Another Happy Mendel90 Owner

Hi Guys

After extensive research I put an order in to nophead for one of his Mendel90 kits on Friday 18th Oct. It arrived at 9am on Monday 21st Oct! I have spent every evening this week and all day today putting it together. The kit is excellently organised and documented. I find written instructions hard going, but with plenty of illustrations and photographs, I didn't run into any major problems. Everything I needed to know was in the manual (imagine that!). The only thing I found myself doing was googling images of other Mendel90's in the wild, to double/triple check I was going in the right directions. Would be good to have a series of photographs to go along with all the CAD views. I danced around in the manual a bit, to fit various parts of the build into the time I had available, but it all seemed to come together fine. Today was final wiring, setup and testing. Only real screw up I had was the female D connector that didn't properly make contact with all the wired when I clamped it shut. Had to manually solder each wire to the back of the connector. Worked fine after that. I simply could not believe that at the end of a week long build, it powered on without fault, first time! Super impressed.

Sadly I fell short of getting a successful test print out today. My first attempt went wrong somewhere. The hot end seemed to take an unusually long time to heat up and the machine started 'printing' before the filament was extruding. I left it a while to see if it would start, but in the end stopped it about 5mins in. Murphy's law, as soon as I stopped it, the filament started to ooze out. I went back and had another quick go, but this time I could not get the extruder to extrude at all. Bed and Hot End both got up to temp and all setup parameters seem to be correct. In the end it was getting late, so I took off the idler plate and ejected the filament. My hunch is that the filament is not feeding through smoothly enough. Will have another play tomorrow night if I get time.

Just wanted to say a huge thanks to nophead and Mary for offering such a top class service all around!

Edited 1 time(s). Last edit at 10/27/2013 08:01AM by IntelliTom.

Ok, so some progress. I managed to get the machine printing tonight. I decided to follow the instructions for slicing a different STL file from ThingiVerse. After going over all the adjustments of the machine and final checks, I ran the print. It seemed to print well but I got quite a bit of lift off from the bed on one side. So I raised the bed temp to 72c and rand it again. This one worked much better with just a tiny about of lift off. So I went back to the included Android file. This started fine but oddly stopped a few layers in, with an error: {ERROR:0 Extruder Switched Off} So I reset everything changed the extruder temp from 190 to 187 and ran it again. The exact same thing happened at the exact same part of the print.

Could I get some tips on how to trouble shoot this problem?

I don't think changing the bed temperature by 2 degrees or the extruder by 3 degrees will make a noticeable difference. To stick properly the bed needs to be squeaky clean. I think vinegar is the most recommended cleaner or although I use acetone or meths. Isopropyl is known not to work and some glass cleaners that contain wax.

I am note sure what you mean when you say you changed the extruder temperature from 190 to 187 because the supplied android.gcode file has it set to 185C. Did you re-slice it?

It seems like maybe the connections to the extruder are intermittent. In particular the thermistor connection. Is the IDC connector fitted correctly and closed with a vice? Try heating the extruder manually with Pronterface with the monitor printer option ticked. Then try waggling the wires to see if any are intermittent on the graph.

---

[www.hydraraptor.blogspot.com]

Thanks for your reply nophead. I think I've obviously not understood how the software works. I'd assumed the temp settings in Pronterface where what set the temps in the print. I take it this is just for pre-heating and testing?

I have some Acetone here, so will try using that on the glass bed. I have read things like hair spray and blue tape are used? Is this worth trying?

Actually I meant to say above. The only major problem I had was the D connector. Somehow it did not make the connections properly when I clamped it closed. I ended up taking the back off of the male connector and directly soldering each wire to the pins. So I can still disconnect it, but it's kind of ugly. This sounds like it may be the problem. Should I get another male D adapter, or using something else?

As for bed cleaning suggestions I find vinegar (for household purposes) to be absolutely great so far. Also dilluted pva (white woodglue) in water in a 1:10 ratio works wonderful too. With acetone and window cleaner I get mixed results.

Yes settings in Pronterface don't affect the print unless you send them while it is printing, or send them while it is uploading to the SD card. In those cases they get merged into the gcode stream.

I don't use blue tape any more and have never had success with hairspray but other people swear by them. I have only used clean glass for PLA after discovering it works.

Quite a lot of people have trouble with the connector. I don't know what they do wrong as IDC connectors should always work and are much faster and easier than soldering 14 wires.

In theory it should work if you have soldered to the IDC terminals as long as they are good joints and not able to short to each other. The plastic may melt though because they are not designed to handle the heat of soldering.

Normally you should use a connector with solder bucket terminals if you want to solder them. They use higher temperature plastic I think. I usually put heatshrink on each wire so it is very tedious procedure compared to IDC. The connectors are a lot cheaper though.

I can supply a replacement connector but you would also need to cut the end of the cable off so it might be better if I send you a new cable with the connector fitted. Also if you try IDC again we need to work out what you did wrong because it is not a hit and miss procedure. It is used in commercial electronics so it is should always work and does in my personal experience.

---

[www.hydraraptor.blogspot.com]

Thanks

I have had some more time on it this afternoon. I downloaded a few small basic things from ThingiVerse and sliced them in Skeinforge with the default nophead settings. All seemed to print fine but still with some warp/lift off. Even after I had cleaned with acetone. I may need to get the dial gauge out again to check level. However, my last attempt was with blue tape and I got a 'perfect' print. No lift off at all. In fact it was a little too stuck to the bed, but I managed to prize it off gently.

The only other issue I have at the moment is that the prints are 'gappy'. i;e when viewed down the Z axis, there are gaps in the layers. Normally at sharp internal corners or think pieces.

With regards to the D connector, at the moment it doesn't seem to be a problem. My guess is that last night's error was user induced. Whilst I agree that the IDC connector should have worked fine, I think I would rather go with a soldered option as that seems to be slightly easier (for me at least).

Very happy so far though. Going to have a go at printing my own 3D creations later tonight.

Skeinforge will only add infill where there is room for it. If you have very thin sections that you want filling then set the number of extra perimeters to 0 to leave maximum room for infill. It will still only fill sections bigger than three times the filament width so anything with a width between 2 and 3 times the filament width will have a gap. You may be able to reduce the width by using thinner layers.

What is your difficulty with the IDC? It should be just a matter of lining the cable up so it sits in the forks of the contacts at both sides of the connector and then closing it completely with a vice. The fact there are only 14 wires for 15 pins means the alignment needs a bit more care. The only way I can see it not work is if the cable is not in far enough to meet the second row of contacts, is not aligned with the top edge, not straight or the back is not fully closed down.

Edited 1 time(s). Last edit at 10/29/2013 12:46PM by nophead.

---

[www.hydraraptor.blogspot.com]

now i am in progress for building a prusa i3, honestly i also be very interest into this mendel 90 type. i want to have both and to tell the diffrence between them. my question is where can i get the kits for mendel 90 and how much will it cost. FYI, i am in china.

Ok I have been further experimenting and seem to have the in fill problem tamed.

Few quick questions.

When my Z axis moves all the way up or down, I get a high pitched whine from the lead screws (worse on the way down). I tried some light application of 3-in-1 oil, but it doesn't appear to have helped. I also have some dry PTFE lubricant, should I try this, or should I be looking at making an adjustment somewhere?

While experimenting, I often get failed first attempts ( where the filament doesn't stick, mainly). I can't work out the best way to 'Emergency Stop'? I've taken to reaching round and flipping off the PSU off and back on, then re-connecting in PronterFace. Is there an easier way?

When loading GCode onto the micro SD card, I'm currently taking it out of the machine and using the included card reader. Is the 'Upload File' under SD Card in PronterFace, basically the same thing? Seeing as I'm connected through USB to the board, should it not show up as a disk drive in windows?

I would like to get to the point where I can enclose the electronics 'bay' to keep things neater. I am also going to look into the OctoPrint that I have heard people talk about.

IntelliTom Wrote:
-------------------------------------------------------
> Ok I have been further experimenting and seem to
> have the in fill problem tamed.
>
> Few quick questions.
>
> When my Z axis moves all the way up or down, I get
> a high pitched whine from the lead screws (worse
> on the way down). I tried some light application
> of 3-in-1 oil, but it doesn't appear to have
> helped. I also have some dry PTFE lubricant,
> should I try this, or should I be looking at
> making an adjustment somewhere?

It is normal for the Z motors to make a fairly high pitched whine. They are spinning much faster than the other axes and so the step frequency is higher. I have never heard any noise from the screws themselves.

>
> While experimenting, I often get failed first
> attempts ( where the filament doesn't stick,
> mainly). I can't work out the best way to
> 'Emergency Stop'? I've taken to reaching round
> and flipping off the PSU off and back on, then
> re-connecting in PronterFace. Is there an easier
> way?

Press the reset button for an immediate stop. Wait for it to reset and then home.

For a less drastic one press the pause button, turn the heaters off and then home.

>
> When loading GCode onto the micro SD card, I'm
> currently taking it out of the machine and using
> the included card reader. Is the 'Upload File'
> under SD Card in PronterFace, basically the same
> thing? Seeing as I'm connected through USB to the
> board, should it not show up as a disk drive in
> windows?

It has the same effect, i.e. adds a file, but is very slow, especially under windows. The USB port on the Melzi, and any Arduino, is connected to a chip that emulates a serial port so it can't act as mass storage device.

>
> I would like to get to the point where I can
> enclose the electronics 'bay' to keep things
> neater. I am also going to look into the
> OctoPrint that I have heard people talk about.

There is a printed cover design on thingiverse. A slot could be left for the SD card.

---

[www.hydraraptor.blogspot.com]

Fantastic. All makes sense. Thanks.

It's weird. I wanted to run my largest print so far tonight. Nothing major, just a 'wavy bangle'. It took three attempts to get it printing properly. The first 2 times the first layer did not adhere. I changed nothing between each attempt, it just worked third time lucky. Bed is just acetone cleaned glass, stock config with no cardboard insulation.

Is this bed too hot, too cold, or something else? Or indeed, is this just an expected 3D printing experience?

Yeah I think it's a bit part of the experience when you try to use just clean glass. Even though I have very good results after the glass is cleaned with vinegar, sometimes the print does lift for now apparent reason. Maybe little things like fluctuation in room temperature or humidity and things like that come into play as well.

It always sticks for me. The only time I had problems was at a Maker faire when somebody opened the windows at the other side of a substantial room on a fairly cool damp day.

The way to increase adhesion is to have the bed much hotter for the first layer, say 100C but then drop it at the start of the second layer. Not sure if you can do that with Skeinforge without hand editing the gcode.

Vinegar may be better than acetone, I have never tried it. The problem with solvents is they can dissolve things out of the cloth you use and leave a residue. I imagine any residue vinegar might leave would be sticky.

---

[www.hydraraptor.blogspot.com]

I don't like the idea of Vinegar. I have good quality Acetone here and the smell from that seems to dissipate rapidly. I doubt that's true of vinegar.


Anyway. I think I made it mad!

Tried using some black ABS from Faberdashery today. I sliced the part using the ABS setting in Skeinforge. I pre-heated the bed and extruder and made sure it was extruding before I started. However, when I came to print the part, it went through the motions, but did not extrude. After playing around with it for a while I noticed that the bed and extruder where not getting up to the expected ABS temps. In fact it looked like it was using the previous PLA temps. How can I check inside the GCode to confirm this? I checked the skeinforge settings and under the ABS profile, 'Activate Temperature' is checked and the temps are up in 'abs territory'. So what am I doing wrong?

I tried using a couple of other slicers, but could not work out the settings on those. That's another post I think.

So I went back to PLA. However, I had problems here too. It was a new colour (grey), but I could not get bed adhesion here either. It was only after running it a few times, I realised that my 'origin' position is wrong. It's about 1-2mm above the surface of the bed (in the Z axis). I'm not sure how this could have happened, but also not sure how to re-set it? I turned the motors off and would it down, but of course it goes back to the wrong position when printing.

Yesterday was an awesome day for 3D printing. Today, not so much

Seems like you have to recalibrate your z home position in the firmware as described in the manual. I'm too wondering how to set the temperatures for different filaments. I think it is set in the firmware also.

Yep it is set in .skeinforge\alterations\start.gcode.

Thanks. So you have to manually change that for every filament change?

Yes. How could it be automatic?

You can of course make new profiles for each type of plastic and reference different versions of start,gcode.

---

[www.hydraraptor.blogspot.com]

Tom, see this thread:

[forums.reprap.org]

Thanks Nilez. I'll jump in on that thread in a sec

Just one query here on the Z head position. So I went back over the manual instructions. I now remember that when setting it up, the 'factory' position was spot on what my 'feeler' gave me, so no need to make adjustments. Now there is a difference of +1.59mm. So I need to reduce the Z_HOME_POS from 203 to 201.5 (rounded). However, I've now realised that I've lost the 'melzi' setting for arduino and so can't select Sanguilolu in the board menu. Is it online somewhere? I must have deleted it, assuming I already had Arduino IDE on my PC.

It is here: [github.com]

Copy it to the hardware folder of Arduino.

Hard to understand how the Z position could suddenly change by that much. Is the machine on a stable surface?

Edited 1 time(s). Last edit at 11/01/2013 04:10PM by nophead.

---

[www.hydraraptor.blogspot.com]

I'm surprised also. The machine is on a solid table and has not been moved or bumped in any way. I double checked that the nuts on the screws where properly seated and they all seem fine.

Quote
IntelliTom
Tried using some black ABS from Faberdashery today

Does Faberdashery sell ABS now? I thought they were PLA-only for sustainability reasons?

As for your Z_HOME_POS: two areas that I found were susceptible to variance are the x-ends after changing filament (sometimes the force of pulling the old filament out is enough to raise one or both ends of the carriage up slightly) and the screws that hold down the heated bed. This is metal on metal, and I found if the screws aren't torqued up securely enough then the vibration can cause them to loosen. My rear left corner screw actually would itself all the way off before I noticed - I have a beautiful time-lapse of it coming off!

oh no, not Faber, it was 3D Filaprint? I think?

I'm not having much luck at all today. So I updated the arduino IDE with the Melzi settings, made the edite for Z_HOME_POS and clicked upload. I got the following error:

==
C:\3d-arduino\hardware\arduino\cores\arduino\IPAddress.cpp: In member function 'virtual size_t IPAddress::printTo(Print&) const':
C:\3d-arduino\hardware\arduino\cores\arduino\IPAddress.cpp:50: error: void value not ignored as it ought to be
C:\3d-arduino\hardware\arduino\cores\arduino\IPAddress.cpp:50: error: in evaluation of 'operator+=(size_t, void)'
C:\3d-arduino\hardware\arduino\cores\arduino\IPAddress.cpp:51: error: void value not ignored as it ought to be
C:\3d-arduino\hardware\arduino\cores\arduino\IPAddress.cpp:51: error: in evaluation of 'operator+=(size_t, void)'
C:\3d-arduino\hardware\arduino\cores\arduino\IPAddress.cpp:53: error: void value not ignored as it ought to be
C:\3d-arduino\hardware\arduino\cores\arduino\IPAddress.cpp:53: error: in evaluation of 'operator+=(size_t, void)'
==

I've googled it, but can't seem to work out what it's referring to?

Use the files provide on the SD card. Arduino is a nightmare of incompatible versions.

---

[www.hydraraptor.blogspot.com]

I wiped the card

The instruction say to copy it first! It's too big to email.

---

[www.hydraraptor.blogspot.com]

What version is it? I tried 1.01 and 1.02 and got similar errors. Have you altered the actual arduino code?

I realise it was silly to delete the arduino directory, but now I'm up the creek...

It is 1.01. I can't remember what I had to change it was more than a year ago.

I can't believe the number of people that lose the contents of the SD card. It's hundreds of megabytes and takes me about 15 minutes to make each one.

---

[www.hydraraptor.blogspot.com]