Temp Sensor Bed Error

[thorvinus 4]

As the title says, my heated bed sensor is gone. And as Murphy's Law predicts, at the worst time ever.

I got a project at work to make some special tools out of some XT to test (I can smell the promotion) over the weekend and in the middle of my first prototype the sensor fails :((

First thought, after some reboots, was the wireing or the connector are the problem, because jiggeling the wires made the readout go up and down.

Resolderd the Pins on the connector, try'd soldering wires directly to the solderpoints on the PCB... nothing, sometimes it would read normal temps again, cause me to put everthing together again (around 3 times ) and fail after 5 minutes of printing again.

So i got frustrated and had a beer

Then after i calmed down, i ran some new wires from the mainboard to the connector and at first it looked good, but about 3h into my second started print (only a half hour more and i would have someting to show on monday) fails again.

So thats for storytime , now i already ordered a new bed (~200€ with shipping is quite a price i have to say) because i wrecked the stickers on the connectors and i need a new one ASAP. Also i have no time, or interest in dealing with sales people (Nothing personal )

And now to my question, can someone tell me the partnumber of that little SMD sucker thats driving me crazy? Or a replacement non SMD thermistor?

I only found the bed itself, the cabel and the PT100 on the extruder on the github files. Maybe i could go out and find one, not that i'am that optimistic about that

Greetings, Igor

(edit:some spelling)

 

[randyinla 14]

My heated bed sensor for my U2 finally died just as well!! It's been acting up for about 10 days now. Thorvinus, was yours for an U1 or U2? Where did you order a new heated bed?

I went through the exact same steps you did, short of running new wires. I noticed the ends of the wires that were screwed down to the bed's connection block were soldered/tinned. Bad idea! After years of wiring up recording studios, I can tell you first hand that you are never supposed to tin the ends of wire used in screw blocks. Ever. The solder doesn't compress under the screw and you end up with loose connections that can arc or become intermittent at best. I clipped off the soldered ends, stripped back the wire and re-attached. I then noticed through magnifying glasses that the four pads of the connector block were moving slightly in place. Cold solder joints! So I de-soldered, cleaned, re-soldered, but they just would not take the solder. So I took the connector block all the way off and soldered the wires directly to the pads on the heated plate. Beeped it out to make sure things were cleanly connected and when I rebooted, I had the same thing as you; The temp would go up/down if I wiggled the wire or tapped on the back of the heated bed. Something is just not right with this design. That was about 10 days ago and I have been printing happily since then. It seemed relatively stable as long as I didn't touch the cabling or tap on the build plate.

Today, I printed a 3-hour ABS part with a cardboard box covering the U2. I let the part cool down on its own, about 1/2 hour, then took it off the build plate with ease. No warping due to the door on the front and the cardboard box on top. About 5 min after I had the part in my hands, not even touching the U2, the display changed to, "Error - Stopped, Temp bed sensor, contact support." I did all the usual things, tapping, wriggling of cabling... nothing worked. I then took out the main circuit board underneath the U2 for the first time ever to see if that end of the cable was loose. The connector beeped out perfectly back to the white SMD part on the heated plate, not a loose solder joint or short. Just for completeness, I checked the wires that go to screw block on the main circuit board and sure enough, the green wires for the extruder head heating element were also tinned (but not the other two white wires...strange). Clipped the green ones and re-screwed in place.

So now everything is back together and I have a dead machine.

Is there any way to reset everything while turning it on? It might start working again. I assume you can run the U2 without a heated bed, no? The heating element and sensor are such rudimentary parts/connections, it's hard to imagine so much trouble! Should be able to totally disconnect it and have the U2 still start up. It would be nice to know if the little white SMD part has blown or if the main circuitry is hosed. I've been running the heated bed around 105 ~ 110 for ABS. Combine that with the heated chamber effect of a front door and cardboard box on top... could it be too hot for the sensor?

One last thought I have that could possibly contribute to this failed sensor state is that I broke my encoder on the front circuit board around the same time the heated bed sensor started failing. (yes, user error. Yes, another one is in the mail) I have a new one coming in the mail, but have the front circuit board detached from the back of the front panel and sitting on the desk in front of the printer. I have connected buttons to the circuit board so I can still control the printer. Could this circuit board be missing a ground connection via being screwed into the back of the panel? Both sides of the white panel material has thin metal sheets that the screws must come in contact with. Could that be throwing off the sensor circuits for the whole box? I suppose I could try connecting the standoffs of that circuit board to the standoffs of the main one and find out for myself.

Guess I'll open a ticket for this heat sensor and see what happens. Crossing fingers...

 

[thorvinus 4]

Its a UM2 and i ordered directly from the Ultimaker Storepage.

I was also running my bed quite hot for the last couple of days (i was at 90-100°C for printing XT), normaly thats not a critical temperature for solderjoints, but who knows the joints could never be perfect and worked for the time... pretty hard to solder that small stuff onto something that large (and if they used Pb-free solder even harder).

I for now plugged a 250Ohm resistor (two of these anti-noise cable extensions for PC Fans in serial) into the mainboard, with a switch to complete the circuit (and to abord if something is going wrong ).

Had all the wires laying around my desk because i was building a pc earlier and it made a *ding* noise in my head

The Board now reads 120°C and is not sending any power to the bed so i can print with PLA on some Bluetape for now, but i hope the "2-5 business days" statement for shipping stays true.

Hope you get your stuff sorted out RandyInLA

My question again... Can someone tell me the partnumber of that little sucker? Is it a KTY82-210?

Also crossing fingers...

 

[randyinla 14]

Well, a quick googling revealed that the melting point for electronic solder is typically much higher than 100c. I doubt it melted due to the heat of the plate itself.

Good idea putting the resistors in there so you can at least print PLA! I might research that a little bit. Maybe use my Fluke to measure the actual temp of the bed too. It would be nice if there was a small LED on the back of the heated plate to know when the power was being applied to it.

Hope you get your replacement fast!

 

[randyinla 14]

Meh. Didn't work for me. I soldered a 220ohm across the SMD part and nothing. Then removed the part and put the 220ohm in its place. Still nothing. Error - Stopped. Oh well. I have a few other things I can work on in the mean time

 

[randyinla 14]

Success! I was reading another thread and learned that different valued thermistors were sometimes used. I connected a potentiometer to the board to hopefully find the proper range for my machine and it worked! 120ohm registered between 42c~43c and 138ohm as 110c. I soldered in a 120ohm and was able to print out a small part in ABS. Went into the tune menu during the print and alternated the desired temp between 40~45 to turn the bed off/on. Was purely guessing as to the actual temp but will try my Fluke meter's temp probe tomorrow to get a better sense of actual temp. I'm sure support will get me sorted out soon, but being able to print in the interim is a welcome relief.

 

[randyinla 14]

My question again... Can someone tell me the partnumber of that little sucker? Is it a KTY82-210?

 

Here is the thermistor part:

http://www.newark.com/ist-innovative-sensor-technology/p0k1-0805-2p-b/sensor-pt100-0805-class-b-0-04/dp/24M9666?ost=1266934

I'm ordering one as a backup.

 

[z-axis-3d 4]

Hi

 

Here is the thermistor part:

http://www.newark.com/ist-innovative-sensor-technology/p0k1-0805-2p-b/sensor-pt100-0805-class-b-0-04/dp/24M9666?ost=1266934

I'm ordering one as a backup.

 

Hi RandyInLA, thanks for that!

May I ask how you know, where you got the info? I am having similar issues to Thorvinus, if I actually measure across the component I have an open circuit. (I was intermittently getting 108 Ohm, but any movement and it jumps all over the place.

My colleague offered to order me a new PT100, and said Element 14 do a PT100A and a PT100B, where the A in more accurate. I expect it is the B as you have suggested, but would like to be sure I'm getting the right thing to save time.

Cheers,

Greg

 

[randyinla 14]

I was referred to that specific part page by Simon from US support and another guy from Ultimaker support in NL.

Make sure to carefully inspect the solder joints from the 4-wire connector to the heated board. All four of mine were pretty bad, intermittent. I desoldered them, cleaned and re-soldered. Also, the wires that go into the screw holes should NOT be tinned! Whenever connecting wire to a screw block, bare wire is the preferred method. Open the panel that houses the main circuit board and make sure the wires going into those screw blocks are also not tinned.

Good luck!

 

[z-axis-3d 4]

Thanks Randy, much appreciated.

I have ordered one from Farnell here in the UK, the one he was looking at was 3mm rather than 2mm, so I was glad to find your post before he ordered the wrong one.

I had already tried most of this, the un-tinned ends was not something I tried, but I did loosen and re-tighten a couple of times. I will take that advice when I re-assemble.

I was really struggling to figure out what was going on until the PT100 failed altogether; so now it's pretty clear from measuring directly on the component. Initially I managed to get a reading on 108 Ohms, I couldn't get it consistently, but now I have a permanent open-circuit.

Thanks again,

Greg