Marlin Temp Readings bouncing all over the place

I.m having a wierd issue.

When I go to print or set the bed and hotend temps, the temp reading is all over the map. The temp will still rise, but itll jump from 32.8 to 31 to 28, back up to 35 to 33.8 to 30.1, back up to 39... etc etc etc.

Very strange. My thermostors seem to soldered and plugged in just fine...

Anyone else every had this problem?

I had a similar problem with dirty input power driving my AREF. I could filter the thermistor input all I wanted, but the sloppy AREF gave bad ADC readings. A 100uf cap cleaned it up.

Sounds like you might need to tune your PID numbers.

---

greghoge.com

HUGE 3D PRINTER PARTS SALE!!!

I had the same issue with my heat-bed, make sure in your firmware that you have the correct thermistor selected.

Quote
kfootball15
I.m having a wierd issue.

When I go to print or set the bed and hotend temps, the temp reading is all over the map. The temp will still rise, but itll jump from 32.8 to 31 to 28, back up to 35 to 33.8 to 30.1, back up to 39... etc etc etc.

Very strange. My thermostors seem to soldered and plugged in just fine...

Anyone else every had this problem?

This will help

[reprap.org]

Marlin on Gen 6.

Your problem has happened to me about 3x over the last two years of printing 6-8 hours daily. Caused by excessive usage, cold solder joints on the PCB, PCB connectors not properly manufactured or insulated, and or stress/strain/breakage in the wiring.

I have had this same problem going on for a couple of weeks now... Again .

While printing... the temp jumps wildly way above my gcode or pronterface temp setting, way below temp settings, erratic temperature control, or just no temp feedback from the PCB back to pronterface (but the thermistor works properly and reports back accurately to pronterface; room temperature. The radical temp fluctuations have killed several prints mid-stream by the firmware preventing extrusion overheat condition or cold extrusion.

As a competent electronic tech of 25+ years...
Many of you young guys, or backyard mechanics type, skip right over the 101/kindergarten level basics needed for an accurate diagnosis, and jump right towards more complicated and convoluted attempted fixes. Often, screwing your machine and or it's firmware to hell because of this.

CHECK your voltages out of the PCB to the components, and continuity of the connectors, wiring and the component in question (i.e. from the PCB to the component; check it all with a reliable non-chinese VOM.

Nine out of ten times it will be a bad wire connector, burnt connection, cold solder joint on the pcb, or a separation/breakage of the wire contained within its insulation, unseen to the human eye... A VOM ALWAYS TELLS YOU WHERE YOU NEED TO GO to fix the problem (if everything checks out electronically with the VOM; then look elsewhere for the problem; like firmware). Its as simple as doing the 101 with a VOM, checking the basics first before jumping to conclusions guys.

For example... I have had this going on for a couple of weeks now... Again.
Last night during a print, I lost all temperature to the hot end, pronterface gave it's temp command to the PCB, and returned with non response as to changing the temp though, pronterface gave the HE full throttle/current to raise the temp; pronterface value @127. So back to the basics this morning to fix this, or sadly having to buy a new Controller PCB.

So this morning, back to the VOM 101...
I checked heat resistor for continuity, the thermistor for continuity and monitor falling or rising readings with blowing a hair dryer on the HE, wiggled HE connections, smooshed and wiggled the wire set from the PCB to the HE, all passed the continuity check, but still no response in temp shown in pronterface. Conclusion I do not have a bad HE (heat resistor or thermistor or bad wire). So I go to the HE PCB connector... disconnect the HE PCB Connector to do an output check from the PCB, As soon as I disconnected the HE PCB Connector, I noticed a little brown discoloration on the underneath side connector itself (not visible when connected to the PCB, immediately I new I found the problem, the brown discoloration indicates a bad wire connection to the PCB Pin Connector. Removed the Pin from the connector, sure enough the wire was basically separated from the pin connector. Re-soldered wire to pin connector, and happy printing again !

You can all save yourself a lot of headaches and frustration, by going back to the basics first, and learn to use a VOM first before wildly blaming everything else under the sun.

You never made any mention of doing your basic VOM 101 diagnosis, or a complete visual PCB board inspection for cold solder joints or burnt discoloration marks first.

Hope this helps.

Edited 2 time(s). Last edit at 03/01/2014 06:05PM by ShawnT98027.

For better or worse we are assuming a coherent mechanical build. I agree with using the proper tools to troubleshoot but before one starts using test equipment I'd suggest knowing the mechanics and parameters under which the machine was designed. For example, unless I know how all the basic parts are supposed to work together and the basics of the properties that make them run (Ohms law, for example understanding potential, current and resistance) using a meter isn't going to do one much good because one may not know how to use it and even then they may not know what to do with even the most basic data. Many (though I can't say most) of the folks that post issues are not experienced in any sort of troubleshooting much less specific mechanical or electrical troubleshooting. Many of the issues posted are due more to not configuring the machine appropriately in the first place rather than an electro-mechanical issue. As for the temp ramping, ruling out a mechanical or interface issue first is indeed the way to go. That said, PID settings will absolutely cause that sort of instability and running a PID autotune as part of commissioning a new build is a best practice. Every issue is not always going to be mechanical or electrical.

I would suggest that instead of ranting on how those less experienced than yourself approach troubleshooting, a more constructive approach would be to offer guidance on the basics. For example how to use a dvm and what a particular reading might be applied to finding a solution.

If your printer is a new build, with a un-adulterated Marlin firmware installed, you still should not have to mess with PID Tuning.
ALL Marlin firmware versions released that I have ran over the last two years, have always been able to maintain +-1c. So PID Turning/fiddling is not necessary.

My approach given to your temp issue, is direct and specific, in this reprap community the assumption seems to be you have basic electro-mechanical skills.

If your machine had been functioning properly before (maintaining a stable temp), and at some point your temp started jumping around, it's not anything to with PID tuning. That is just a ridiculous thought, and makes no sense whatsoever (as firmware coding does not magically/dynamically change by itself, causing your printer a temp stability problem).

ALWAYS do the basic electro-mechanical testing before blaming firmware, PID Tuning, etc. You must think logical before jumping to conclusions. Basic ability to perform a continuity test and voltage check is critical knowledge one needs to build, diagnose & repair, and maintain any 3D printer.

I get sick when I see people giving advice out of their you know what, that really have no clue as to what they are talking about, or how any machine functions. Then someone takes their bad advice, and totally messes up their machine.

BAD advice like PID Tuning a previously good running printer, to fix a temperature stability problem is unbelievable (and not in a good way). This causes people like yourself to chase your tail around and screw things up, and waste your life away fiddling with things that have no bearing on your current problem (temperature stability fluctuations).

Oh as for 'vegasloki'... oops! I dated myself, VOM verses DVM, both do the same basic testing, so don't let other people confuse the matter.
Funny 'vegasloki', not sure how long you been around machines or how many you have designed and built (and not just simple 3D printers; as they are about as basic as any machine gets), Most if not all machine problems are electro-mechanical related. as software or firmware does not degrade or change over time like physical components do. Unbelievable !

There are a lot of electronics websites and publication that can quickly teach you the basics.

I recommend learning some very basic VOM/DVM diagnostic tests (for continuity and voltage checking)...
but most of all, think logically... and enjoy Happy Printing !

Edited 1 time(s). Last edit at 03/02/2014 02:57AM by ShawnT98027.

Hey guys,

To update:

I agree with Shawn above about PID tuning. It cannot be the problem as this printer was working perfectly fine a few days ago.

I am new to this for sure, but I did do a continuity test with a VOM and it seemed fine, but I re-soldered the connection just to be sure.

I just unplugged the thermistors from the electronics completely, and the temps STILL fluctuate rapidly (just at lower temps). What does that mean exactly?

Edited 1 time(s). Last edit at 03/03/2014 06:05PM by kfootball15.

It doesn't sound unlike the noisy input power problem I saw using a poor-quality ATX power supply driving a breadboarded set of electronics.

What electronics are you using?

What thermistor table/configuration?

How rapidly is rapidly? (10s of seconds could be physical tuning problems, sub-second isn't physical noise, its electronic.)

Could the rapid fluctuation have been happening during the perfect operation a few days ago?

Assuming your thermistors are still open-circuit, the pull-up resistor should be making the thermistor input follow VCC/AREF, and the ADC should be delivering a nice, constant 1023, and the temperatures you see should be the last entry in the thermistor table you are using.

If you are still getting fluctuations with a completely disconnected thermistor, I would suggest you are getting noise into your Analog-Digital Converter somehow. If your schematic lis like [reprap.org] and you don't have a thermistor connected, there is a 3Hz RC filter (1/2*pi*4.7K*10uF) connected directly to the input. If noise is getting into the reference side of your ADC, which is hard-coded to be VCC in Marlin, then the ADC could jump around. Before the OVERSAMPLENR stuff, 10 counts of noise at 30C could be about 5C of noise.

How clean is your power? If you have something like RAMPS, are your capacitors working?

On [reprap.org] @2012-11-27, I was stupidly putting unfiltered ATX +5SB power into my chip. I looked at it with an oscilloscope and saw 50mV P-P 50kHz noise going into my chip. I filtered VCC with a 220uF cap and it cleaned up the ADC considerably.

Quote

kfootball15

I agree with Shawn above about PID tuning. It cannot be the problem as this printer was working perfectly fine a few days ago.

I am new to this for sure, but I did do a continuity test with a VOM and it seemed fine, but I re-soldered the connection just to be sure.

I just unplugged the thermistors from the electronics completely, and the temps STILL fluctuate rapidly (just at lower temps). What does that mean exactly?

Hello kfootball15,

Frustrating, I know...

If your HE Hot End (heat resistor and thermistor) and the wiring check out with a continuity test... Then your good to go with these components.

Next I would... disconnect your HE wire connector (only the heat resistor wires; leave thermistor wires connected) from the PCB, power-up your machine, send the command to the printer to heat (e.g. via pronterface), check it to see if you have steady voltage (as tested with a VOM) coming out of your PCB from HE pins. If so, this does not mean you are in the clear with the PCB being good. Its' very possible if the PCB is outputting stable voltage to the HE pin connectors under little/no load (i.e. not attached to the heat resistor in your HE), that your PCB fails to produce a stable current to the HE under load (i.e. a compromised PCB connection that can not handle any real load).

At this point... I would pull the PCB, break out your magnifying glass, and carefully inspect your board for cold solder joints/cracks in the soldered connections to the PCB pad/via etc. Also look for any signs of brown discoloration, as this indicates also a bad unstable connection and will soon lead to erratic behavior and or PCB board failure. I HAVE SEEN THIS HUNDREDS OF TIMES OVER THE LAST 25+ Years, and is generally a simple fix of cleaning-up and re-soldering any bad connection found. Remember the HE heat resistor is a direct short, and this cause the heat resistor to heat, this load takes a heavy toll on all components within that circuit.

NOT Surprising, that your temp stability showing up more at lower temperatures.... lower temps (I assume your talking about low temps proceeding to your designated higher print temperature), well at this stage of the game your PCB is being told to put maximum throttle/power/voltage/load to the circuit to bring the HE to temp, to your meet your gcode or command (e.g. via pronterface Heater-Temp-Set). If you are using pronterface, you know that a cold HE being heated to print temp, shows a value in pronterfaces status (Monitor Print window) of @:127, once the HE reaches print temp, the PCB reduces the power/voltage/load to a minimal level to maintain the desired set print heat temp, and you will see much lower value than @:127. There is no magic or mystery in these 3D Printers/or any machines or electronics... it's all a matter of understanding them and using logical thinking to resolve issues.

Depending on the PCB Circuit design... one cold solder joint could cause a host problems. Identifying a cold solder joint Google Images - Cold Solder Joint. All your solder joints should be smooth, shiny and fully filled (not dull or crystalline in appearance, or having any hairline fractures or fill voids), I have seen PCBc that the whole board has cold solder joints or just one, either way it's trouble. I suspect your problem is a bad connection or cold solder joint on your PCB Print Controller.

I feel your pain, as I have been there myself.

Regards,

Shawn

Edited 2 time(s). Last edit at 03/04/2014 03:11PM by ShawnT98027.

Quote
ShawnT98027

Quote

kfootball15

I agree with Shawn above about PID tuning. It cannot be the problem as this printer was working perfectly fine a few days ago.

I am new to this for sure, but I did do a continuity test with a VOM and it seemed fine, but I re-soldered the connection just to be sure.

I just unplugged the thermistors from the electronics completely, and the temps STILL fluctuate rapidly (just at lower temps). What does that mean exactly?

Hello kfootball15,

Frustrating, I know...

If your HE Hot End (heat resistor and thermistor) and the wiring check out with a continuity test... Then your good to go with these components.

Next I would... disconnect your HE wire connector (only the heat resistor wires; leave thermistor wires connected) from the PCB, power-up your machine, send the command to the printer to heat (e.g. via pronterface), check it to see if you have steady voltage (as tested with a VOM) coming out of your PCB from HE pins. If so, this does not mean you are in the clear with the PCB being good. Its' very possible if the PCB is outputting stable voltage to the HE pin connectors under little/no load (i.e. not attached to the heat resistor in your HE), that your PCB fails to produce a stable current to the HE under load (i.e. a compromised PCB connection that can not handle any real load).

At this point... I would pull the PCB, break out your magnifying glass, and carefully inspect your board for cold solder joints/cracks in the soldered connections to the PCB pad/via etc. Also look for any signs of brown discoloration, as this indicates also a bad unstable connection and will soon lead to erratic behavior and or PCB board failure. I HAVE SEEN THIS HUNDREDS OF TIMES OVER THE LAST 25+ Years, and is generally a simple fix of cleaning-up and re-soldering any bad connection found. Remember the HE heat resistor is a direct short, and this cause the heat resistor to heat, this load takes a heavy toll on all components within that circuit.

NOT Surprising, that your temp stability showing up more at lower temperatures.... lower temps (I assume your talking about low temps proceeding to your designated higher print temperature), well at this stage of the game your PCB is being told to put maximum throttle/power/voltage/load to the circuit to bring the HE to temp, to your meet your gcode or command (e.g. via pronterface Heater-Temp-Set). If you are using pronterface, you know that a cold HE being heated to print temp, shows a value in pronterfaces status (Monitor Print window) of @:127, once the HE reaches print temp, the PCB reduces the power/voltage/load to a minimal level to maintain the desired set print heat temp, and you will see much lower value than @:127. There is no magic or mystery in these 3D Printers/or any machines or electronics... it's all a matter of understanding them and using logical thinking to resolve issues.

Depending on the PCB Circuit design... one cold solder joint could cause a host problems. Identifying a cold solder joint Google Images - Cold Solder Joint. All your solder joints should be smooth, shiny and fully filled (not dull or crystalline in appearance, or having any hairline fractures or fill voids), I have seen PCBc that the whole board has cold solder joints or just one, either way it's trouble. I suspect your problem is a bad connection or cold solder joint on your PCB Print Controller.

I feel your pain, as I have been there myself.

Regards,

Shawn

Thank you so much!!

I was able to find the problem and you were right! A faulty solder joint that was hidden by some shrink wrap.

Thanks so much Shawn!

Quote
kfootball15

Thank you so much!!

I was able to find the problem and you were right! A faulty solder joint that was hidden by some shrink wrap.

Thanks so much Shawn!

EXCELLENT !... glad I could be of help.

Best Regards & Happy Printing !