DIY Moon Phase Dial

[Gina]

I've ordered some more DS3231 High Accuracy RTC Real time clock modules - due to arrive Tuesday.   I want a couple for other clocks too.  At least they don't cost much   Meanwhile, I'll alter the sketch to run the clock without.  Can't set it to real time but can test the motor and sound deadening measures.

[Gina]

Not having much joy here   Checked that the Nano didn't produce any unwelcome voltages on the stepper driver and then soldered in the new one.  No movement from the motor so I checked the voltages around the ciccuits and the driver was only getting a couple of volts on the +5v pin also found the Nano getting hot.  Switched off and disconnected the wire to the stepper driver +5v logic supply and powered up again.  Checked volts on +5v Nano output and it was 7v   I'm wondering if I should use a separate voltage regulator to provide the +5v from the +12v.

Thing is, the clock had been running for several days without any problem other than too much motor noise - the Nano built-in regulator seemed perfectly happy.

[Stub Mandrel]

[Icosahedron]

Recently I had a problem with running a Nano demanding 120 mA from a 12V supply. The backlight on the TFT display would start flickering within half a minute after power-up as the regulator became unable to shed the heat. I rectified the problem by inserting a 1N5338B 5.1V 5W zener diode in series with the supply, dropping the input voltage to 7V.

The Arduino specification does specify 12V input but these boards we use are sure cheap.

 

[MarkyD]

Other than sensors which use the onboard regulator I supply a separate regulated +5V supply to power ancillaries (LCD's RTC etc).  I also limit the Vin to the Arduinos to +9V.  If using +12V the onbaord reg has to shed 7volts multiplied by the current supplied in Watts as heat. 3.5 Watts if a max capacity.  It's only 4 Volts drop with a 9 volt supply with 2 Watts at max

[Gina]

Thanks folks Good point about the volts drop in the internal regulator with 12v supply.  It could be that these cheapo Chinese clones vary from sample to sample - some cope with it but some don't.  A separate +5v supply does seem a better idea.  I should have some 5v regulator chips.  I've decided to build a completely new board anyway.

[MarkyD]

If you want a pcb making Gina I can make one for you if you give me the Schematic and part numbers of the components etc for sizing and pin layout

[MarkyD]

34 minutes ago, Gina said:

Thanks folks Good point about the volts drop in the internal regulator with 12v supply.  It could be that these cheapo Chinese clones vary from sample to sample - some cope with it but some don't.  A separate +5v supply does seem a better idea.  I should have some 5v regulator chips.  I've decided to build a completely new board anyway.

I use these MP1584's from the same guy on ebay we get out nanos from.

http://www.ebay.co.uk/itm/5-pack-MP1584-miniature-Power-supply-DC-3A-adjustable-LM2596-alternative-/161811570603?hash=item25acb89fab:g:vfwAAOSwLVZV5Kdc

[Gina]

They look generally useful - I've ordered a 5 pack - thank you

42 minutes ago, MarkyD said:

If you want a pcb making Gina I can make one for you if you give me the Schematic and part numbers of the components etc for sizing and pin layout

Thank you for the offer Mark   I'll make it up on stripboard first though to check it works properly.  I may end up making more than one of these as I have other clocks in mind.

[Gina]

Regarding powering the Nano :-

Quote

Power:

The Arduino Nano can be powered via the Mini-B USB connection, 6-20V unregulated external power supply (pin 30), or 5V regulated external power supply (pin 27). The power source is automatically selected to the highest voltage source.

I presume this means I could have one +5v regulator powering the Nano and other 5v devices rather than connecting the Nano to +12v (or less) by its 6-20V unregulated external power supply.  Anyone done it this way?

[MarkyD]

I power mine with a 12V - 9V to Vin and then 12V - 5V to power ancillaries.  If you use Vin you will need 6V or more as the on board reg has a volt drop of approx 1 volt.  If you use the 5V in you bypass the on-board regulator but I have only done this when programming the LCD which then gave comm conflicts so I had to disconnect the 5V between the two.  All GNDS are common

[Stub Mandrel]

I've powered a nano off 5V, it used the 5V on the USB connector IIRC

[Gina]

Thanks both   I had a feeling there might be some reason for not using the +5v pin for power input.  I have a vague idea of reading something about this some time ago.

One other point...  When I had the previous setup with +12v supply feeding Nano, ULN2003AN and motor driver module then Nano +5v output to power the RTC module and logic power on the driver module, the Nano wouldn't program with the 12v power on.  I reckon this may have been due to power funnies.  I really want to be able to upload a new sketch to the Nano without powering down or removing the Nano from the clock.  I'm sure this should work.

[Gina]

Found this on the Arduino message board 

Quote

USB Problem with Arduino Nano 3.0 With External Power Applied

Jan 17, 2012, 11:19 pm

I am unable to get a Nano 3.0 to be recognized by Windows USB port when I apply 5v to the regulated input or 12v to the VCC input.  All my Nanos behave this way.  I've seen this problem described in other posts.  Apparently, this issue may be specific to the Nano.  I disabled the 5v power line on a USB cable, but the Nano received no USB power when itis powered by my circuit board.  So it seems like Nano can get power from its USB, but USB can't get power from Nano.  I haven't reviewed the schematic, but I'm guessing there's a diabolical diode in there somewhere. 

<<< snip >>>

I remember this now - I've had the problem before   I think I got round it be taking a USB lead and stripping the cover off to get at the wires.  Cut the +5v wire and connected it to the 5v supply on the circuit board, powering the Nano from the PSU rather than the computer.  That should cure the overheating problem too.

[Gina]

USB lead doctored.  Tested with bench PSU to red wire and Nano Gnd and powers it fine.  Running test sketch Blink.

[Gina]

I'm still having odd problems if I plug the Nano into my clock circuit board.  I've very carefully looked over it with a high power magnifying glass for anything silly such as solder bridges and can see nothing wrong.  The only thing connected is the ULN2003AN chip.  Now I've been thining about this and using this chip for driving the three lines to the LED light strip seems like overkill - using only 3 of the 7 driver sections.  I think I might change to discrete transistors and resistors.  The load is purely resistive so no need for catching diodes - just 3 resistors driving 3 transistors.

[hughgilhespie]

Hi Gina,

My 'recipe' for powering Nano's is to always use the Vin pin and feed it about 8 volts. My main DC PSU puts out 14 volts and I think this is too high for the Chinese Nanos. I use a cheapo buck converter to drop to about 8 volts for Vin to use. I like to use Vin as that means the 5 volt to the Nano comes via a non-switching regulator so it should be a cleaner supply. I have one Nano that is a bit funny when I connect it to a powered USB hub to upload a new sketch, the overall current to my various obsy boxes jumps from around 900 mA to 1700 mA. The Nano won't run when the USB is still attached but performs normally after the USB cable is disconnected. I have no idea what is happening but, as it works, I am not too worried. The other Nanos in my system, 3 of them, all work as expected with a permanent USB connection.

What I enjoy about electronics and computing is that everything is so very precise and completely predictable!!

Regards and good luck.

Hugh

[Gina]

Ha Ha !!!  Thank you   All the Nanos I've tried won't upload if power is connected to Vin.  The voltage regulator I shall use for the +5v supply (dropping from 12.23v) is analogue - LM317T so no switching noise.  I have used a number of these for various projects with success   I see my stock is running a bit low so I'll order some more.

 

[Gina]

I couldn't find a piece of stripboard big enough to build a new circuit so I've stripped down my existing board and equipped it with sockets for the main components viz. Nano, RTC module and motor driver module.  The ULN2993AN has been removed and will be replaced with resistors and transistors.

[Gina]

The Nano is now working fine when plugged in both uploading and with comms to the Serial Monitor.  Running the RTC test sketch and also my clock sketch has shown that the RTC module is dead.  The sketches hang at the point where the RTC is asked for the time viz.

    setSyncProvider(RTC.get);   // the function to get the time from the RTC

If this statement is commented out the sketch continues reporting "Unable to sync with the RTC" and then other things.

Quote

 Moon Clock v2.03 - NEMA16 microstepping
 testpoint 1
 Unable to sync with the RTC
 testpoint 2
 Moon Clock v2.03 - NEMA16 microstepping
 Moon Clock v2.03 - NEMA16 microstepping
 Setting hands to 12 o'clock
 Speeding up 2000
 Speeding up 4000
 Slowing down
 Moving hands to 1:0 using 232000 steps - including 1536 extra steps

This means I can't properly run the clock until I receive the new RTC modules (Expected tuesday) but I can test the motor functions and maybe sort out the quieting. So next step will be to plug in a motor driver module and connect 12v.

EDIT...  I should add that the reason the sketch runs right through that far is that the Hall sensors are disconnected and the sketch thinks the hands are already at 12 o'clock.  And the reason it's setting the clock to 1 o'clock is that it's set to BST and the default time without RTC is midnight.

[Gina]

Some progress to report - motor now running again

[Gina]

Component layout.

[Gina]

The diagram above is wrong - here's the amended one with resistors for the LM317T voltage regulator.

[Gina]

LM317T voltage regulator added to circuit board plus the two resistors to give +5v output - 330R & 1K.  Calculated output voltage taking the resistor values and chip parameters is 5.13v - voltage measured with my DMM on test was 5.17v.  That's within spec.  I have run +5v power to RTC & motor driver modules (pink sleeved wires) but not connected the USB power yet.

[Gina]

Connected up the USB wire and carefully checked the board over for any problems like solder whiskers and nothing found.  Plugged in the Nano and applied power.  The LM317T started getting hot and showed 5.15v comopared with 5.17v on no load.  Something evidently wrong but goodness knows what.  This project has got at me again!!!   How on earth can it be fine with +5v supplied from my bench PSU but not +5.17v supplied from the LM317T??  To add further to my woes, my bench PSU has stopped working - it blew its mains input fuse