French makers here !

[Jean-Philippe]

Hello Quentin !

Your design is… amazing. It’s another breakthrough divide by two the costs of a printer !

And this looks ! is great !

I’m printing parts for a French maker and for mine is on the rails too !

But the point is : I’m founding my web shop in France. Opening could be this winter. Please, contact me by mail, I want to help you with the howtos and manual, and translate it into French.

[Quentin Harley]

Thanks JP. Will do.

[florent]

I’m the french maker JP is printing parts for !

Can’t wait for those to be printed ! Will try to upload some pics during the construction.

++

Florent

[florent]

Hi there ! I’d like to upload here a picture of my Reprap Morgan in building, but I don’t know how to do it. If somebody knows…

Anyway I got another question, because the mechanical part of the reprap is soon to be finished. Here it is : I bought the stepper driver DRV8825 md20a, and the nema motors 0.9, but I fear a lot that my soldering skills aren’t good enough to allow me do the hack on the stepper driver. ( soldering some pin to avoid a problem).
The thing is, what if I don’t do the soldering for the moment ? Can I still try to use my reprap, even with less good quality of printing ? As I understand it, I can use the motors, but it would be just like having bought 1.8 motors instead of 0.9.

Is that right ?

[florent]

(BTW, JP, if you need help for the translation into french, I can help)

[Quentin Harley]

Sorry to hear you got the md20a. Depending on your electronics choice, you might have to do two modifications on it. The enable line is not active when not defined, and would then not work at all. If it does work, you are wolcome to use it as is, and get someone to help you with the smoothing modification later.

I use the md20b models, and will soon use a south-africa spinn on ot that might not even need the smoothing mod.

Lets see.
Q

[florent]

Hello Quentin,
I have a RAMPS 1.4, does it add/remove problems ? 🙂

Regarding the modification you recommend (the smoothing one https://www.morgan3dp.com/?p=321 ), is there any way that it can be done not by soldering directly on the pin of the chip ? It’s so tiny ! Anyway I’ll try if I can’t get it work properly.

Second question, a couple of days ago I found on a website the second modification I may have to do ( soldering a resistor between two pin). This one seems doable for me, but I can’t anymore find the webpage that explained it ! HAve an idea ?

Thanks for all !

Florent

PS: My Morgan is going well, I think the mechanical parts are almost done, though it is not first class assembling :s… We’ll see !

[Quentin Harley]

I use RAMPS 1.4 – a great board.

You can use the drivers without the smoothing mod, but I would recommend doing it. Find a youngster with good eyesight an d steady hands – and use thin transformer wire to connect it. My telecoms wire was much too thick

For your second question, it seems you have to connect SLEEP and FAULT lines with the resistor

http://www.pololu.com/catalog/product/2132

Have great fun assembling your machine! Photo’s are expected when you are done…

Cheers,
Quentin

[florent]

Hi ! Thanks for the answer, I’ll try soldering this afternoon. Quick questions :

– What software do I need to test my motors ? For example I connect one motor to my RAMPS, connect al this to the power supply and to the computer, but then how can I try to move the motor from the PC ?

– Do we need one or two nut to go on the Alpen leadscrew ? I got only one maybe I lost the other… If that is the case I’ll start my reprap with a theaded rod, to see if this work.

Thaks !!

Bye
Florent

[Quentin Harley]

You can make it run well enough with the nut at the bottom. You can print yourself another nut to add later.

There is a ramps test you can run. Just make sure the arms are in the center of the print area, and the bed somewhere halfway.

[florent]

Hi ! Thanks again !

I tried to test one motor with this file

http://reprap.org/wiki/File:RAMPSTestCode.pde

that they recommend to test the ramps1.4 :

http://reprap.org/wiki/RAMPS_1.4#Pre-Flight_Check

Still, my motors isn’t working. I tried turning the pot to a higher value, still nothing.As I only have one motor connected to my ramps, I was wondering if the endstops should be connected, if maybe the board prevents any movement of the motors if the endstops are not connected. Is it ?

[Quentin Harley]

Ramps test moves the motors regardless of limits.

If it is not working, the disable signals may be active. Did you solder in the resistor between SLEEP and FAULT?

[florent]

Yes I soldered a 10k resistor between those two. But do I need then not to connect fault with the board ?

It doesn’t seem to be heating or to be giving any current to the motor… I am SCARED.

Maybe my soldering is not good enough….

INCREDIBLE !

As I was writing I accidentally I touched the board , and then it started working ! I should have a problem in the connection to the power supply, bad contact or smthg like that. It’s a relief to see that the 10euros chip is not dead 🙂

[florent]

It’s still not perfect, though. It should turn 5 sec in one direction and 5 in the other, but it sometimes fails, turn less than the good time in one direction and go back in the other. Some king of step skipping, I guess…

[Jean-Philippe]

I’m sorry but I didn’t start my morgans yet … Mail me If I can help with advices !

[Quentin Harley]

Check for dry soldering links on the Ramps. I had similar problems initially.

[Jean-Philippe]

Hi Quentin !

Do you know Megatronics V2 boards ? it support many good features and I had no trouble to run it on my custom Reprap (a very big prusa i2 totaly reworked)

I’ll try to use new pololu DVs on it. Just a bridge to had and it works normaly.

Did you modified by yourself the Marlin or Marlin carries the kind of motion ?

[Quentin Harley]

I use a modified Marlin that you can find on in my Git repositories. You can give it a shot, should work.

A future task for me would be to try and incorporate Morgan code into Marlin master. This will make keeping up with future hardware easier to do.

[Jean-Philippe]

I’ll probably need to compile the modified marlin of Megatronics V2 with yours.

I’ll keep you in touch a soon I I’ve startedto build my own Morgan.

[florent]

Silly me ! I had inverted two wires in the motors wiring ! It works just fine now. I can’t wait the for the first print, even if it’s not for tomorrow, I guess. Btw, how can I upload some photos on this forum ? Do I have to put them online ? How ?

Another question : to do the calibration, what software do I need ? I downloaded the ultimaker software (named Cura on my Ubuntu) but I can’t find the windows where you can “control your reprap by puhsing buttons or something like that”. Is Cura just for printing or for configuration also ??

Thanks again for all the answers Quentin ! That really helps a lot !

[hoff70]

My Morgan is not functional yet but I use Pronterface on my other printer. It connects to your board the same way you upload firmware and it has buttons you can use to jog the machine. Once you bring up the Pronterface program and plug into your controller it should connect.

Cura is a program that creates Gcode files that run the machine from .stl files. Quentin uses “Slic3r” to generate his Gcode so you may want to install that to use for generating your Gcode. Should make your first prints easier since you can use his settings.

I use Cura but getting it working is a PITA because it doesn’t agree with certain firmware flavors. Since your new I’d say stick with Slic3r for now.

Check this out: http://reprap.org/wiki/Printrun

I use the free service photobucket to host pictures but flickr is popular as well.

Good luck \o/

[florent]

Okay thanks a lot ! I’ll try to install it. Right now I’m trying to make my screw for the extruder with a metal file, cause I don’t have a tap… I’ll tell you how it goes. I’ll try to put the pictures, too.

[florent]

On the first picture you can see my “nice” soldering of the 10kohms resistors on the drv8825 md20a. Turned out it wasn’t that difficult. The smoothing hack is gonna be harder, though.
The second picture is the opto-endstops I bought. Are three of them enough ? I count two for X/Y axis, but for we need two others for the Z axis, right ? Maybe by putting it in the middle of the course with two triggering piece I can only need one…

[Quentin Harley]

One for Z is enough. Are you going to attach flags to the wheels for that?

[florent]

Yes, yes, yes and yes I am. Is it a bad idea ?

I’m happy, my filed-extruder-screw seems to be working. We’ll see for the accuracy later on.

[Quentin Harley]

Did not mean to ask four times… Network is a bit slow today.

Good progress,
Q

[florent]

Thanks for reposting the video, Hoff70 🙂

Yes, good progress indeed ! I’m on it full time these days !

I think I found what the tightening cone are for : aren’t they usedwith the help of the cap to “block” the plastic tube in which the filament goes ?

[florent]

Speaking of plastic tube, I bought a random plastic tube, not finding the PTFE tube. Is that a problem ? PTFE is -heat proof or something like that ?

[florent]

I think I have a problem with my J-head resistor. It does not fit tight in the hole of the head. There is a small gap, maybe of one millimeter between the two diameter (the one of the resistor and the one of the hole).

Is that a problem ?

[florent]

OK I found the answer. Kitchen foil.

[Quentin Harley]

A mm may be problematic, but there are ways around it. Fill the gap with thermal paste, and block the ends carefully with kapton tape. You want the heat from the resistor to be lead away from it, otherwise while the j-head may reach the temperature, the resistor might be much to hot, and fail earlier than intended.

I do this with all of my hotend resistors, and touch wood, have not lost a single one on my machines.

[Quentin Harley]

Foil is also ok if it is tightly wound.

[hoff70]

And about the bowden tube…

The “cone” is what holds the tube in place when the knurled nut is tightened. You’ll want the outside diameter of your tube correct for this to hold securely.

You’ll also want the inside diameter of your tube sized correctly for the filament you plan on using. Too tight and you’ll have problems with the bowden popping off at the extruder end. Remember that the extruder hobbed bolt/other will deform the filament a bit so if filament directly off the spool is a little tight it may be too tight once it passes through the extruder.

You also don’t want the fit to be too loose either. If there is too much clearance between the filament/Bowden ID then your retractions will have to be pretty long to straighten the filament before movement at the extruder is transferred to the nozzle. Then when the extruder reverses retraction the slack will need to be taken up as the filament moves side/side within the bowden before movement happens at nozzle… Picture an old fashion choke cable or throttle on a small engine.

I don’t imagine heat resistance is much of an issue but it should be slippery inside and well sized.

Hope this helps. Just some of my experiences coming to grips with the Bowden tube setup \o/

[florent]

Hello ! Thanks for the help ! As I had a small piece of ptfetube, I put It in series with my lrss-heat-resistant tube. Seems to work when I turn the wheel of the extruder by hand. But then I got to check my motor, because it’sonly whistling when I try to extrude filament, and the filament doesn’t move. Maybe the current Is Not sufficient ? What do You think ?

The kitchen foil seems to be ok, I mounted and tested the j-head and pronterface seems to read temperature well.

My next question regards the Martin software. Is the conversion of the coordinates done in the arduino ? I think it is, but when I move one one axis (x or y), I have only one motor turning, though they should be both in action to trace a straight line. Is There something in the arduino code to comment/uncomment that I may have missed, needed for the morgan only ? well.

[florent]

Thanks again for the design, the help ! 🙂
Bye !

[Quentin Harley]

Right…

Which marlin did you get? I would recommend getting the armlevel branch from my git repository. It has the scara reverse kinematics needed to get your morgan moving properly. Get the link on the first post on the home page of my site.

Regarding the motor not turning, it may be one of many things
– temperature
– speed too high
– current too low.

Use a multimeter, and set 0.8v on the potentiometer of the x and y drivers, and 0.6v on the Z and e drivers. A good start.

[Quentin Harley]

PS – none of the motors will turn before the first homing has been completed. Safety.

[florent]

Hello !

The problem is that I don’t know where to mesure the tension on the potentiometer. It’s so tiny :s

Anyway I installed the 2 endstops for X/Y now, but when I press homing, the arms just move abruptly, then stop abruptly, then start to whistle and do not move till the endstops. At that moment I think that the reprap is not responding anymore, because when I send commands like “read the temperature”, I just get no response. After a certain time though, I think the homing is done, because I can move on X, but then it just goes all the way till it blocks itself because the angle between the two arms is too big.

[florent]

I really know nothing about Reprap

[florent]

After doing it again I think I can’t move , it just does it by itself. If I look to the connection on my motors they are not in the same sens, I’ll change that. Which makes me think it is actually doing the homing movement. Still when it passes the opto endstops it’s not stopping. OK I try again.

[florent]

Ok now the homing seemed done, even if only one of the endstops was activated. Then I could move the arms (gently) but as soon as I pressed the home button again, the arms abruptly changed positions, really fast, and then stopped.

[Quentin Harley]

To measure the voltage put black probe of the meter on a ground spot. I use the power connector. Put the red probe on the screwdriver turning the small pot. The screwdriver now acts as the probe.

Make sure both arms are moving together towards the home position like in my youtube video. Also check the endstops before you start. Reset the electronics, and move the arms towards the home position while sending M119 every now and then. As the endstops line up it should trigger the stops on the terminal in pronterface. If not, fix it first. It could be that the magnets are inverted.

Sorry about the delay. Hectic day at work and not over yet…

Q

[florent]

It looks like I have a homing problem. I tried several things this afternoon looking at this wabpage :
http://forums.reprap.org/read.php?146,125056,137250

Still I couldn’t manage to have the homing done properly, that is, the arms won’t move to the position where the endstops are activated when I home the reprap. I will try again tomorrow.

[florent]

OK thanks Quentin, I’ll try that. But I don’t have any hall-effect sensor, I opted for opto ones. (uhuh) They give a 5v on the signal wire when you pass the flag between the 2 captors.

[florent]

OKay first I’m stupid. The z endstops was in the place of the X or Y. I put the Zendstops in the right place, but which arm is the x and which is the Y ? So that I can put the endstop in the right place ?

Then, with M119, it tells me that the endstops are triggered when the flag is not on it. Shouldn’t it be the opposite ?
That is, the flag should trigger the endstops when the homing position is reached. Right ?

[Quentin Harley]

Should be fine then. Checking with M119, do the MAX endstops get triggered?

[Quentin Harley]

Optos, right…

It triggers when low currently. Change the config file in the firmware.

[Quentin Harley]

Configuration.h, line 288 or thereabout.

X_ENDSTOPS_INVERTING should be false. Same for Y and Z.

[florent]

Ok here’s what I get when I connect the printer to pronterface. I post it so that you who knows what happens usually can maybe detect something weird happening.

Connection en cours…
start
L’imprimante est connectée
echo: External Reset
Marlin 1.0.0
echo: Last Updated: Oct 9 2013 19:58:26 | Author: (qharley, Morgan SCARA armlevel)
Compiled: Oct 9 2013
echo: Free Memory: 4742 PlannerBufferBytes: 1232
echo:Using Default settings:
echo:Steps per unit:
echo: M92 X215.00 Y215.00 Z100.00 E450.00
echo:Scaling factors:
echo: M365 X1.00 Y1.00 Z1.00
echo:Maximum feedrates (mm/s):
echo: M203 X300.00 Y300.00 Z300.00 E45.00
echo:Maximum Acceleration (mm/s2):
echo: M201 X400 Y400 Z400 E8000
echo:Acceleration: S=acceleration, T=retract acceleration
echo: M204 S9000.00 T3000.00
echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum xY jerk (mm/s), Z=maximum Z jerk (mm/s)
echo: M205 S0.00 T0.00 B20000 X10.00 Z10.00 E5.00
echo:Home offset (mm):
echo: M206 X0.00 Y0.00 Z0.00
echo:PID settings:
echo: M301 P22.20 I1.08 D114.00
ok T:0.0 /0.0 B:0.0 /0.0 @:0 B@:0
ok T:20.6 /0.0 B:0.0 /0.0 @:0 B@:0

[florent]

I feel like it doesn’t detect the endstops while homing:/

[Quentin Harley]

After changing the code, what does M119 report?

[florent]

Do you ever sleep ? o_O
🙂

SENDING:M119
Reporting endstop status
x_min: open
y_min: open
z_min: TRIGGERED
>>>m119
SENDING:M119
Reporting endstop status
x_min: TRIGGERED
y_min: open
z_min: open
>>>m119
SENDING:M119
Reporting endstop status
x_min: open
y_min: TRIGGERED
z_min: open

I guess the triggering of the endstops is ok now. I tried to change the 1 to -1 in this block

// ENDSTOP SETTINGS:
// Sets direction of endstops when homing; 1=MAX, -1=MIN
#define X_HOME_DIR 1
#define Y_HOME_DIR 1
#define Z_HOME_DIR 1 // Bed home to top (0)

to

// ENDSTOP SETTINGS:
// Sets direction of endstops when homing; 1=MAX, -1=MIN
#define X_HOME_DIR -1
#define Y_HOME_DIR -1
#define Z_HOME_DIR 0 // Bed home to top (0)

That doesn’t solve the problem…
Thanks for the help Quentin ! But maybe I should’nt ask for it at night, you should have the right to rest sometimes…

[florent]

Ok I did some changes (like inverting some directions…) and I just can’t get it to work. Here is what happens :

When I connect the reprap, I can read the temperature, I can send M119 to see if the endstops are triggered. Then when I press homing, the reprap starts homing, but when then it passes the endstops without stopping, and keeps going till it cannot anymore, because the belt is preventing it. Just like it doesn’t see the endstop, or even if it does see it, keep on going anyway. That is why I then tried to disable the pull-up resistors, but nothing changed.

Also, when homing, I can’t anymore require information on temperature or whatever. If, when homing, I send M 119 or I request the temperature via the button in pronterface, I don’t receive anything.
I’ll try some other tests, but if anybody had an idea, I’m open to suggestions…

[Jean-Philippe]

plop. I did not followed all the trouble.

You seems to have a configuration matter. Probably nothing bad.

or a reversed connection maybe.

turn your wheel as far as possible from the endstop, then press homing and trig one by one all endstop on the machine.

If no one stop the homing, then, change the #DEFINE ?_HOME_DIR -1

OH !!! YOU MAKE A MISTAKE in your conf ! you forget the space between DIR and -1 ! your marlin probably just don’t recognize the Define ligne so !

Call me if you need help, I can help you on the phone if you prefer !

You’re machine is nearly good, hold on !

[florent]

Plop.

Doesn’t change a thing. I can’t call you, I’m back in Italy for my studies… :s

[florent]

I’m gonna make a new thread so that everybody can look at it more easily.

[florent]

Ok my homing issue is over, I can look forward now.
So the homing is doing ok, but they just describe a circle when I send M 360, and doesn’t stop. Maybe some position problem. I tiped M 114 after homing and it returns me this

SENDING:M114
X:-35.00Y:-60.00Z:235.00E:0.00
SCARA Theta:114.64 Psi+Theta:241.12
SCARA Cal – Theta:114.64 Psi+Theta (90):36.47
SCARA step Cal – Theta:273.87 Psi+Theta:302.13

Any idea ? Sorry again for all my questions, I’m a huge beginner in Reprap…

[Quentin Harley]

What type of motors (1,8, 0.9), and which microstepping did you select?

[florent]

I have some 0.9 motors. Regarding microstepping, do I had to parametrize something ? I did the M 92 X430 Y430, but that’s it.

[Quentin Harley]

Remind me, 16th or 32th uStepping?

[Quentin Harley]

Try X215 Y215 to see what happens…

[florent]

Hello ! Long time no see, I’m taking back the Morgan after some time away of it.

So, homing is going fine, but when I send M360, the machine is going crazy (speed much more fast than during the homing) and above all, both arms are moving. Shouldn’t just one arm move ? And then stop ? My arms just move both and then stop(they are at the limits of their belts) and draw strongly the belt trying to force it further than it can go.

I feel like the machine is in pain :s

Any idea ?

[Jean-Philippe]

Nope. I’m still stuck and don’t have time to start my own Morgans. But I’m providing now 3 kits so we will progress big steps by big step.

Quentin, If I dev modifications to adapt to european pipe size, I will give it to you.

[florent]

I tried X215 Y215 it doesn’t change anything.

I’m thinking maybe that pronterface is sending some bad stuff to morgan when I connect and initialize it. Here is what Pronterface send :

start
L’imprimante est connectée
echo: External Reset
Marlin 1.0.0
echo: Last Updated: Oct 12 2013 17:17:25 | Author: (qharley, Morgan SCARA armlevel)
Compiled: Oct 12 2013
echo: Free Memory: 4739 PlannerBufferBytes: 1232
echo:Using Default settings:
echo:Steps per unit:
echo: M92 X215.00 Y215.00 Z100.00 E450.00
echo:Scaling factors:
echo: M365 X1.00 Y1.00 Z1.00
echo:Maximum feedrates (mm/s):
echo: M203 X300.00 Y300.00 Z300.00 E45.00
echo:Maximum Acceleration (mm/s2):
echo: M201 X400 Y400 Z400 E8000
echo:Acceleration: S=acceleration, T=retract acceleration
echo: M204 S9000.00 T3000.00
echo:Advanced variables: S=Min feedrate (mm/s), T=Min travel feedrate (mm/s), B=minimum segment time (ms), X=maximum xY jerk (mm/s), Z=maximum Z jerk (mm/s)
echo: M205 S0.00 T0.00 B20000 X10.00 Z10.00 E5.00
echo:Home offset (mm):
echo: M206 X0.00 Y0.00 Z0.00
echo:PID settings:
echo: M301 P22.20 I1.08 D114.00

maybe the problem comes from that ?

[Quentin Harley]

Ok. Go for the extreme.

M92 X100 y100
M500

This should make it move only a little bit.
We can see how much, and then adjust the steps accordingly.

Let’s alleviate the pain

[florent]

OK. So thank you, it’s much slower now, doesn’t feel like suffering. Still, either when I press the jog on +Y of 0.1, or on X, or M360, it looks like it’s always doing the same thing, that is: turning both arms at the same speed up to the physical limit.

Had you time to look in the echo file ? I was thinking that maybe my pronterface is setting some bad parameters at the beginning.

[Quentin Harley]

So it still keeps on going?

How does homing compare to the homing in the Youtube vid?
After homing, what does M114 return?

[florent]

Yes it still keeps on going.

Homing is going like in the video, I think. Both arms turn counterclockwise until the endstops are activated.

M114 returns
SENDING:M114
X:-35.00Y:-60.00Z:235.00E:0.00
SCARA Theta:114.64 Psi+Theta:241.12
SCARA Cal – Theta:114.64 Psi+Theta (90):36.47
SCARA step Cal – Theta:127.38 Psi+Theta:140.53

[Quentin Harley]

Looks right.

So, on M360 both arms turn clockwise until the end of the belt?

[florent]

Yes, exactly. Also, they are going fast again.
Yeeey :s

[Quentin Harley]

Ok.

Tell me about your motors and the stepper drivers, and uStepping settings you are using again.

I use 0.8deg (400 step) motors, with 32thuStepping for instance.

[florent]

So :

I have 4 motors 0.9deg Nema 17.

For the stepper drivers, I bought some Pololu md20a, and on those I soldered a 10k resistor between slp and flt, and that’s it, I plugged them on the RAMPS.

As for the ustepping settings, what do you mean by that ? Do I have to change something on the hardware ? On the configuration file ?
I thought the settings were done with the M92 command only.

[Quentin Harley]

I assume you have all the jumpers installed under the stepper drivers, right? That would make it 32uStepping. If not, install them ASAP

Interesting… I have no idea why the arms should overdrive.

[florent]

Mmmm… What jumpers ? o_O I’m such a noob, sorry ! So how do I put those jumpers ? I didn’t get some with the drivers. If I solder the 3 pins together would it be ok ?

[florent]

Or can I first have it run without the jumpers ?

[Quentin Harley]

Without the jumpers you have no uStepping. The motors are probably very very loud, right?

This is what you should have…

[florent]

Argh ! I don’t have those ! Maybe that’s the cause of the all problem ?!

They weren’t given with the RAMPS, can we buy it easily ? Or can I make something to make up for them ?

[Quentin Harley]

Do you have the pins at least?

If so, you can simply turn some thin wire around the 3 pairs of pins to make the contact. It does not draw any current.

You can get the jumper tabs from old hard drives, old cd-rom drives, some old PC motherboards, or from your local electronics shop.

Now everything makes sense again!

[Quentin Harley]

PS: Those arms really must have been flying!

32 times faster than it was supposed to… Wow!

[florent]

Amen ! I will join the pins with wires before buying some jumpers. Can I join all the six together or should I join them 2 by 2 ?

Thanks again ! I see this as a break through for my Morgan ! 😀

[Quentin Harley]

You can connect all six together – will be fine.

Nice to see you getting ahead!

[Jean-Philippe]

You can pick up jumpers from old computer motherboard 😉

[florent]

OKaaaay ! I pushed a tiny piece of metal between all pin and it seems to work fine ! The printer is MUCH slower which is perfect, and it seems to stop at some point ! I’m gonna run the calibration and do some tests, I’ll tell you soon !

[florent]

Ok so I run M360, and try then to move the arms to make theta arm parallel to the platform, but before it is actually parallel, there is a point where I cannot go further. Looks like a endstops except that when I do M119 before the point I receive :

x_min: TRIGGERED
x_max: open
y_min: TRIGGERED
y_max: open
z_min: TRIGGERED
z_max: open

And when I do it where the arms won’t move further, I receive the same.

(I feel like I’m getting stucked at EVERY steps of the build :s)

[florent]

Wait, I may have figured it out. 🙂

[Quentin Harley]

Your steps are now too low… but I guess you figured that one out yourself it seems.

Depending on your motors and uStepping setting you have to choose a starting value for the steps.

Should be close to 430 for X and Y on your current settings.

[florent]

OK Thanks !

[Steve]

Hi Quentin – pardon me . . .
Is the first 200, the X – the arms left to right action?

I ask because I got a larger rectangular heated bed (300×250) and need to figure where the changes are made. I want the long way to be left-right across the front.
Thanks

[Quentin Harley]

…So, to sumarize, Change X_MAX_POS 200 to 300, Y_MAX_POS 200 to 250

and

SCARA_Offset_x 100 to 150

[florent]

Hey !
I have so much trouble with the calibration that I’ m wondering if my arm-motors are switched. Mine are in this configuration :

– the motor controling the 22 copper pipe (theta arm) is wired to Y on the ramps
– the motor controling the rod is wired to X on the ramps

Am I doing it right ? And then how can I control that they turn in the right direction ?

Thks !

[Quentin Harley]

You are right. Motors are switched.

X: Theta (copper pipe)
Y: Psi ( threaded rod)

Just swap the two plugs on the ramps board.

[florent]

🙂 I feel like I’m doing every mistake there is to be made.

Hey is there on this website a scheme of the Morgan seen from the top where the angles (theta and psi) and also the reference position for X, Y are shown ? It would be helpful to try to do the calibration by myself. If I can’t, I’ll just ask you 🙂

[Quentin Harley]

It is on one of the chinese buid forum topics,

But it is also on the new calibration flowchart.

[Quentin Harley]

PS look in the Calibration guide link at the top

[florent]

Hello !

1) The calibration is going OK now for the arms, I changed the value of the position of the homing from -35 to 35, and it goes fine now.
/ Manual homing switch locations:
// For deltabots this means top and center of the cartesian print volume.
// Scara: x centered, y 0
#define MANUAL_X_HOME_POS 35//-35
My former value was too far from the actual position, I guess.

2) So I am now starting to calibrate z, and I think this is going to be ok( even though I have a threaded rod instead of the lead screw for now, so I have to higher a lot the steps for Z)

3) I have some problems with the extruder, I think my tube (which is not a ptfe one) is too soft and sticky ( it is a tube like the ones for aquarium pumps), so the filament is hardly going through it. Is there any way to use this extruder without a ptfe tube ? I was thinking of something like a shower hose… Would it be a bad idea ?

4) I broke the z-motor coupler, so I have started looking for a place to reprint it in Milan, or I’ll just buy smthg like this :
http://reprapworld.com/?products_details&products_id=198&cPath=1655

That’s it for now !

[florent]

A last question : do I need really need a heated bed or can I do some “OK” prints without it ? So that if I buy some pieces I’ll just buy it in the same order.

[Quentin Harley]

Great news. I suppose it makes perfect sense that you should enter the correct (or close to) home offset start position manually. I think I will add this to an updated version of the calibration flowchart. It is not something that you would change during the lifetime of the machine (unless you take it apart) and the offset correction can actually correct for quite large offsets successfully.

You will 100% have to replace the tube. PTFE (Teflon) tube may be expensive, but the extremely low friction and relatively low elasticity is what you need to make it work. Remember that the filament also get roughed up in the extruder, making it bite into the soft tube even more…

For the print bed:
You can get away with not using a heated bed if you get yourself some of the 3M blue painters tape and print only in PLA. It works very well, but will need sectional replacement every 3 to 7 prints. I print too much to bother with the maintenance of the print surface, so the wide Kapton and the heated bed makes it much easier for me to keep keeping on. If you stick to PLA the Kapton surface can literally last Months.

I have separate aluminium build plates for ABS that I coat with cheap hairspray when heated.

[florent]

Hey ! I bought two of those stickers :

http://reprapworld.com/?products_details&products_id=469&cPath=1629

I have no aluminium plate but I have wood and also a cheap plastic plate (not so much heat-resistant I think). Can I still stick it on the plastic plate or should I stick it on the wood ?

The ptfe tube has arrived, also I bought this M8 motor coupler

http://reprapworld.com/?products_details&products_id=198&cPath=1655

Maybe I have to drill in my motor axis. Anyway I’ll try to do it tonight.

[florent]

Hello !

Yesterday has seen my first printing tests ! Everything seems to go fine (I have to re-do the calibration of the bed), except that the filament is popping out of the nozzle not only by the small hole but also from the thread… I think that my problem may be the speed of the extruder motor, which seems to go fast. What should (approximately) be the speed/steps if my motor is in 1/32stepping mode ?

Also, I am not sure where to place the configuration files in Slic3r. I’ve put them in the Slic3r folder so that I now have all the folders bin-dll-installer-lib-res-print-filament-printer together in the Slic3r folder. Is it ok this way ?

Thanks again ! I think I’m not that far from my first print ! 😀

[Jean-Philippe]

Aïe ! this is a BIG problem.

DO you try PLA or ABS first ? beware with ABS, you must don’t heat to much, or destroy the JHead plastic part. Try to heat at 220, get out the plastic wire, turn off heat and unmount the nozzle to clean it.

When it’s done, use PTFE plumber tape (make 3 or 4 turns around the nozzle screw) and replace it into the Jhead. Tighten it with tools. Then, place tje heat resistor and 100k thermistor.

I have destroyed 2 Jhead because of thermistor ejection. My tip is to drill a little bit more the themristor hole and to roll up tiny kapton tape around just the nozzle (NOT AROUND the plastique part /!\)

Good luck. And don’t forget to always have et Pololu and a a JHead.. in case of…

[florent]

OK ! But I cannot see how this could destroy my thermistor … For the moment it’s just that I have to clean it everytime. I will try with the PTFE tape on the thread of the nozzle before screwing it.

But I really think the problem (also ) that my extruder motor is going too fast. Then the flowrate of PLA going out of the nozzle the right way is too small compared to the flowrate the filament bring, thus this induce pressure in the nozzle,thus leading to the nozzle leaking…

How can I “calibrate” the speed of my extruder motor ?

Also, when the hot end is heating at the beginning of the print, should the filament be into it already ? Or should it still be in the cold end, and just when it starts printing, the filament starts going down in the hot end ? (Mine was in the hot end from the beginning )

[Quentin Harley]

I have been luck thus far… Not a single damaged jhead so far.

It can be that the head was not properly tightened in the assembly.

Read and do what JP said. I also like to work in a drop of silicon heat sink paste into the hole for the thermistor and the heating resistor. The kapton around the brass/aluminium head of the jhead is also a must.

[Quentin Harley]

Being in the hotend already is normal. Leaking out the sides, not so much…

[florent]

Aaaaaaand I destroyed the thread in the nozzle holder. The nozzle fell when I was heating everything… Guess I’m good to buy another one, now

[Quentin Harley]

Better do the filament calibration before the new hotend comes…

😉

[florent]

Turns out that when I ask for 100mm of filament it goes out only around 90. So it shouldn’t be the problem.

Do you think I can have this one work before I get the new one ? I was thinking of a small screw inside the nozzle holder to maintain the nozzle.
Not a good idea, huh ? But anyway the nozzle holder is unusable…

[Jean-Philippe]

The probleme can’t be the motor speed. The nozzle MUST be sealed.

I did not say Thermistor is destroyed, but not in place and don’t measuring the right temp. Of coper wires not insulated in short circuit.

[florent]

Hey !

So I will buy a new hotend, and I’m thinking of buying the whole package again so that I have some spare ( thermistor,…) What do you think of this one :

http://www.ebay.fr/itm/RepRap-Bowden-Extruder-Hot-End-Kit-/221318376820?pt=LH_DefaultDomain_3&hash=item33879a6d74

or this one :

http://cgi.ebay.fr/ws/eBayISAPI.dll?ViewItem&item=291029827475&ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649.

The first seems a bit heavy for a Morgan, but it has a fan. What do you think ?

[Quentin Harley]

I use the J heads with much success… Would go for it ever though it is more exspensive. The all metal hotends still needs some development in my opinion, and they are heavy. Mounting on the morgan arm would also be a challenge…

[florent]

Finally received the new hot end !!!! Right in the middle of the study month for the exams of my university… Yeyyy !:/
I tell you when I have time to try !

See you !

[Quentin Harley]

Well well! Happy birthday Florent’s Morgan.

It is magical when you watch the first print unfold. You chose a tricky print to boot!

Cheers,
Q

[Quentin Harley]

I don’t think the problem lies within slic3r…

It could be a Z-offset problem.

Enter the following in order:

M206 Z0
M370
M373
M500

G28
G1 X100 Y100 Z50 F10000

Now Jog the z until the nozzle just touch the bed.

M114
Record the Z position, and change the sign (eg. 14.4 should be -14.4)

Enter this into
M206 Z(value)

M500

Now, perform the bed level auto correction (M370) section. It should now function correctly.

[mk3v]

Well, I followed exactly what you explained…

The z offset I had was very absurd but I did it, just to be sure.

And at M371 the bed encountered the nozzle.

I think my problem is more mechanic.

I use a M8 threaded rod as Z-screw with two nuts, one at the bottom and one I inserted in the hexa hole a bit tighten against a spring.

How much does is have to be tighten in this configuration ?

During the trip from Z-home to Z10 the motor speed changes several times and sometimes it goes down a bit while it was going up, then it goes up again…

Could it be a motor or a driver problem ?

And the sound the motor makes while jogging with pronterface is completely different from the sound while travelling in G1. Is it normal ?

Or is it simply my chinese deal extreme RAMPS ?!?

This is enough questions for now…

[Quentin Harley]

It sounds like you have speed issues here.

Remember that the firmware is configured for the fast lead screw, and the stepper cannot tun fas enough to enable the same speed movements using M8 threaded rod lead screw.

What are your steps per mm on Z?

[mk3v]

First I caclulated 400 x 32 x 1.25 –> 16000 steps/mm

Then with the first step of the Z calibration I had about 10200 steps/mm.

[mk3v]

Oh I was mistaking when calcuting, I had to divide by 1.25 instead of multiplying.

So I’m OK with something around 10200…

[Quentin Harley]

Now it should be fine

[mk3v]

Hello,

I’ve solved a lot of little problems I had.
And now I’m pretty satisfied of my first print tests.

But there is still one thing that appears on prints like the calibration part, it’s a gap between the perimeters and the fill-in, but just on the right side of X:

http://hpics.li/014c7e3

It’s weird and I don’t know where it’s coming from.
I will try to print it not on the center of the bed, just to see what happens.

But if someone knows what it could be ??

Just let me know 😉

K

[Quentin Harley]

It has to do with hysteresis. There are a couple of things that can cause this:
– friction between the Theta and Psi drive shafts
– flex in the motor – try the new motor mounts
– flex in the internal drive shaft – make sure that the threaded rod cannot rotate easily on the nuts of the internal pipe. You can use threadlock or even superglue to make sure of that.

Hope it helps

[florent]

Hey !
At last I have time to take back the Morgan project. So I decided to work on the main issue I had when I first tried the reprap.( I only did one print “correctly”, the one on the video.)

So, the extruder motor would go push the PLA in the PTFE tube, and this tube would follow the PLA because there was too much friction or because the plastic could not go ahead. After dismouting the tube, nozzle, it turns out that apparently it is very hard to push the PLA in the J head. I was thinking that the Jhead might be not hot enough, but setting it to 230 it seems the same. I think my nozzle is 0.4mm large, and my PLA plastic filament is 3mm large. MAybe the problem is that too much filament is sent through the tube and so the nozzle cannot extrude that much. What can I do ? Reduce the speed of the extruder motor ?

[mk3v]

Salut Florent,

I think I have the same problem…

I tried to print the new motor mounts, the short one went correctly, but the print of the tall one failed at half-height.
I think it’s due to the fact that I maybe have a bad J-head and the heat goes up in the PEEK part and the PLA get melted over time during the print and when enough length of PLA is melted is sticks very hard to the PTFE tube.

So I think we have to cool down the PEEK part of the J-head during print or Quentin will have teach us another trick to get through that problem…

A+

K

[Quentin Harley]

I have had a couple of brand new (real) jheads at 0.4mm nozzle size blocked. Get yourself a 0.4 pcboard drill bit. It has a thick shaft, and can be turned by hand, and drill out the offending nozzle carefully. After cleaning I did not have trouble.

Regarding cooling the PEEK, if you have the cooling fan mounted on the top platform like I have it on my machines, this is not a problem at all.

3mm into 0.4mm nozzle is a lot of pressure. If the problem remains, I will think of a way to solve the issue.

[mk3v]

Atually, instead of florent I have a 1.75 mm filament for a 0.35 nozzle.

The problem is I don’t really have to clean the nozzle but I need to get back all the filament out the PTFE tube and there is always about 5mm of molten PLA at the end of the PTFE tube (I will try to take a picture).

Then I increase the hotend temp to 170 / 180, I put the tube with the “blocking” PLA in the hotend a few seconds.

At this time I can push the “blocking” PLA with fresh filament.

But at the next long print the problem comes again.

[mk3v]

Here’s what I was talking about :

http://hpics.li/757202b

[florent]

I think the problem is more the pressure than the blocking of the J head. When I try to push the filament by hand through the Jhead, the melted plastic goes out from the nozzle, but the filament goes in really slowly.

Would a 3mm Jhead work for a 1.75 mm filament ?

I’ll try again tomorrow and see what happens.
Thanks for the answers!

[Quentin Harley]

To adapt 3mm nozzle to 1.75 is a workaround at best. Rather get a 1.75 nozzle if you want to go that route. That said, I never had any issues with my 3mm systems, except that 3mm is sometimes hard to find here in South-Africa.

I did run it on 0.5mm nozzles though – less pressure.

My new 1.75 filament nozzles are 0.4, and the difference in terms of slic3r alone is noticeable.

The other thing you can try is to insert M6 nuts onto the PTFE tube and tighten the cap over it. This should combat slip

[florent]

Hey !
How can I slow down the speed during a print ?
🙂

[Quentin Harley]

Send the gcode M220 S(percentage)

So, if you want 80% speed, issue
M220 S80

[Quentin Harley]

You can type it into the pronterface terminal during a print, and it will take.

[florent]

[florent]

hi five !

[Quentin Harley]

Nice!

[florent]

Ahah I’d say more than nice ! 😀

Enoooorme !

[mk3v]

Hey nice work Flo !

I was wondering if it’s possible to decrease the speed of the M371 move, and also for M372.
I think the speed for my hardware is too high…

Quentin, is there a way to do that ?

K

[Quentin Harley]

The speed is dependent on your settings, regardless of hardware. Your steps per degree might be too high, causing it to move too quickly (and hurting itself in the process…)

What is you current steps/degree?

You can find it by typing M503 in pronterface – M92 values.

[Quentin Harley]

… and also tell me about your hardware.

Stepper driver type, jumpertabs installed for microstepping, stepsize of motors (1.8 or 0.9)

[Quentin Harley]

Oh, I see… You are using threaded rod…

Let me see.

You have to limit the maximum speed of your Z-axis using M203 Zxxx

Default is Z300, but you will have to experiment to find your correct max speed.

Try:
M203 Z10
M500

[Quentin Harley]

Let me know what you get so that I can add that number into the calibration guide

[mk3v]

OK so I can put the setting I found on pronterface for jogging Z axis correctly…?

I am at 20 mm/min when doing the bed auto correction and it seems to be fine.

[florent]

Hi there !

My morgan is starting to print relatively (for me who built it : incredibly) well, but I recently discovered a small problem on z calibration.

e.g. I print a cube with dimensions : x=20mm, y=20mm, z = 20mm. The x and y dimensions are, after measurement on the printed cube, correct, but the z dimensions would be 15mm or so.
Maybe I can change a setting in Slicer to fix the prob ? So that he knows that my slice thickness is smaller than he expects, and thus plan to do more slices.

What do you think ?

thanks for all, as usual !

Bye,

Florent

[Quentin Harley]

You will have to check the complete Z calibration, starting with the steps per mm.

IT should be spot on…

[florent]

Mmm I tried adjusting the step calibration like you said, and effectively I find that my steps per mm should be 10204 instead of 7500. (I use a threaded rod) The problem is that when I then record 10204 and try to do the plate calibration, when I move to the initial bed position, the bed wants to go higher than the hotend! So I have to stop the motor before hitting it…

The calculation is (100/(calculated distance)) * (current Z step/mm), right ? Not 100/(calculated distance * current Z step/mm).

Now that I see this problem of calibrating Z, it occurs to me that I already had this problem before, and I think I never actually did the tuning of Z step per mm…

[Quentin Harley]

My other post disappeared… Lets try again.

Clear your z calibration completely.

M206 Z0
M370
M373
M500

G28
G1 X100 Y100 Z100 F10000
Jog the bed up to zero, and then issue
M114. Take the Z value, change the sign, and enter into M206.

Now, do the m370 371 372 373 adjustment again…

[florent]

OK. Should I also try again the step/mm for Z??

[Quentin Harley]

Absolutely!

The reason Z overshot is because you did the calibration before checking or setting Z steps, and the machine thus thinks you have a longer Z axis.

Do it in order, and all should be fine

Q

[hfisselier]

I’d like to build one too. I’m reading a lot to prepare myself. I’m reading you.

[Quentin Harley]

Welcome to the ever growing French Morgan club!

Looking forward to your pictures as well.

Q

[florent]

Quentin ! It did it again ! But I think I may know why : at one point in the calibration, the z-motor go fast and the axe blocked a few times, so maybe the firmware isn’t aware that it blocked… Anyway I can reduce the speed during the calibration procedure ? If this is not this I don’t know what it can be. I’ll start the whole procedure again this Weekend ! :/

[mk3v]

Hi !

I think you’re using a threaded rod just like me.

So you have to change the max speed of the Z motor.
Just experiement with pronterface… I’m using 60 mm/min for joggning Z axis in pronterface.
It gives 1mm/sec to load in the firmware with M203, but the hardware can go a bit faster, I will tell you later which value I put in the firmware.

K

[Quentin Harley]

You can get faster movement with threaded rod, but you will have to make the acceleration value much smaller for the Z access. It will then start slow and speed up, so as to not lose the motor causing it to skip steps harmonically.

I would love to get those max speed values for threaded rod, so that we can add it to the start-up guide

Q

[mk3v]

OK, I never tried to tweak the acceleration value.
I think my speed value is around 2.

I will try to modify the acceleration values this evening and I will let you know what I find.

Quentin, could you tell me which value to start with for the acceleration ?

[Quentin Harley]

What is your current Z acceleration?

[mk3v]

I’ve got 400 mm/s2 for X,Y and Z.
Maybe I should reduce acceleration on every axes due to my hardware (400 steps motors and 1/32 ustepping) ?

[Quentin Harley]

300mm/s2 is more than enough for X and Y, and Z you can take down to 50 for the experiment.

[florent]

Hey there,
Quentin when you say jog the bed up to zero that’s jog the bed til the hotend touch the bed, right ?

[Quentin Harley]

Yes.

[florent]

Peeerfect my Z calibration is ok now ! I took it from the beginning and it’s going fine!
I’ll try to post a photo of a small part I printed for the kitchen robot of my brother!

[florent]

[florent]

The next step for me is to try harder things ! Thanks a lot Quentin for all the help ! And also the others !

I recently got hold of a Microsoft Kinect, and I heard we can do some 3D scan with it ? Any software to recommend ?

[Quentin Harley]

Excellent news! It is great when the machine starts giving back

Regarding kinect, there are many software solutions for the kinect on windows – many are free, but the better ones are not.

I bought a Sense scanner, based on the same technology as kinect (prime sense), but with different lenses. It works great on larger subjects, and people.

Lots of fun…

[mk3v]

Hi,

I still have some problems with large parts.

I was wondering how the level auto-correction works.
Since it seems that my nozzle doesn’t move on a flat plan but on a curved shape.
If I move the nozzle on the same X line : at Y0, I could be 0.5 mm far from the bed ; at Y100, I’m just touching the bed ; and at Y200, the nozzle is again 0.5 / 1 mm far from the bed.
And I don’t think my glass bed isn’t flat, I tested it on a marble bench at work…

So does the auto-correction works with this kind of default ??

[Quentin Harley]

When you do your Z correction, how many point does it make you program? 25 points is the default – takes a while but should take care of any problems.

If not 25, issue it in this way:
M370 X5 Y5

You can also go for more than 25 if you are very brave. I recommenf keeping it uneven numbers. You can also have different numbers of rows and colums if you find your problem lies in a certain plane:
M370 X3 Y5

[mk3v]

OK, I will try to increase the numbers of rows and columns.
But do you really think it can correct my default of curved shape ?

Or do I have another mechanical problem…?

[Quentin Harley]

It should. How many rows and columns did you use?

[mk3v]

I was using the default grid 5×5.
Maybe I will try 7×7 or 8×8.
It will be long, but maybe it’s worth it…

[mk3v]

Hello there,

I just wanted to point out a strange thing :

I needed to print a set of herringbone gears for the 3mm extruder.
But with le latest STL’s the small gear is inverted, so I checked the instructions on the last reprap magazine and it’s not working the way it’s shown.

I think the small gear has been modified, because in the kit I bought to build my morgan, the small gear is fine…

Could it be possible to get the old one back ?

[Quentin Harley]

The small gear should be installed with the gear to the inside and the ring with the grubscrew to the outside.

Yes… you will see that in the mag the small gear is the right one from my repository – It will not mesh that way round.

I will have to chat to Paulo about fixing that. I let it slip through…

[florent]

Here are some photos ! And a video of me testing the motors.

[hoff70]

It’s alive \o/

http://s932.photobucket.com/user/tnerolf_duagir/media/video-2013-10-01-16-21-55_zps5a5f48c7.mp4.html

Your link didn’t post. Hope you don’t mind that I re-posted the link.

[Quentin Harley]

The reason why it stops, is that it is bound by software to remain in the 200×200 print area of the heated printbed. IF it goes out too far, it simply ignores any further commands unless it is one that will bring it back into the printable area.

When you increase the steps per degree, it will move the motor further before it thinks that it is out of bounds. If you can just get it far enough to be parallel, it will work. The next step will give you the actual steps.

[Quentin Harley]

X is right to left across, and Y is forwards and backwards from the rod.

In other words X0 Y0 would be the top right corner if you look at the machine from the front.

To adjust the print bed size, look for the following in the configuration.h file:

// Travel limits after homing
#define X_MAX_POS 200
#define X_MIN_POS 0
#define Y_MAX_POS 200
#define Y_MIN_POS 0
#define Z_MAX_POS MANUAL_Z_HOME_POS
#define Z_MIN_POS -10

After that you have to make sure you have the arms lenghts entered correctly if you do not use the standard 150mm

// Length of inner support arm
#define Linkage_1 150000 //um      Preprocessor cannot handle decimal point...
// Length of outer support arm     Measure arm lengths precisely, and enter 
#define Linkage_2 150000 //um      define in micrometer

And finally the new offset towards the Tower

// SCARA tower offset (position of Tower relative to bed zero position) This do not need to be accurate
// Homing offset calibration takes care of real homing position offset.
#define SCARA_offset_x 100 //mm   
#define SCARA_offset_y -65 //mm

[Quentin Harley]

To calibrate the extruder do the following.

1. Pull the filament back into the bowden, to somewhere in the teflon tube, visible just after the extruder. Mark the spot. Using pronterface or the like, extrude exactly 100mm. You may have the disable the cold extrusion prevention first : M302 also remember that no motor movement will be allowed without running homing first…

Measure exactly how far it moved, and then calculate the new steps/mm.

100 / (actual distance moved) * (current E0 steps/mm)

You can check the steps/mm by running M503 and looking at the values next to M92

Store with M500, and repeat to check.

[florent]

Ohmygodohmygodohmygod.

http://s932.photobucket.com/user/tnerolf_duagir/media/video-2014-01-09-22-13-49_zps15ebd76b.mp4.html

It still needs some tuning, but I guess I can say that it is working, now.

[mk3v]

Hi guys !!

My build is finished…

I think calibration is OK.

But when I try to print the SCAL part with the settings found in Morgan Master, the bed goes up but it doesn’t stop at zero but it tries to go further up…

I think it’s a param in Slicer… but wich one ?

FYI I tried to print from an SD card and the smart LCD thing…

Thanks for your help.

Kevin

[mk3v]

Hi guys !!

I think I finally solved my problems and finished the calibrations.

I used the ini files in morgan master to configurate slicer.
But my first print is a mess, there is a lot of parameters in slicer and I don’t really know wich one to start with.

Here’s a pic of this print :

http://hpics.li/75e45a3

Hope you can guide me a bit with this…

Cheers,

K

[mk3v]

I have 400steps motors and 1/32 usteps.
My M92 value for Z is 10244.79 calibrated with yours method…

EDIT : and a M8 screw

[Author]

Posts