Gina

The two more Duet WiFi kits arrived this morning so I am now the proud owner of three of these

Printed 40t drum gear and see I've made a mistake Not much of a problem, can alway redesign the gears. Adequate centre spacing is 70mm (not 80mm). Back to the drawing board then... Can't have 40t and 30t because that gives a 4:3 ratio and 20x4/3 = 26.67 steps. No good! Checking with the 40t gear I've just printed shows that 45mm radius will be alright giving 45t and 25t at mod 2. Try 45t and 25t ie. 9:5 (that rings a bell ) 20x9/5 = 36 (tick). Using the same gears for the months gives 200/12 x 9/5 = 30 steps (tick). Q.E.D.

If I make the 42t and 35t gears mod 2 their centres would be 77mm apart giving 3mm spacing between motor and drum. To return this to 5mm means the centres want to be 80mm apart meaning the the radii will need increasing by 80/77 and slightly more than mod 2. 42x80/77 = 43.64mm and 35x80/77 = 36.36mm. Here are screenshots of the gears in SketchUp.

Now to calculate the number of steps required for driving each drum. 3D printers use 16x micro-stepping whereas full stepping would be better for the calendar. The RAMPS board has plug in links to set the micro-stepping so it's only necessary to remove these links to change to full stepping. The stepper motors have 200 full steps per revolution. For the date this works out as simply 20 steps to advance either digit (assuming the units digits are equidistant around the drum and the tens match). For the months there are 12 positions corresponding to 200/12 = 16.67 steps if using 1:1 gear ratio - awkward But with a 5:6 gear ratio one position would be 200x6/(12x5) = 20. The drum could have a 42t gear and the motor a 35t gear to give the right ratio. The weekday drum could have the words spaced the same as the date using 7 positions out of the 10. It isn't necessary to have the spacing equal around the drum - a stepper motor can go in both directions. That means the weekday drum can use equal 40t gears like the date and 20 steps to advance the weekday. At the end of the week the drum/motor can either advance by 4 days or go back 6.

I shall not be using the 3D printer firmware (sketch) in the Arduino Mega but a sketch which I shall write specifically for this purpose. 3D printers do not use Real Time Clocks but the RAMPS board does provide for extensions and the RTC module can be added on a small daughter board plugged onto the extension pins on the RAMPS. There is an Arduino library for use with the RTC which decodes the data to provide year, month, date, weekday, hour, minute and second so we have the data required for a perpetual calendar immediately. I already have experience of the RTC Library in my Moon Dial Clock to automatically set the time from its RTC.

To drive the motors I plan to make use of the electronics recovered from a 3D printer which I've replaced with an upgrade circuit board. As mentioned above, this consists of an Arduino Mega 2560 plus an Arduino shield called RAMPS (RepRap Arduino Mega Pololu Shield) which provides drivers for up to 5 stepper motors. It also includes electronics which aren't needed for this project but there is far more wanted than unwanted and it will save me designing and building my own circuit board. Stepper motors do not provide any feedback for their position so other methods are used to set a starting position. In 3D printers these are called EndStops and mostly use micro-switches, though anything that will switch when a certain position is reached will suffice. For this project I propose to use magnets and Hall effect switches which provide a very effective solution. The RAMPS happens to provide exactly what we want, allowing for up to 6 EndStops. This project will use 4 - one for each drum.

Next is the drive. I could use 100mm and 50mm gears giving a 2:1 reduction ratio. 50mm & 25mm radii. Trouble with that is the drum gear teeth will protrude beyond the drums. Also, this leaves virtually no gap between motors and drums. With the motor against a drum, distance between centres = 75mm. With say 5mm gap the distance between centres becomes 80mm. Gears could both be 40mm giving 1:1 and no reduction - worth a try. To get some reduction the drum gears could be 45mm and the motor gears 35mm giving a 9:7 ratio but that's little better than 1:1 so might as well go for 1:1 and save hassle.

The present drums are 100mm OD so 314mm circumference. Number of digits/words per drum is 7, 4, 10, 12 (weekday, date tens, date units, month) so worst case is 12 giving a maximum height of 314/12 = 26mm. 25mm high letters for the month then. For symmetry it probably makes sense the use the same for weekday. The date digits could be the same size or slightly bigger at 30mm (not a lot of difference really). The digital clock and calendar has 50mm time digits and 25mm calendar. The calendar digits are marginal at the distance I normally view it. OTOH the main problem with the digital clock is contrast which is poor even in daylight. I have just placed the calendar drum unit on the bookshelf against the wall and a 25mm high digit is pretty clear. I should have some 25mm or 1" letters and numbers but can't find them so ordered some more. Self-adhesive vinyl.

Crimping tool didn't go small enough for the power ferrules so just used bare wires - seems firm enough. I can crimp the other connectors with narrow nosed pliers as I usually do with these (and similar Molex) multi-way connectors. My order for BAT85 Schottky diodes has arrived so I can connect the Z probe. Here's the wiring diagram.

I seem to vaguely remember doing that too - can't remember if it was a RAMPS board though.

That's what I usually do but the Duet wiki warns against that, saying the connections heat up and melt the solder. But I've used that method for high current connections without any problem. I'm running 15A with a 24V PSU on my Titan printer and with a separate bed heater feed on my Mini printer too, the Duet won't be taking that much current. I guess they are thinking of cases with high current motors and bed heater supplied from the Duet.

Back to the Duet. Have crimp tool but the smallest slot is too big for the power ferrules - maybe try a blunt pair of wire cutters... or not bother with ferrules the bed power is separate.

Decided NEMA 17s are too big for this job - I have 3 NEMA 11s and a NEMA 14 I can use instead. With this arrangement I should be able to put the calendar below my Moon Dial Clock if I move the clock a few inches higher up. That will leave room on that wall for my weather station analogue wall display, which can make use of the NEMA 17 motors.

A CNC machine is a possibility for a future project.

The 3" dome has arrived from China (earlier than expected) and looks a much better option that the 4" ones. Several advantages. Here is a new design for the casing top to take this dome.

I shall need to find my crimping tool - I thought I knew where it was bit it's not there I saw it when I was clearing up. Anyway, I still have wiring to sort out. It's very difficult to decide the best place for the control board as wires come from all directions. I guess it really doesn't matter and probably best to just plonk it in a reasonable place and just wire up. Probably best to keep the power input wires short (ish).

I was looking at the hang printer - might have problems with the geometry though. I think I might be able to work out the theory but how to incorporate it into the printer control I don't know. I have an idea of how the geometry may be modified to Cartesian but this would be as complicated as building a true Cartesian printer though probably using less material. I think a hang printer may only be really useful if it could be arranged with a large print bed and probably heated too.

Now printing a clamp for the PSU to hold it where I couldn't drill a hole in the casing because the circuit board was in the way.

How many 3D printers have you got Julian?

Have correct size acrylic sheet, 3m thick for back plane which I've attached, together with PSU, to the frame. Then attached Duet board to back plane with four M4 standoffs. Both PSU and Duet board are outside the main printer volume and separated from the heat from the heated print bed by the back plane. The Duet board is spaced 25mm away from the back plane which should allow good circulation of cooling air. Amazing how long these simple things seem to take these days! Now need a cuppa

Now running wires/cables and planning positions and fastenings for PSU and Duet board.

I might have been a touch rash but I found Marlin had it's bad points. Seems to me the Duet is better thought out but OTOH Marlin has a much bigger user base, at least at present.

OK if you don't mind Dave it might be better to keep any newer posts with what has already been posted. I'm easy either way...

Thank you I too was of the opinion that "if it ain't broke don't fix it" but I've been thinking of adding a Raspberry Pi for remote control of my Titan printer as I've been finding the way it works at present is a chore. The USB connection to my laptop is unreliable. Changing things in Marlin is also a chore. Windows editors don't work well for editing config.h in Marlin so any time I want to alter things I have to mess about with connections to use Linux for editing. So I've already decide I want the Duet on the Titan. I guess it's quite possible that I may find disadvantages with the Duet but it seems pretty unlikely. Much of the Titan is already running on 24v and I want to change to a 24v 60w hotend heater as well as print bed. It uses large size NEMA 17 motors and I think the Duet may run these better. Changing to Duet on all my printers won't be a total extra cost as I can make use of the Mega/RAMPS for my weather station wall display.

I've done it There have been so many times in the past when I've hesitated about buying things then later when I decided to buy, the item was out of stock. 3D printers are rather less expensive than astro gear Paid with PayPal balance from selling unwanted astro.