. The volts had to be rather high to get the power out.

John

-