Thermistor, thermocouple or PT100?

As we will have the choice (with daughter boards) of using thermistors, thermocouples, or PT100 probes what do you experts recommend for a largish printer with potentially longish leads (say 1 to 2 metres)? Accuracy and reliability being more important than low cost. Thanks in advance.

... for a project with super high resolution/precision measuring of temperatures down to 0.002 degC I've used PT100 sensors with shielded 4-wire cable to the sensors (two for the 'measuring current of 1mA, two for the voltage drop) - so lengths of up to 5m didn't change the reading.

But could be a bit 'overshooting' for your application

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Yes could be a bit overkill. From memory of the pics that DC42 posted, I think the daughter boards only have 2 terminals per sensor so probably only support 2 wire PT100s but an accuracy of +/- 1 degree would be plenty good enough for the hot end and bed I'd have thought.

Edit. Just looked again and it looks like there are indeed 4 terminals per sensor, not 2.

Edited 1 time(s). Last edit at 06/15/2016 01:36PM by deckingman.

Quote
deckingman
Yes could be a bit overkill. From memory of the pics that DC42 posted, I think the daughter boards only have 2 terminals per sensor so probably only support 2 wire PT100s but an accuracy of +/- 1 degree would be plenty good enough for the hot end and bed I'd have thought.

Edit. Just looked again and it looks like there are indeed 4 terminals per sensor, not 2.

The RTD daughter boards do have 4 terminals. However, as the only RTD sensor we know of that people are actually using is the 2-wire E3D one, we have linked them in pairs on the board. So you will need to cut the two PCB traces between the pairs of solder pads to use a 4-wire sensor. If you need to revert to using a 2-wire sensor, you can put a solder bridge across the gold pads.



Edited 1 time(s). Last edit at 06/15/2016 03:38PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
deckingman
As we will have the choice (with daughter boards) of using thermistors, thermocouples, or PT100 probes what do you experts recommend for a largish printer with potentially longish leads (say 1 to 2 metres)? Accuracy and reliability being more important than low cost. Thanks in advance.

Here are some pros and cons of the various temperature sensors:

Thermistors: Inexpensive, but not suitable for high temperatures. OK at the temperatures used to print PLA, ABS and PETG but not for materials that need 300C or more. Not very accurate, but that isn't too important because the ideal temperature depends on your hot end design and filament, so is to some extent a matter of trial and error anyway.

Thermocouples: Suitable for very high temperatures. Accurate, provided that you run the thermocouple wires all the way back to the interface board. You need to insulate the thermocouple junction from the heater block, which requires sleeving or insulation suitable for high temperatures. You can get encapsulated mineral insulated thermocouples, but they generally come in very long cases. I have talked to a thermocouple manufacturer, and I can get some more suitable ones made if there is enough demand.

PT100 sensors: The 4-wire ones are very accurate. The 2-wire ones are less accurate because the resistance of the connecting wires degrades the accuracy, but still quite accurate. Additional resistance will increase the reading, which is safe, and also not hard to spot at room temperature. E3D sells a 2-wire PT100 sensor, along with a heater block for using it with their E3Dv6 hot end.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

... any resistive sensor will be much more accurate with 4-wire connection - and any 2-wire type can be used in "4-wire" mode too

It's simply driving the sensor with a constant current and measuring the voltage drop as near at the sensor, as possible - so simply solder 4 wires to the sensor(two at each pin).

One problem with the PT100 could be the smaller resistive change per centigrade - you'll need a high accurate ADC to get the best from it.

I've used high-speed 24Bit-ADC's for my sensors ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
VDX
... any resistive sensor will be much more accurate with 4-wire connection - and any 2-wire type can be used in "4-wire" mode too

It's simply driving the sensor with a constant current and measuring the voltage drop as near at the sensor, as possible - so simply solder 4 wires to the sensor(two at each pin).

Good.point.

Quote

One problem with the PT100 could be the smaller resistive change per centigrade - you'll need a high accurate ADC to get the best from it.

I've used high-speed 24Bit-ADC's for my sensors ...

The MAX31865 chips we use on the RTD interface board for the Duet WiFi contain the accurate ADC.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

... the MAX31865 is a good choice for 'normal' temperature measuring and controlling - my conditions were then a bit more crazy -- so we've built a pretty complex TLD-sensor around the ADuC847.

It was not only the accuracy - the device measures thermal conductivity of a gas mixture and then calculates the concentration of this mixture with up to some ppm accuracy -- and the measuring range is per software selectable for 16 different mixtures and the calibration curves can be preset with polynomics of up to 6th order (e.g. for Ar in O2 or Ethylene in Air) ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

hey DM, I have around 4M of wiring to get to the thermister on the hotend of the Delta and my readings are consistent with a 3rd party measurement of nozzle temps. All i ensured was that I kept the loop resistance of the wiring to a minimum..... any run over a couple feet use decent 20/18/16 AWG depending on current draw.

This is for my new (still planned) printer and one of the objectives is to make it as future proof as possible. With that in mind, I've discounted thermistors as I may want to print at elevated temperatures at some point in the future. Thermocouples seem to be potentially problematic with wiring and insulation. Which leaves me with PT100. Initially, the printer will be using a Diamond hot end (although this may change at some point in the future) and the hole for the temperature sensor is 2mm diameter, 10mm deep. I've asked E3d for the dimensions of their probe. I've also found this [uk.rs-online.com] which fits the hole perfectly, has 4 wires and a measuring range of -70 to +500 deg C. which looks to me like it would fit the bill.

... I have some cylindrical PT100 from Heraeus with 2mm diameter but didn't find them at their site.

Here's another 'round' type with 1.8mm diameter - [www.heraeus.com]

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
VDX
... I have some cylindrical PT100 from Heraeus with 2mm diameter but didn't find them at their site.

Here's another 'round' type with 1.8mm diameter - [www.heraeus.com]

Thanks Viktor. It'll probably be easier for me to source them here in the UK though.

Quote
deckingman
This is for my new (still planned) printer and one of the objectives is to make it as future proof as possible. With that in mind, I've discounted thermistors as I may want to print at elevated temperatures at some point in the future. Thermocouples seem to be potentially problematic with wiring and insulation. Which leaves me with PT100. Initially, the printer will be using a Diamond hot end (although this may change at some point in the future) and the hole for the temperature sensor is 2mm diameter, 10mm deep. I've asked E3d for the dimensions of their probe. I've also found this [uk.rs-online.com] which fits the hole perfectly, has 4 wires and a measuring range of -70 to +500 deg C. which looks to me like it would fit the bill.

My PT100 sensor from E3D is 3mm diameter and 15mm long. That RS one looks like a good choice too, if 350C max temperature is high enough for you.

Edited 5 time(s). Last edit at 06/16/2016 08:54AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

... with two PT100 similar to this RS-types I've done some "max temp" measuring - I've clamped two of the sensors back to back, used one as heater, the other as temp-sensor - got heating and measurind up to +700 degC before stopped ... so should be usable reliable with 500 degC

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
dc42

Quote
deckingman
This is for my new (still planned) printer and one of the objectives is to make it as future proof as possible. With that in mind, I've discounted thermistors as I may want to print at elevated temperatures at some point in the future. Thermocouples seem to be potentially problematic with wiring and insulation. Which leaves me with PT100. Initially, the printer will be using a Diamond hot end (although this may change at some point in the future) and the hole for the temperature sensor is 2mm diameter, 10mm deep. I've asked E3d for the dimensions of their probe. I've also found this [uk.rs-online.com] which fits the hole perfectly, has 4 wires and a measuring range of -70 to +500 deg C. which looks to me like it would fit the bill.

My PT100 sensor from E3D is 3mm diameter and 15mm long. That RS one looks like a good choice too, if 350C max temperature is high enough for you.

Yes, got a reply from E3D confirming those dimensions - too big to fit the Diamond unless I drill the hole out bigger. Ref the RS one, spec says 500C max. Where did you get 350 from? Did I miss something?

Quote
VDX
... with two PT100 similar to this RS-types I've done some "max temp" measuring - I've clamped two of the sensors back to back, used one as heater, the other as temp-sensor - got heating and measurind up to +700 degC before stopped ... so should be usable reliable with 500 degC

Thanks again.

Quote
deckingman

Quote
dc42

Quote
deckingman
This is for my new (still planned) printer and one of the objectives is to make it as future proof as possible. With that in mind, I've discounted thermistors as I may want to print at elevated temperatures at some point in the future. Thermocouples seem to be potentially problematic with wiring and insulation. Which leaves me with PT100. Initially, the printer will be using a Diamond hot end (although this may change at some point in the future) and the hole for the temperature sensor is 2mm diameter, 10mm deep. I've asked E3d for the dimensions of their probe. I've also found this [uk.rs-online.com] which fits the hole perfectly, has 4 wires and a measuring range of -70 to +500 deg C. which looks to me like it would fit the bill.

My PT100 sensor from E3D is 3mm diameter and 15mm long. That RS one looks like a good choice too, if 350C max temperature is high enough for you.

Yes, got a reply from E3D confirming those dimensions - too big to fit the Diamond unless I drill the hole out bigger. Ref the RS one, spec says 500C max. Where did you get 350 from? Did I miss something?

The headline and specification on the product page say 500C, the product details section says 250C, and the datasheet says 350C. Take your pick! I've reported a similar error for the version with 1000mm leads to RS.

Edited 1 time(s). Last edit at 06/16/2016 09:12AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42

The headline and specification on the product page say 500C, the product details section says 250C, and the datasheet says 350C. Take your pick! I've reported a similar error for the version with 1000mm leads to RS.

Well b*gger me. I just ordered a couple too. I think the next time 'er in doors does a pyrolytic oven clean, I'll bung one of them inside and see if it survives, and if not complain to RS

Am I OK to just smear these RS PT100s with thermal paste and shove them in the hole? They'll be retained the same way as the thermistor (basically a cable tie). BTW, the hole in the Diamond is 1.9 mm not 2.0 mm as per drawing - out with the drills then

Edited 1 time(s). Last edit at 06/18/2016 05:22AM by deckingman.

Don't use ordinary thermal paste, it isn't intended for those temperatures. Just put them in the holes and secure them in place with cable ties (not to close to the heater blocks of course) or whatever.

Edited 1 time(s). Last edit at 06/18/2016 06:07AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hi DC,

The thermnal compound is the stuff that RepRap.Me supply with their hot end and looks exactly like this that E3d sell [e3d-online.com] so I'm guessing it ought to be suitable.

The instructions for fitting a thermistor are here.

[reprap.org]

Just wanted to check that one could fit at PT 100 the same way. Would you still recommend not using the thermal paste? Will it still give me an accurate temperature reading?

Thanks

... we used a silicone thermal 2K-glue filled with micro-glassspheres, but it was suboptimal - you can try with a normal 2K-slicone and fill it with graphite or metal powder -- but embedd the sensor only up to the blue glass bead, so the conductive filling won't cause a short ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

Quote
deckingman
Hi DC,

The thermnal compound is the stuff that RepRap.Me supply with their hot end and looks exactly like this that E3d sell [e3d-online.com] so I'm guessing it ought to be suitable.

The instructions for fitting a thermistor are here.

[reprap.org]

Just wanted to check that one could fit at PT 100 the same way. Would you still recommend not using the thermal paste? Will it still give me an accurate temperature reading?

Thanks

If the thermal paste is intended for high temperatures - unlike the usual thermal paste for use between heatsinks and semiconductors - then it should be OK.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I had another look on E3D's site and found the assembly instructions for the E3D v6 where they state...

"The thermal compound should be spread evenly across the threads of the heat-break, only on the cold-side of the heat-break that screws into the heat-sink. The compound should not be used on any of the threads on the hot-side of the heat-break.".

I think I trust E3D more than RepRap.Me and as the supplied paste looks exactly the same as that used by E3D, I'll assume that it isn't suitable and ignore what RepRap.Me say. For now I've just inserted the probe in the hole with no paste- it's a pretty snug fit.

Thanks for the help.

Hello,

Where can we order the RTD board ?

If you mean the RTD board for the Duet WiFi, it's one of the add-ons you can select on the pre-order page. If you mean a RTD board for use with a Duet, see the wiki page I created on the subject and search eBay for MAX31865. The board I tested with comes from Hungary.

Edited 1 time(s). Last edit at 06/29/2016 02:42AM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hello,

Can we use these models of PT100 ?

[www.aliexpress.com]

and this one

[www.aliexpress.com]

Any PT100 sensor should work with any RTD interface board, provided that the can is electrically insulated from the sensor element.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].