How to tell if this material is ABS or PLA? (or other?)

Hi people ;-)

Just got 8KG of different colored plastics - some ABS, some PLA, from a supplier who I've just found - I pulled a nice red spool of ABS out of the box, setup the printer to test a 50mm high 10x10mm "tower" to calibrate the temperatures (by starting at around 230 degrees and lowering by 5 degrees every few mm of height)

Now, it started out VERY very 'runny' - liquified... I quickly went to 200, then 195 and so on until I hit 170 (at which point it was still seemingly liquid!) my machine stopped extruding and at first I thought "damn! I've reached the temperature where the extruder physically can't cope" - but it turns out I actually hit a software limit - the machine will not extrude below 170 degrees..

2 questions...

I thought being ABS there is no way it would still be extruding at such a low temperature? Does this seem normal? I'm wondering if I have PLA instead? Googling has turned up "PLA snaps easily and ABS will bend" and it does bend, turns white when bent - seems like ABS?

And

there must be a way to lower the minimum extrusion temperature so that I can extrude at say, 150 or 160? I really can't imagine it needing to go so low. Infact from what I've read, colours seem to have a higher melting point than natural and my natural ABS only copes with around 200 as a minimum when printing this object, so I wouldn't think it would need to be so low but as I said, it still seems liquidy at 170?

Any input is appreciated ;-)

Thanks in advance!

JazzyB

It does very much like to be fan cooled also :-/ (PLA +1)

This could simply be because it's coming out too hot though...

Sounds like PLA and you can lower the minimum extrusion temperature in the config of Marlin or put M302 in the start gcode and it will disable the limit entirely.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Ah that's right - allow cold extrusion ;-) I'll try that first, I'm honestly terrified of flashing the firmware cos it's "melzi" electronics

- Whaaat?

- exactly... If it were ramps or something there's tutorials everywhere but I don't know enough about it to be confident... Apparently it's the same process but it's still making me a bit nervous...

So M302 it is, for now ;-)

Any other tests I can do to check? I've heard pla dissolves in caustic soda and abs does not, but I haven't got any caustic soda...

*runs off to google*

Yeah, no, it's not really a household item...

Scratch that question - I just pulled out a roll of filament marked PLA - it's an entirely different beast... My assumption now is that my thermistor is reading the temperatures incorrectly - every material I use requires fairly low temperatures - so quite possibly the setting of 170 is really a lot higher and just being reported as 170...

But it's obvious to me now the difference between PLA and ABS - the PLA seems to take a lot longer to "set" and therefore (as I've read everywhere) it loves cooling... it's much shinier once extruded compared to ABS and at high temperatures it just turns to pure liquid and oozes like crazy...

AND it's a lot harder, meaning I put PLA into my ABS tuned printer and woooow there's a lot more plastic coming out... needa retune for the PLA - the other material was very likely ABS and just required a lower temp...

If you dip a piece in acetone or nail polish remover it will get soft and gooey really quick if it is ABS and only just kind of dull if it is PLA.

You can also try burning some (note this does give off dangerous gases) and if it smells sweet like syrup then it is PLA, if it smells strongly like plastic it is ABS. But I am sure there are exceptions to this rule because of all the different mixes of plastic but it in general it seems to be true.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Also .... as a quick test, try some rapid rubbing of the filament on your clothing (to create heat by friction) and then putting it under your nose. ABS has a VERY distinctive "plastic" smell.
Alan

Lol - I read somewhere that "burning abs smells really bad" and thought "yeah, but, compared to what? Burning most things smells pretty bad!?" But that clarifies - a sweet syrupy smell... Cheers ;-)
And the acetone did make it go gooey and melted it - it's abs, I'm sure now - does firmware allow for thermistor calibration? Might have to flash eventually...

All firmwares have methods for changing the thermistor. Marlin, Sprinter and Teacup use predefined look up tables that are defined in the config file by selecting the one you have from a list. Repetier allows your to enter the values from the datasheet and it generates the new table. So what you need is to find out what part number thermistor you have and set it correctly and that should help with your temperature readings.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

This was all in a kit from the folks at "reprap pro" - bought from them to try and avoid things like this... I will ask the question of them... The electronics board is a purpose built one (based on arduino) designed by them... So you'd kinda hope they got the thermistor table close to correct!

Spewing its a table... Was thinking it'd be a simple coefficient like e-steps or something...

Cheers again, appreciate you answering all my questions ;-)

I am now well confused as this started out as "Just got 8KG of different colored plastics - some ABS, some PLA, from a supplier who I've just found...."

Are you saying that the supplier was RepRapPro who are based in Bath UK who supply kits for their Huxley and Mendel designs? If so, I believe that they supply PLA from Faberdashery with all of their kits and that the electronics they use is the Melzi, a Sanguinololu derivative which has screw connectors and all stepper drivers on board. The firmware for the RRP Huxley which I have - downloaded from the RRP GITHUB repo - is consistent with Faberdashery PLA.
If you are trying filament from other suppliers, there is absolutely no way of telling what its optimium extrusion temperature will be - it is all down to experimentation based on what the supplier recommends.

Alan

Edited 1 time(s). Last edit at 01/01/2013 03:24PM by Alzibiff.

I'll clarify:

Faberdashery ONLY sell PLA... so the ABS was from the "supplier I've just found" - who is "Weistek", from somewhere in Asia (ordered from my workplace so I don't have the details on my right now as I'm on summer break)

I was under the impression that the Melzi was an arduino derived, purpose built board, as somewhere in the RRP instructions for firmware it states "if you haven't already, download and install the arduino drivers" and this was enough for a n00b such as myself to think "aaah, so it's based on arduino" - but if the Sanguinololu requires the same drivers then yes, i may be mistaken there!

The RRP board has etched on it:

"Design: Joe Mosfet & Adrian" (who I assumed was Adrian Bowyer, founder of reprap pro) so I figured they'd experimented, found a working circuit and rather than shipping boards which are capable of much more than driving a reprap, they narrowed down the electronics to be a purpose built board JUST for driving the reprap pro models.

The discussion started out as "do I actually have ABS? or is this PLA?" because the temperature was SOOO far under what I've heard ABS can be extruded at as a minimum that I was starting to wonder if it was infact ABS or if they could have mixed up and sent me PLA

Then I moved onto a roll which was MARKED PLA - and this was far different to the roll marked ABS (so i feel the original roll was infact ABS)

BUT - all of the material I'm using (including original materials from faberdashery AND another roll of ABS I aquired here in Australia) require very low temperatures - the RRP host software (pronterface) provided on the RRP Git hub (I THINK) is a slightly modified version of skeinforge (modified profiles only, sorry, not the guts of the slicing software itself) as it has profiles such as "PLA HUXLEY" built in...

The pronterface software has 2 "basic temp settings" on the interface, "PLA - 205 Deg" and "ABS - 250 DEG" - obviously to be taken lightly, because as you say, different materials, colours, and providers will all have different melting points...

BUT - with ABS I find anything over 220 is far to hot... and with this new roll (I'll call it "Red ABS") a temperature of 170 is still very hot! it's really liquid!

And the roll of PLA I have - forget it! 170 and it's like water - it's WAAAY too hot - and RRP "say" 205 degrees, a lot of folks on here say 185, even as low as 170-175, but for me, that's obviously not even CLOSE - I think from the results I'm getting I'm looking at more like 130 being a good temp... which makes me ask - are my thermistor calibrations waaay off?

So the discussion transitioned from "what sort of plastic is this? how do I check" - to now being - "ok I think it WAS ABS, but I can't print with it based on my current settings, possibly because my thermistor calibration is wrong - can I change this and what would be involved"

I have a lot of mechanical engineering knowledge, some materials knowledge, am great with COMPUTERS, but not necessarily electronics and firmware upgrades on a board I have no idea how to troubleshoot are a little scary at the moment! so I'm just finding out if that could even be what my problem is and how I could determine that it is/isn't the issue!

thermistor tables are one thing, but if there is an issue with the thermistor or something else in my temp-reading setup, then re-flashing the unit with the default table will be of no assistance - I think what I really need is a way to actually measure the temperature OR, at least for now, I can use the M302 command to allow me to try different temperatures and at least for my setup, I can print using the temperatures I find work! if printing at 50deg works well, well now I can simply do that (even thought it may ACTUALLY be 180deg or so) at least I can print, and if all my PLA likes low temperatures around 50 and my abs likes temperatures around 120 (for instance) then I know, I've got a major calibration issue!

Cheers!
JazzyB

Furthermore, Alzibiff,

Was up late printing something and thought I'd read up on the Melzi (which IS a derivative of an arduino leonardo, as I first thought)

It has the same chip as a Sanguinololu and is compatible with its firmware... but wiki says it's based on the Arduino

[reprap.org]

Edited 1 time(s). Last edit at 01/02/2013 09:42AM by JazzyB.

It's the Melzi with Leonardo that is based on Arduino. The Melzi Ardentissimo (which is the popular version) isn't. It is a Sanguino derivative.

---

[www.hydraraptor.blogspot.com]

Wiki needs an update then, or I've read it wrong? It introduces the "Melzi" saying:

Melzi is designed to be a complete set of RepRap electronics that should be very cheap to mass produce. It is based on the Arduino Leonardo (Francesco Melzi was Leonardo's pupil).

I figured being on reprap.org that the information would be correct!

It contradicts itself by saying it's compatible with sanguino firmware and uses the same chip, but I'm not going to say I know all about these things - if it's actually sanguino then I've been misled! Sorry ;-)

Well I read the same page and understood it as I said above. You can always change if if you think it is unclear.

I think the original intention was the Leonardo version but the Ardentissimo was created for backwards compatibility and the Leonardo version withered and died.

---

[www.hydraraptor.blogspot.com]

Seems like you've got a ton of responses but I'll toss in my 2 cents.

ABS CANNOT melt @ 170 deg. It doesn't even begin to soften yet, and PLA is near liquid at this temp. That's really your tell right there...
I've been able to print with as far up as 230 with PLA, but with a TON of cooling and an open window to the wintry cold. I find that the head temperature reading is highly inaccurate unless you have a controlled environment.

Pla is near odorless, and when ABS starts to melt, your nose will wrinkle!

The plastics also have completely different elastic properties...try applying quick sharp pressure to the filament...does it whiten and stretch? Does it just break cleanly with little stretching? If it stretches a lot, it's ABS. If not...PLA. This might be hard to test with a smaller filament diameter.
Good luck!

I've determined that my temp sensor is way off - pity that I'm mechanical and software minded, but not electrical, which is where I suspect the problem lies :-( I was printing with stuff I know to be ABS today, previously it liked no lower than 200 and now it's around 130-140 before it stops extruding... At 200 its so liquidy, haven't been able to use the machine over the past few days, got on it today and was so disappointed :-( any tips beside "disassembled the wiring to the thermistor"? I'm sick of pulling this thing apart already, I've rebuilt it so many times! Used it in the garage on a 41 deg day the other day and the extruder gear got so hot it warped and the captive nut slipped... Was easy to fix, just heated it and squashed it back to shape... So, been avoiding use on hot days!

Anyways, yeah the material probably is abs... Just my electronics r stuffed it sounds like...

I have never managed to make ABS too "liquidy" it goes from like chewing gum at 150C to paste like > 200C and starts to smoke and burn > 260C. Even at ~ 280C it still extrudes paste like and builds rock hard objects that are brown and stink.

Try dunking it in hot but not boiling water. If it goes rubber like it is PLA. ABS should remain firm.

Some Melzis have 10K resistors on the thermistor inputs instead of 4K7 so need different thermistor tables. They will read low with the wrong table.

---

[www.hydraraptor.blogspot.com]

How mathematically minded are you? :p
If you can get yourself an accurate environment temperature sensor you can take that data and use it. An Infrared thermal gun can get you true heated bed temp..you also can calculate your mm/sec extrusiuon rate based on fall distance of 100mm nozzle above plate, and since plastics cool at a constant rate in a static temperature environment...well..you get the idea.

You could basically map what your actual temperatures are without knowing the real temp of your extruder but it's kind of involved. What you would basically be doing is establishing a ratio between what your machine thinks your plastic is and what you've established it has to be based on empical data (which should include your info on the diff types of plastics & their melting points.) I realize there is some varied information on these forums about the "right temperature" to print at, but as you're finding out because there are so many variables, it's very difficult to get the actual number.

The best advice I could offer for this without as you say drilling a hole and sticking your thermistor where it SHOULD BE...well;
Try visually establishing what is optimal temperature based on printing things, do many small tests, and ignore what things "should be" and "are reading." I imagine it's sort of like baking with a semi broken oven! I have had a TON of issues related to heat, and have found that doing small things to elmininate the variables has helped drastically.
Ie; fixed bed adherance issues, made enclosure to trap heat / eliminate direct drafts etc

Lol... Was thinking of doing that - I have an ir thermometer, buit's actually quite hard to get a true reading from the tiny extrusion! For me anyways...

I'm mathematically minded for sure, would be easy to map temperatures and create my own ratio/formula for temperatures BUT... - it WAS right on day 1... So it's changed, and the thing is, if it's changed, then it may (and likely will) continue to change... If it was never right to begin with I'd try this, but I fear I'll make up a table, trust in this table, and then it will change on me again - I should have pointed out that it has CHANGED to be incorrect, it was once very close to correct, so something (most likely physically) has changed, my guess would be resistance on the thermistor wiring perhaps...

Might get the multimeter onto it tomorrow...

Like the idea tho ;-)

10k thermistor is reading 5.5k (wiki [reprap.org] here says it's meant to read around 10k while connected up) so - something's amiss... pity for me it's NOT a short, or a severed wire... cos I could easily find and fix that... I think the thermistor may have failed? cos if not I've no clue where to start looking...

funny, right now it's saying it's 32deg while at room temp (which is about 32 deg today...)

I've emailed reprap pro asking their opinion, but I'm really hoping it's not a failed thermistor cos these guys haven't designed the hot end to come off the machine too easily...

If it DOES have to come off I'm sooo putting in a RJ-45 plug/socket so I can just 'unplug' the hot end next time... (history tells me there will be a next time)

It should measure 10K at 25C. If your room temp is 32C it will read a lot lower. In the UK it normally reads higher as rooms are rarely 25C.

---

[www.hydraraptor.blogspot.com]

I think it's faulting... Extruded abs at 130 deg today... Emailed RRP for support, might get a new thermistor...

Bummed!

What would you say is normal at 32 (could have even been 40 or so in the hot garage that I'm printing in at the moment) Nophead?

I thought to be at 5.5k it would have to be a very high temperature? Googling actually shows that perhaps you are right - at 40deg it might be half the resistance that it is at 25deg (around about what I was measuring)

But it must be wrong!?

Edited 1 time(s). Last edit at 01/07/2013 08:25AM by JazzyB.

No resistance against temperature is a logarithmic function see [hydraraptor.blogspot.co.uk]. Looking at my graph for a 10K thermistor 5.5K would be about 40C.

Seems like your hot end temperature is out though.

If your garage is that hot you wont be needing a heated chamber! My Mendel runs at about 40C inside an insulated box but that is made from ABS parts. If your ambient is 42C you only have about 13C headroom before your PLA parts soften. Motors easily give more temperature rise than that as you have found.

Edited 1 time(s). Last edit at 01/07/2013 04:32PM by nophead.

---

[www.hydraraptor.blogspot.com]

Is it possible the thermistor has come out of its mounting hole so it is not reading the actual temperature of the heater block.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

Nophead - F-me dead - that blog post of yours went WAY over my head! You're a scientist of sorts? I'm a CNC machinist at a mechanical engineering place, I can build the hell out of any machine, but understanding the stuff on that page is beyond me!

Not that none of it makes ANY sense, I understand you're saying it's APPROXIMATELY a logarithmic curve, but that even the approximation is not 100% correct? (correct?)


And yeah - lol - no turkey bagging for me in the garage for the next month or two :-P it's unbearably hot in there, for me AND my machine - i've told myself, only printing at night from now on!


Sublime - Hmmmm..... :-/ I should look into that - since ambient temp seems to be 'relatively correct' - it's possible that it has slipped out! although the reprap pro thermistor configuration seems very "fixed" in its position, I won't rule it out! I'll check that tonight - it definitely hasn't slipped out the FRONT of the thermistor "hole" but it may be poking out the back and I haven't noticed!

*fingers crossed* If that's it, I will be so stoked... I wiggled all the connections between thermistor / melzi board yesterday with the multimeter attached to see if there was some difference in the resistance, perhaps it was slightly shorting, or slightly severed or something - but it was all rock solid at 5.5k while I was doing that... no loose wires or badly soldered connections I don't think...

Will definitely check that the thermistor is still central - will be such a "duuuh!" moment if it's not :-P

Cheers again guys! Appreciate your efforts ;-)

Quote

You're a scientist of sorts?

More of an engineer. I studied computer engineering at university and spent a career working with embedded systems.

Quote

I understand you're saying it's APPROXIMATELY a logarithmic curve, but that even the approximation is not 100% correct?

Yes it is only an approximation. There is a much better one I use nowadays: [hydraraptor.blogspot.co.uk]

I agree with Sublime, if it reads the correct ambient temperature and your ambient is a long way from the 25C value it implies the thermistor is working and the firmware has the right table, so it looks like a poor thermal connection to the hot end.

---

[www.hydraraptor.blogspot.com]

Thermal paste? Like, the type they use for CPU heatsinks? or does this get too hot for that stuff?

Just thinking ahead to when I get the problem fixed, this may make it even more accurate :-)

I'm positively ITCHING to get home and see that this thermistor is poking out the back of my heater block :-) please please please! lol...

Shoot...

That wasn't it...

And at ambient temperature, the bed and the heater block thermistor's both return (around) the same temperature... it's only when heated that the thermistor reads incorrectly, but it's definitely incorrect at higher temperatures, just turned the machine on and pushed it to 200 before, smelled burning abs... turned it down to 170 and tried to print 10x10x50 high tower, came out looking like a fat grub... just too hot... (and my esteps r a bit off now cos I've changed materials without recalibrating)

so... bad thermal connection, but not because it has 'popped out' of the heater block, it is a rep rap pro hot end (if there is another more correct name for this PLEASE let me know because I read about all these hot ends and it seems like mine is very "unique"... )

Looks like this: [reprap.org] (may need to scroll down a bit)

I should pull it apart... I know... I guess I will - no other choice!

_____________________________________________________________________________________________

So one more separate question for you guys since you're listening ;-)

Anyone had good/bad/any experience with using a RJ-45 connector to wire up the ribbon cable to the hot-end? I want to be able to detach it easily and if I'm going to go ahead and take the wiring off now, I might as well solder on some RJ-45 plugs/sockets while I'm at it - or will this cause some other sort of headache? I'm assuming if that little ribbon cable can take the wattage involved that the RJ-45 will be able to handle it too?

Seems like a good idea to me, otherwise the hot-end cannot leave like a 300mm radius from the machine without REALLY un-wiring the whole thing, which is a big pain!