I have the Medium Crazy Gear Clock been runing for over 5 months with no problems now it won't keep correct time losses about 6 minutes in 24 hours the second hand rotates correct no loss of time there i have timed it.
Hey Norm, I was having a problem with my own clock I just finished, but I think our problems might be similar.
My minute hand wasn't moving at all, because I think that the spring that I used for the gear 5 assembly is too soft, and therefore the front 2 gears move independently.
To troubleshoot it I lined them up, and a few minutes later the inner one (I think it's called gear 5 30) moved, but the one in front of it stayed at the same spot.
Yours might be slipping slightly, losing a few minutes every day.
A simple thing to do is I guess just tape them together and see if it still does that. Just make sure not to try and set the clock when they are taped together, because you might break something!
There are really only a few ways for the new controller to lose time.
The first possibility is the stepper motor skipping steps because the TMC2208 driver current is too low. It can be solved by turning the TMC2208 potentiometer slightly higher.
The second possibility is the Arduino Nano internal clock drifting with a seasonal change in temperature. The algorithm uses two delay values, one slightly fast, and one slightly slow. The algorithm moves the proper number of steps for the second hand to move one second, then checks if the precision real time clock has advanced by one second. It toggles between the fast and slow delay settings to lock onto the precision RTC. These fast/slow delays use the builtin Arduino Nano delay routine. If the Arduino Nano delay routine has drifted by about 1% or more, it might lose the ability to lock onto the RTC. A new algorithm could solve this.
A third possibility is the precision RTC failing. The algorithm would revert to the slow delay time if it thinks it is ahead of the RTC, for example if the RTC has stopped being recognized.
There are a few ways of determining if your system is experiencing any of these possible causes. The first possibility would still show the second Arduino Nano LED blinking to indicate that it is locked to the RTC. The serial debug monitor would also show a mixture of "+" and "-" characters. The second and third possibilities would show a solid LED that never blinks. Distinguishing between them would require looking at the serial debug monitor in the Arduino programming IDE.
It looks like you purchased pre-programmed boards. Have you ever used the IDE (Integrated Development Environment)? A screen dump from the serial debug monitor would answer a lot of questions.
I will send you new good parts since you are a Patron, but I would like to understand what has changed to prevent locking onto the correct time. Plus, I may need to write an updated algorithm.
Steve I checked all that it works perfect, I even put another controller that I have and it does the same thing, it has got to be something in the gears I believe. I reset the clock this morning and so far it is correct, so I don't know. I am thinking about buying more pre-pr0igramed boards. Where did I buy them from?
The gears cannot slip. The minute hand should rotate exactly one time for every 60 rotations of the second hand.
The only part that might slip is the friction clutch, which is a fourth possible way for the clock to lose time. This is a mechanical problem that can usually be solved by stretching the spring in the friction clutch. Also, make sure the gears passing through the front dial are not binding. The symptoms would include the second hand tracking a reference clock exactly, but the minute hand losing time randomly. The first possible cause of the stepper motor skipping steps would have the second hand get out of sync with a reference clock.
The questions are:
Is the LED blinking on the clock that is losing time? It is inside the base, so the clock needs to be opened up to check. Usually the stepper motor cable can be passed through the USB hole in the base and the electronics are temporarily placed outside the clock.
Is the second hand tracking exactly relative to a reference clock? This would help distinguish between possible cause #1 or #4.
Try to hold off ording them until we figure out what is happening in your board, in case I need to include extra parts to fix your old boards. Etsy likes to have parts shipped within 1-2 days.
Problem solved it was the clutch spring.
Hey Norm, I was having a problem with my own clock I just finished, but I think our problems might be similar.
My minute hand wasn't moving at all, because I think that the spring that I used for the gear 5 assembly is too soft, and therefore the front 2 gears move independently.
To troubleshoot it I lined them up, and a few minutes later the inner one (I think it's called gear 5 30) moved, but the one in front of it stayed at the same spot.
Yours might be slipping slightly, losing a few minutes every day.
A simple thing to do is I guess just tape them together and see if it still does that. Just make sure not to try and set the clock when they are taped together, because you might break something!
Hi Norm,
There are really only a few ways for the new controller to lose time.
The first possibility is the stepper motor skipping steps because the TMC2208 driver current is too low. It can be solved by turning the TMC2208 potentiometer slightly higher.
The second possibility is the Arduino Nano internal clock drifting with a seasonal change in temperature. The algorithm uses two delay values, one slightly fast, and one slightly slow. The algorithm moves the proper number of steps for the second hand to move one second, then checks if the precision real time clock has advanced by one second. It toggles between the fast and slow delay settings to lock onto the precision RTC. These fast/slow delays use the builtin Arduino Nano delay routine. If the Arduino Nano delay routine has drifted by about 1% or more, it might lose the ability to lock onto the RTC. A new algorithm could solve this.
A third possibility is the precision RTC failing. The algorithm would revert to the slow delay time if it thinks it is ahead of the RTC, for example if the RTC has stopped being recognized.
There are a few ways of determining if your system is experiencing any of these possible causes. The first possibility would still show the second Arduino Nano LED blinking to indicate that it is locked to the RTC. The serial debug monitor would also show a mixture of "+" and "-" characters. The second and third possibilities would show a solid LED that never blinks. Distinguishing between them would require looking at the serial debug monitor in the Arduino programming IDE.
It looks like you purchased pre-programmed boards. Have you ever used the IDE (Integrated Development Environment)? A screen dump from the serial debug monitor would answer a lot of questions.
I will send you new good parts since you are a Patron, but I would like to understand what has changed to prevent locking onto the correct time. Plus, I may need to write an updated algorithm.
Steve