LaserCut Mendel by Tech Zone Remix

Hey guys, got my TZ laser cut mendel all assembled now electronics and all working with a few errors that im trying to hammer out. Meanwhile Im waiting for some PLA to be shipped so i can start the calibration process. My question is for those with the TZ gen3 electronics Ive been getting an error with:

20:46:57 Send >>Received: Serial Error:Recieved a LineNr code without a Checksum code or Checksum without LineNr 
20:46:57 Send >>Received: Resend:21 with parameter 21
...
20:47:02 Send >>Received: Serial Error: LineNr is not the last lineNr+1 
20:47:02 Send >>Received: Resend:33 with parameter 33

does anyone know whats casuing the error? Ive check google and some have fixed this error with an older version of repsnapper which runs on linux only. While that would be great im using a windows machine to run the repsnapper. Ive checked all the wires cables and all seem good so im a little stuck.

Let me know what you guys think, thanks in advance!

EDIT: i forgot to mention even though im getting these errors repsnapper is still able to start the print, i have no idea what the outcome would look like but all the axis work and move like they should when i test it with a 10mm square, extruder motor works, heater works.

Edited 4 time(s). Last edit at 05/25/2011 12:14AM by ECSuyu.

I'm running linux myself so I'm not a lot of help to you. I know I saw that error before but I couldn't reproduce it to figure out exactly what was causing it. Are you running the repsnapper off this page?

chase82 Wrote:
-------------------------------------------------------
> I'm running linux myself so I'm not a lot of help
> to you. I know I saw that error before but I
> couldn't reproduce it to figure out exactly what
> was causing it. Are you running the repsnapper off
> this page?

ya im using the Kulitorum version for windows from that
page. All the "dry runs" I've been doing until now had been
ok until yesterday which threw alot of lineNr errors for
checksums. could it be that i edited the gcode to manually
control the extruder controller PWM which caused a checksum
mismatch?

I haven't got far enough into this yet to know for sure but that doesn't sound like it should effect serial coms checksums. If it was working fine and you didn't change firmware or software then I'd be double checking all my cables. Maybe somebody with a little more experience with the software could shed some light on it.

USB cable too long, too cheap? Wrap the entire length of the USB cable in tin foil. Report back if the problem persists. If I recal, the USB connection on the techzone machines is carried on a ribbon. THis is not ideal.

Its a USB to TTL then ribbon so it'd only be 19200 baud. But still the ribbon doesn't really fit well in the connector provided. I found it to be the biggest culprit in buggy electronics. Even when my extruder acts up half the time it's that stupid connector.

Not having the proper tool to assemble the cables that TechZone provided, I simply wire-wrapped usb-serial converter on mine. It seems to work reasonably well so far.

Edit: even with the checksum/lineNR error i can still print, i realized that once the print starts i dont get any communication between the temperature reading and the repsnapper software. The temp will show once again after the print is finished.

Got the filament and got some calibration prints done,Im not sure which is causing the issues, it looks like either extruding too much plastic or the Z axis doesnt move up enough when it steps up between layersim trying to figure out where to adjust the Z axis step and where to find the gen3 firmware. The X and Y axis prints spot on with maybe +/-0.5mm if not less. Any help would be appreciated. Ive added 2 pictures maybe people with more printing experience can tell me which one they think is causing the extruder tip to run into the layer it had just laid down

10mm cube


40mm flat (never let it finish)


Edited 1 time(s). Last edit at 06/03/2011 03:09AM by ECSuyu.

@ECSuyu - It's better to disable temperature updates while printing.

Also, you need to give the previous layers get a chance to cool down and harden before laying the next layer down. If you're printing a small part like a 10 mm cube, you can either slow down your feed and flow rates or print multiple parts together. I've found that turning on the cool feature and selecting slow down works fine. PLA is worst than ABS. Is this printed with PLA? What feed and flow rates?

I was getting ugliness like that when I first started trying and just ended up going straight to skeinforge to generate the gcode to print in repsnapper. What do you have your extrusion width and layer thickness set to?

The temperature won't update while printing. That's typical. If you keep an eye on your output LED for your heater though you can see it blinking on and off. Once you send the heat on command the electronics regulates the temp from there not the software.

@brnrd - thanks for the reply, I still need to figure out how to use SF40 as the files i was using to print was gotten from the public file section of the 3dtoolchain website provided by techzone. Which from my understanding has SF running in the background to do the conversions of STL. I still have to figure out how to convert solidworks STLs to proper gcodes.

do you guys mind with me what you guys used to calibrate? I think it might be the 3dtoolchain that is messing things up as i just took a ruler to the Z axis and when i moved it 10mm it moves 10mm so im starting to think it might be what you guys have said. the flowrate is the issue.

edit: I also realized when i send a reverse command via repsnapper it gives me a M81 error. Im guessing thats repsnapper isnt good at sending reverse codes. Perhaps this is why im getting some blobs forming when the z axis moves up.

Edited 1 time(s). Last edit at 06/03/2011 11:34PM by ECSuyu.

It seems I've reached the point in my build that many others have, where everything but the extruder is working. The extruder motor whines, and locks into position, but doesn't rotate.

Using repsnapper to monitor the hot-end temperature doesn't seem to work - it's always at 0°c. However, watching the MOSFET LED, it appears to be keeping the hot-end from overheating - things haven't burst into flames yet, at any rate.

As far as I can tell, it's all wired correctly -

mainboard SDA -> extruder D10
mainboard SCL -> extruder D9
mainboard RS485-A -> extruder RS485-A
mainboard RS485-B -> extruder RS485-B
mainboard ground (from the JTAG connector) -> extruder ground (on the serial connector).

Anyone have any suggestions for things to try?

I'd like to update the firmware in both the mainboard and extruder controller - both to make sure I'm running the latest, and to give myself some debugging options, but it seems that TechZone's branch of the extruder firmware never made it back into the "official" source.

Is there any way we can get TechZone's firmware changes merged into the main branch?

Edited 1 time(s). Last edit at 06/04/2011 12:48AM by raldrich.

raldrich - if you go into the communications tab instead of the interactive control tab, you should see the response to M105 commands from the Mendel. You should see the temperature there. Does it report 0 or the expected value?

brnrd - it's reporting 0 in the comms log - although it did seem to report the correct temperature for a short time.

Do you think when the driver led is blinking its getting actual values and not other times? Does it seem like it's running 100% of the time when you receive 0 as a response or does the led blink fairly regularly?

As long as your nichrome wire stays at 6.8 ohms you shouldn't see temps much hire then 260 to 280. At least from my experience. So you shouldn't burst into flames.

The MOSFET LED blinks pretty regularly, and spends most of its time off, so the temperature control part of it seems to be working - it's just it's method of reporting the temperature that seems to have broken down - although, since both are carried on the same wire, I'm unsure how one could break without the other going with it.

Happily though, I've gotten the extruder motor to run - it chatters like a 5 year old, thanks to the hack that's running it, but it does spin.

Now if I can just get TechZone to send me a properly machined hobbed bolt, I'll be ready to try printing something - is the bolt they sent me, while shows what it's supposed to look like - the problem with the one they sent me is that its diameter is too large where the hobbing is, and it forces the filament so far out of place that it binds up completely.

chase82 Wrote:
-------------------------------------------------------
>
> As long as your nichrome wire stays at 6.8 ohms
> you shouldn't see temps much hire then 260 to 280.
> At least from my experience. So you shouldn't
> burst into flames.

In my experience, around a threaded nozzle without a heating block or a small heating block, this resistance of nichrome wire will easily exceed 280C if given 12V continuously. My old tips used to smoke when this happened.

BTW, those using TZ themocouple AD should make sure that the bug that improperly uses a byte variable to read the temperature if fixed in their firmware. If not, it reads 256 as 0, 257 as 1, etc and the power applied continuously on the tip. The result is an overheated or smouldering tip.

raldrich - if repsnapper is telling you that the temperature is 0 in the communication log, then the extruder controller is probably reading 0 for some reason. I don't know how your tip can get to the right temperature in this case. Have you tried pushing the filament through the heated tip manually? You should be able to extrude this way. If not, then you're tip is probably not hot enough.

brnrd - I really don't know how it is that it seems to be working. The temperature reported by repsnapper is 0. When set to 200, the tip gets hot enough that I can extrude ABS (by hand) with a fair amount of effort. When set to 220, I can extrude with considerably less effort.

The MOSFET LED comes on solid for a time when the heater is turned on, blinks (mostly on) for awhile and eventually settles to blinking (mostly off).

It all appears to work, except that the temperature shown in comms is zero. I suspect a firmware issue.

ECSuyu:
I have had very good results using netFabb to fix Solidworks stl files
[www.netfabb.com]
Tinker around with the repair settings you will get good stl files.
As to your z-Axis blobs I had problems with this when my z-axis was moving to slowly. Maybe this is your problem?? I'm not sure why you are getting error messages for reversing the extruder however...
My z-Axis is set to 130 in skeinforge settings.

What Temperature's are you guys using for printing the Techzone PLA?? I'm running at a skeinforge setting of 245 which works pretty well but I think it is a reason for my excessive warping even with a heated bed...

BTW I'll bookmark this forum and check it regularly, I got the techzone machine working pretty well now so hopefully I can help some others out. Building this machine is difficult technically and due to the fact that necessary information is scattered all over the net .

Edited 1 time(s). Last edit at 06/05/2011 10:06PM by tcool1008.

raldrich Wrote:
-------------------------------------------------------
> brnrd - I really don't know how it is that it
> seems to be working. The temperature reported by
> repsnapper is 0. When set to 200, the tip gets
> hot enough that I can extrude ABS (by hand) with a
> fair amount of effort. When set to 220, I can
> extrude with considerably less effort.
>
> The MOSFET LED comes on solid for a time when the
> heater is turned on, blinks (mostly on) for awhile
> and eventually settles to blinking (mostly off).
>
> It all appears to work, except that the
> temperature shown in comms is zero. I suspect a
> firmware issue.

Can you post what the communication log shows when it returns the temperature? I agree that it's regulating the temperature of the tip properly. My guess is that you're getting RS485 communication errors between the MB and EC everytime repsnapper tries to read the temperature with the M105 command. This is the problem I had with the TZ thermocouple AD. It took too long to read the temperature that it doesn't leave enough time for RS485 io to happen. I have switched my electronics back to thermistor partly because of this.

brnrd Wrote:
-------------------------------------------------------

> Can you post what the communication log shows when
> it returns the temperature? I agree that it's
> regulating the temperature of the tip properly. My
> guess is that you're getting RS485 communication
> errors between the MB and EC everytime repsnapper
> tries to read the temperature with the M105
> command. This is the problem I had with the TZ
> thermocouple AD. It took too long to read the
> temperature that it doesn't leave enough time for
> RS485 io to happen. I have switched my electronics
> back to thermistor partly because of this.

22:50:56 Send >>Sending:M105
22:50:56 Send >>Received:T:0 with parameter 0
22:50:56 Send >>Receivedk T:0 B:0 with parameter T:0 B:0**************************************

I've no idea whether that indicates a comm error or not.

I've done some more experimenting - built a sketch to test the OneWire communications directly with, and found that the chip seems to be working.

@raldrich - Your motherboard firmware might not be compatible with the extruder firmware. It doesn't sound like an RS485 issue if you're able to set the temperature. What versions are you using for the MB and EC?

@brnrd

brnrd Wrote:
-------------------------------------------------------
> @raldrich - Your motherboard firmware might not be
> compatible with the extruder firmware. It doesn't
> sound like an RS485 issue if you're able to set
> the temperature. What versions are you using for
> the MB and EC?

I honestly don't know - the firmware that I got it mostly working with was burned in by techzone, and they don't seem to want to share their sources in any structured way (i.e. subversion or github).

I'm currently working on a redesign of the temperature control portion of the extruder firmware while I wait on parts. So far, I've made an abstract "Thermometer" class which will be overriden separately for each type of thermometer (thermocouple or thermistor), and which handles interpolation of the temperatures. I've also implemented a DS2760_Thermocouple class to manage the techzone thermocouple.

Once I've gotten it all working well enough to share, I'll set up a github for it, and start a thread in the experimental firmware forum.

I'm having a problem with the Extruder Controller board. I am using a thermistor (100kohm) for temperature sensing, and cant get it to note different temperatures. With all settings on the thermistor set according to specs, I get a constant 163 C, regardless of whether the thermistor is plugged, shorted, anything.

Currently running Makerbot EC firmware 2.6 for steppers w/o relay board on the EC, and 2.4 on the motherboard. When I use reprap firmware I can connect to RepG using 19200 Baud settings, but nothing is controllable.

Ideas?

The thermistor has been tested with a known good makerbot and works correctly.

raldrich Wrote:

> I honestly don't know - the firmware that I got it
> mostly working with was burned in by techzone, and
> they don't seem to want to share their sources in
> any structured way (i.e. subversion or github).
>
> I'm currently working on a redesign of the
> temperature control portion of the extruder
> firmware while I wait on parts. So far, I've made
> an abstract "Thermometer" class which will be
> overriden separately for each type of thermometer
> (thermocouple or thermistor), and which handles
> interpolation of the temperatures. I've also
> implemented a DS2760_Thermocouple class to manage
> the techzone thermocouple.
>
> Once I've gotten it all working well enough to
> share, I'll set up a github for it, and start a
> thread in the experimental firmware forum.

Try the extruder firmware that I modified and posted in this thread: [forums.reprap.org]. I used the 20100719 firmware and added the AD calls from Techzone. I also fixed the their bug. It also supports a heated bed with a thermistor. They told me that their firmware was based on an earlier version.

brnrd Wrote:
-------------------------------------------------------

>
> Try the extruder firmware that I modified and
> posted in this thread:
> [forums.reprap.org]#
> msg-77252. I used the 20100719 firmware and added
> the AD calls from Techzone. I also fixed the their
> bug. It also supports a heated bed with a
> thermistor. They told me that their firmware was
> based on an earlier version.

From what I could tell, both versions have a lot of debugging code in them - it appears that it retries communication with the OneWire library several times, and then throws away whatever result it had gotten, and reads it all one last time. Both sources looked like they were spending an unnecessary amount of time reading the thermocouple.

Since I'm probably going to be waiting on parts for most of the week (I let the magic smoke out of a motor driver ), re-writing the extruder firmware gives me something to do

Quote
raldrich
From what I could tell, both versions have a lot of debugging code in them - it appears that it retries communication with the OneWire library several times, and then throws away whatever result it had gotten, and reads it all one last time. Both sources looked like they were spending an unnecessary amount of time reading the thermocouple.

It reads the chip at least twice: once to measure reference junction and the other to read the thermocouple voltage. Good luck. As I wrote earlier, I gave up on the Thermocouple AD because it took too long to read the temperature through the serial connection between the AD board and the EC. I hope you can speed it up.

raldrich Wrote:
-------------------------------------------------------
> brnrd Wrote:
> --------------------------------------------------
> -----
>
> > Can you post what the communication log shows
> when
> > it returns the temperature? I agree that it's
> > regulating the temperature of the tip properly.
> My
> > guess is that you're getting RS485
> communication
> > errors between the MB and EC everytime
> repsnapper
> > tries to read the temperature with the M105
> > command. This is the problem I had with the TZ
> > thermocouple AD. It took too long to read the
> > temperature that it doesn't leave enough time
> for
> > RS485 io to happen. I have switched my
> electronics
> > back to thermistor partly because of this.
>
> 22:50:56 Send >>Sending:M105
> 22:50:56 Send >>Received:T:0 with parameter 0
> 22:50:56 Send >>Receivedk T:0 B:0 with parameter
> T:0 B:0**************************************
>
> I've no idea whether that indicates a comm error
> or not.
>
> I've done some more experimenting - built a sketch
> to test the OneWire communications directly with,
> and found that the chip seems to be working.

@raldrich - What i found with the TZ serial cable is it is sometimes pretty finicky. I have my machine setup and though it works. the temperature feedback will not always show. When i set a M105 sometimes i will get a T0 however for me if i send it again a few seconds later it'll show the proper reading. Also make sure all the wires are in good order by taking a multimeter to the 2 ends of each wire on the serial cable. When i was getting a constant reading of 0/2000 my solid blue wire stopped working and now that its fixed i have communication again for temperature update.
I wouldnt trust the heated tip without the extruder controller giving feedback to the host software. Also make sure you have the update temperature in whichever software you are using turned on.

ECSuyu Wrote:
-------------------------------------------------------

> @raldrich - What i found with the TZ serial cable
> is it is sometimes pretty finicky. I have my
> machine setup and though it works. the temperature
> feedback will not always show. When i set a M105
> sometimes i will get a T0 however for me if i send
> it again a few seconds later it'll show the proper
> reading. Also make sure all the wires are in good
> order by taking a multimeter to the 2 ends of each
> wire on the serial cable. When i was getting a
> constant reading of 0/2000 my solid blue wire
> stopped working and now that its fixed i have
> communication again for temperature update.
> I wouldnt trust the heated tip without the
> extruder controller giving feedback to the host
> software. Also make sure you have the update
> temperature in whichever software you are using
> turned on.

I'm certain that the problem isn't with communications between the extruder and the controller - all of the temperature control is done by the controller, and only works if the temperature can be read using the OneWire library. If OneWire isn't working, your heater won't work either.

That said, the extruder heats quickly enough that even if you only were able to read the temperature 25% of the time, it probably would still give the appearance of working - but my rig always shows 0 for the temperature, so if OneWire wasn't working, I would expect full power at the heater.

I'm about ready to take my temperature control rewrite out of its test harness, and put it into the 20110509 firmware - I'll let you all know how it goes.