Why? do we use thermistors the size of pixi-dust –> an alternative

I know the little ones have a low thermal mass, but they are such a pain to use, crimp, and place. I have been using the below Axial version for a few months now and can only recommend everyone to give it a try the next time you make a hot-end. So much easier to use and much more rugged.



One in my high-powered hot-end here -



You can buy them from Farnell - [uk.farnell.com]

I’m still using the pixi-dust sized ones for the heated bed, but only until I have used them all up.

Rich.

---

[richrap.blogspot.com]

Hi Rich,

How are these thermistors holding up for you?

The leads on my 'pixi-dust' thermistor broke today, so I am probably going to try these axial thermistors next. Would be interested if you have any feedback after using them for a couple of months.

They work really well, I have 5 fitted to hot-ends, and one is on a heated bed, I have never had a single problem with them.

I still use the Makerbot 100K thermistor as it's easier to fit to a J-head nozzle, it's slightly bigger and has thicker leads than many of the other ones.

---

[richrap.blogspot.com]

Agreed about the pixie dust sized thermistors. Having snapped a lead off one I bought, I switched to an axial :

RS axial thermistor

from RS in the UK, min order a pack of 5.

Thermistor is referenced here:
mini extruder info inc thermistor tables

inc. some look up tables (although my printer is now getting results using the default in sprinter without changing the table). Same type on the edge of my HBP.

Mike

Edited 3 time(s). Last edit at 04/02/2012 03:04PM by 3D2B.

Hi Rich. Hi Mike. Thanks for the feedback.

Good to know these alternatives are working out well. (I was skeptical of using the mini thermistors from the outset.)

The main difference is the axial one is +/-10% whereas the beads are +/-5%. 10% at 240C is +/-24C of course and that is like chalk and cheese as far as extruding plastic is concerned.

I always calibrate every thermistor I use against a thermocouple, so it makes no difference to me. It seems most people just download a table from the web. That is fine if you develop your own temperature settings by trial and error (which is pretty much what you have to do with new plastic anyway), but no good for exchanging setting with other people, or even between two identical machines.

---

[www.hydraraptor.blogspot.com]

Even if the tolerance is +/-10% the error is still going to be relatively consistent over the range the PID is trying to hit isn't it? It's not going to shift +/-24C over the course of a few degrees.

No it will be a consistent percentage offset for a particular thermistor but it could be worst case 24C different from what you think it is. That (plus different thermistor positions) is why people on these forums keep stating wildly different temperatures for extruding PLA, for example.

---

[www.hydraraptor.blogspot.com]

The first thing you should do with any new assembled hot end is calibrate it's temperature inside the nozzle. I use a little digital meter with a Ktype thermocouple inside the hot-end, then set the host software to a 150 deg C temperature and record the readings in 5deg steps, they are usually 3-5 degrees out of what you set it as long as you have a table that's supposed to match your thermistor.

Then you can either tune the table or just remember that 192 set is actually 188.

With a hot-end you only need an accurate table between 160 and 250 so don't worry too much about the low-end being out.

It's a little harder with a heated bed, especially if you are using glass or a mirror surface, reading the temperature can get tricky if using IR meters. And again calibrate your bed table between 40 and 150. And I would recommend placing the thermistor for the heated bed on the edge or corner, not in the middle.

---

[richrap.blogspot.com]

Even if a setup can quote an accurate temperature value for its melting chamber it still needs to be calibrated to find the best performance for any given filament. I don't think having an absolute temperature reading is all that important in the grand scheme of things as long as you can relatively adjust the temperature consistently.

Hi Mike.

3D2B Wrote:
-------------------------------------------------------
> Agreed about the pixie dust sized thermistors.
> Having snapped a lead off one I bought, I switched
> to a radial :
>
> RS radial thermistor
>
> from RS in the UK, min order a pack of 5.

You meant axial (rather than "radial") right? I ask because I am probably going to order these. The component Rich referred to is cheaper per unit sale, but unfortunately it has to be shipped from the US so incurs a 15-20 euro shipping charge (whether you order them from Radionics or Farnell).

The specs are slightly confusing. The radionics website states that the diameter is 11 mm, but it is in fact a D0-35 package => diam. is circa 2 mm, right?

yes, it's 2mm diameter and about a 5mm long body, leads are about 0.8mm

---

[richrap.blogspot.com]

Cool. Thanks Rich. I've gone ahead and ordered from Radionics anyway.

dslc Wrote:
-------------------------------------------------------
> Hi Mike.
>
> 3D2B Wrote:
> --------------------------------------------------

> You meant axial (rather than "radial") right?

Yes sorry my error. Axial is correct, have edited to correct.

>
> The specs are slightly confusing. The radionics
> website states that the diameter is 11 mm, but it
> is in fact a D0-35 package => diam. is circa 2 mm,
> right?


agreed with Rich's measurements for approx diameter & length. The leads 0.4mm diameter on the ones from RS .

Mike

Edited 1 time(s). Last edit at 04/02/2012 03:05PM by 3D2B.

This is the axial thermistor I've been using for the last couple of years and have had no problems. 5% tolerance too.

HONEYWELL S&C / FENWALL - 135-104LAF-J01 - NTC Thermistor
[nz.element14.com]

Cheers,
Len.

It appears that they are available in 1%, 2% 5% and 10%: [uk.mouser.com]

1% would eliminate the need for calibration.

---

[www.hydraraptor.blogspot.com]

Ouch... €20 postage on 5 micro-size thingamabobs worth €13. Ebay to the rescue, I guess (hope)