Member postings for SillyOldDuffer

I think the die threading Brass correctly proves the Aluminium rod is the problem.

Most likely it's probably a soft alloy rather than H30. Any tendency of the Aluminium to tear will be worsened by a fine thread like 40tpi. Andrew suggested tearing in the first answer and I think he's right. Cure: buy an Aluminium alloy known to be machinable rather than hoping what's in the junk box will do!

For threading I recommend a designer Tapping Fluid rather than a plain Cutting Fluid or what's to hand. I use CT90, because it's what my local emporium stocks. It works well.

Some sort of fluid is essential when threading Aluminium. Almost anything is better than nothing, but meths doesn't lubricate and it evaporates quickly. WD40, Paraffin and other light oils work well as Cutting Fluids on Aluminium, but neither handles high-pressure well, so less suitable for threading. Never used Trefolex, but should be suitable. Not keen on old-wives methods like Bacon fat - although they work moderately well they're a stinky bio-hazard.

No cutting fluid will fix a fundamental problem like the wrong alloy.

Final thought, failure to cut a decent thread may have prevented a worse problem - the fixing failing under load. If the alloy is torn by the die, it means the thread will be weak and liable to strip. Not good in a braking system or for holding on aircraft wings!

Dave

How about a line-borer?

This UK company sells or hires them, and may do so in Europe. (Their website is multi-lingual.)

I'm sure line-borers are available on the Continent too. No idea how much they cost or how hard they are to set up though.

Dave

I made a stupid mistake, twice!

The item is 400mm length of steel-conduit, turned down to 16.9mm diameter so that collars can be clamped to it. (It's a mandrel, the idea is it will position a pair of wooden discs turned to fit inside a length of 4" diameter PVC drain pipe so the pipe end can be machined smooth and true between centres.)

Having successfully turned the steel pipe to size and made four fitting collars with grub-screws, I officiously decided to mill a shallow flat along the pipe for the grub-screws to grip. My milling vice has a jaw with a small V in it, which I judged adequate to hold the pipe. Wrong! The helical cutter pulled the pipe out of the V and gouged deep into the pipe before I noticed. Thinking I'd forgotten to tighten the vice, I reset the job and tried again. Far worse, because this time I ploughed on regardless, fully confident all must be well. It wasn't.

Lesson learned, practice what you preach. I know full well that solid work-holding and paying attention are important. And over-confidence is akin to foolishness.

What really hurt is it took less than a minute to file a perfectly good flat on the pipe with it held in a bench vice.

Another painful lesson, don't use a powerful milling machine for silly little jobs just because you happen to have one!

Having made a mess of a simple job, I'm thinking of starting a YouTube channel called BaldiHacks specialising in videos full of unconscious workshop goofs and sins. Certain to gather millions of loyal followers all innocently imitating my every blunder! Bound to become a highly paid influencer!

Dave

You're not trying hard enough John!

Dave (Bodger First Class, mentioned in Despatches, not in a good way...)

More likely there is nothing for small-birds to eat. Intensive agriculture generally removes habitat and bumps off most of the insects by spraying. Most UK land is managed, which is good for us but not the wild-life. Global warming is altering patterns too.

Dave

Does the die still cut Brass? Phosphor Bronze is tough stuff, so maybe the die had a hard time and is now blunt.

Otherwise, I'm with Andrew in suspecting the Aluminium. Pure Aluminium is very soft and doesn't machine well. Many Aluminium alloys cut poorly because they're race-tuned to be extruded, bent or stamped, not cut. The difference is marked. Whilst being cut, especially threading, the softer alloys tear, deform, smear and stick to the cutter rather than cut. Horrible.

If a cutter is choked with Aluminium that won't come off when poked, it dissolves in Caustic Soda solution. (H&S warning: read the label!)

Dave

Can you give examples Vic?

Pure Tungsten Carbide is about 30 times harder than HSS, so in theory no contest! Compared with Carbide, HSS is Cheesium. A few possibilities:

• carbide tooling is often a percentage of Tungsten Carbide powder inside a softer matrix. This reduces hardness in favour of some other desirable property such as strength, elasticity, toughness, sharpness or cheapness etc. (Only a BF buys £££ solid carbide when a cheaper cutter does exactly the same job on soft materials.) Possibly the carbide tooling looked at isn't the hardest possible.
• Maybe not comparing like with like? There are several different scales and sub-scales: Vickers, Rockwell, Brinell, MOHS etc.

In my experience HSS is no match for carbide, though it tries to put up a fight. Still good stuff: HSS can be ground into different shapes, including very sharp, and I find it less fussy than carbide for getting a good finish on difficult materials.

Dave

Can you let us know quotes?

It's cost versus value question, where the answer depends on circumstances.

My workshop has a basic vertical hobby mill. It's mostly used for light detail work, not hogging metal. Not having a horizontal mill means paying to sharpen horizontal cutters has no value to me. Might be worth having a big helical cutter sharpened for a special job, but I rarely use them. Those I bought with the mill 'justin case', are still sharp, and a few have never been used. This is why experienced machinists don't recommend buying sets - there's a risk of wasting money on tools that never get used.

Same problem with buying second-hand and finding it comes with a box full of random cutters, many blunt. Financially, there is no point in having them sharpened unless they're going to be used.

There's an argument in favour of owning tools 'just in case'. It's time saved when a new job suddenly requires a particular tool and there's a painful delay whilst one is ordered and delivered. Sod's Law guarantees that urgently needed tools are always out-of-stock.

Trouble with buying 'just in case' is it costs a fortune, takes up space, unusual tools are hard to find in deep storage 10 years later, and grieving relatives will probably dump the whole lot into a skip when the time comes.

I guess most Model Engineer's target our tool-spending on things that are definitely needed rather than stocking up on the off chance. Bargains are always welcome, but some thought goes into 'is this worth it?'

The value of sharpening cutters is strongly influenced by what I do with them - light detail work at gentlemanly hobby rates! Cutters last a long time in my workshop because I don't cut a lot of metal quickly. Others work their machines much harder, and value of sharpening becomes more evident. Professional workshops have to work quickly and efficiently, time is money. Amateurs are between two extremes. Some collect tools and never use them! Others are as busy as a busy commercial operation. Most are in the middle, and I suspect tool sharpening is a low priority.

Not resharpening is something of an industry trend. In the good old days most machine shops resharpened in-house. These days that's unlikely because third-party specialists do the same job cheaper and better. In the smaller sizes CNC cutters are often single-use, replaced when worn. They're recycled rather than resharpened. I think this is partly economics and partly because resharpening alters cutter size, bad news when a machine auto-changer expects cutters to be standard sizes.

I'm not sure resharpening is high value to most Model Engineers. I've been a cynic since reading the accusation in an old ME magazine that almost every workshop in the land had an unfinished Stent and Quorn under the bench! In my case, most of the time, it's easier to buy new. I'd think differently if my workshop took a week to blunt a big cutter costing £100 plus that could be resharpened in a week for £20. Then I buy 4 of them, so one was always cutting, with one out being resharpened, and two in reserve. In this scenario the initial investment is £400, plus £1040 per year on sharpening. Assuming each cutter can be sharpened 5 times, I also buy one new cutter per month (£1200). But I'm not in that game - my needs are met by occasional orders for a few new cutters.

Dave

Below about a thou, various tricky factors start causing trouble, and they get progressively worse:

• Temperature - a rod warmed to 40°C is measurably bigger than the same rod at 20°C. If temperature isn't accounted for, you ain't measuring tenths!
• Calibration becomes vital. Second-hand gauge blocks and ancient micrometers aren't good enough.
• Technique is difficult and tiny flaws destroy accuracy.
• Although a large diameter dial helps, don't expect much from dials, and especially not small ones. Dial marks are usually indicative rather than accurate: the operator is expected to use a micrometer, calipers, or DTI
• The brand-new lead-screw of an ordinary lathe is unlikely to be tenths accurate and older lead-screws will be more-or-less worn too. The accuracy of fine-cuts is fundamentally limited.
• Tiny saddle, headstock and other alignment errors show up
• Work and tools both bend under cutting pressure, making it likely that cuts are not all of equal depth end-to-end. (Ever noticed that tools often remove more metal when reversed out of a cut?)
• Scale causes the relative surface finish to go to pot! As the size of cut is small compared with the lathe's automatic feed-rate, fine cuts leave spiral grooves rather than a smooth surface. Looked at with a microscope after a few passes, the surface resembles knurling. Is a tenths micrometer measuring between peaks, or has the anvil crushed them towards the valley bottoms? Dunno: unless the surface is good, the tenths are uncertain

What this boils down to is that lathes (and milling machines) struggle to take cuts that are both fine and accurate. Tool-room lathes and jig-borers do better, but they're elaborately engineered and maintained to hold accuracy. They have obvious limits too.

The answer to high accuracy isn't a lathe. Industry are fond of grinding machines but these are expensive. Fortunately small workshops can get good accuracy and surface finish with manual abrasives, notably emery papers,and lapping. Making pistons, the lathe is used to get close to size, but turning stops a little above target. After that a succession of ever finer abrasives are used remove unwanted metal, not cutters. Unlike cutters, correctly used abrasives improve the surface as well as removing metal. The main problem with abrasives are removing metal evenly, and the amount of time it takes.

Larger pistons don't go for a very tight fit: instead the seal is provided by piston rings which can expand slightly to fit. Although cylinders are bored very accurately in big engines, I've got the impression surface finish is even more important. There is no 'Running In, Please Pass'. Au contraire, I've built 5 model engines and they all needed a lot of running in! I spoiled several pistons by rushing the finishing stage. Patience in the workshop is not one of my virtues.

I rarely work to better than ±0.05mm and avoid high-accuracy if I can. I stay with innocent measuring because high-accuracy is dangerously additive. 'Chasing Tenths' could easily take over my life. As Chasing Microns is even more difficult, only real men dare go metric!

Dave

Swings and roundabouts! My experience of POTS wasn't good. Crossed-lines, snap-crackle-and-pop, wrong-numbers, long wait to have phones fitted, party lines, no choice, having to pre-book long-distance and international calls. In an emergency, no call-box within miles, and if one was found it had been used as a toilet and/or vandalised. Restrictive practices that had somehow confused national interest with jobs for the boys, old-fashioned, reluctant to change, dreadful support for computer communications. and all of it highly expensive compared with the rest of the world.

All systems have shortcomings. Modern telecomms have fixed most of the above, but new technology isn't always used appropriately. To reduce taxes, emergency services have been extensively centralised, allowing many local police, ambulance, and fire stations to be closed in favour of large modern facilities. Results are mixed, for example don't expect 999 operators to have any local knowledge. Last time I dialled 999 the police didn't recognise any of the placenames: I had to provide a Grid Reference. I hope they have Google now!

I'm afraid seeing the past through rose-tinted specs is an age thing. I often have stop myself and ask "come on Dave, is that really true?" The answer is usually 'no', but I was young, optimistic and in good health, and most of the changes were inflicted by me on the old-folk. Anyone else remember thinking how silly their elders were, wanting needles for their HiFi, and no idea who the Rolling Stones were? Now I'm old-folk myself, and right about everything, I don't want anything to change. Pressing Button B worked perfectly well in the past, I think, so no-one needs a Smart Phone...

Dave

Dave

I often think Model Engineers should be sent on a Management Accountancy course to learn it's about value rather than price. Value is is some combination of cost, benefit, and time. Price is obviously important, but it's far from being the only factor.

Whilst price dominates decision making In ordinary life, and many small businesses are successfully run the same way, there's a point at which other considerations come into play. A relative encountered this painfully: he ran a small-business, which he enjoyed thoroughly leading from the front, him and a couple of mates. After the business grew to employ about 15 people, he hated it! Rather than doing the practical work he enjoyed, he found himself managing people, money and assets, all of which were much harder to do than he'd imagined, plus lots of responsibility, and all the aggro. Tragic, but in a business beyond a certain point, the need to manage people, money and assets mean the financial game is often played to different rules. What's worth the money, has to take more than price into account.

In that context, I spent £427 last week buying not very much metal. (Anyone who hasn't bought Brass recently is in for a shock!) Not a sound business decision, but I enjoy Model Engineering, and the metal will keep me amused for at least a year, I hope! Not everyone would have made the same purchase.

Back to mending earth movers, part of the cost of a repair is off-set by the cost of downtime. Machines that work of a living out of action for a few months whilst parts are shipped from the US or Japan cost a fortune, making local repair an option even if repair costs more than a new part. Time is money. Judging by the videos, it appears repairs are about half replacement price, and the work is done quickly compared with importing from abroad. $20000 dollars Australian is a mere £10000.

I've learned a good deal about Hydraulic rams from the videos. They're a mix of high and low-tech. Outside the cylinder, simple and chunky parts are highly stressed with cheap bits designed to break before anything more expensive. Bearings, pins, eyes and end parts all take a hammering, either broken off, bent or badly worn. There's much scope for repairing the chunky parts with a big lathe, big milling machine, big hydraulic press, and lots of skilled welding. Inside the cylinder, the piston and seals aren't too difficult, but the aforementioned damage can make it very difficult to get inside. Changing worn out piston seals - easy on a bench - might start by chopping the ends off to gain access and then welding the whole lot back together again.

Hydraulic cylinders and rods aren't made of ordinary mild-steel. Cylinder steel arrives pre-bored, skived and honed to size. Piston rod is delivered accurately ground to outside diameter. Both are induction hardened up to a depth of about 10mm. Not cheap.

If they have to be replaced, they're machined as necessary to fit the end-parts. Getting through the induction hardening is fun, and after machining the end-parts are then welded on. Not seen a video where either a critically damaged rod or internal cylinder surface is repaired. I guess that's too difficult.

The most accurate work seems to be making pistons. They're turned to a diameter, I guess about 0.05mm, but then grooved to take the O-rings and other seals. The grooves are cut micrometer accurately. Some of the milling is accurate, but much isn't, 0.05mm rather than 0.01mm.

I'm not sure what the 'prescribed standard' is. The parts being mended are treated very roughly and aren't safety critical. I think the only requirement is that repairs last long enough to cover the cost of doing it, plus a margin.

I recommend watching the videos.

Dave

I'd bet the farm on it being a BA thread - extremely common for small fasteners, especially telecomms, electromechanical, instruments, electrical etc. Highly likely in a British made mechanism of that era. Although supplanted by metric in new products, BA is still widely available; the internet is your friend!

Dave

I'm intrigued! I've come across at least 3 variants of the 13A plug and socket, but never one where the cable came out the top!

Is it a plug or a wall-wart power-supply? The latter often have the output cable on the top.

Sometimes it's a good thing - ordinary plugs don't go into this socket because lower cable fouls the base. No problem for the wall-wart.

Or, the top exit blocks another socket:

Clear proof of the inherent cussedness of things!

Dave

Edited By SillyOldDuffer on 25/06/2023 09:34:33

Above a certain organisational size, 3 Grand is tiny money. My boss once remarked he wouldn't get out of bed for less than £100k. He was being flippant of course, but even small organisations work in millions.

On the shop-floor, individuals might think they only cost the firm what's in their pay-packet. Not so! The firm also pays National Insurance, Rates, Tax, Heating & Lighting, Sewage and Water, for the Building, tools, materials, training, paid-holidays, sick-leave, injuries, and much else.

How much above pay-packet people cost the firm depends on the job, but the average is about double salary. So having a busy digger out of action for any length of time costs a fortune, so well worth having someone like Curtis on hand, even though I sure he charges big money for his services. John Stevenson did this kind of work too - not easy, high-skill, working fast, having the right tools, understanding materials, working around problems and having enormous cojones!

I would have given up after trying WD40!

Dave

PS I hope everyone watched the out-takes at the end! Much 'Industrial Language' when the job and/or making the video went wrong! I suspect getting that nut off was much was much harder to do than expected.

My advice is not to fret over Brushed vs Brushless. It's true Brushless is 'better', but the extra performance isn't a game changer. Brushed motors may be a notch less desirable than brushless, but they're still a good choice for a machine tool requiring good torque, low vibration and speed control. Main disadvantage is the brushes wear out, which will happen quickly if the machine is pushed hard, but in my experience they last well provided machines aren't pushed too hard. (Hint: hobby machines should never be hammered! )

The main thing I regret about getting into the hobby is the amount of time I wasted dithering about which mini-lathe was 'best' before buying one. With hindsight, I should have just got on with it. I learned more about good and bad by using one than by all my theorising.

True you might do better by shopping around, but the smaller hobby mills are much the same design, with the price reflecting moderate improvements or omissions rather than tool-room quality versus  heaps of junk. Avoid buying cheap direct from abroad though: if by mischance a lemon turns up, UK consumer protection may not apply to a direct purchase or be impossible to enforce.

Though it does what I need, my WM18 has several limitations. It's worst sin is being too small, which has to be forgiven because I don't have space for bigger. I can live with that!

Dave

Edited By SillyOldDuffer on 23/06/2023 18:58:57

I haven't yet.

I bought MT because at the time MT tooling was more common than R8, and because it allows some tools to be shared between lathe and mill.

In practice most of my milling is done with ER collets so I rarely change between MT tools.

Main problems with MT, I think, are:

1. The need to undo a drawbar means it's not as quick change as later systems (I don't care)
2. Failing to tighten the drawbar means the tapers might spin and be damaged (not a problem for me yet)
3. Over-tightening the drawbar, especially if a cold tool is inserted into a hot spindle, can jamb it in place. The tapers can also stick if left in place for a long time, perhaps due to very slight corrosion or oil gumming up. Removing a badly stuck taper is difficult!
4. Tapping the drawbar sharply with a dead hammer to unstick the taper causes many to assume bearings must be damaged. Actually, although MT has been in wide use for about 150 years, there's very little evidence of this. Nonetheless people believe it. My mill is 7 or 8 years old, bearings are OK.
5. MT relies on a friction grip, which is very strong, but other systems provide positive locking. These allow more power to be put into the cutter, probably more than a hobby lathe can cope with, but accidental breakages of tools and holders become more likely.

Of these only the possibility a seriously stuck taper worries me. So far I've avoided it by not leaving MT tapers screwed tight for long periods - I undo them every six weeks or so and give them a wipe.

I think the main advantage of R8 in amateur hands they don't jamb. Plenty of tooling available for R8 now, so one of my original reasons for choosing MT has gone. Nor is it common for me to swap MT tools between mill and lathe.

For what I do MT is fine, R8 avoids a potential problem but in practice MT vs R8 makes no difference in my workshop. Be aware that not everyone thinks R8 is the best of all possible systems! I'm happy when tools do the job, and don't upgrade them unless there's a specific benefit.

Dave

I hope enough of the wreck is recovered to find out what went wrong, otherwise we'll soon be into conspiracy theories and red-herrings!

As to the technical cause, my guess is failure at a joint. The submersible seems to have consisted of a carbon-fibre cylinder sealed at both ends by Titanium domes. One the domes was fitted with an Acrylic Window, which maybe doubled as a hatch, possibly the other end was fitted with a hatch, or maybe an entire dome had to be taken off to get inside. However it was assembled, there are a number of risky joints. My chief suspect are the joints between the carbon-fibre body and the Titanium ends, closely followed by the Acrylic Window. Under pressure, Acryllc, Titanium and Carbon-fibre flex differently, and I expect the design included O-rings or similar to take up the slack. O-rings are pretty good, but...

If one trusts news reports, ahem, the owners appear to have deliberately exploited operating in international waters as a way of avoiding regulations and keeping costs down. (I've no problem with that sort of experiment unless they take paying passengers, or involve bystanders.)

We may never find out exactly what went wrong. As the firm were operating outside any national jurisdiction, I don't know who will organise and pay for the enquiry. which - done properly - will be expensive. There will be something, but may be skimped.

If there is an enquiry, I'd expect it to identify an accident chain, that is a series of decisions, possibly none critical in themselves, that combine to cause a disaster. It is ensuring accident chains cannot form that make H&S so tedious - common sense is hopeless at detail.

I'm guessing but, the Titan chain might go like this:

• An individual sees an opportunity to make money by offering affordable trips to see the Titanic. Big personal commitment to an interesting and exciting project, so maybe 'my baby' resulted in a 'can-do' mindset that ignored awkward concerns. Such chaps often surround themselves with 'yes-men'.
• Costs have to be kept low, forcing the need for:
  • a simplified vessel, forgetting that value engineering is the most difficult type of engineering in the world.
  • an avant-garde design using novel materials operating near the edge, that are difficult to inspect, with limited experience of their behaviour in an unusual environment .
  • minimised bureaucracy, maybe by evading the normal checks and rules.
  • maximum use of off-the-shelf components, which function with ordinary reliability. (Not daft, but their suitability in a special case needs careful evaluation)
  • relying on a small number of talented optimists to do the design with limited facilities rather than the large well-equipped and informed teams who normally develop high-tech vessels, whose work is Quality Assured.
  • deploying and recovering the vessel with simple methods that risk damaging it in small ways. Easy to forget that a trivial bump on the surface might become a horrendous stress raiser at 5500psi.
  • Fail to employ competent operators and maintenance staff.
  • Minimise maintenance and delay repairs
• Historically, there have been many examples of engineers under-estimating fatigue. Doesn't seem likely that parts made with a x10 safety factor will fail in normal operation due to mild flexing, yet it happens. Pessimists worry about fatigue, optimists assume hefty lumps will be fine!
• Even with costs trimmed to bone, building and operating a submersible is expensive, so it's probably been financed by loans requiring regular interest payments. Horrible things happen to the company and owner if the money isn't sorted, and the urgent need for assets to make a profit often encourages businesses to take unwise risks. Such as taking passengers before the vessel has been adequately tested.

Add all that together, and the risk of an adventurous enterprise going wrong is high, unless serious effort is put into de-risking the project throughout. I wonder if Oceangate did a risk assessment, and if so, what it says. I've worked on projects were management gambled high-risk/high-impact events wouldn't happen, and we often got away with it. In the face of instant pain it's very tempting to take risks for emotional reasons: people gamble all the time. In my world the risks were financial, but I've often wondered how a safety risk would have been managed - I'm certain some of my colleagues would have tolerated fatalities.

Bottom line is, if you take a risk, there's always a chance it will go wrong, and it pays to manage the possibility. Very little in life is completely safe, and we have to put up with it.

Dave

I don't know what to do with the 3 internal earths either, but it wouldn't hurt to earth them all at one end.

1.5mm2 should be fine for 2.2kW - it can take about 14A buried in a wall without overheating.

Simple answer, see Robert, Mike and others. Unfortunately, doing everything properly guarantees results, but is expensive.

There's no sure fire alternative to doing it properly. The technique is to unscientifically add layers of protection until the problem is controlled. Results are uncertain because exactly what's causing the trouble varies case-by-case, and difficult examples often end up by forcing the victim to do the job properly in several painful steps.

On the other hand with a bit of luck simple change might fix it, so worth trying a few. Things to try:

• The source is the VFD:
  • Change the cable, because it's cheap and easy, and the type already used isn't particularly good at shielding interference. (It's better than nothing, but not good enough)
  • The VFD is a component, with a only thin plastic case providing minimal physical protection, and no shielding. They're meant to be enclosed in an outer box, ideally made of earthed metal. The box protects the VFD from dirt and physical accidents, it protects the operator from shocks, and it helps contains the radiation that's zapping the DRO. It also allows the screened cable above be terminated correctly, important because simply dangling wires through a hole will leak. A box is relatively cheap, and can be home-made, but it means a pedant will have to be wired in. The pendant has a forward/off/reverse switch, an emergency stop, and a speed control, also in a box. Bit of a faff.
  • The VFD probably isn't fitted with a filter, which prevents EMC escaping into the mains, and from there getting into other equipment. Good filters are expensive, so I wouldn't rush to fit one unless putting the VFD in a shielded box doesn't fix the problem. I've thought of fitting a washing machine filter to the mains input. Far from ideal, but it might reduce the problem to 'good enough' level. If a filter is fitted, it works best inside a shielded box.
• The DRO is guilty too - it shouldn't respond to EMC.
  • Its leads should all be shielded. (I think they are)
  • It helps to keep the DRO and VFD cabling and the electronics well away from each other, certainly not running wires parallel in the same conduit.
  • All cables should be as short as possible
  • The DRO power supply should be filtered (but this is cheaper than a VFD filter).
  • Ideally the DRO should be in a shielded metal box
• The next bit should be only be followed after consulting an electrician. Mains safety earths are often complete poo when dealing with EMC. Various shortcomings, for example a safety earth wire might be routed all round the house before it connects to ground, turning the earth into an antenna blasting VFD muck with gay abandon rather than shorting it to ground. To remove EMC, the earth lead should be as short as possible, A separate earth might help, but don't install one unless certain the extra earth compatible with the electrical safety earth. Be embarrassing to get the DRO working and then burn the house down. I'm not a qualified electrician.

Might help to listen to the VFD with a battery powered portable medium wave radio. Close to, VFDs emit a harsh roar covering most of the band in a series of broad peaks and troughs. The radio can reveal if interference is on the mains cabling, or on the motor cable, or the VFD, which helps target the problem. Sorry to be a Jeremiah, but it will probably be all of them.

Good luck,

Dave

I confess to procrastination. Fearful of what looking at my iffy clock is doing, I've left it alone for 8 days with compensation switched on. Reason for my lack of moral fibre is I've run out of ideas. I don't mind problems I can solve, but not having a clue about what to try next is demoralising! The forum ought to have a Topic where members suffering the trauma of yet another workshop disaster can go for tea and sympathy.

Anyway, a short power-cut this morning stopped the clock, so I girded my loins, gritted my teeth, stiffened my lip, thought of England, and downloaded the log.

Results not too bad. After 8 days my clock was 0.51 seconds fast when the power-cut stopped it. Not truly wonderful because as this graph shows it wandered on the way. The blue line should follow the green line:

Looks as if the compensation isn't quite right, because the wander shows temperature bumps. I'm not too upset because this new graph shows a reason. It takes the clock longer that I expected to gather enough pressure and temperature data.

The graph plots the three factors used to compensate the clock. The red line is the intercept, and the green and red lines are the temperature and pressure coefficients. The three lines should become parallel, and it can be seen that they don't approach that until day 5. And the regression results are wildly wrong during the first 3 days.

Explains at least part of the wander because the clock can't keep good time until the compensation coefficients are correct.

Pity the power failed. After 8 days the lines still aren't completely parallel, which means more temperature and pressure data is needed. 746758 samples weren't enough!

I'm going to restart the clock with the latest coefficients and let it run for as long as possible, thunderstorms permitting, to see if the wandering is reduced. Meanwhile, with morale restored somewhat, I'll make a start on building the vacuum containment.

Dave

My mum has an Android Galaxy 8, so I'm mildly familiar. It doesn't have what I would call a desktop. All you can do is run an App, by default on mum's machine that's the browser, which is sawn-off!. Works well for most purposes, but in comparison with Windows, Mac, and Linux, Android 8 is very constrained. Partly I think because its implemented for simplicity and security, and partly because early tablets and smart phones didn't have the horse power needed run greedy applications. Android isn't a general purpose computer yet and it looks like you've hit one of the things it doesn't do.

Android and the hardware have advanced considerably since 8, so a new tablet might work. An expensive experiment if it doesn't though!

My sympathies - why is nothing ever easy?

Dave