Member postings for Andy Stopford

They used to do this at Tonbridge with the Hastings Line DMUs - when the inevitable letters of complaint were printed in the local rag, a BR representative would explain that the engines were worn out and would be unstartable if left to cool down overnight.*

Of course being left ticking over all night glazed the cylinder bores, making the engines even harder to start, and making them produce steam engine-like clouds of smoke. It wasn't limited to DMUs. BR was so cash-starved in its latter days that apparently they couldn't afford batteries for the big single unit diesels, and wrecked the engines by leaving them idling all night. Of course everyone involved with the motive power side of things knew this was a stupid thing to do, but ... politicians.

*I seem to recall it was said that the Sulzer engines in these had negative clearances when cold, so the donkey engine, which was plumbed into the same cooling system had to be started a couple of hours in advance to warm the engine up so it could be cranked over.

I got one last week - my impressions so far are pretty favourable.

I've tried it with C, MicroPython and CircuitPython (the latter is a version of MicroPython put out by Adafruit, potentially useful if you're planning to use the kind of break-out boards they supply)

At the moment everything is all very new, and as SOD says, things will probably be easier when the Arduino toolchain is available. I managed to get a Neo-6M GPS module to work (fairly easy - I found a MicroPython library for it (https://github.com/inmcm/micropyGPS), and an MCU6050 accelerometer/gyro (much harder figuring out how to use the i2c bus, but I've found these tricky to get working even with a pre-made arduino library).

i used Thonny for the MicroPython stuff - it's OK, but a little idiosyncratic.

If I can get it to work with a 480x320 TFT touchscreen it will be really handy for a project I'm working on, though this could be a real challenge - with a bit of luck someone cleverer than me will come up with the necessary magic.

edit: Looks like someone's on the case with the TFT - https://github.com/Bodmer/TFT_eSPI/discussions/934

Edited By Andy Stopford on 04/02/2021 20:04:00

Nigel -

You're right in that there is some concern that widespread adoption of ground source heat pumps might freeze the ground under very highly concentrated housing and reduce the pumps' performance, but I don't think it is regarded as a problem for individual installations - if you've got a garden to put your heat exchangers under, you'll probably be OK.

David Mackay has information and calculations on the subject in Sustainable Energy Without The Hot Air:

https://withouthotair.com/c21/page_152.shtml

Just to add to SOD's explanation, thermal efficiency for a Carnot cycle engine depends on the difference between the peak temperature (say boiler temperature in a steam engine) and the temperature at the end of expansion (the exhaust temperature in most applications):

efficiency = peakT - endT / peakT

multiplied by 100 if you want it as a percentage.

This is just the theoretical efficiency, in practise there'll be all sorts of other losses, and I've presented it in a form which idiots like me can , hopefully, understand. It does show why diesel engines, say, with a very high combustion temperature are more efficient than steam engines, where the practical limit on maximum steam temperature is quite low.

Oh, and don't think you can get better than 100% efficiency by making it exhaust at below 0, the equation only works if you use Kelvin.

It would seem that Sainsbury's trialled it in 2016:

https://www.coolingpost.com/uk-news/sainburys-to-trial-dearman-engine/

No details as to how the trial went, unfortunately.

Most of the descriptions of how it works are mangled by journalistic Chinese whispers; it appears to be a piston engine which uses a warm water/glycol mixture to flash liquid nitrogen to gas, i.e. the water/glycol provides the 'hot' part of the thermodynamic cycle and I'm guessing that it's in the form of steam, since there are mentions of it being compressed prior to N2 injection, and that's suggested by this diagram;

https://revolution-green.com/piston-engine-driven-expansion-liquid-nitrogen-liquid-air/

There was a more expansive discussion on this on Radio 4 yesterday:

https://www.bbc.co.uk/programmes/m000r368

The salient fact is that this is for a specialised purpose - refrigeration plants on lorries (as some have already noted), so simple Carnot efficiency may not be too relevant. Their is some discussion on the programme about the energy used to produce the nitrogen (disclaimer - I was busy at the time so only half listening).

According to the programme, 'a major supermarket chain' is converting their lorries to use this. If so, I very much doubt if they would be forking out in something which doesn't work.

I bought a CJ18 from Amadeal about a year ago, and I've been pleasantly surprised at how good it is for the price.

It needed a little fettling, as others have remarked - namely the cross slide gib strip was poorly machined and completely the wrong cross-section. I though of trying to get a better one from Amadeal, but decided it would be better to make a new one, which greatly improved the cross slide location. I also scraped the sliding surfaces.

As Andy Gray mentions above, the calibrations on the cross and top slide dials are absolutely maddening with 0.025mm/division, and with major lines every 10 divisions marked 10, 20 etc.. 10 what?

As discussed in this thread:

https://www.model-engineer.co.uk/forums/postings.asp?th=162227

I made new dials with diametral calibrations - a very worthwhile modification, and fitted roller thrust bearings to them which make for a smoother action.

At some stage I'll fit angular contact bearings to the headstock, but it works well the way it is.

Of course second hand industrial machines offer better value, but I don't have the room - I doubt the lady in the flat below me would appreciate having a Holbrook B-type arriving in her bedroom.

I can't imagine what you could do to an L5 to bend the spindle (at least without gross damage being visible on spindle nose, headstock casting, etc.).

Has it been taken apart and the taper roller bearings incorrectly re-installed?

Ha! should have thought of that!

@Nick - I have no excuse, I made it myself...

Talking of which, I think I'll shut the computer down now and see if the beer-tasting situation has improved; if not, I believe I still have a bottle of vintage port (strangely, sweet things seem unaffected)

Cheers, chaps!

I received an Xmas greeting from the NHS this morning in the form of a positive Covid test result.

Having Covid for Christmas is definitely the most Grumpy Old Man thing I've done in ages; best time of the year for it in my opinion, so, Bah!, Humbug! to all (I just wish beer didn't taste peculiar at the moment).

You can't really compare processor speeds for different generations of computer - a modern processor with the same on-paper specs will perform far better than an old one.

None of the programs you list should tax even the humblest of modern computers (having multiple browser tabs excepted - they can really hog memory).

Don't be seduced by the multiple processor cores thing - most programs can only use one thread at a time. From your list, I should think the only one that might use multithreading is the CAD one if it does ray-traced rendering, and LibreOffice Calc apparently does if you enable it.

Also: I recently did a clean install, as opposed to an upgrade, of the operating system on my 9 year old Fujitsu laptop, and it now boots faster and generally works far better than it has for a long time. This is using OpenSUSE 15.2 with KDE desktop which must make comparable demands on the hardware to Mint.

So you might as well give a new install a try - you might be pleasantly surprised, and save yourself forking out for a new machine.

Interesting. I didn't know about Reverse Lend-Lease, I suppose I always had some vague assumption the Americans just paid for the British stuff they used.

re. the glider into IMC video - wow, sobering stuff. For those interested as to how they got themselves into this situation, there's an analysis here:

https://www.youtube.com/watch?v=NpyqiAp2TFY&feature=youtu.be

The Nazis had a curious infatuation with wacky super-advanced gadgets, see all the projects for advanced aircraft, jet aircraft (often multiple designs to do the same job), ultra complicated engines, radio control, even flying saucers for goodness' sake.

Most of these weren't even wanted by the services who were going to use them, and never got beyond the prototype stage, but they represented an ongoing drain on resources.

It's said (haven't got a ref to hand, but its an oft-repeated view) that the expenditure on the V-Weapons was comparable with the Manhattan Project, but it was too late to do anything to significantly affect the outcome, and strangely there was little enthusiasm for building the device to put on the top of the V2 which would turn it into a real game-changer. Dodgy Jewish science and all that.

Although the Allies had their share of wacky ideas, they were kept under control, so that the vast majority of design and production resources were devoted to churning out effective, good-enough designs. As Stalin said, "Quantity has a quality all of its own"

As for the reason for this blindspot, it might be down to romantic delusions of 1000-year Reichs bristling with Wagnerian weapons, but another factor might have been that they didn't have a formal system of reserved occupations, so that if you were a geeky, techy youngster you had a very real prospect of being sent to the Russian Front, no matter how good your abilities at science/engineering, etc, and no matter how lamentably un-military your character.

So the reason for some of this stuff may well have been down to kindly departmental heads putting in a good word to the authorities about young Hans' flying saucer ideas and saving him from a pointless fate at Stalingrad.

For anyone who can't view the older mags in the Linux version of Vivaldi, you need to L-Click the padlock icon in the address bar, choose Allow Flash and reload the page. You may have to do this each time you logout/in.

You may need to have Pepper Flash installed; a script to automate this can be found here:

https://gist.github.com/ruario/215c365facfe8d3c5071

This may work for other Chromium-based browsers.

I agree with the above, always clean brushes thoroughly, and don't expect them to last forever (though you can of course demote them to less exacting duties). For both artistic and other purposes, the synthetic ones are excellent now, and cheaper than sable or whatever.

I'm just posting really, to add that acrylic usually succumbs to meths -you could probably also use IPA (no, not the beer, you drink that whilst the brush is soaking).

What's the problem with speed limiters?

Society has decided, for excellent reasons, to legally restrict the speed at which we are allowed to drive. Many people break these laws, most get away with it, some get caught, some have crashes, some injure or kill others.

Excessive speed also wastes fuel, increases pollution, etc.

Why do otherwise law-abiding people want to have the option to indulge in this particular lawlessness? I note someone made a comment about 'Honest citizens' vs 'Pikeys, new age travellers' . Hmm - not too sure what the difference is (New Age Travellers? This isn't 1985 you know).

'So, Andy' you might say 'you've never broken the speed limit then?'

Of course I have, but I've also driven many thousands of miles in speed-limiter equipped HGVs, and I can't say I ever felt that my God-given Right To Speed was being infringed (even though variations in power output, load and gearing make the speed limiter a lot more tiresome than one on a car would be).

Though I must admit, when the company mechanic in a futile attempt to cure an engine problem, disabled the speed limiter on the 17 tonner, it was great late at night on the way home from a Removal Horrorshow to open it up to a meteoric 75 mph.

I've brazed copper using the standard brass rods (I don't know the precise formulation - it was a long time ago), and borax as flux.

It was easy, no problem getting enough heat.

One job involved fixing a wood-fired water heater that someone had tried to repair with a stick welder and steel filler rods(?!). I had no problem building up a nice thick fillet to bury the resulting lacework.

You'd have to check with those who know about suitability for any sort of pressure vessel, though the flanged joints on the boiler of my espresso machine (a La Pavoni P67, for those who are interested) appear to be brazed. That only runs at about 1 bar though.

The piston/turbine compound was the Nomad, the Dagger and Sabre were the H24s.

Some great info on these and many others here:

https://oldmachinepress.com

For a general introduction to C++, This is pretty good:

https://www.learncpp.com/

Note though that the Arduino uses a special sub-set of C++, and some things are missing, like the ability to find out the type of a variable, which is a pity because this can be a very confusing aspect of microcontroller programming.

You need the 7-segment display with i2c backpack - that way there are only two data lines to connect up. They are readily available from ebay and look really good. Unfortunately I've only used one with a Raspberry Pi zero, so I can't give you the C code to make one work (I used Python for the Pi, and the libraries are different anyway).

Hopefully the following link to Adafruit's information and instructions won't fall foul of the forum rules:

https://learn.adafruit.com/adafruit-led-backpack/

edit: Sorry, the ebay ones are SPI rather than i2c; they should be similar to get working, but for simplicity, it might be best to use the Adafruit ones - if you're in the UK, Pimoroni sell them - they cost a bit more than the ebay ones, unfortunately. 

Edited By Andy Stopford on 20/11/2020 20:27:24