By continuing to use this site, you agree to our use of cookies. Find out more

Member postings for Clive Foster

Here is a list of all the postings Clive Foster has made in our forums. Click on a thread name to jump to the thread.

Thread: Screw cutting BSP threads
22/07/2017 17:38:06

Myford leadscrew accuracy is said to be 2 thou per foot or better. 1/10 th per inch is same order of magnitude so probably of no great importance in practice. Especially if the errors subtract.


Thread: woodworking
22/07/2017 17:26:45

On a similar task, although maybe my pallets weren't quiet as heavy as yours, I found a long handled pallet breaker did the deed. Being too tight to buy one used found bits to make my own ending up with a nearly 4 ft long handle cos thats how long the most suitable piece of material was. 5 or 6 ft of scaffold pole should shift anything!

Friend Roddie cheats by chainsawing the joined parts out and raking the nails out of the ashes after burning.


Thread: Conversion table
16/07/2017 01:16:15

When searching for near equivalent sizes the best and most nearly comprehensive list to be found on the internet is that originally due to Andy Pugh. It presents a goodly selection of threads in simple nominal outside diameter size order. All threads are given in both imperial and metric units regardless of the official system. The original has been extended and re-formatted for easier reference by other workers.

Original text format file here :- **LINK**

Prettier Excel format version here :- **LINK**

PDF version arranged to print out on A4 pages here :- **LINK**

Don't recall where I found the one I keep on the computer. In Excel format and styled like the print out from PDF version except for a having the thread type designation data at the top. Its a single unpaginated list with only one row of column headers. Using a horizontal split screen display the designations and column headers at the top can be fixed with the sizes showing in a second scrolling window. Much easier than a long list.

Certainly not my re-arrangement because I always put "harder" breaks every 5 and 10 lines in long tables to facilitate reading across.


Thread: Pulleys and pulley wheels?
15/07/2017 13:13:48


At the quick and dirty end its probably easier to address the problem by using outside diameter of pulleys and outside diameter of the belt. Won't be dead accurate but I find it better than results from the proper calculators when fed with wrong numbers due to incorrectly specified components.

The proper calculator formulae use belt pitch length and pulley pitch diameter. Pulley Vee angle varies for smaller pulleys made to specification which doesn't help.

Effective belt length is the length about the effective outside diameter of the pulley sheeve at a specified tension and wrap as defined in the specifications. Which is rarely the actual diameter.

Inside diameter is frequently the one given as the nominal size printed on the belt. Probably chosen as being the easy one to measure. Put the belt on a pair of flat pulleys under suitable nominal tension and measure the separation. Add half the flat pulley diameter to the separation. Nominal length is double that result.

I'm not sure that the folk who make belts and pulleys really understand it either :- **LINK** . Frankly if I wan't that much of a headache I'd rather get it the old fashioned way via a bottle of Glenfiddich!

No great help with your pulley problem tho'. Get two of appropriate diameters and allow enough wiggle room to fit. When you get into the lower end of the market there is no telling what you will actually get in pulleys or belts. For one job I ended up with had three pulleys of very different diameters but same nominal size. A section, maybe 3/4" variation from largest to smallest. On another three different lengths of A section belt with same number covering three "proper" sizes. This went down really well as I trying to buy 4L section.

This link is helpful in converting nominal length to outside length :- **LINK** .


Thread: UK company supplying knobs, handles etc?
15/07/2017 12:04:19

Banbury Plastic Fittings **LINK** came up trumps for me when I needed a slightly unusual sized knob with threaded stem for a Creusen grinder tool rest. Other folk had too big or too small. Decent range of handles, hinges, cable management and other stuff.


Thread: mt3 to r8
14/07/2017 17:05:13

I'm surprised that no supplier offers a limited set of ER collets comprising only standard milling cutter shank sizes for pure milling machine use. Cost savings should be appreciated by typically impecunious Model Engineering / Home Workshop types. It would be no great problem to offer expansion sets to cover the full range of native imperial or metric sizes should the purchaser eventually decide to use them on the lathe as well.

I prefer the Clarkson style treaded shank cutters for un the appropriate chuck milling in my Bridgeport. I do have an ER32 set on an R8 taper, purchased following magazine advice when I got an R8 square column mill. Almost unused since I rapidly bought the Clarkson set-up following practical advice from an experienced person. Along the way I picked up native R8 collets making the ER system redundant. No use to me on the lathe as mine has a native 5C spindle. A full set of x 1/64 th imperial and half set of x 1 mm 5C covers all my needs.

Never been that impressed from an engineering viewpoint by the ER system due to the monster tightening torque needed to ensure accurate results. The can hold accurately at lower torques, especially when used with minimal compression on material whose nominal size matches that specified on the collet, but if you want specified accuracy everywhere in the range book torque figures must be followed. You won't do that with the tiddly sheet steel spanner in the set.


13/07/2017 10:54:53

If the machine is also sold with an R8 fitting then there is almost certainly room to re-machine the spindle to take an R8 as Mike suggests. Seems very unlikely that the main spindle dimensions would differ between MT3 and R8 versions. If an R8 option was offered maybe look into getting an R8 spindle as a spare part so you can do a straight swop. Machining is going to be tricky unless you have a decent size, very accurate, lathe with appropriate tooling and its not going to be a fast job. Objectively you could probably earn more than the cost of a spindle, if such is available, in the time it would take you to do the job.

Was asked to look into similar issue some years back on a spindle that really was a bit too small OD to put an R8 in. Concluded that the only way to do the job was to make up a permanent insert with most of the R8 contained in a cylindrical section fitted into the bored out spindle with a projecting end made larger than spindle diameter to carry the taper. Probably added about 1/2" or so to the effective spindle length with part of the taper running into the main cylindrical body. Sanity prevailed and the idea never got past the drawing board, well CAD programme actually, so things like fitting and fixing never got looked into. Shrink fit the mane body and weld round the end I guess. Finish machine in situ.


Thread: Top slide stud too short?
12/07/2017 14:31:12

There are good, albeit somewhat arcane, engineering reasons why its correct to have the tool post bore rather larger than the stud diameter with a short, top hat style, bush to both span the difference and apply clamping pressure. Its also correct to have a large diameter annular relief in the centre of the tool post base rather than making it flat. Typically the diameter of said relief is around half the length of the block sides.

Probably the most obvious example of why this arrangement is a good idea is to consider fitting the post to an elderly machine whose stud is no longer completely straight and slide top no longer perfectly flat. On smaller machines it doesn't take much more than occasional careless over-tightening to pull the top slide surface up a little where the stud threads in. Due to the relatively coarse threads involved such pulling will almost certainly shift the stud out of perpendicular too. Relatively small studs are easily bent a bit too.

Clearly if the tool post is bored a close fit on the stud and its base made dead flat it becomes very hard to get a firm, rotation free, seating of the tool post. Resistance to rotation being very important with a QC system because not only do all the loads feed to the slide via the post but also the tool tip is considerably offset from the post centre line so has considerable leverage. With the oversize bore and adapter bush at the top small deviations from perpendicular or small bends in the stud will tend to be pulled out when the nut is tightened down. Obviously the securing force will not be quite as great or evenly distributed as it will with a perfectly perpendicular and perfectly straight stud but small deviations can be accommodated in an adequately satisfactory manner.

Similarly a flat bottomed tool post on a distorted or damaged surface may only have proper contact over a relatively narrow annulus surrounding the stud. Not good for resisting rotation as the contact area will be small and the tool leverage large. With the relieved base not only is the contact area much larger, even if the actual width is no greater than before, but its a lot closer to the tool tip correspondingly reducing the leverage. Even if the actual mounting surface on the slide is a little uneven, so not all of the available surface on the base to the tool post is actually in proper contact, the actual holding area will almost certainly still be much greater than with the flat base contacting only around the stud.

Many folk, myself included, consider a thin washer of stiff card or aluminium alloy underneath the tool post gives a useful improvement in grip and allows a lower torque to be used on the mounting nut. Need a new washer every couple of years but thats no great issue. On some of the smaller machines high torque on a QC tool post mounting stud can distort the slide enough to impede its operation. Not enough to lock things up but can certainly destroy the nice even feel so important for precise work.


Edited By Clive Foster on 12/07/2017 14:32:33

Thread: Quick change tool post query
11/07/2017 22:25:12

Not a lot you can do about it with a conventional QC tool post system. Overhung tool, round the corner load transfer and relatively inefficient part turn wedging fastenings rather than conceptually more secure screw threads are inherent and inescapable features of the concept. As ever with effective devices correct component selection and proper installation allow you to engineer your way round flawed concepts but objectively its less than ideal.

Choose a good quality toolpost of the correct size and you should be fine, My practice is to set the top slide at around 25 - 26° offset from parallel to the cross slide. This improves the load path via the gibs. The common practice of setting top slide parallel to the cross slide puts a primarily rocking couple via the dovetails into the cutting load paths. Setting at a shallow angle gives a more direct path for a significant portion of the loads. It also reduces direct loading on the screw, offsets the slide handles so the don't get in each others way, which can be very important on small lathes, and is ready for screw cutting duties if you use the zero-2-zero method. Win-win-win-win arrangement.

A QD two way block system will give a more direct load path but, for some reason, such devices never caught on commercially. Pretty easy to DIY if you can come up with a simple part turn release device to hold the block down via the tool post stud instead of the usual nut. An interrupted thread, breech block style would work. I proposed a rotating stud cross drilled for a tommy bar arranged to engage in a castellated hollow nut screwed into the tool block in my design. Never built due to changing lathes and obtaining an adequate number of Dickson posts and holders along the way.


Edited By Clive Foster on 11/07/2017 22:31:14

Thread: 19 TPI
10/07/2017 16:48:03

Ooops. Need new glasses. John and Jason are right. Sorry I mis-read my reference.

Clive Brown has the answer. 38 tooth gear on gearbox input with standard 20 tooth stud gear with 28 TPI setting on the gearbox gives 19 tpi.


10/07/2017 15:29:43

If your lathe has 28 tpi on the gearbox easiest way is to double the size of the stud gear or halve the size of the gearbox input gear. For a SouthBend 9" I think doubling the stud (input) gear size is the recommended way as the gear required is one of the standard set for a non-gearbox version. Also part of the conversion set for metric threads.


Thread: inverter stick welder
07/07/2017 13:51:13


Budget seems reasonable to get a perfectly satisfactory welder. Hard to know what the right price is with so many known brands coming up £50 or more off.

Main things I'd look for are a decent warranty from a known brand or supplier who is going to be around should you have problems. Inverters are intrinsically less rugged than buzz boxes. Overheating is the usual reason for early demise so look carefully at duty cycle. Specifications are very misleading. 50% duty cycle doesn't mean half the time welding and half the time waiting for it to cool down. Actual welding time will be rather less.

For example my Fronius specification is for the given duty cycle over 10 minutes. Then you have to wait until it cools down. Flat out at 140 amps 10 minute specification is 35 % duty cycle. Cooldown time if you do red line it is maybe 20 minutes before the watchdog circuit lets you start again. But the fan will still be going so its not cooled right down. Now thats a full on industrial welder rated 100% duty cycle at up to 100 amps and even at 120 amps you don't really have to think about duty cycle. Fan may be running permanently tho'. Had to think when I used it at 140. That sort of toughness and safety margin comes expensive.

Gotta accept that with a lower end, i.e. affordable, machine you need to be a bit more careful in use. Doesn't help that many suppliers inflate the actual capability of the machines with an unrealistically low duty cycle at maximum output. Doesn't help that there is no time specification either. Sheer mass of a hefty heat sink may give a minute or two flat out before it warms up enough to endanger the electronics but then it takes a long time to cool back down. Noodling around E-bay I see a few likely suspects for this trick. Nobody specifies cool down times.

20% duty cycle over 10 minutes is only a couple of minutes welding. Realistically anything under 40% duty cycle in the range of currents you tend to use means a serious risk of trespassing outside the safe operating area and into the will it / won't it go pop region. No one puts a stop watch on their welding time and I guarantee that how ever carful you intend to be you will overstep the mark at times.

If you possibly can find one with 100% duty cycle at the maximum current of the rods you will normally use. 80% will do for the upper end of currents but you will need a bit of care. Helps that inverter welders work fine with lower currents than buzzboxes on a similar rod size. No experience of them but the Draper 64533 INV146 seems to specify well with 40% duty cycle flat out.


06/07/2017 22:15:42

I have one of the older Fronius Transpocket 140 units. Superb welder, gives me the illusion of knowing what I'm doing, lightweight and very compact. Her ladyship has bigger, heavier, handbags. Aimed at the on site market so dual voltage 110-240 input. Eye-wateringly expensive when new, current versions merely painful.

If you are used to a normal buzz-box stick welder the two big differences are reliable welding at low currents the ability of a DC inverter welder to hold a very long arc if you aren't decisive about breaking contact. Mine will easily pull an 8" long arc. Ignore the bottom end imports, get a good brand. Very bottom of the range ones have a well deserved reputation for unreliability.

The R-Tech combined MIG / MMA looks excellent value for money but, inevitably, is a bigger box than mine.

Mine lives happily in a customised mount replacing the top box of a plastic tool cart thingy from Lidl. Three drawers for rods and bits. Large flap opening space below with room for the my welding helmet, welders apron and gloves. Maybe takes up 10" by 15" floor space. Wheels and handle make it easy to move up to the job. So nice to be able to keep everything together.


Thread: Norton Quick Change Gearbox Removal
04/07/2017 12:51:40

Great that you have it out.

Rather than simply replacing the pin with another 2 mm once consider re-emgineering things to avoid similar sheared pin jamming issues in future. I'd drill the holes out to 3.5 or 4 mm and make a new pin with shallow, taper sided grooves turned into it at the joint line. Threading tool with the tip rounded off worked for me. Half mm, maybe a touch more, deep should do fine. That way if it shears the break line and distortion will be contained in the holes so no smear to jam up. May still have a bit of twisting'n shifting to do to line things up before it can be punched out but that should be relatively easy. Light knurl on the collar ends held fine for me. Brass or aluminium alloy is often recommended for such pins. Never tried myself.

If confident that there is no shear risk I use rollers out of scrap needle roller bearings as pins. Keep a stash primarily to use as dowel and alignment pins.


04/07/2017 10:25:17

A pin through a cross hole is common practice when joining such shafts to a bored to fit collar. Frequently the pin is made small and / or weak so that it shears when overloaded to protect the drive train. Unfortunately if a plain pin is used it can smear at the shear line between the shaft and the collar instead of breaking cleanly. If this happens things tend to twist a bit and jam up solid with the remains of the pin somewhat out of line with the holes in the collar.

Given the small size of you joining pin I suspect this has indeed sheared at the joint line and smeared across so drilling out won't help. The smeared material takes up clearances and holds in against movement in all directions. Odds are it sheared one side first then the other after bending the pin slightly. The bend makes the smearing effect worse.

Best way to release things is to twist the shaft relative to the collar bringing the pin back into alignment with the hole so it can be driven out. If you can get things moving can be possible to feel the smeared pin moving back into alignment with the holes for driving out. 50/50 chance if the four similar jobs I've done are typical. Anoint with plenty of plus gas or similar releasing stuff, heat, try, re-anoint, wait and have another go type job.

Solid clamp on levers to hold the shaft and collar for mutual twisting will be more secure and less damaging than grips. Ideally alloy on steel as this holds better than steel on steel..


PS:-  KWIL posted whilst I was typing.  Interesting picture and smeared pin theorry may not apply but that one looks to be slightly different design to yours.

Edited By Clive Foster on 04/07/2017 10:28:35

Thread: Feeds/Speeds/Tool Geometry with fast feeds
28/06/2017 11:18:36

Your Keighley is essentially a Mk1 Denham Junior so the threading chart shown on the 7 th picture down here ; - **LINK** probably applies.

Looks like you need to alter the gears on the banjo. If that chart is right you have 30 - 70/80 - 90 as your drop gear train which is also the setting for 21 TPI. Probably 20 - 90/49 - 120 or 30 - 120/80 - 70 would be more appropriate for feeds giving 216, 180, 54 and 224, 112, 56 cuts per inch respectively. Translates to 4.6, 6.5, 18.5 and 4.5, 8.9 and 17.9 thou per rev. No reason to stick with the chart of course as you can manipulate gears at will.


Thread: Milling in lathe
27/06/2017 19:14:57

Chuck gripping geometry. Lathe chuck grips directly sideways with no endwise force derivation so the end mill stays put.

Drill chucks draw the jaws down an inclined plane to derive the gripping force. Endwise push as when drilling increases grip so long as the drill doesn't slip. Endwise pull as can occur in milling reduces grip by moving end mill and gripping jaws out up the inclined plane until clearances are taken up. Grip is reduced which may be enough for the end mill to come loose. When cutting you have a combination of sideways vibrations and endwise push'n pulls repeating two or four times as fast as the spindle is turning. For maybe minutes at a time. No wonder the thing tends to shake loose.

Theoretically collets could have same sort of problem but the gripping area is much larger, clearances in the drawbar thread much smaller and tightening forces from the drawbar higher so generally tends not to happen. Can be a problem with small collets on small machines tho'. Especially with hand tightened drawbars. My old BCA was a forever pulling out cutters if normal cuts were attempted.

If screwed shank cutters in Clarkson et al holders try to slip they screw themselves outwards pushing the collet harder against the end cap increasing grip until slipping stops.


Thread: Combination squares
26/06/2017 19:47:14

A good brand machinists combination square will have been made to better than 10 thou per foot accuracy. Brand new old line Starrett, M&W, B&S square and protractor sets made to maybe ± 2 thou per foot nominal. Time was a good combination square was an expensive piece of kit and had to be made well to persuade a craftsman to buy it. Not so much now.

Big box and import? How wide is your pencil line? If it came from B&Q an ordinary framing square will be better.


Thread: Hacksaw versus Bandsaw
26/06/2017 17:42:46

That Rapidor Major linked to by Martin is pretty much the same as the one I use except for mine having the fixed square cut vice rather than the pivoting version shown in the pictures. Fixed vice is slotted to take the blade so it grabs both sides of the material. Mine just works without fuss or hassle. G&M price seems a bit high. The Major is a lot heavier than the open framed light version. Major needs a trailer or transit type van to shift but I bought my light one home in a SAAB 900 combi. Had to take the legs off. Main saw part is a two man or engine hoist lift. Darned if I can see any significant functional or engineering differences between the major and the light one I also have. Major has a built in coolant system and thats about it.

I'd not be worried about wear on a Rapidor or, for that matter, any of the other typical British industrial types with slide / guide ways symmetrically disposed relative to the blade. I imagine the types with a heavy cast arm above the sliding bow might be subject to derangements of the cut control and blade support hydraulics which might be troublesome to fix. The Rapidor is a very simple device and probably pretty much immune to wear out in any normal lifetime. Simple plain bearings and pivots throughout easily fixed by re-bushing and / or new shafts if it really has been abused. Can't see any issues with changing the square slider bars either. Should be easy enough to find an economically priced example £45 for one sans motor, £100 for one with seem about the going rate.


Thread: Car problems
25/06/2017 21:01:36

Bottom line is that you always pay for servicing. On top of that with a new car you pay depreciation, with an old car you pay maintenance and in between you pay some of both! Now the body killing rust bug is pretty much an endangered species there is no real mechanical reason why almost any half decent car shouldn't do starship miles. Electronic side of things on very modern stuff is another issue entirely.

Best way I know of looking at it is to consider things in 100,000 mile chunks. Once you get up to 100,000 or a similar target mileage how much could it cost you in expected replacements over the next 100,000. Compare that to how much it will cost you to swop your bolide for another that will do another 100,000 with normal servicing and minimal maintenance. Choose the right vehicles and the answers can be very surprising. But you've gotta do your homework. Lots of cars out there with expensive weaknesses that are essentially terminal soon after 100,000. There are good reasons why the miles limit of those lease with a big balloon payment at the end (which we aren't really going to explain properly) plans are so low. If you do go for starship miles option you have to save into a fixit fund. Either to actually fix or to fund replacement.

Being a grumpy low mileage type old fart I run a late Range Rover P38. Comfortable, tough, simple reliable, easy to fix and cheap on parts if need be. At 5,000 miles a year V8 thirst worries me not.



Edited By Clive Foster on 25/06/2017 21:02:18

Email News - Join our newsletter

Love Model Engineering? Sign up to our emails for the latest news and special offers!

Latest Forum Posts
Support Our Partners
Expo Tools July 14
Advertise With Us
TRANSWAVE Converters
Eccentric Engineering
Allendale Electronics
Reeves 2000
Subscription Offer

Latest "For Sale" Ads
Latest "Wanted" Ads
Get In Touch!

Do you want to contact the Model Engineer and Model Engineers' Workshop team?

You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.

Click THIS LINK for full contact details.

For subscription issues please see THIS LINK.

Digital Back Issues

Social Media online

'Like' us on Facebook
Follow us on Facebook

Follow us on Twitter
 Twitter Logo

Pin us on Pinterest

Visit the Model Engineer
Exhibition website

Model Engineer Exhibition