arduino mega256 / Ramps 1.4 re setting

Hi all, my i3 clone with auto levelling induction probe and reprap discount smart controller is posing me great problems. I can move all axis individually home all with G28 level the bed from smart controller or pronterface with G29, heat up the diamond type nozzle manually extrude plastic but when I try to send gcode, problems, the G28 and G29 at the start of the code work perfectly, the bed and nozzle heat up to temp and the nozzle moves to the print start position and then no error message the arduino resets it's self, the gcode is lost and the heaters turn off and the smart controller screen show printer ready. Can some of you experienced guys please help. Thanks in anticipation.

Potentially a power issue, not enough current? Causing a reboot.

Thanks Dust, that is something that I had not thought about, I am using an ATX power supply that says on the label the 12 volt rail is capable of 13 amps, I am not sure how I can check if it does, In my estimation the heated bed extruder and motors should be using less than that but perhaps I am wrong?. Since posting this original message I have been looking very closely at my gcode and slicer settings and suspect that some of the feed speeds are on the high side so I inserted M300 codes into the file to try to see where it was falling over and I was amazed how far down the file the machine is reading as it prints the first few lines and so my problem may not as I suspected be at the beginning of the file. So my conclusion is to create the simplest of file possibly a shallow rectangle and check the code from end to end and try that. Thanks for your reply and input I will certainly bear it in mind and work out just what current the printer is drawing .

If you have a standard 200x200 heated bed that alone uses 11amp
+5 amps for everything else

+4 amps for good measure (as power supply labels aren’t known for being accurate)

So I would very much suspect nothing like enough power.

Hi Dust,
Thanks for your suggestion, I have been on holiday so have not replied earlier I took head of what you say regards power and since my return I connected my 13 amp supply to the 11 amp connection and a second 7amp ATX supply to the 5 amp side but that did not alter the issue, trying to print a 20mm test cube the G28 and G29 codes worked fine the nozzle heated to 240 deg and then the nozzle printed the first full 4 sides single bead and stopped dead then locked out as before could this still be under powered? I tried reducing feed speeds way down to a crawl but still the same. I have ordered a replacement Ramps board to ensure mine is not faulty so I will wait for it to arrive and try all these things again and report what I find. As a fleeting thought could the arduino megga have a faulty processor? now I am clutching at straws???.

Try a print without the heated bed to confirm/rule out insufficient current availability....
could be a duff mega2560, could be electrical noise, could be a number of other things. Rule out the PSU, RAMPS & Mega 2560 by substitution

try a print from the SD card without the USB cable connected to rule out comms issues over the USB

Have you got any forced air cooling over the RAMPS/Mega2560? is it getting too hot?

Thanks dart16 for the reply, I will try all of your logistical steps as soon as I have the time. The electronic noise idea is one that I had not thought about but does sound logical and so I intend to address this one, for instance the probe, heater fan and thermister wires all run side by side though I suppose that this is normal so I am going to make twisted pairs of all that I can and I was wondering if laying a bare copper wire grounded at one end along side or in the midst of the wire bundle would help to capture and ground any generated noise ?. Nothing seems to be getting hot not to the touch anyway and the exhaust air from the power supply is blowing over the boards but that also is something I could fix. I am awaiting the delivery of a new ramps board to eliminate that possibility and if all else fails I will order another Mega 2560.

Thanks dart16 for the reply, I will try all of your logistical steps as soon as I have the time. The electronic noise idea is one that I had not thought about but does sound logical and so I intend to address this one, for instance the probe, heater fan and thermister wires all run side by side though I suppose that this is normal so I am going to make twisted pairs of all that I can and I was wondering if laying a bare copper wire grounded at one end along side or in the midst of the wire bundle would help to capture and ground any generated noise ?. Nothing seems to be getting hot not to the touch anyway and the exhaust air from the power supply is blowing over the boards but that also is something I could fix. I am awaiting the delivery of a new ramps board to eliminate that possibility and if all else fails I will order another Mega 2560.

Hi folks, I have just had a thought my mega 2560 appears to be taking its 5 volts from the USB and not from the ramps board or power supply could this be why it keeps resetting its self due to insufficient amperage if so how do I give it more power

hi all,
I have now replaced the Mega 2560 (clone) and now the printer operates as a stand alone as well as when connected to a computer with Pronterface, great, ah but my problem still persists I am now convinced that the 5 volt rail is overloaded, the smart controller screen, levelling probe and third stepper driver would seem to be starving the megga's processor resulting in it re booting and returning to the printer ready status when code is running, has any one else had this problem ?. The smart controller takes its supply from the interface board the probe takes power from aux 3 and the 3rd stepper driver takes its pulse step and direction from aux 2 and 5 volts from the live rail of the limit switches. Pronterface or smart controller either will perform G28 and G29 commands and my G29 routine ends with the nozell returning diagonally from back right corner to front left home corner and both will move any axis individually whilst the heater and bed are heating I know that the new 20 amp power supply is not starving the 12 volt rail but it does not have a 5 volt connection from which I could power the probe or smart controller.
When trying to print code the printer goes through the G28 followed by the G 29 followed by the heating command when the heating is done the head moves rapidly to print start possion, stops dead and the processor re sets and the screen shows printer ready all code is lost. can someone please help.

Yes, maybe your 5V collapses when all motors are moved at the same time. To check that you'd need a DSO.
My Octopi, Mega & RAMPS are powered from the 12V PS through a simple 5V regulator, forgot the name right now. But if you have some electronics addictive around you, (s)he wil probably have some 5V voltage regulator at hand, or take a look at Ebay.

If you have a full graphic display, these can be a bit greedy on 5v supply, try running the printer with the display un-plugged.

Thanks guys for your replies, if I aquire a 5 volt supply how should I power the smart controller away from the interface board? I can power the probe that is straight forward but the smart controller is a puzzel

If you use a externall 5 volt supply you must remove D1 on the ramps board, then you can connect the new 5v supply to a unused end stop 5v 0v pins

I removed the mosfet on the Mega that switches between USB and DC power, this way my Mega can only be powered by the external power and not from USB. The LM2596 DC convertor is good to step down a 12V or 24V supply, but if you are using an ATX then you should have a 5V supply ready made and available.

thanks for your reply, but due to earlier suggestions I have swapped out my ATX supply for a 12 volt 20 amp supply with no 5 volt tappings

Hi all, here I am again getting more and more frustrated. Now having updated to a 20 amp power supply, removing D1 and connecting an L2596 2amp 5volt supply to the unused limit pins so I am sure that I do not have any power problems on either 12 volt or 5 volt rails but still have re-setting and kill errors messages after the G28 and G29 commands have completed and heating is up to temperature the head moves to start position then either stops dead and re-sets to printer ready status or moves along the print path for a couple of seconds and then shows the error Kill called with a request to reset. I have set the print speeds and acceleration speeds to snails pace but I cannot overcome these errors, please can someone help.

Please post exact error message

Hi Dust, thank you for your patience as I said the printer rapidly moves to the start of the print displays on the smart controller "heating done" and then either line 1 "Err: MAXTEMP " Line 3 "PRINTER HALTED" Line 4 "Please reset" Or line 1 "KILLED" line 3 "PRINTER HALTED" line 4 "Please reset" Or the screen goes straight from the bottom line displaying "Heating" to a complete arduino reset showing printer ready on the bottom line. As I have said in previous posts my last printers all very basic and used the sanguinololu 1.3a board with little or no problems, now that I have upgraded to arduino mega 2560 clone and ramps 1.4 and added a fixed inductive probe and a reprap smart controller I am venturing into a totally different field. In an attempt to trace the problems I have replaced both the arduino Mega clone and the ramps shield together with changing the ATX power supply for a 12 volt 20 amp and after removing the D1 diode powered the 5 volt rail with a 5 volt 2 amp voltage reducer connected to the spare limit switch pins. As I am retired the time spent with all this does not matter I put it down to adult learning ah ah, but I do not like to be defeated without at least trying so if you could suggest where else to look I will be truly grateful, could it even be a configuration.h problem? As a second though I looked at the pronterface log after trying to use the same gcode, the error is as follows.
Traceback (most recent call last):
File "threading.pyc", line 808, in __bootstrap_inner
File "threading.pyc", line 761, in run
File "printrun\pronterface.pyc", line 1006, in statuschecker
File "printrun\pronsole.pyc", line 860, in statuschecker
File "printrun\pronterface.pyc", line 998, in statuschecker_inner
File "wx\_core.pyc", line 14619, in __getattr__

Edited 1 time(s). Last edit at 07/18/2017 01:17PM by papakel.

Is it a new printer build? Have you tried new stepper drivers? Perhaps there is an issue with a driver, shorting out as it warms up internally or feeding back current through the Mega. Maybe the extruder motor has an intermittent short.

Try to pair all wires close to their respective grounds, they don't have to be twisted but optimally within 1cm without other wires crossing in-between. Power cable should only be 2-3ft or so long with a separate feed for the bed.

Also try redownloading Marlin from scratch just in case there is a bit error lurking somewhere.

"MAXTEMP" is your hotend getting to hot or the more likely the thermistor wires on your hotend shorting together

This doesn't explain the arduino reset however...

Thanks guys for your replies, however if I had some sort of short on the steppers would that not happen during the g28 and g29 routines and also when manually extruding filament ??.
And also I recently ran PID autotune for 240 degrees and entered the Ki Kd and Kp numbers and saved them to eprom.
I noticed in my configuration.h file that my PID_FUNCTIONAL_RANGE is set to 10 would that be right?.
I am running temperature set to 240 with my Max temp to 250 deg, i.e. 10 deg higher than wanted, is this too close to the PID_FUNCTIONAL_RANGE ?? should I change one or the other.
On a separate note a wild suggestion was made to me by a computer technician, much as your earlier thoughts Dust, he thought that the processor was being starved despite the 2 amp feed to the 5 volt rail, and suggested fitting a separate 960 Ma supply directly to the +5 volt and Ground of the mega chip , THIS idea put the fear of --- up me but do you guys think he has a point? something is upsetting it, as it happens I have a said supply so your views would be appreciated

Hi guys, I am glad to report that my resetting problem has now been overcome, the thing that appears to have been the problem all along was the SD card, that is to say that I set the sd spi speed to the lowest setting in configuration.h file, all that my feeble mind can think is that the arduino must use the SD card as memory dump for the processing and if the full speed was too high then that may be why it reset being unable to retrieve or send data, anyway the resetting problem stopped happening. Now that I can keep the machine running I just need to fine tune all the other bits, the Z-offset and the virtual tools are my first jobs. Never the less I would like to offer my greatest thanks to all members for the help you have offered along the way, and hope that I can ask you all again with further problems. Many thanks

I see you have found something that works, thought i would place some of my fixes out.

I have the RepRapDiscount Smart Controller the one with the Click controller and the SD card.

If the cables for my LCD/SD are routed too close to the Stepper drivers, I get all kinds of errors.
"Err: MAXTEMP ", Filament run-out sensor led lighted "meaning it has filament", but Controller ejects filament, filament run error, saying change filament,
The LCD Menu displays garbage on different lines.

So I enclosed the Controller and LCD into separate shielded enclosures.
and Routed the Cables as far away from driver chips as possible.

As a simple test I had the LCD running straight to the Front, lying on a microfiber cloth my errors went away.
When I placed the LCD back on top routing the Cables alone the Side of the Drivers errors returned.

My Setting
#define SPI_SPEED SPI_HALF_SPEED

I can duplicate the errors on a moments notice. All I have is this strategy myself.
4 Controllers 4 Displays all have the Same Errors, All can be fixed with shielding and proper cable routing.

I have tried 3 different power supply's including a non-switching Ham Radio 12 Volt 30 Amp highly filtered supply same issue with all.

Edited 7 time(s). Last edit at 07/27/2017 06:04PM by Roberts_Clif.