PT100 reading 2000°

User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

PT100 reading 2000°

Post by Aflac »

I am trying to set up a PT100 on my 0.8.5 system, I have the 1.12a dc42 firmware.
I attached a picture of my wiring.
for testing purposes i set it up on heater 2, So in my config file i have
M305 P2 X200
M563 P1 H2 D0

however it always reads 2000°, which i understand is what it puts if its not getting any data.
Does anyone have any ideas what I am doing wrong?
Attachments
IMG_20160525_093509.jpg
User avatar
Jimustanguitar
ULTIMATE 3D JEDI
Posts: 2631
Joined: Sun Mar 31, 2013 1:35 am
Location: Notre Dame area
Contact:

Re: PT100 reading 2000°

Post by Jimustanguitar »

I've been fighting with a thermocouple board on my Duet. I posted about it on the RepRap forums, and the consensus was that the unshielded leads of the thermocouple were seeing EMF from the heater lines that they run down to the hotend with. Interesting that I don't see the same behavior on my other Rambo Max which has the leads travel twice as far, but still a valid point and fair conclusion.

I tried two different thermocouple amplifiers, different thermocouples, etc, and was still getting the same results. The 'correct' solution would be to locate the amplifier board at the hotend and minimize the distance that those leads travel, then just run the 5 leads from the board to the Duet in the bundle with the bowden tube. After fighting with it for a while, I decided it was easier to just use a plain old thermistor. Not as accurate (which is arguably a moot point anyway) but I haven't had a temperature fault since.

David (DC42) did comment in that thread that an apparent open connection to the thermocouple, or a drop in communication from the amplifier board would both appear as 2000c to the Duet. Hopefully I'm retelling that correctly.

Perhaps the behavior that I've been seeing with my thermocouple is similar to your PT100 experience?
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

yeah i read that thread, thanks,
however if you look in my picture the hotend its attached to is actually sitting right there, i dont want to hook it up to the printer until i know its working, so every wire is as short as possible and it does not get close to any other lines at all, it looks like i have a wiring problem, but ive looked a ton of times and i cant figure out what ive done wrong. mine is a constant 2000° so its not getting any communcation from the board i think.
Xenocrates
ULTIMATE 3D JEDI
Posts: 1561
Joined: Wed Sep 23, 2015 2:55 pm

Re: PT100 reading 2000°

Post by Xenocrates »

Have you checked those jumpers for continuity? I assume that you have the channel defined for it correctly (X200 in this case, looks like) I might try switching the CS pin to 26, which would want X201. I might also try using shorter jumpers between the boards (From what I can find, 4Mhz SPI wants a max length of 6 inches.) I'm firing up my own Duet with PT100 soon, and it looks like we have an identical board (I ordered the one DC42 linked, and presume you did as well). So if I encounter the issue, I'll let you know (I also presume that you have at least 11.A, which fixed an SPI initialization issue.)
Machines:
Rostock Max V2, Duet .8.5, PT100 enabled E3D V6 and volcano, Raymond style enclosure
Automation Technology 60W laser cutter/engraver
1m X-carve router

Sic Transit Gloria Mundi
01-10011-11111100001
dc42
Printmaster!
Posts: 454
Joined: Mon Mar 07, 2016 10:17 am

Re: PT100 reading 2000°

Post by dc42 »

When you issue the M305 command with X200 in it to enable the RTD channel, is an error message displayed, in the gcode console tab of the web interface, or in your USB host program?
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

No, i get no error message.

just to test i commented it out of the config and then manually typed it in the web interface g-code console and got no error message.
dc42
Printmaster!
Posts: 454
Joined: Mon Mar 07, 2016 10:17 am

Re: PT100 reading 2000°

Post by dc42 »

A few suggestions:

1. Your photos show 5 or perhaps 6 wires going into the terminal block of the RTD board. How do you have them connected?

2. If you choose a different RTD channel (say X201 instead of X200) and send the M305 command from the console to select it, do you get an error message? If I select a non-existent RTD channel, this is what I get:

[img]https://dl.dropboxusercontent.com/u/19369680/NoRTD.png[/img]

If you get that message when you specify the wrong channel, but not when you specify the correct channel, that implies that the Duet is communicating successfully with the RTD board.

3. If you have a multimeter, measure the voltage between the two centre terminals of the terminal block on the RTD board. It should read about 0.28V if everything is connected correctly and the Duet has configured the RTD board.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

i have 6 wires, 3.3v grd, SDO CS SCLD and SDI, wired i think per the instructions. but im guessing i must have a problem there, i just dont see it.

I am not sure what im doing differently now, but now when i try m305 p2 x200 i get the same error you are showing in your screen shot....well with heater 2 instead of 1.

when i measure those center two terminals of the RTD board i get 2.014 V.
User avatar
Windshadow
Printmaster!
Posts: 526
Joined: Mon Nov 30, 2015 6:35 pm
Location: Mid Coast Maine

Re: PT100 reading 2000°

Post by Windshadow »

this is going tosound a bit nuts but I was taught my first electronics, when I was about 10 years old, by a former WW2 radar tech when i was attempting to make a 1949 vintage DuMont tv set (it was then 1960 or so and he was running his TV repair shop and he had an assembly manual for the that old tv so i followed the instruction replacing popped caps and discolored resistors etc testing tubes from other scrapped sets etc. but one whole section, I think it was an IF stage, all the wiring had been chewed up by rats in the barn where he put his junk sets like the DuMont he gave me, in case he needed it for parts.

any way i followed the manual and circuit diagram rewiring it and it did not work but i was sure i had done it step by step.... he told me to take it all apart, but step by step, reading the instructions and diagram backwards as i did it and darned if i did not find my wiring error.... and i got the old set working...

when i was writing stuff in college i recall being told to read what i had written backwards out loud as a form of self proof reading i wonder if it is the same principal.... and no i don't proofread much any more :D
from the take it for what is worth department
dc42
Printmaster!
Posts: 454
Joined: Mon Mar 07, 2016 10:17 am

Re: PT100 reading 2000°

Post by dc42 »

Aflac wrote:i have 6 wires, 3.3v grd, SDO CS SCLD and SDI, wired i think per the instructions. but im guessing i must have a problem there, i just dont see it.

I am not sure what im doing differently now, but now when i try m305 p2 x200 i get the same error you are showing in your screen shot....well with heater 2 instead of 1.

when i measure those center two terminals of the RTD board i get 2.014 V.
If you are getting 2.014V then your PT100 is not connected between those terminals, or it is open circuit. That explains the 2000 reading.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

When I ordered my PT100 they originally sent me one that did not fit in the heatblock, so i got a second one that does fit. turns out that one is bad, i disconnected it and hooked up the other one (that dosnt fit) and bam works perfectly, so I must have a bad sensor or sensor with bad wiring.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

So I took a dremel polisher to the PT100 that didnt fit and made it fit.

Got everything set up and it is working well....most of the time, every once in a while the temp jumps back up to 2000 degrees, just for a second, but its enough to shut down the heater and ruin a print. sometimes i make it through a whole print sometimes i do not. I have not found a pattern or reason to it yet. I read a thread that had something similar with a thermocouple that suggested interference, but it seems like it would happen a lot more frequently with that. its a very intermittent up to 2000 then right back down to the real temp.

Any ideas on this one?

Also my spell check wants to turn thermocouple into thermonuclear and I found that humorous.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

Ok, here is a screenshot of the error
As you can see it happened 4 times within a few minutes. after a long period of it not happening.

The first two happened with things still running, but the second two was after everything was shut down so interference from other lines seems unlikely to me.
Attachments
TempSpike.JPG
TempError.JPG
IMBoring25
Printmaster!
Posts: 616
Joined: Wed Mar 18, 2015 1:11 am

Re: PT100 reading 2000°

Post by IMBoring25 »

I have no experience specifically with the PT100 board, but that looks like intermittent contact somewhere in the circuit. Reseat plugs? Add strain relief? I've sometimes needed to cut off and re-crimp connectors.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

Ill check all the connections, which would make sense if it happened when things were moving, but it still happens with everything stopped so what would cause it to momentarily lose a connection, and if it was a connection would it not be lost for longer, its pretty much an instantaneous thing, bad then good.
Xenocrates
ULTIMATE 3D JEDI
Posts: 1561
Joined: Wed Sep 23, 2015 2:55 pm

Re: PT100 reading 2000°

Post by Xenocrates »

I know Jim had experience with this with a thermocouple board. DC42, is is practical to add a setting to ignore temperatures which don't persist for a settable amount of time? Or is there already something like this available?
Machines:
Rostock Max V2, Duet .8.5, PT100 enabled E3D V6 and volcano, Raymond style enclosure
Automation Technology 60W laser cutter/engraver
1m X-carve router

Sic Transit Gloria Mundi
01-10011-11111100001
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

So I completely separated the pt100 line from the others and took out all extra length and connections, It's not pretty and just runs right off to side instead up into the top and down a tower.so all possible sources of interference have been minimized as have the connections. And I still got the problem, it took longer than it normally does, making it about 3 hours into a print when it had been doing it at about 40 minutes.

But then again it made it completely through one last night so it could just be random.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

Are there any data logging functions? It would be good to see how often this happens and under what conditions, It's hard to just stare at it and hope.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

For a few days I've been printing at 205°, and did not get the problem, got through multiple long prints with itout it stopping, today I went back to 250°, the same temp I was getting the problem at before, And it happened about 30 minutes into the print.

So it seems to only happen when I'm at a higher temp. That is strange.
dc42
Printmaster!
Posts: 454
Joined: Mon Mar 07, 2016 10:17 am

Re: PT100 reading 2000°

Post by dc42 »

Aflac wrote:For a few days I've been printing at 205°, and did not get the problem, got through multiple long prints with itout it stopping, today I went back to 250°, the same temp I was getting the problem at before, And it happened about 30 minutes into the print.

So it seems to only happen when I'm at a higher temp. That is strange.
I haven't quite finished making changes for the 1.13 firmware release, so I'll take another look at the RTD code. It already allows for a small number of dud readings in succession before reporting an error, but I'll check that code is working and see whether it is appropriate to increase the number of dud readings allowed.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

so it turns out my theory on it being the high temps was wrong, it happened again last night, 6 hours into a 7 hour print @ 205.

DC,
Thanks for looking into it. your support here and in general is nothing short of amazing, it makes me glad I switched to the duet after my smoothie died.
dc42
Printmaster!
Posts: 454
Joined: Mon Mar 07, 2016 10:17 am

Re: PT100 reading 2000°

Post by dc42 »

I have taken a look at this. For some error types, the firmware allows several attempts to get correct readings, but for others it flags a temperature fault immediately.

When a temperature fault occurs, there should be a message written to the GCode console in the web interface and/or the USB port and PanelDue if you have one, giving the error details. Please can you look for this next time the fault occurs, and let me know the full message.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

We are out of town for the weekend, I'll try a print Monday when I'm home and let you know the results.
User avatar
Aflac
Printmaster!
Posts: 146
Joined: Thu May 28, 2015 11:32 pm

Re: PT100 reading 2000°

Post by Aflac »

this time it was almost funny, I ran a print to get the error to post it, and it happened with literally 10 seconds left in the print.
Attachments
tempError2.JPG
User avatar
Jimustanguitar
ULTIMATE 3D JEDI
Posts: 2631
Joined: Sun Mar 31, 2013 1:35 am
Location: Notre Dame area
Contact:

Re: PT100 reading 2000°

Post by Jimustanguitar »

And I never got anything in the console window when I was fighting it, just these pop-ups.
[img]http://i.imgur.com/UIXtPdn.png?1[/img]


That was a version or two back, and definitely used a different version of the web interface... I haven't checked for errors in recent history.
Post Reply

Return to “Duet”