Member postings for Roger Hart

I agree with Michael. I think you should have a very good chance of rescuing the optics. Old lenses were glued with Canada balsam or similar, this discolours over time and does not like damp. First get the lens out of its brass cell making careful notes of distance and orientation. Special spanners may need to be made, look out for little locking screws etc. Once out mark lens pair with a V in pencil so you are certain which way it goes back and which surface matches with which.

If lens does not fall apart soak in methylated spirit for 24 hours (assumes lens is glass!) when it should slide apart. Clean up with meths. With luck lens components will not be too etched by mould etc.

I have never re-glued lens components - too nervous. I have put back together using 3 very small slips of cooking foil to stop the lens components touching. The cell hides them from view. Other people may know how to re glue and what with. The books Amateur Telescope Making do contain the detail using Canada balsam - looks pretty fiddly and sticky. Probably a modern UV setting clear glue would work but the permanence of that stuff makes me nervous.

Promised myself no more timewasting projects or buying junk.... But for £2 got this Parrot Bebop drone. No battery no propellers. The GPS module works OK (didn't take long). but this thing has 2 cameras and after time wasted on web I reckon they are both some sort of CSI interface a la raspberry pi.

In contravention of my injunction not to get diverted down dark alleys - has anyone any idea of what connector is shown. 7mm wide by 4mm and 24 pins. I feel a breakout board coming on. Slapped wrist time again.

Trying to make a black terminal for an old AVO without spending any money.

So far I have a bit of brass with a 2ba thread, a collar and a head 12.25mm dia and 8mm deep. I also have a bit of acetyl with a suitable cavity bored in it.

The first attempt at sticking brass into acetyl with Gorilla Glue was a failure, it practically fell out.

Once reasonably secure I plan to turn down the outside, polish up a bit and part off and put AVO back on shelf.

Any ideas how to glue - or other methods

Hi again. No I am not sure about an upper eccentric adjustment. But I think it is worth looking very closely in that area. There is a hole directly above the carriage handle - is this an oil hole or for a grub screw for an eccentric sleeve adjustment?

I no longer have my ML10 and cannot find a detailed enough drawing. Not much to lose by removing the carriage handle and disk and taking a look. The lathes.uk ML10 site and pictures give a hint.

If I were designing that thing I would have put in some kind of adjustment - just 'feels' as if it ought to be there - so just a hunch - could be wrong, I often am....

I had an ML10 and the mesh gears looked a bit worse than yours but they did not 'skate'.

Before spending money on new gears I would have a very good look at the eccentric adjustment top and bottom. Are you withdrawing the locking grub screws far enough to get a good enough rotation?

You can get the whole apron assembly off fairly easily. Undo the righthand lead screw handle and nut and washers, slacken collar at left hand inboard end and pull the entire lead screw out from the left hand side. Then undo a couple of allen cap screws holding the apron and voila.

The imperial leadscrew is 8tpi, easily measured. Take a look inside the change gear housing to see the thread/wheel chart. That will tell you if you have a metric or imperial machine.

bon chance

I too worry about the intrinsic quality. Is this one of those 'Napoleon's Hat' clocks? If it is the wheels and plates were never polished, they were as stamped out and never meant to be looked at.

Nearly 60 years ago I had a Saturday job in a clock shop and got the mucky jobs - cleaning clocks. For a Napoleon's Hat it was strip down, mark the going, striking and chiming wheels for identification and relative position. Then clean in 'benzine', dry off and brush with french chalk on a clock brush then peg out the holes. Rebush or hammer the holes as required. Clean out the spokes with a 3 corner graver, brush the wheels and reassemble.

For a nice Grandfather or carriage clock it was strip down, mark the going etc then polish all the bits with Brasso with a selection of really mucky crusty clock brushes and an electric polisher with a bristle wheel. Used to get covered in Brasso and black fingers. A cloth wheel and rouge for the plates and barrels except the front plate. Once all polished it was wash in benzine, re-do the grimy bits and rewash then dry off, french chalk etc.

Springs were anchored on a nail and cleaned with brasso while stretched out then benzined etc.

I was allowed to reassemble grandfather clocks but carriage clock bits were made perfect then put in tissue in a box and handed to No 2 Clockmaker who would moan like hell if any muck was found. All this for 10 bob and I hate the smell of Brasso to this day.

I have used the ammonia stuff and it is OK but does not seem to make the bits as bright. I hear some professionals make noises about 'preservation' as a cover for not doing too much. As it is quite easy to break a pivot etc I can understand. Also apprentices cost a bit more than 10 bob these days.

So have a good look, was it ever properly polished. That makes a big difference. BTW the 'benzine' was probably not really benzine but common or garden petrol, loads of it under the bench. Elf n Safety was not a big thing back then. TBH I think Fairy liquid and hot water would be OK and SWMBO's toothbrush.

One of those grandfathers was a big heavy job with latches holding the plates together. No 1 Clockmaker said - this was made by Tompion lad, be careful.

I agree with the DC motor option, 4 AA cells will easily drive that.

I have made a few gyros and can say what does not work. One I made from an old disk drive motor, used a turnigy model plane motor driver + a 555 pulse generator and camera batteries. All in a double gimble. The idea being to show the slow rotation of the earth. Big snag was getting all that clutter on board the gyro to balance dynamically. It would always drift too much. Also machining the 3 inch wheel was not so easy for vibration-proof running. Also such a gyro would need a much larger diameter wheel in brass, the clutter away from the gyro (slip rings etc) and better symmetry.

Next try used the same wheel with ex disk drive ball bearings and a string pull. The string pull was no problem provided good string was used. You can easily pull 18" of wrapped string in 0.1 seconds. But the ball bearings I used did not seem any better than traditional point-and-cup bearings.

So far as I can see the gyro shown does not have a large wheel, the frame looks heavy in relation to the wheel size. It may run for 20 mins but I doubt it stands up all that long. By comparison an old style child's toy gyro had a 3" wheel made of lead (brass painted) with a fairly light steel frame. The key issue is the moment of inertia of the rotating wheel - large == better and large == expensive especially in brass and a lightish frame if the thing is to stand up. The question is what is the gyro for. Navigation gyros tend to be cylindrical lumps whilst demonstration gyros need most of the mass in the rim if they are to stand up. The two requirements are different.

Is there anything different about the gyro you intend to make? Good luck and enjoy.

@SOD, liked the palindrome, put it into my list of pythons. Reminded me of back pages of Sci Am way back.

Anyone know how the CMM2 produces VGA. Is it some kind of fast loop does it use dma? Just curious, not really sure if I want to spend too much time on a VGA gadget, but I have a vga screen that has been kicking round the shop for too long.

I am a pretty dim programmer and I prefer simple and obvious with just a dash of elegance. Back in the day 'an instruction saved was an instruction earned' when you mapped your program round a disk to keep the speed up.

Those days are gone thankfully, the SAMD21 I usually use has around 32K of data space and some 256K of code space all for a few quid. You can afford to be a bit profligate unless you are trying to squeeze a quart into a pint pot. You can afford obvious, let the machine take the strain. Mercifully the Arduino platform hides the nasty business of handling the ARM peripheral port multiplexer. Other platforms make you see the gory details and strong men/women are known to take fright at the sight.

I like the code improvements from SoD, he points out the tradeoff between obvious/fairly obvious/terse code. Except that I would be tempted to (kindly and respectfully) strangle terse programmers (not really).

Right now I am wrestling with a Python written by some academics who seem to believe writing comments and clear indenting is a mortal sin. Perhaps in academe tab characters and comments cost money - who knows.... Its so bad I even drew a flowchart, not done that for many a year. Snag was I flogged my templates on fleabay years back.

Here is the approach I used. not original, pinched off the WEB. Can't remember where.

First a mapping from logical to hardware pin numbers. Then convert a number into an 8 bit wide pattern then bit bang the bit pattern. The convert number part gets around logical anding and testing a 32 bit integer.

// define pin Logical pin / Hardware pin
int pin0 = 0;
int pin1 = 1;
int pin2 = 2;
int pin3 = 3;
int pin4 = 5;
int pin5 = 6;
int pin6 = 7;
int pin7 = 8;

void Send_8_bit(int byt)
{
int d=0;
int binary[8] = {0,0,0,0,0,0,0,0};
{
//convert byt to bit pattern
while (byt>0)
{
binary[d] = byt%2;
byt = byt/2;
d++;
}
// write 8 bits to port pins
if (binary[0]==1) digitalWrite(pin0,HIGH);
else digitalWrite(pin0,LOW);
if (binary[1]==1) digitalWrite(pin1,HIGH);
else digitalWrite(pin1,LOW);
if (binary[2]==1) digitalWrite(pin2,HIGH);
else digitalWrite(pin2,LOW);
if (binary[3]==1) digitalWrite(pin3,HIGH);
else digitalWrite(pin3,LOW);
if (binary[4]==1) digitalWrite(pin4,HIGH);
else digitalWrite(pin4,LOW);
if (binary[5]==1) digitalWrite(pin5,HIGH);
else digitalWrite(pin5,LOW);
if (binary[6]==1) digitalWrite(pin6,HIGH);
else digitalWrite(pin6,LOW);
if (binary[7]==1) digitalWrite(pin7,HIGH);
else digitalWrite(pin7,LOW);
}
}

void setup() {
// set up 8 bit port
pinMode(pin0, OUTPUT);
pinMode(pin1, OUTPUT);
pinMode(pin2, OUTPUT);
pinMode(pin3, OUTPUT);
pinMode(pin4, OUTPUT);
pinMode(pin5, OUTPUT);
pinMode(pin6, OUTPUT);
pinMode(pin7, OUTPUT);
}

void loop() {
// put your main code here, to run repeatedly:

int byt;

byt = 0;

while (byt < 256)
{
Send_8_bit(byt);
byt++;
delay(50);
}
}

Much good advice here. Start small, the diagnostic led is a very very useful tool. Crib and copy, there is almost always something close to your needs out there. The Arduino tool is very useful, learn how to use its diagnostics - the serial monitor tool lets you print stuff from inside your programme. A simple 'got to here' message will help a lot.

Slowly and by small steps move from 'what works' to something new. One pitfall is displays, they often require a library to work. I bought a cheap display only to find the standard Arduino library would not drive it. But a bit of hunting around on the web found the necessary.

Simple program fragments are your friend and good comments are also your friend.

I think pure electric cars are a development dead end. Hybrids are a not entirely honest alternative but at least represent a development route for changing primary fuels or energy storage technologies. The real problem is the primary fuel and its replenishment.

Electric cars have been in development for almost 200 years. Improved a lot in the last 20 years, but think how much the ICE improved from nothing in 1876 to a workable car in 1886. The snag is that we don't really have the basic physics (chemistry is physics) to make a fast chargeable/big capacity/long life battery.

Even if we did the idea of pushing 350KWh + worth of power down a little cable in three minutes or so is a bit daunting let alone taking a spanner to it. The thought of mending an electric farm tractor in the middle of a wet ploughed field suggests that vehicle mechanics of the future will need to remember the 'no old bold mechanics' dictum. The alternative may be a serious risk of frostbite. What joys.

Progress is the exchange of one inconvenience for another.

Had trouble with 20W led bench lamp flickering, been that way a long time, finally swapped lamp and chopped open the offending item. Cut away the grey goop potting stuff and...

Inside is a typical power supply based around a BD3216 chip. A close look shows the mains fuse ended in a dry joint with a black burn mark. Also L3 the buck inductor is missing, it came away with the potting - it too seemed a bit cooked. The mucky soldering on the led assembly is original.

From a quick trace of the circuit the transformer is part of the Vcc power supply and feedback loop.

Now to think of a use for it...

Why not build one. Try klabs.org/history/build.agc

Apologies for useless calculation. With a very thin rod this setup looks more like a Foucault pendulum and those who set up that type go to a great deal of trouble to control wobbles and uncertainties. Even then Foucault pendulums are liable to start following a figure 8 pattern unless controlled.

I do feel that some of the trouble lies in the lack of a conventional suspension constraining the degree of freedom. I also worry about interactions between an iron electromagnet coil and the iron pendulum - symmetry and residual magnetism. I would be thinking to keep the electromagnet very symmetrical and possibly air cored.

Interesting project, good luck with it.

Whoops again stick to 3.78Hz. Blame scruffy notes.

Whoops 0.387Hz.

I know very little but FWIW I think this is a single ended tuning fork. Had a poke around and no convenient formula came up on the web. Lord Raleigh did one but can't find it. But Wiki has a formula for a conventional tuning fork.

Taking E as 100GPa, rho as 1.5g/cm^3 and the radius of the rod(s) as 3mm and taking the Wiki round rod option for a 240mm long unloaded rod I get 3.78Hz resonance, other options are available.

Now I don't think losing one rod makes a big difference (!!) but the loading lump on the end will drop the resonant frequency, but not smart enough to figure out how much. What bothers me about this design is the round rod - it has no favoured direction of vibration - could go anywhere. Takes us into Sturm-Liouville territory where chaos reigns.

So, a rectangular rod and much heavier stiffer mountings is my feeling - or buy a rubidium oscillator.

looks a decent pump, if a bit small. Should get to about 1mm mercury on a small tight chamber.

On the inlet side the vanes sweep oil+air round and compress the air to be exhausted. There is no valve on the inlet side. On the outlet side there is usually some sort of reed valve. The old slow pumps had a neoprene flap, I have seen a steel reed valve on faster pumps. These sometimes break and don't work.

There is sometimes a bleed valve to let air into the pump on the inlet side to help degas the oil and 'give something to chew on'. Usually this can be screwed up tight shut.

The vanes run immersed in oil but not too much - say 3/4 full. Ordinary 20/50 oil works OK, Edwards oil v pricey.

The drive shaft sometimes has a simple oil pump - waggling rod in off centre groove to pump bearing oil. Take care to put this back right. Otherwise, if the vanes are not stuck there is not much to go wrong. Gaskets not usually used but some older pumps used a lick of shellac on the mating faces.

Slightly worried about running a clock in a vacuum chamber, outgassing etc. Not at all easy to get a really good seal, not really a vac once and forget job. Still - enjoy.

As said in ATM books, high vacuum is a technique that makes figuring mirrors look easy - and I agree!

True high vacuum is in the 10^-5 and beyond region, think turbo pumps etc.

I am afraid I have never used this technique. I looked into making an STM some time back and remembered the technique.

The nearest I have come to making needles is the watchmaker technique of:-

fit brass wire in pin chuck

pull out bench drawer and file shallow groove in top edge

lay wire in groove

file and twiddle to get long thin point, use very fine file at end

You 'LINK' technique looks like it should work and from a credible source - and simple too. Good luck.