So as its says, as soon as i connect 2 MAX modules, DCF receiver goes crazy and i cant lock on the signal. It works sporadically with one module but as i need at least 4+ modules, i need some sort of solution.
I am aware that all sorts of electronic interference will affect DCF receiver, interesting thing is, i already have 7 adafruit 4 digit 7 segment displays with backpack connected and everything works great until i connect MAX modules.
MAX modules work as they should, i have reduced brightness of the displays and they dont heat up any more, also reducing brightness seems to help with the DCF signal but i can't lower the brightness more as the displays would be too dim.
MAX modules have 10uF electrolytic capacitors each which should help reduce ripple and noise but doesnt seems to have any improvement.
My setup is as follows" Arduino UNO has Udo Klein's superfilter and DCF module (from ELV) connected to it, it then exports filtered signal to MEGA board. Both boards share the same LM2596 Regulator Module, both boards as well as everything else shares the same GND.
I know i can perhaps try with external DCF antenna with shielded cable far away from my setup, but lets leave that as a last resort and try to eliminate noise if possible.
Question here is simple (hope it's the same for solution) is there anything i can do to eliminate MAX interference with DCF module?
Need to just tidy up my worktable and will take a photo. What i did now is cut an old USB cable and use it to connect DCF module and UNO, at 1m distance, signal as clean as it can get, will start connecting MAX modules now and see what happens.
Each MAX module has 0.1uF and 10uF capacitors, if thats what you meant?
Added 2 MAX modules and the signal is still very clean and i get a sync right away. So DCF receiver at cca 1m away from MAX modules and shielded USB cable does the trick, but still not a solution i am happy with.
Is it too much to ask to have DCF receiver very close to my rig, so once finished i can have it mounted just outside of the clock box? Otherwise there will be cables going out every which way.
The Max7219 is notoriously noisy. The Max7221 is supposed to be quieter, but otherwise the same. So, incidentally, are switching power supplies such as buck converters.
The DCF77 receiver part in on the Uno so that should be well away from the Mega which is driving all those displays. The Uno should be dedicated to running the DCF77 receiver / filter and have good power supply. Good here probably means a bit more expensive but it is very difficult to measure objectively. Just do some sort of experimentation.
The cable between the Uno and the Mega should signal and ground (best twisted together) which no other interconnections.
OK, so i not only have a problem with MAX7219 but with buck converter as well, if i got it correctly... 7219 is integrated in some of the boards but i also have blank boards so will try fitting 7221 and see if that helps.
I did try to connect UNO to separate power supply with no improvement, although it stayed closed to MEGA and DCF module in the same position, some 30cm from the setup and displays.
The idea here is to fit both MEGA and UNO in the same box with everything else, including displays, as one device/clock enclosure. Plan was also to mount DCF module just outside the box, but not 1m away...
Interesting thing is, when i use adafruit displays with backpacks, i have no interference whatsoever.
Reading more about the issue, everyone is talking about decoupling power rail between DCF/Uno and displays... 10uF, 220uF, 1000uF...
so, should i put a cap between 5V and GND from buck converter to UNO or from buck to displays?
NOw i have power and GND from buck going directly to MAX modules and from buck to UNO, so basically power and ground for both UNO and MAX are on the same rail...
The problem is that the multiplexing frequency of displays and their driver chips, and the oscillator frequency of buck/boost converters can all lie in the same (low) frequency range of the 77.5kHz DCF transmitter. Hence interference is a big risk.
yes, aware of the frequency. now, i may have a workaround which is not so great but it works, i can simply tell all displays to turn off at, lets say, 03:00 am and turn on again at 03:30am, in this 30min it would have enough time to sync and keep the clock running accurate.
