PT1000 pullup resistor?

[kharrisma]

Hey Folks!  Now that the Klipper is happy with my config files and appears to be fully functional, I'm wondering about a short article I read concerning the pullup resistor best used to realize the full accuracy potential of a PT1000 thermistor.  Typical default resistor is usually a 4.7K (?), and the Nitehawk toolboard in my machine has a 2.2K.  The article stated that the best possible resistor to use with a PT1000 is a 1K, 0.1%.  TINY little thing, an 0402-size SMD on the Nitehawk.  Not the easiest things to find, either, as everyone wants to show you plenty of ONE% resistors.  I find myself wistfully recalling the days when Alta Vista was the premier search engine, one that actually paid attention to your search terms and fully utilized Boolean operators, which are largely ignored by today's search engines... and don't get me started on Amazon's or Home Depot's search engines.  Ask for anything, and you'll be shown everything, *including* the kitchen sink!           I feel so much better now!        I read the article on the TH3D site, here:  PTt1000 article  It's brief, and I'm not taking it as a sales pitch, since it's talking more work, and they aren't selling the resistors.  Just wondering about the technical accuracy, or necessity for, that 1K resistor??

[PFarm]

Maybe @ChicagoKeri can help you with this.

[Penatr8tor]

This is just my opinion and experience. Take it with a grain of salt. If you like this sort of rabbit hole then you go buddy!

I have 2 printers running 1st gen Rapido hotends with standard 104NT4 thermistors, one printer running a Rapido 2+ and one running a Goliath hotend. Both are equipped with PT1000's.

The big question for me whenever there's a potential rabbit hole to explore is...

1. Is it worth it?

2. Will it improve my prints?

3. Will it let me print with higher temp materials?

Based on my experience...

PLA? 1=Nope 2=Nope 3=N/A

ABS/ASA? 1=Nope 2=Nope 3=N/A

PETG? 1=Nope 2=Nope 3=N/A

TPU? 1=Nope 2=Nope 3=N/A

Exotic High Temp Materials? 1=Yes, 2=Maybe 3=Yes

These are just my shoot from the hip observations.

Normally when I do an upgrade to a new hotend... I'll choose the one with the best thermistor etc but is the thermistor driving my purchase? No, I just don't print with those materials because I don't need to for the most part and If I'm going to go down a rabbit hole... I generally tend to go down the wrong ones. 

 

[therm_virtual0y]

I have a Phaetus Rapido V1 running a PT1000 that plugs into a BTT EEB2209 tool head with a 4.7k ohm resistor.  I have no issues with temp at all.  I have never tested it with a thermocouple so If it’s exactly accurate I can’t say.   I just calibrate the filament to the system and get great results no matter what filament I print with.  The filament temperature calibrations work on two different printers, one with a regular thermistor, and the prints turn out the same.  
 

If I had another printer with the same tool head board, I’d like to test it out just because.

good luck and don’t forget to have fun!  (Not my quote) lol

[kharrisma]

Much appreciated, guys.  I wanted it so I could print with nylon and/or other hi-temp materials, but that's down the road apiece, 'til I build an enclosure.  From what I gather, the PT1000 works well enough with standard pull-up resistor values, but the potential for closer accuracy is the deal with the 1K resistor - - ain't necessary, just opens things up a bit.  Not sure how accurate the temps need to be, really; likely it's a case of 'close enough for government work.'  If it doesn't work out for some reason, I have a bagful of 'standard' 104NT thermistors I can switch to!   

[ChicagoKeri]

Hi!

I just did this!  The research was confusing.

 

It seems that Klipper has built in tables for at least the following temp sensors: Generic 3950,  ATC Semitec 104GT and PT1000. 

E3D, who made my PT1000, states that it will work with a standard 4.7k Pullup. 

 

It seems that my toolboard, the Mellow SB2040 was designed for use with either a standard 100k  Thermistor or a PT1000 on the same input, having a separate input for PT100 sensors.

The standard Hotend TEMP input is a 2-wire connector, JP4, with a standard 4.7k pull-up resistor as default.  This serves for the standard 100k Thermistor types.

If one uses a PT1000 sensor, a jumper, JP2, just above JP4, must be installed and Printer.cfg updated to "sensor_type=PT1000" and "pullup_resistor=1000" under the Extruder section.  The nomenclature indicates that one might get a gross error of -180C if the jumper is not installed.

So what does this jumper do?  it shunts out 3.7k off the 4.7k total pull-up resistor leaving just 1k. One would normally leave the jumper off for standard thermistors.

Other systems I've seen also use a standard 4.7k pull-up.... But not the Nighthawk!

So, it seems that your Nitehawk board has a fixed 2.2k pull-up resistor. Perhaps just entering "pullup_resistor: 2200" and "sensor_type_PT1000" in printer.cfg under extruder will suffice ?  Will Klipper do the math?  I think, probably?

 For some reason LDO appears to think you should enter in a series of values to define your thermistor such as;  "temperature1: 0, resistance1: 32116.0, temperature2: 40 resistance2: 5309,... etc.  Unfortunately, their sample printer.cfg settings only include values for a generic 3950 thermistor.

I suspect that you might have to find and enter the appropriate values for your PT1000.  

It seems that Slice Engineering publishes these values for their PT1000 sensor. It appears that E3D does not publish these values for theirs.

 

Clear as Mud!

 

[midmadn]

I have a Rapido 2F UHF with a PT1000 on my new LDO V2.4

I'm also running the Nighthawk 36 toolhead board.

This is what I ended up with in my printer.cfg

---------------------------------------------------

heater_pin: nhk:gpio9
sensor_type: PT1000
sensor_pin: nhk:gpio29
pullup_resistor: 2200

----------------------------------------------------

Upon printing a temp tower - my findings are exactly the same as my other machines with lower grade thermistors. Best temp for Polymaker and Ambrosia ASA/ABS 255-260.

Therefore I am satisfied with the settings and performance of the PT1000. No rabbit holes needed.

Jack

[kharrisma]

Keri, when I tried first "defining" the PT1000 with three resistance/temp values, then declaring that defined thermistor in the configuration, it gave Klipper a fit.  Wasn't until I got rid of the definition and declaration, and simply stuck "PT1000" in sensor_type and 2200 for pullup resistor, exactly as midmadn said, that things started working better.  Apparently when you try to tell Klipper something it already knows, it gets testy!    I can see how that jumper works - - slick way to implement different resistor values for different thermistors!  I don't think such strict accuracy is required for quality prints; likely using a PT1000 for printing with plastics is overkill, like using a shotgun to deal with mosquitoes, given that printing temp ranges stated for any given filament are closer to ballpark estimates than to three decimal-place precision figures.  I'll forget about the precision and just take advantage of the greater range.

16 hours ago, Penatr8tor said:

If you like this sort of rabbit hole then you go buddy!

Yeah.... this is exactly the kind of rabbit-hole I'm far too prone to get sucked into!  I get caught up in the technical side of it and forget about the 'diminishing returns' aspect.  It's great fun when you finally figure things out and everything works, but if it makes no practical difference in the end, it's just for fun (or "funstration," as my partner-in-crime likes to call it, given the dearth of accurate and *complete* information.)  I think consumer/hobbyist-level 3D printing is still in the "wild-west' stage, much like phonographs were back in the day, when each manufacturer did their own thing, records played at different speeds for different makers, etc - - took quite some time for any industry-wide standards to become established.  Still, this printer is serving it's purpose; I spent three decades learning to troubleshoot and repair some fairly complex equipment, and hated the thought of losing those hard-won skills in retirement.  For me it's more about tinkering with the technology than actually printing useable and/or fun stuff.

[mvdveer]

8 hours ago, kharrisma said:

For me it's more about tinkering with the technology than actually printing useable and/or fun stuff.

Music to my ears - there are other like me, Im no longer alone. Hey wifey - come read this!