Gen 1 electronics problem: low extruder heating + firmware build fails

Hi,

I have almost finished my RepRap but the extruder doesn't heat up (you can still touch it) and the thermistor gives a value of 67' Celsius (which seems about right). Holding the soldering iron against the extruder head heats it up and it indeed registers a higher temperature as it should. So it's not the 10K thermistor / Analog2digital measurement problem (yet?).

The extruder didn't just heat through (software) PWM, because it's off for a long while and then pulses for a short while, repeats, and so on, and so on. This long period's without heating seemed to me a bug that might have been fixed. My next step was to compile the firmware from SVN since this could very well be newer.

If anyone has a compiled hex of extruder 0 and 1 (with the above issue fixed), that would also solve my problem, though I prefer to be able to control the source code of the controllers.

Instructions were very clear and worked like a charm (I repeat them here):

=============
1) Prerequisites

To build the Reprap firmware, you will need a recent version of sdcc,
and a recent (late 2007 or newer) version of gputils. Do not just use
whatever your Linux distribution provides, or one you find somewhere
on the Internet... please follow the instructions below.

1.1) C compiler and development tools

# Obtain gcc, binutils, subversion, bison and flex
sudo apt-get install build-essential subversion bison flex

1.2) sdcc

# Obtain the current sdcc sourcs from subversion and cd into them
svn co [sdcc.svn.sourceforge.net] sdcc
cd sdcc

# Configure it for pic14 target only
./configure --disable-mcs51-port --disable-gbz80-port \
--disable-z80-port --disable-avr-port \
--disable-ds390-port --disable-ds400-port \
--disable-pic16-port --disable-xa51-port \
--disable-hc08-port --disable-ucsim \
--disable-sdcdb

# Make and install it (into /usr/local)
make && sudo make install

# Verify it is in your $PATH
which sdcc # Should output "/usr/local/bin/sdcc"
sdcc -v # Should output "SDCC : pic14 2.7.4 #4988 (Jan 1 2008) (UNIX)"
cd ..

1.3) gputils

# Download the source tarball of gputils 0.13.5 from SourceForge
wget [downloads.sourceforge.net]

# Unpack it and cd inside it
tar zxf gputils-0.13.5.tar.gz
cd gputils-0.13.5

# Configure, make and install it (into /usr/local)
./configure && make && sudo make install

# Verify it is in your $PATH
which gpasm # Should output "/usr/local/bin/gpasm"
gpasm -v # Should output "gpasm-0.13.5 beta"


2) Configuring and building the Reprap Firmware

With all the development tools in place, a normal autoconf approach
can be used to build the firmware.

# Configure and build the firmware
./configure && make

=============

Except that at the last step, configure gave the following problem:
[skipped most of the output]
checking for an ANSI C-conforming const... yes
checking whether time.h and sys/time.h may both be included... yes
checking for working volatile... yes
checking sys/select.h usability... yes
checking sys/select.h presence... yes
checking for sys/select.h... yes
checking sys/socket.h usability... yes
checking sys/socket.h presence... yes
checking for sys/socket.h... yes
checking types of arguments for select... int,fd_set *,struct timeval *
checking for select... yes
checking for strtol... yes
configure: creating ./config.status
config.status: creating Makefile
config.status: creating config.h
config.status: config.h is unchanged
config.status: executing depfiles commands
checking whether sdcc and cc understand -c and -o together... no

I'm using Kubuntu Hardy Heron (8.10).

If the heating problem is solved, I think I can start printing!! So any help is greatly appreciated!

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

Erik,
Have you run the heater profiler? [reprap.org]

---

[www.hydraraptor.blogspot.com]

Yes Chris, I ran the heater profiler, but it ups the power level (at least the 'desired power level') without becoming more hot than the max 67 degrees Celsius that I've seen. With the power levels of 15 and 30 (i think) it does stabilize at lower temperatures (40 and 50 or so). At the highest power levels, I can see the heater led being off most of the time and than blinking for a few seconds. So it really heats only part of the time even with high power settings.

Anyone's got a .HEX file of extruder0? (and extruder1 if you're at it)

Thanks!

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

Okay, now it seems like it is a feature rather than a problem in the firmware. If I disconnect the heating coil from the PCB, the led stays on the whole time. Otherwise it's off most of the time, and there's a single blink somewhere in this cycle. At this moment, there's communication as well (tx/rx leds blink).

Of course after reading the source code there's no such 'feature'... I expect that the thermistor lead touches that of the nichrome wire. I'll check to see if this is true, if so, I CAN START PRINTING!!!

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

Nope, I checked. It's not that (the thermistor wires and nichrome wires are not shorted). What is it that I do not understand?

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

If I recall correctly, the Gen 1 firmware has an auto-cutoff based on the extrusion temperature. It's supposed to prevent overheating but I've seen it do some odd things before. What's the value of your Extruder0_ExtrusionTemp setting?

Hi Steve,

The Extrusion0_ExtrusionTemp(C) was left at the default of 130. I know that it should turn of when the temperature is reached (and that the PWM's duty cycle is proportional to the temperature difference (it is, right?), but heating less when you need more seems a little strange.

Any ideas?

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

I've been debugging the electronics of the extruder for another two days now. In the mean time I've managed to make the situation worse (converted a stepper PCB extruder PCB and back) and the first didn't work anymore. Now, after replacing the PIC I seem to almost be back to the problem that I can't heat well.

Last time it was the heater that didn't turn on, now the temperature that is displayed is -273 degrees Celsius. I have achieved the absolute lowest temperature possible with my heater. Very dubious indeed. Anyone has any hints? My thermistor in the setup gives me about 10K at the end of the wires where it goes to the PCB. The PCB shows no visible shorts or damages. Changing the temperature of the head also does nothing (soldering iron or putting the heater connector directly on the powercomms 12 V ports), it changes the resistance (checked with multimeter) but doesn't change anything else. I've checked again that the thermistor is short with anything else, but there's nothing anymore that it could be connected to... The ceramic disc capacitor of 10 nF was replaced which didn't help. I'm out of ideas!!

Days ago I was so close to actually printing......

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

A reading of -273 could indicate a really small thermistor resistance (e.g. a short or near-short) or it could be that c3 is open or too small.

What value of c3 are you using?

My extruder would not go above 57' C but i changed c3 from 10000pF to 100000pF today and it works fine now.

see here:
[forums.reprap.org]

With the 100nF capacitor it worked better. Now I still have the problem that it more or less stops heating in cycles when the power setting is very high.

Now, the hot nichrome wire just disconnected because the solder had gone fluid. I had a heatsink on top at first to prevent the heat going up too much, but

Perhaps I'm printing with a bit of a difficult material (HDPE). I need pretty high temperatures, otherwise either the PTFE rod is pushed out of the clamp or the motor hardly rotates or the screw doesn't grab the fillament well enough to push it through.

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]

I had a lot of difficulty with HDPE too. The only way I found to get around that high-temperature cut-off problem was to switch to a 100k thermistor.

I also had problems keeping the nichrome soldered. Ended up using screw terminals instead.

For the PTFE barrel slippage, I used wood screws through the top end and a hose clamp on the bottom end. Not sure who came up with that but I know a lot of people have done the same.

I'm kind of stuck at the moment. I've gotten to the point where I think HDPE is probably impractical but I haven't gotten around to buying a different kind of filament. I'm leaning toward ABS but I've been too busy/distracted/lazy to actually get some.

Yes you need 100K thermistor for HDPE, the 10K one has too low a resistance at HDPE temperatures for the PIC output to be able to provide the current to charge the cap.

You need to be somewhere in the range 200 - 240C for HDPE.

I used high temperature solder which melts at 300C (I think) to connect the nichrome. Solder does not stick to nichrome so you need a good mechanical connection as well.

HDPE is right on the edge with the current extruder. The JBWeld heater insulation disintegrates eventually and the PTFE barrel loses its strength and tends to let go of the barrel, even with a pipe clip.

ABS needs similar temperatures to HDPE.

---

[www.hydraraptor.blogspot.com]

> needs a good mechanical connection as well
Yes, I had that on the nichrome (used a clipped lead of a resistor), but the wire that connected that let go because it was only soldered. I don't have 300C solder but I've rearranged it and now I think it will hold.

Nophead, I really love all your experimentation with all these materials. I knew in advance that HDPE would be hard to get right, but it was the only thing that was available at a good price. I'll definitely try the 100K resistor. Reading all your posts make it feel like I'm experienced myself, while I've only tried HDPE yet.

Right now I'm really thinking that that's why the extruder stops heating in cycles. I have a 100K thermistor now that I just boiled and put in ice water to measure b0 and b1 of y = b0 + b1x. It's just a little big to fit in a small pit in the extruder nozzle, so I hope that it will respond fast enough to get stable temperatures.

---

Regards,

Erik de Bruijn
[Ultimaker.com] - [blog.erikdebruijn.nl]