Catalyst Mendel continued development

While the Catalyst may not be the printer suited to every need, it does fill an important niche.

So in order to leverage some community innovation, where do we want to discuss improvements? How about here.

Y-Motor Mount Lower Plate
One piece I needed to modify extensively was the base plate of the y-motor mount. Until I reduced the thickness of the plate significantly, my printer rocked on the bolts. I also cut into bottom face so that the bolts would be slightly recessed.

Proposal
A mounting bracket which clamps the rails. This would replace both the top and bottom plate of the y-motor mount.


Negative X-Axis
With the current construction the x-motor extends over the build surface. If the motor could be moved off the surface, then one of the barriers to increasing the build area would be addressed.

Proposal
Modify negative z-axis mounting plate to extend further from center. Replace the bolts connecting positive z-axis mounts with longer bolts. Replace x-axis hollow rod with longer hollow rod. Cut a channel for z-axis threaded rod in x-axis motor mount. Cut holes through x-axis motor mount to incorporate carriage bolts.

(Minor) Drawback
Increased machine footprint. But that's all build space and then some.


Y-Axis Mounts
These simple little clamps are nice, but often the bolt is too close to the rod and I can't get the bolt in. The outer edge is too thin to sand, and I don't want to widen the rod's hole because that would loosen the build surface.

Proposal
They need to be a little bigger.


Pinch Wheel Feeder - Location
Should we concede that the feeder wishes to run on a leash around the printer?

Proposal
Add an actual leash so that when you move the printer the feeder will move with it. Twine the wiring around the feeder tube so it won't get tangled. Redesign the mount into a more stable shape with a broad base.


Ease of Assembly
Proposal
Every component should have identifying text printed into it. Ideally, also: assembly location. Conventions might be established, such as hatch patterns on parts in contact with benchtop.

Drawback
A lot of work. What standard? URL? Build info?


Board Mount
Proposal
The board mount does not leave enough room to mount board and arduino on it. The upper edge could be slightly shorter (2 inch or 50 mm) to more easily meet the frame. The board mount could provide channels and holes for ties for the wires.

My comments about the issues listed, and then some...

Buzz

Y-Motor Mount Lower Plate

My solution was to place the plate on the table with the motor on top. The bottom plate is not attached to the rest of the assembly in any way, shape, or form. I too saw the issue and did not want the printer rocking on the bolt. The motor is held in place by an angled bolt at the back and washers in the front. (A picture will help explain...)

Negative X
Yes, the X axis eats into the build space. This is an issue that should be addressed. I am wanting the Catalyst to print me parts for a larger non-metric unit.

Y-Axis mounts
Agreed. Enough said :-)

Pinch wheel
I am wondering if anyone has gotten the pinch wheel to feed reliably. I have doubts concerning the precision (slippage) considering the gasket provided to me already mounted on the motor shaft was off center. I like Wade's geared extruder and would suggest changing the catlyst feed mechanism to something similar to that.

Assembly
One thing that would help a ton is exploded drawings of the various assemblies. Watching a 10 minute video to do a 3 minute assembly is not efficient. Looking at an exploded diagram to see the intended assembly is a big improvement, IMHO. Also, clear images of the assemblies would help.

Board mount
A flat panel across the frame would be more usable as you could screw what you wanted where you wanted it. This could also leave room for the heater circuit and sensing circuit.

Circuit diagram
Currently, there are two different diagrams for the heating and sensing circuits on the web site. Which is correct? Suggested layouts and pictures would help.

Z axis top mount
If the frame built with equilateral sides and parallel sides, the Z-axis top mounts are not far enough out. The suggestion in the assembly video is to crank in the bottom vertexes. This adds tension to the entire assembly, adding to the potential to crack vertexes. Suggest adding longer bolts to the top. (Again, I'll add a picture of how I handled this).

Z axis bottom mount
The method of holding the z motor needs improvement. Either a different method for clamping, or strengthening the tensioned parts when tightening the bolt.

Software
The PD website have about 5 different versions of the firmware. It is unclear what the differences are between those versions, let alone the differences between the official reprap version. I am currently trying to get the official version to work but running into issues, I believe dealing with the interrupt issue. The PD website states that the one version "also has correct interupt (sic) fix, that reprap offical (sic) version does not have yet". I am looking to isolate this "fix" and see about incorporating it into the main version. I like to stick with the main version, with mods to that vs. the unknown version with unknown changes.

Re: Pinch wheel gasket
I noticed that the gasket was off center. If you look on the non-beveled side, you will see that there is a portion of the gasket cut away. I finished the gasket by shaping the cut away into a 4-point star and the gasket centered.

Re: Top Mount
I think that the top threaded rod of the frame should be longer so that we can eliminate the couplers and replace them with bolts for spacing. This would be more consistent with the rest of construction, but would remove the radial symmetry of the design. This would allow more leeway for further modification.

Perhaps the feeder could mount on one side?

Re: Top Mount

That is essentially what I did. I took a threaded rod and cut it into 3" pieces and threaded those into the couplers.

Mounting the feeder on one of those wings is one of the options I am considering.

Build Surface
The bolts as provided seem to extend overly close to the extruder. Quick fix is to use shorter bolts

I would like longer table mounts, preferably gripping notches cut into the outer edge of the table. I found a pre-cut piece of Lexan at HomeDepot which I will use to replace the build surface.


BTW, I will of course make any STLs available. None of these seem particularly challenging designs. The table mounts I think will require the most novel design. I believe that a mechanical means can be found to replace the springs in the table.

Bowman Feeder Mount
The mount is difficult to insert considering the movable motor.

Solution
Zip tie the nut in place to the feeder. Experimentation required.

Edited 1 time(s). Last edit at 04/05/2010 05:39PM by KrisC.

Table Mounting Bolt Collisions
While trying to trace the bounds of the print area, the screws which secure the x-motor collide with the screws which support the table.

I had already replaced the table screws with shorter screws. I reversed the screws so their heads are visible. Then I placed a bolt to hold the x-motor screws above the surface. The screws are not as tight as I'd like, and they no longer prevent the extruder head from contacting the table, but that is the limit stop's job anyway.

Hi guys.
I received my package week or so ago.
Here are some pics after I finished accounting and organizing parts (most of them)
Sorry for bad quality, took them with my Nokia5800XpressMusic




As far as I checked most of parts are there, some nuts and washers are missing due to hole in package
Haven`t checked the printed parts, as it`s not so easy to do due to lack of partlist etc.
Will start my build somewhere next week and will be posting my progress.
-------------------------------------------------------------------------
So as I understand James is missing, and he doesn`t reply to emails ?
About the extruder design, as far as I have red in these threads there is no way it would work well t.i. give decent results ?
Is there anything else I should know ?

The extruder should work ok. It just takes a lot of tweaking of settings.

I made a parts list on the reprap wiki. Search for catalyst.

Hand written parts labels? Wow, couldn't come off the money for a $40 label printer? Glad to see you got your stuff!

---

repraplogphase.blogspot.com

Easy... he is improving. My bags were regular thin bags that were taped shut. No labels, hand-written or otherwise. The change to zip-lock bags was needed as almost all of my bags had split open.

Buzz

No this is after I sorted and labeled everything myself. They arrived in regular taped thin bags, many of them split open, same as buzzard192.

Edited 1 time(s). Last edit at 04/22/2010 04:34AM by joblafors.

Can someone help me with my Arduino ?
I followed instructions on Arduino homepage but cant get Blink example to work.

OS: Windows 7
My settings:


ERROR

Quote
Error output

Binary sketch size: 896 bytes (of a 30720 byte maximum)
C:\Users\Valters\Desktop\arduino-0018\hardware/tools/avr/bin/avrdude -CC:\Users\Valters\Desktop\arduino-0018\hardware/tools/avr/etc/avrdude.conf -v -v -v -v -patmega328p -cstk500v1 -P\\.\COM3 -b57600 -D -Uflash:w:C:\Users\Valters\AppData\Local\Temp\build2959345295801215766.tmp\Blink.cpp.hex:i


avrdude: Version 5.4-arduino, compiled on Oct 11 2007 at 19:12:32
        Copyright (c) 2000-2005 Brian Dean, [www.bdmicro.com]

        System wide configuration file is "C:\Users\Valters\Desktop\arduino-0018\hardware/tools/avr/etc/avrdude.conf"

        Using Port            : \\.\COM3
        Using Programmer      : stk500v1
        Overriding Baud Rate  : 57600
avrdude: ser_open(): setting dtr
avrdude: Send: 0 [30]   [20]
avrdude: Send: 0 [30]   [20]
avrdude: Send: 0 [30]   [20]
avrdude: Recv:
avrdude: stk500_getsync(): not in sync: resp=0x30
avrdude: Send: Q [51]   [20]
avrdude: Recv:
avrdude: stk500_disable(): protocol error, expect=0x14, resp=0x51

avrdude done.  Thank you.

I try pushing reset button before uploading but nothing works.
Posted this in Arduino forum, but no help jet.

KrisC Wrote:
-------------------------------------------------------
> While the Catalyst may not be the printer suited
> to every need, it does fill an important niche.
>
> So in order to leverage some community innovation,
> where do we want to discuss improvements? How
> about here.



Do you have some website stating what "Catalyst" is, who is developing it
and what the status is?
This thread seems to assume that you already know all that.

---

-------------------------------------------
* homeprototype free 3d design repository
* Blog
* Google+

Catalyst is the name James Villeneuve gave to his Mendel variant.

his site
reprap wiki

You can also search the forum for our ongoing discussions.

With the 4-axis controller james provides..

I have set my hardware stops at the axes.. But when I home the device the hardware stops are hit.. As supposed but then when you give the command to move away from the stop the hardware stop prevents the motor from running..

The stop decouples the motor from the current.. you can feel almost no resistance when the stop has hit..

Can we hook the stop up to the arduino directly.. I asume that the reprap software detects it and stops driving the stepper.. this way the current won;t get decoupled by hardware. but then the question is.. is that arranged in the firmware.. Still have trouble with witch firmware to start or perhaps do my first initial runs in Mach3 or simular...

I also have some other hardware laying around.. the reprap boards from kimberly.. perhaps that is a better option!


please comment if you have suggestions of comments ...

(Still waiting for my 100k thermistor, due to the vulcano eruption in iceland my mail is delayed !!)

with kind regards,

Michael

Hey all,

I've got the bot assembled, but I'm not quite sure about getting the steppers hooked up & tuned. I have the Z axis motors hooked up to the driver board in parallel, and the driver board is still configured to defaults (as per the manual)

What's odd, is the test software doesn't move the axes unless I have the board set to full stepping (even though the arduino firmware I'm using from James's site says to use 1/4 stepping)

Any thoughts on how the board should be tuned for current/PFD voltage/step sizes?

Many thanks,

Jes

Try wiring up one of the other steppers and trying it. When I first tested the Z it seemed not to move, but then I realize it was just moving a small amount and also the axle failed to connect with the x-mount.

No go, all 3 axes are already hooked up, and they all behave the same way unless I have them at full step. Just a loud buzz, with no motion.

I've tried all other stepping options, but haven't tried adjusting the current or PFD yet. As I said, these are still set to the board's default.

I presume we all got the same steppers, 1.6 ohm, 1.8 deg/step

Are the motors installed or loose? They might be binding? I am not certain as I didn't have that problem. You don't have wires reversed do you? Have you tried reversing only 1 pair of wires?

I did actually try reversing one pair. That made no difference in how it worked when set to full stepping, and no difference to how it didn't work at any other fraction of full stepping.

The entire kit is assembled, with the exception of the extruder. None of the motors are loose, and I'm not sure that they'd be binding if they all function the same at full stepping.

I'm going to check the arduino source to make sure that it IS configured for partial stepping. I'm used to PIC assembler, but haven't bothered to look too closely at the arduino firmware yet

I found a datasheet that's pretty close to those steppers here

The 103H548-5040 seems to be the closest, however the resitance of the coils is a bit off.

I haven't had the chance to tune my controller board yet, but hopefully in the next day I'll get to and post my results.

Jes

Ok, so I did some reading and matching up the specs in the datasheets and had a couple interesting findings.

1) I goofed when reading the dips. (ok maybe that isn't interesting) The board was actually configured for 1/8th steps and not full steps. It's full stepping that wasn't working, and according to the arduino source, this appears to be as designed.
2) The Z-Axis motors, being hooked up in parallel will consume twice as much current as the other steppers. I've tuned the board to provide 1.0 amp to the Z axis, while the X and Y axes appear to work fine at the factory default of 0.5 amps
3) The pinout on the motors is A+ A- B+ B-, while the pinout on the controller board is A+ A- B- B+ This means that they should not be wired straight through, but should be wired with either pair swapped. In my case I've rewired my motors so that the B pair is swapped.


In case this helps anyone, when facing the controller board with the stepper terminals at the bottom, each terminal corresponds to the stepper wires in this order: Orange, Blue, Yellow, Red

This is based off the datasheet I posted earlier, which isn't for these specific steppers but hopefully the wiring colors are consistent.

Regards,

Jes

What is the 'best' software to use?

Looking at the PD site several versions of software are available.. I have tried them all (I believe) but what is the best one to use without MACH3 or 2.. Just the plain arduino version..

So far I can move the bars and get the extruder to heat up .. but moving it 100mm is actually just moving 32mm so I am off *3 :-)

But where do I change this.. Tried it in the settings of the host software but that doesn't do much.. and the parameters.h of the firmware contains so many settings :-)

MENDEL is 1 etc.. I cannot find it as of yet..

So if anybody has a 'working' firmware with some correct settings. please tell me where to download. and then from that point I hope I can contribute :-)


EDIT Seems like I've got it under control now..

X Y and Z axis are up an running and moving the correct amount.. But needs a lot of tweaking!!! But getting somewhere in the end :-)

Edited 1 time(s). Last edit at 05/02/2010 05:43AM by mhensen.

ok.. software seems running.

And also the design exposes it's faults.

Like in the video I have quite a lot of ooze.. although I haven't been able to do more then 1 layer I see the oozing..

This can be resolved by removing the bowden cable as this is too soft material so it builds up a lot of pressure inside.

2nd is the pushing mechanisme.. A nice design but the components need to be accurate to work correctly. As I was missing some parts mine is not accurate!!

3th is the way the heater works.. it is bit-bang and the temp goes up 20-30C . No PMW..

I already ordered a wade extruder type so that would resolve the bowden issues as the the extruder and heated end can be combined without cable..

The hot end will be replaced by a variation that can be controled by PWM but then I hope someone can help me on this design with a mosfet in the form of a schema!!!

Hope this answers some questions about the design.. the x,y z is ok.. it is just the extruder section that needs some more attention!!

I received my Wade extruder.. I think this thing is essential to get the catalyst going.. The power of the profounddevices extruder is way to low.

I guess this step is worth every penny.. (thanks Rusty)..

It also is the best piece in my complete mendel and the only item that fits!
Now I have a luxery extruder, next up is a new heater circuit.
The current schema's can work but his circuit is giving me an temp of above 400C before I can extrude.. I will go for a more tested circuit ( no I never have been over the 300C as my thermistor is still ok!!! :-)


The rest of the Catalyst is a mess with wires and with tyreps.. (don't know the word:-)

So all in all.. How much as I like hobbying around I guess more documentation, accuracy and testing should have been done before selling.

As far as I see no catalyst has been printing correctly??

I tried the new software 2day but I have removed it too as the pinout is completly different and no docs are there on how to hook al up....
I personally am not looking for the fastest machine in the world but a machine that does work, and can print.. even at slow rates .. and then you always can improve.. So James, if you read this.. just a working version of the software in a low speed is all we need.. speeding up things is something that needs to be in both hard and software.. not just multipliers and higher baudrates..


But in the end I am positive in running the catalyst as a repstrap for my own mendel, I hope I can figure out all the software and spawm my own version of the formware that works, but as said before.. I hope it will just work and print decent objects instead of wanting to have the fastest mendel in the world, affin. with that extruder I cannot even run at the slowest speeds.. (the wade will, but that you need to be able to print it.. . chicken and egg story :-)..

A brand new mendel is what I am after, but self printed!! .. and looking at the quality of the wade extruder... wow .. hope to ever reach that build quality!!!


So this is my feeling about it all.. I have spent my 2 weeks of building but now 'real' extruding have been possible due to quality issues and loads of missing parts..

I hope that in the coming week I will make some progress and get it up and running. If this is the case I am willing to help out wherever I can.. in building the catalyst.. based on my own experiences ....

Edited 2 time(s). Last edit at 05/04/2010 04:36PM by mhensen.

Hey all. I'm running with the "Recommended" firmware from James's page, and have just gotten the extruder more or less working.

I'm still trying to better understand how to tune the Reprap host software to talk to the catalyst. (Currently running version 20100308) There don't seem to be many settings that apply to the pwm for the heater, and setting the feedrate for the extruder doesn't seem to have the desired effect. Instead of making it feed slower, it's making it feed for a shorter burst before stopping. Has anyone had much luck tuning the Reprap preferences? (not the arduino firmware, I think the Recommended code seems to be mostly correct)

Also, I've noticed that the host software seems to get stuck at a particular temperature reading, and that restarting the software seems to clear this up. I've read that it is known to be unstable on java 6, and I'll probably try 5 soon to see if this clears up.

Is anyone else this far in testing their catalyst?

-----------------------------------

An update, it appears that running the Arduino IDE while running the Reprap host leads to sporadic conflicts. Who knew?

That aside, the heater and thermistor circuit is working well. Instead of using the relay, I've made a simplified mosfet driver based on the reprap pwm heater driver. The only issue I'm having with the heater now is that the lag between the heater coming on and the thermistor detecting the difference is causing it to race a bit.

One other note, the extruder doesn't seem to have enough torque to feed ABS. I'm going to try machining a metal drive pulley which, with the smaller diameter should give me 3x the torque. This will instead rely on a spring loaded bearing to apply pressure to the feedstock, which I'm hoping can be hacked together without too much trouble.

------------------------------------

Another update, I've rebuilt the extruder barrel this time using Stove Cement instead of the Kapton tape to encase the heater wire as well as the thermistor. This seems to have almost eliminated the lag I mentioned before.

I've also made the described change to the extruder, now using a home-knurled gear driving the feedstock, and a spring-loaded arm that presses the extra bearing into the gear. I don't have any pictures up as of yet, but if anyone is interested I'd be more than happy to post a few. The only issue I've seen with this setup, is the wire starts getting torn apart by the worm gear if trying to feed it too fast. This was tested before I rebuilt the heater barrel as i mentioned above, and hopefully I'll have it ready to test later today. I suspect, due to the size of the orifice and the viscosity of the ABS that I may have to print at a slower speed, but I'll post more once I've tested this new setup.

Ciao for now,

Jesse

Edited 1 time(s). Last edit at 06/21/2010 02:39PM by AphoticJezter.

Hey all, time for a much overdue update on my progress. The hardware seems to be working well, however finding firmware that performs reliably has had me hung up for a while now. I've found that the Recommended firmware does not run reliably and as a result I started exploring alternatives.

The Arduino that comes with this kit, while upgraded to the Atmel 328P does not run the official firmware. Nor does the lions share of firmware seem to work on this platform. A few solutions to this have come along, including Triffid Hunter's firmware, as well as a few others. I've even been in contact with a fellow by the name of Jeremy who has successfully been printing using Skeinforge and a separate fork of the official firmware at [github.com]. Let me say that with a little time and effort, I'm very confident that a working setup using the 328P is possible.

That being said, I'm looking for something that will work with as little modification as possible and have purchased an Arduino Mega2560 which I'm shoehorning into my current setup. This should let me use the official firmware, and also provide plenty of room for expansion. Perhaps in time, I will even go back to the 328P and use the Mega for another project.

Anyone else having any success with this kit?