Member postings for John Haine

Thanks for that Michael, as you say the response time at ~0.5ms is rather long and also variable which would increase the "escapement deviation". I think in practice it will be easier to just shield the opto from too much ambient light.

Michael, it's the GP1A57HRJ00F.

I have been chasing the source of what seemed like excessive noise in my pendulum measurements and here's a quick summary.

1. There is definitely a problem with the rod being slightly bent by the initial impulsing method (recall that there is a small bar magnet in the top of the rod that twists in a transverse magnetic field produce by a pair of coils). This seems to be a sort of "double bass string" mode. The 6mm rod is too flexible.
2. Even when not impulsed run-down measurements were rather noisy. Trying to cure this I first replaced the rod with a 10mm dia FC tube, then replaced the slotted vane used by the sensor with a plain 5mm pin - this inverts the logical sense of the sensor signal so a suitable adjustment was needed in the measuring system.
3. Eventually I think I have established that the irreducible uncertainty of the time measurement is a few microseconds - I was getting about 5us rms variation in daylight reducing to ~3.5us in darkness, clearly the sensor is somewhat affected by daylight.
4. Knowing the amplitude of swing we can estimate the rms position error sensing by the opto, which comes out to ~0.25 microns in the light or 0.16 microns in darkness. This is the same order as I actually measured on these devices and mentioned here in a previous post.

So I think my conclusion is that one will get an irreducible measurement error of the order of a few microseconds rms using the Sharp optos. To put that in perspective the expected time error over a year due to that would be about 14ms which is probably insignificant compared to other time errors for a pendulum clock.

At a pinch, use a countersink.

Maybe you could avoid pins altogether? Use a "wireless power" setup (chips and coils readily available) to transfer power through the plastic pipe, put the processor inside the enclosure, Bluetooth or Wi-Fi to get the data out?

How are you generating the pulses? If the stepper stalls on switch on it may indicate that it needs and acceleration phase. On my xfeed I use grbl for control and the acceleration comes for free.

Try rolling them across a softish flat surface while fitted in the tool. Soft wood, plasticine? See what kind of pattern they imprint.

Sorry I didn't realise they still sell it - seems to have the old software though.

There's a Makespace in Cambridge (google finds it), but it costs £40/month to join. They have a small Roland gantry router, I did think about joining specifically to be able to engrave a clock dial, but at that price it would be cheaper to pay someone else to do it! Also when I looked the machine seemed to be regularly out of order.

As far as I know the Denford Micromill has been out of production for several years and like all the Denford cnc Mills now will have no support. Mechanically their own machines were bulletproof but the electronics very first generation. With modern controllers and drivers, and possibly a stepper motor upgrade, they can work very well but support will be an issue. Sir John used to upgrade Denfords in university workshops by fitting a one box Chinese controller, described on a thread here somewhere with great success I believe.

The problem with a cnc machine for a makespace is its complexity and need for support over the long term. A mechanical lathe or mill almost by definition can be maintained by a reasonably competent user but throw some power electronics, cnc control software and an operating system into the mix and it becomes a different kettle of fish. For many maker type jobs a small cnc mill will be too small. A decent small router could be a better choice, but one that doesn't need modifying out of the box to be able to do good work.

Apply force to handwheel via a spring balance.

And tapping!

You need to measure how much torque is needed to turn the handwheel and choose accordingly, allowing for any gearing you put in as well.

I made a collet box for ER16 collets by CNC machining in a block of MDF, with a matching block that sits on top to "close" it. To reinforce the material and proof it I soaked it in several coats of Ronseal wet rot treatment - this is a penetrating "lacquer" that soaks into the material and dries and hardens. Not sign of any corrosion after 10 years. May work on other woods too.

Ballscrews might be a good upgrade if not already fitted. For the X (I assume you mean left-right) axis, it looks like the electronics are bolted to the back of the rear rails. Removing that and fitting a piece of 6mm plate right across, bolted into the edges of the side "cheeks" would be a lot more rigid. A Z axis assembly fabricated from alloy plate would be more rigid than plastic if you want to do routing.

Even on a proper mill pure soft aluminium is horrid to machine, much better to use a machinable alloy (6082?). A material which is a delight to machine is Corian, get offcuts from an up-market kitchen fitter. Good for making precise parts that don't need to be very hard or strong. There is a material not unlike MDF called Valchromat which uses a melamine resin rather than urea formaldehyde. It routes very well, doesn't blunt cutters like MDF, is through-coloured, and not nearly so unfriendly to breath the dust. Great for things like spoil boards, small sacrificial mounting plates etc. Expensive to buy an 8x4 sheet so try to get offcuts. I actually got a sun-bleached sheet almost free that had been in a showroom, I also made a clock case from new stuff (blue), I had the components CNC routed and took away all the offcuts too, I seem to have a lifetime supply at least for small projects.

Basically the rpm of a typical DC motor depends on the number of turns, the magnetic circuit design, and the applied voltage. So a motor that runs at 10,000rpm on a 24v supply will very nearly be generating 24v back emf with no load. At 12v the same motor would run at 5,000rpm. Controlling by PWM, you take the "duty cycle" which is the ratio of on time to on+off time and multiply the supply voltage by that fraction to find the effective voltage. So a 24v 10,000rpm motor supplied at 24v with 50% duty cycle will run at 5000rpm.

You may be lucky enough to find a belt and pulley combination that matches the shaft spacing. If not then an effective belt tensioner can be made using a ball race with a bush loctited in place that has an off centre mounting hole. Bolt this with a suitable spacer off the end plate and adjust the tension by setting the position of the race. I used this arrangement on my mill cross feed, there's a photo on here somewhere.

Edited By John Haine on 02/04/2023 14:22:26

Somewhere I read that the very latest atomic clocks can detect the difference in time running at head height compared to foot height due to gravitational time dilation. Whether your feet or your head gets older first I can't recall!

Cousins is a respected supplier to the horological trade, they should be able to advise.

The best things to read about suspension springs are Philip Woodward and Kenneth James. PW wrote some articles in HSN that also reference James, HSN 1998 No. 5 and 2001 No. 4. A chap called Alan Emmerson has looked at solving the exact equations for a compound pendulum on a spring suspension, it is a horrendous problem and the answer is of little practical interest. As far as I can make out one of the main conclusions is that there is an additional pendulum mode where it rotates around a transverse axis at a frequency determined by the moment of inertia and the spring rate and dimensions.

Watch and chronometer balance wheels have carefully designed "terminal curves" to make them isochronous. Philip Woodward also looked at the theory of this and wrote a 6 part article in HJ.

Compared to springs, pendulums are easy!

Nice drawing - essentially a Quorn type I think? The Quorn has an 8mm taper that matches a standard watchmaking type of collet I think - these collets are quite hard to find and expensive. You can of course just use a plain tapered mandrel but if you were thinking of fitting collets as well maybe consider an ER16 nose? You could use it with a drawbar for wheel mandrels and have the option of collets.

Just check for equal phase to phase resistances.