Extruder Switched off MINTEMP triggered

Mendel 90 has been working perfectly and printed a 4 hour job last night but when trying to print another from the same file the job stops after a few seconds with the subject error.
I have attached a word document with the screen shot that shows the dip in temperature and then the recovery along with the text. Obviously the dip to zero for the extruder is the trigger but also the bed temperature dips. The first layer is printed at 200C and the rest at 195C. After the dip the temperature recovers to 195C and not the 200C. Also the SD printing byte ratio races on even though nothing is happening.
The problem is repeatable on jobs that i have run before and using different computers. I think this eliminates Slic3r, Pronterface and computer. The only thing i haven't changed is the Arduino firmware on the printer itself. I don't want to do this as at the moment i can't find the file!

MINTEMP is caused by bad or broken connection for the thermistor, which means it's getting disconnected. From your description, it's intermittent, so some wire/connector wiggling may be needed to find the problem.

Thanks for the reply.
Both the bed temperature and the extruder demanded temperature drop and recover at exactly the same time. You can see from the attachment that the actual temperatures don't collapse which indicates the thermisters are OK, at least to my inexpert view. If i run the same print a number of times it stops at exactly the same point.
Do the error messages i provided give any clues?
I checked all the connections and did the wiggling but things remained the same.

Check the power supply connections to the Melzi are good and and tight.

Check that there isn't a partial short to the extruder thermistor from the heater caused by burnt plastic on the hot end heater block.

---

[www.hydraraptor.blogspot.com]

If the printer also stops moving and it happens after the same amount of time, it's acting like the power supply is shutting off.

If you're using a RAMPS controller, the MEGA could be having problems.

I've checked all the terminal connections and cleaned the extruder thermistor. Unfortunately one of the wires came detached and so i changed the thermistor.
The first print attempt failed as reported in my first input but subsequent attempts don't get that far.
Connecting the printer with the SD card loaded resulted in the screen shot, "Initial Switch On".
This gives an error that the extruder is switched off; but it should be as i've not started a print.
If i progress to print from the SD card despite the above message it progresses and starts to ramp up the extruder temperature but not the bed temperature; not usual. See screen shot, "SD Card Print 1". The extruder set temperature drops to 195C which it should only do after the first layer and the bed temperature starts to rise.
I then left the printer running from the SD card and and whilst nothing physically happened it eventually came to an end. See screenshot, "SD card Print2".
The extruder was hot as was the bed.
This print should take over 4 hours but the code was raced through in 8 mins!
I've now found the Marlin firmware, i think, which i modified as i had to do a rewiring job on the board under the guidance of Nophead. So if all else fails i could try reloading this. I also have a spare board which i could swap but i'd prefer a simpler solution and to understand the problem.

I presume it rattles through the gcode ignoring most of it because of the MINTEMP error that requires a restart. There is no point trying to print until you fix that error.

Put a voltmeter across the extruder thermistor terminals and check it it behaves erratically. What voltage does it read when it gives MINTEMP?

---

[www.hydraraptor.blogspot.com]

When i set the extruder temperature to say 200C the temperature graph rises steadily and there is a corresponding steady reduction in voltage to .55VDC.
When I set to SD print and the MINTEMP error appears the voltage goes to zero. After a little while it recovers and it is currently showing 2.4V at 115C with no erratic movements.

Something that might be important. I don't do that much printing but some while ago there was a period of erraticness on the temperature graph. The temperature would rise smoothly to 200C but the print would not start because the indicated temperature would swing rapidly, say, +or- 10C. Clearly the extruder temperature would not be doing this. Eventually it would stabalise and off it would go. I checked the wiring and continuity but could find no faults.
This hasn't manifested itself in my recent prints.
Nophead; You may remember that i damaged the chip on the board you supplied which made a rewiring necessary. In pins.h I also changed #define TEMP_0_PIN 7 to ~define TEMP_0_ 1.

Erratic temperature swings are usually caused by molten plastic getting onto the thermistor leads, oxidizing to black and becoming partially conductive.

In your case where you have blown an analogue input pin it is conceivable that more of the chip is damaged such as the analogue multiplexer or the ADC itself. I haven't seen that before though.

Are you saying it works with manual control but not with SD? That would be odd. Does it work manually if you also turn on the bed and the motors?

If the voltage goes to zero you should get a MAXTEMP error. MINTEMP would correspond to very close to 5V. I would also check the 12V at the Melzi terminals is stable, it shouldn't drop below about 11.4V. It would also be a good idea to check that the 5V line is stable. You can get that from the VCC jumper in the middle of the board.

Stable at 115C when set to 200C would imply the heater isn't working as it should be full on under those conditions. The symptoms are all very confusing.

Edited 1 time(s). Last edit at 07/22/2016 03:10PM by nophead.

---

[www.hydraraptor.blogspot.com]

Another possibility is a firmware pin conflict between the SD port and the thermistor pin. Has it ever worked from SD since you remapped the pin?

---

[www.hydraraptor.blogspot.com]

I've switched on the printer but not connected it to Pronterface and the 12V supply reads 12.4V with the 5V supply on JP15 reading 5.01V. Both are rock steady.
Now connected and bed and extruder temperatures ar at RT; 27C. Both above voltages unchanged with hot end and bed thermisters voltages at 4.77V.
Set bed temperarure to 60C and it climbs more or less linearly to set temperature and remains steady. Thermister voltage at 60C is 4.07V. All other voltages steady.
Set hot end temperarure to 200C and it climbs more or less linearly to set temperature and remains steady. Thermister voltage at 200C is .54V. All other voltages steady.
Left with both heaters on and hot end varies less than +or- 1C.
Start print from SD card and temperatures rise normally and countdown OK. MINTEMP message, printer stops and thermister voltage definitely drops to 0. After about 30 secs voltage re-appears but lower as hot end is now cooling.
Repeated process but measured jumper voltage and this remained rock steady at 5.01V.
Repeated process but measured 12V on Melzi voltage and this remained rock steady at 12.2V.
Tried printing directly from computer and got this screenshot but nothing happens. I never print directly as i've had drop outs and so the screenshot may be normal.

Missed your last message. Yes it has always worked from the SD card both before and after I re-mapped it. The re-mapping was in 2014.

> MINTEMP message, printer stops and thermister voltage definitely drops to 0.

That has be baffled. Dropping to zero implies a short but that should give MAXTEMP error. The only other way it could go to zero is if the pin was defined as an output and driven low. If you haven't changed the firmware recently then that can't be the case.

Do the motors ever move? Perhaps a short between a motor and the thermistor? What does your start gcode look like? Perhaps type it manually and see which command causes the MINTEMP.

---

[www.hydraraptor.blogspot.com]

I once more switched the printer on and i got the MAXTEMP error for the first time and the thermistor voltage was 0V! Picked up on your thought re a short between motor and thermistor so i wired the thermistor directly into the Melzi and it seems OK.
The machine is running at the moment but without fan or filament and has been doing so for five minutes with all voltages holding steady.
I'll re-assemble and confirm all is well.
Thanks again.

All's well. I did a 4 hour print yesterday and it came out well. There is still some oscillation of the hot end temperature but that may be because i'm using terminal blocks to slave the cabling. In the course of tidying up the cabling i also found that the fan wasn't responding to M106 and so wired this directly to the Melzi. That now works OK. Seems like the termination of the ribbon strip or the PCB are causing the problem.
Thanks for your help once more.