Member postings for Peter Cook 6

Assuming a 40:1 device, 9 degrees is one rev of the handle. 38' is 25.333 degrees on the handle. One hole on the 15 hole plate is 24 degrees. Using that would give an error of 2' i.e. .36'. If you need more precision I would mark a disk of paper with a 25.33 angle and use that to position the handle.

Quick Google got me What Is The Pressure Per Square Inch (PSI) Of Natural Gas In A Home? (homeupward.com)

This suggest that the pressure in the final leg of the supply is measured in inches of water (1/4psi). So a gas pipe buried a few feet down subject to a foot or two of water would not have enough pressure to resist.

Doesn't answer the question of how or where it got in - there must have been a weak spot somewhere.

Dave, the new behaviour persuades me that the double hump was an artefact of the pendulum rod & suspension spring behaving as a double pendulum. With the wider swing, the suspension spring stiffened up to the point where the carbon fibre rod would bend.

With the new low arc, the forces never get to the point where the carbon fibre rod bends, so it acts as a rigid rod.

You say the hand just hangs down at 6 o clock. This is a diagram of motion work

Do you have the two connected gears on the right of the diagram. One is driven by the pinion on the minute hand. the second drives the hour hand via its gear. You don't mention them.

Sorry postings crossed.

Edited By Peter Cook 6 on 30/11/2022 20:23:18

Dave, there is an interesting simulation of a Double pendulum at Double Pendulum (bestofallpossibleurls.com)

If I set Mass1 and Length 1 to the minimum, and Mass 2 halfway and Length 2 about three quarters. Then set Phi & Theta to values that put the two rods in a straight line about 10 degrees off vertical to give the initial impulse, then the main bob looks like a simple pendulum, but it looks as if the period varies. The bob certainly does some odd things.

I think what you are seeing (assuming the code is OK) is the pendulum behaving as a double pendulum with the suspensions spring as one flexible point, and the pendulum rod itself as the other.

Reducing the stiffness of the suspension spring should help as the pendulum rod itself is probably bending as the suspension spring "stiffens" as it bends.

Edited By Peter Cook 6 on 28/11/2022 18:34:05

Dave - sorry - I got myself confused between positions and transitions - but the idea of the pendulum being out of beat stands.

I would get the negative response - You spend too much time in there as it is!!

Dave,

One way a pendulum could appear to behave like that is if it is slightly off vertical (out of beat), and you were comparing the timing of the left-right swing with the right-left swing.

Looking at the numbers (0.86sec period = about 180mm pendulum length), it seems that you are correctly counting full cycles, i.e. left-right-left.

If you are, then you need to drop the next left-right transition before starting the cycle again otherwise your system or code could be alternately timing left-right-left, and right-left-right which could give you the results you are getting.

PS a second thought - could the pendulum with its spring and flexible rod be behaving as a double pendulum. They do weird things! In regular clocks the pendulum rod is very rigid compared to the suspension spring. Sorry my math is not up to analysing a double pendulum.

Edited By Peter Cook 6 on 28/11/2022 13:52:35

Have a look around the NAWCC forums. NAWCC Forums

There is a specific forum for 400day and Atmos clocks. There is a lot of useful information and advice to be had - bit like here really!

Just a thought!

Are the porcelain tiles glazed or unglazed. Unglazed porcelain is fairly soft - probably too soft even for a marking out table. If they are glazed, the hard glaze layer will be fairly thin and rubbing down high spots will potentially break through.

Agree with Hopper.

Statistically, conclusions based on sample sizes of one are considered rather unreliable. But they make great newspaper headlines.

To test Dave's suggestion, can you tighten the screw back down again. If not then either the key is the wrong size, or the thread in the cover is stripped. 30 turns should have brought the screw all the way out. 2.5mm key suggests M5 x0.8 thread. 30 turns should have shifted the screw 24mm!

Possibly an M2.5 screw which needs a 1.3 key, although I would have thought you would have struggled to get a 2.5mm wrench in that size hole.

If the grub screw is undone, then the cover should slide off the splines, but as others have said they get tight with the gunk.

With an idle half hour, I played with the beam calculator I have used before.

Assuming the plate (620 x 400) is 75mm thick, and that granite has a Youngs modulus of 70,000 MPa* (about 50% of that of cast iron and a third of that of steel**), the calculator suggests that

With the plate treated as a beam supported at the 20% and 80% points (as per Staretts) of the 620mm length, a point weight of 1000Kg (a Tonne) dead centre will cause the centre to depress by 0.0027mm. At less than 400kg point load the calculator gives up and says no deflection (probably less than 0.001mm).

For any reasonable hobby use I suspect anything that supported the plate reasonably well would be good.

*Granite :: MakeItFrom.com

**This surprised me somewhat when digging around. A cast iron surface plate is apparently stiffer than a granite one!

Edited By Peter Cook 6 on 12/11/2022 18:57:14

They have a range

Color sensor | Toshiba Electronic Devices & Storage Corporation | Asia-English (semicon-storage.com)

but the device you have may be a prior generation and/or the device id you are seeing may be for the camera unit rather than the sensor chip.

USB ID Database give Vendor ID 3034 as Toshiba Visual solutions - now known (since 2021) as TVS REGAZA . Owned since 2017 by the Chinese Hisense group.

Nearest product I can ID is

TCD2566BFG | Linear Image Sensors | Toshiba Electronic Devices & Storage Corporation | Asia-English (semicon-storage.com)

Would support the light colour. Mine is clip together (now off!) white plastic tiles. The light colour

a) makes small things easier to spot

b) encourages cleaning up more often.

You only need the slider and the electronics. I used the bits from a set that the plastic case had broken, and 3D printed a replacement housing.

Works well as a cross-slide DRO.

PS Get a set that retain the zero when they switch off. The ones that reset to zero when you switch them back on are a pain as a DRO.

Edited By Peter Cook 6 on 07/11/2022 23:04:34

The motor looks OK if it is man enough for what you want.

One caution. The first controller Huub links to (AliExpress, it's the one I used) has a 10V DC output for signalling. The motor which Michael links to specifies 5v for the Hall effect sensors, so you would need a 10v to 5v step down to drive the hall sensors.

The Stepper online controller linked to provides 5v for the Hall sensors so would be directly useable.

My motor specifies 5-20v for the hall sensors, so I simply used the 10v output to drive them.

Without the pictures it's hard. to tell, but I suspect you have bought a throttle pedal based on a Hall effect device. They are used on e-bikes etc, because they are far less susceptible to dirt and wet. Red is +5v, Black is ground, and the green is a voltage signal (usually 0.8-4.8v).

Whether it works will depend on what the controller expects. If it has a 0-10v pot, then it will be incompatible, and you may have blown up the chip in the throttle pedal.

Not an expert - but have been doing the same research. Looking at APT's offerings, they seem to do 55 & 60 degree tips with "partial profile" which will cut a range of pitches, and a wider range which cut the full profile (presumably including the crests) and for which there is a different cutter for each pitch.

I assume for production work you would choose the full profile tips, but for jobbing and hobby work, the partials make more sense.

Hope this helps.