I have been assembling a RepStrap Seedling with the Generation 1 PIC electronics (with the goal of using that Seedling to then build a Darwin where I'll use Gen 2 and Zach's cool Sanguino). Recently, I have started to build the extruder kit I ordered from Bits from Bytes, but there appears to be a possible incompatibility for temperature measurement. Which (if any) of the following scenarios can work?

1. I have a 100K thermistor like the one from the RRRF store. However, the extruder improvements page at [dev.www.reprap.org] suggests that extruders like the one from BfB need a K-type thermocouple, because they are too hot for thermistors. Can a thermistor still be used for this extruder (embedded in the fire cement), albeit within a limited temperature range?

2. If a K-type thermocouple is required, can it be used with the PIC electronics (Extruder Controller v1.2)?

3. If I have to use a K-type thermocouple and Arduino/Sanguino for the extruder, what is the minimal set of Gen 2 boards I can build and network with my existing Gen 1 boards for my RepStrap Seedling?

I look forward to eventually using Sanguino and Gen 2 for the Darwin, but hope to preserve my Gen 1 PIC board investment and effort for the Seedling. Thanks in advance for any advice you can offer.

I've also got PIC electronics and a BfB extruder, what I've done is embed a 100k thermistor and a K type thermocouple in there. I'm planning on using polymorph and relatively low temps initially. If/when I move to ABS or something similar I'll move over to using Gen 2 stuff and the thermocouple.

I was worried about the curing process, where the barrel gets heated to temperatures way above what the thermistor can measure, the thermistor seems to be ok. I've only checked room temp level readings though, so I may be wrong.

For the PIC to Gen 2 move; I think it may be possible to reuse most of the PIC boards with some extra headers and firmware changes. The hardest aspect seems to be adding current limiting to the stepper motor boards, which I'm doing at the moment to use them with the BfB steppers. It's reminding me why I like digital electronics a lot more than analog ;-}>

Rapid electronics do a 300

---

Ian
[www.bitsfrombytes.com]

... why didn't you use a PT100 as temp-sensor? - I use the 'naked' types with the blue glass-sealing: [de.wikipedia.org]

It gives a nearly linear range from 0

Edited 3 time(s). Last edit at 09/18/2008 04:05AM by Viktor.

For myself, I'd only heard of thermistors and thermocouples.

A bit of reading wikipedia later....

Resistance temperature detectors (RTD) / platinum resistance thermometers (PRTs) seem like a great solution:

- linear resistance change with temp
- No isothermal blocks (not that the current design uses them)
- No precision amps / signal conditioners needed
- Wide temperature range

From what I've found they seem to be more expensive than a thermistor, but cheaper than a thermocouple (especially if you factor in the electronics needed). I found one at Farnel, not the cheapest place, for

peteredworthy Wrote:
-------------------------------------------------------
> ...
> Am I right in
> thinking that a mA scale constant current device
> in series with the RTD would produce a voltage
> across the RTD proportional to it's resistance &
> hence the temperature?

yes, with a constant current the voltage drop over the resistor is temp-proportional and linear too ...

Viktor

PS: i'm measuring with special microfabricated platinum-resistors and parallel PT100's temperatures between 40

Edited 2 time(s). Last edit at 09/18/2008 07:19AM by Viktor.

peteredworthy Wrote:

Quote

I found one at Farnel, not the cheapest place, for

---

[www.hydraraptor.blogspot.com]

Don't bother, I found it here: [uk.farnell.com]

I had to search for PT100 instead of RTD or PTR.

I think they are actually cheaper than the Epcos high temp thermistors I have been buying from RS.

---

[www.hydraraptor.blogspot.com]

They are harder to use than a thermistor as much smaller resistance changes, usual practice is to use a wheatstone bridge and amplifer so it's right back to the same level of complexity as a thermoouple IMHO.

Nophead look at the Rapid offering it's useable to 300

---

Ian
[www.bitsfrombytes.com]