Major help needed - Multiple issues!!

This is for general discussions. Also FAQ can be found here.

Moderator: speedake95

Post Reply
stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Major help needed - Multiple issues!!

Post by stratogavster »

Hello All,

Hope there is some help out there because i'm about to throw my printer out the window....

Summary of issues are as follows:

1.) Prints never complete - sudden death between 2 and 4 hours
2.) Extruder randomly stops allowing filament through
3.) Regardless of setup / slicer prints 7 times out of 10 start in the same but wrong location regardless of programming - Never the intended location is used and pre-prepared gcode never seems to start in the same place consistently either.


My setup:

Dell M4600, Win7 64bit, 16GB RAM, Fully up to date, Not connected to internet normally, no AV to bog it down. Mac is my main machine so PC is devoid of crap. Power-profile customised to prevent all the usual bad things from happening - mainly USB suspend mode being disabled. Generally well maintained slick powerhouse for CAD and rendering...and for 3D printing?

Have a modern Mac laptop but can't leave that running for hours on end as it's my main laptop for working etc. So little use in testing that.

Actions taken:

1.)
- Thought active USB suspend mode was the culprit - Seems not as the issue remains!!!
- PC has almost zero background running tasks and is not used during printing.
- Have latest Repetier Host as of 15/11/2012
- Tried pre-prepared gcode, self-sliced code via SFACT and Felix profiles, and KISSslicer.
- Swapped USB ports
- Many PC reboots

No repeatable results or improvements found.

Suspect: Faulty RAMPS

2.)
- Endless tweaking and really have found a the best 'sweet-spot' with this finicky extruder. Extruder simply blocks randomly. Pliers and a push helps the filament through then it's back to normal....for how long who knows, but it always happens.
- Tried extrudes at different speeds from startup - all work well. Though do have blockages sometimes on manual extrude. Frustrating!!
- Tried manual long purge before print start (though always does pre-print dump)
- Tried temps from 195 to 215
- Using Felix supplied filament
- Used K-type temp sensor on the a barrel between the hotend tip and the next body section (about a 2mm gap). Barrel surface temp reports back as about 5 - 7 degrees lower than the software reads. Could mean nothing as the core temp could be spot on.

Suspect: Faulty Hotend tip and or reported temp is not really the junction temp at the plasticiser section, hence the filament is too solid to pass.

3.)
- Tried all types of gcode configurations - preprepared, custom, SFACT, KISSlicer, Csutom CAD files, Stick CAD files etc etc....
- Many reboots
- No pattern - but 7 out 10 times the print makes an error during the perimeter line and then prints over it.
- Some prints sometimes have zero X movement for the first few seconds then the print continues as normal until sudden death a few hours later.
- Have tuned all stepper pots perfectly, all consistent and no skipped steps ever.

Suspect: Faulty RAMPS


Need help please!! I don't want to throw my printer out the window it was expensive! and after weeks of owning it i've never printed anything yet!! I've covered all mechanical and software variables possible - I think

:(

satoer
Posts: 436
Joined: Mon Oct 01, 2012 8:00 pm
Location: Netherlands

Post by satoer »

2.) Extruder randomly stops allowing filament through
- Have tuned all stepper pots perfectly, all consistent and no skipped steps ever.
Image
Are you sure you adjusted the steppermotor perfectly? Try to turn the extruder steppermotor pot (right below in picture) a tiny-bit clockwise to give it more power.
Have you measured the hot end temperature with a infra-red thermometer?
Have you mounted the heatsinks on de steppermotor controllers with 2 component thermal GLUE? (and not with the supplied thermal paste)? Maybe it’s an overheating problem with the steppermotor controllers. And make sure the heatsinks do not touch the electrical pins
My Felix 1.5 printed designs on Thingiverse
E3D V6 Hot end upgrade for your Felix
Tweezer bracket for your Felix
Led bar upgrade for your Felix

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

Hi Satoer,

Yes have tweaked the stepper drivers quite a lot and have found the sweet-spot to be exactly where expected. Certainly not getting any missed steps, and the thermal output of the drivers seems to be managed well.

As for measuring with an IR temp sensor, i've not got access to one, but generally they are quite inaccurate as they are calibrated to a generic surface reflectivity - the hot end is shiny Alu so i'm not sure they would give an accurate reading of the core temp. Though eve if the temp is off by +/- 10 i'd still expect extrusion as the PLA has a large temp range.

As for stepper heat-sinks, i've literally just bought some thermal adhesive to stick them on as the current solution is kind of poor.

I did investigate the stepper driver heat issue, and I resulted to creating an ad hoc heat-sink that was absolutely huge to remove the temperature issue from the equation....sadly no change.

Though after saying all the above i've just managed to get a single 14min print to complete successfully....it printed in the in the correct location....shame it took 3 goes!

Next test is a pre-dfeined gcode from the forum

Kjetilei
Posts: 334
Joined: Fri Mar 16, 2012 8:26 pm
Location: Stavanger, Norway

Post by Kjetilei »

Hi stratogavster,

Sorry to hear about the continuous problems with your Felix!

My bet would be that you received a dodgy RAMPS/Arduino board or stepper driver. It could also be caused by the drivers being tuned too low resulting in overheating stepper motor(s).

Try to use your Mac with Repetier Host and a pre sliced gcode file from Guillaume (there are several around this forum) to eliminate the PC from the equation.

Regarding the extruder jams: can you hand feed the filament to get some plastic extruding when it has jammed? If that is the case I would think that the problem is not the hot end but somewhere in the stepper/RAMPS setup. I had a lot of issues with my first hot end - an early release. When that jammed the plastic was stuck and hand feeding did not help.

Have you contacted Guillaume regarding advice/solutions/replacement part(s)?

satoer
Posts: 436
Joined: Mon Oct 01, 2012 8:00 pm
Location: Netherlands

Post by satoer »

Okay, so if the steppermotor controller isn’t the problem and the steppermotor isn’t missing any steps., and the temperature of the hot-end is also in the good range… have you adjusted the tension on the fillment bearing? Maybe that needs to be a bit more. Have you tried dismantling the hot-end? The aluminum parts should be firmly screwed in the peek (plastic part). If there’s room between the peek and the upper part, heated fillmament can come between this space and jam the fillament.

The other problem with the moving starting point looks like some kind of calibration problem. The printer calibrates the starting points every time you start a new job. Have you put some aluminum foil on the sensor-blocking arms? Does the led on the optical sensors goes off when the sensor blocking arm goes inside the sensor?
My Felix 1.5 printed designs on Thingiverse
E3D V6 Hot end upgrade for your Felix
Tweezer bracket for your Felix
Led bar upgrade for your Felix

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

Hi Kjetilel,

Indeed the problems go on!! but I hope to sort them soon....

Really am thinking a dodgy RAMPS is a big factor now as i've covered everything countless times :(

Haven't tried my Mac with the printer for a while as I had never intended to use it long term as my Mac is my everyday computer.....but as suggested I think it's a good trouble shooting tool. Though that said I forgot to mention in my plea for help that I had ordered an LCD / SD controller from Reprapdiscount which i'd not managed to try to get working.....this is another part of the equation I think i'll attempt to get working in an effort to solve other issues.

Had not pursued the LCD / SD controller as I had thought the last thing I needed was another set of variables....but don't have much to loose now!

As for the extruder, it seems there is a blockage internally, the drive gear obviously fails to grip, but if i grab it with thin-nose pliers and give it a really strong push I can overcome the blockage with my force, there after the stepper can take over....

***
Crazy idea now...

Further to the blockage issues - and to somewhat 'quality' issues in general....

This might be a strange question / concept for some but.....I'm assuming the internal profile of the extruder at the point where it reduces in diameter to 0.35mm is that of a standard drill bit nose?....Also any ideas how long the 0.35mm hole is?

The reason I ask such a strange question is this....(maybe this is for another post somewhere else)

I saw an incredible video related to flow / flow analysis and curvature continuity in 3D CAD (sorry geeky stuff related to what I do in my non 3d printing life), and there was an awesome example of a sand-filled-hour-glass...stick with me...

The demonstrator went on to prove that the shape of the glass was critical to get continuous flow, prevent blockages and thus have reliable timing.

A glass with a straight wall was made to show the contrast.....the standard hour-glass shape has the exact curvature required to allow the correct flow, and thus no blockages....the straight walled hourglass kept blocking

Now blue sky thinking here.....Considering the small scale inside the extruder tip does anybody think this theory would apply....???

An internally profile cut to promote smoother flow.....it just seems a drastic change from 1.75mm to 0.35mm

Maybe crazy theory.....one for testing perhaps?


***

Anyways....back to my testing....Mac and then LCD / SD controller


To Satoer,

Haven't tried improving the opacity of the opto-sensor 'fins' but think i'll give it a go.....as you say I think my issues lies in the job start calibration.....though all seems normal.

I really do think i've got a dodgy RAMPS or opto-sensor :(

Kjetilei
Posts: 334
Joined: Fri Mar 16, 2012 8:26 pm
Location: Stavanger, Norway

Post by Kjetilei »

I don't think I would bother with the LCD/SD board just now. Try the Mac and see if that changes anything. Most likely not :(

Extruder: I don't know how the hole is drilled out but it seems like Voodoo with many opinons about lenghts and step downs of diameter etc.
I forgot to ask if the stepper extruder stops when it jams or if it continues and "chew" out a piece of the filament?. The Jamming could indeed be caused by to little tension on the extruder arm. Try to tighten it firmly (you might hear the stepper start skipping if you tighten it too much).

From my experience with the sensor fin not triggering the opto sensor (due to translucent plastic) you will hear loud noises from the belt/pulley when the axis hits it's mechanical limit (bumps into something and can not move any further but the stepper continues on since it has no knowledge of it). If you do not hear any such sounds I would not believe the opto sensor to be the problem. You can easily find out if the sensors works ok by triggering them prematurely by blocking them. I use the tweezer (carefully) ;)

Could yo describe the "sudden death" and starts in different location" problems. Does all steppers stop to move but the LEDs on the RAMPS and opto sensors stays lit? Does it vary on witch axis the object "moves". Does it affect the z-axis also?

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

Have an update.....some mixed news...

Current opinion - Broken RAMPS & Dell M4600 poor USB implementation causing some issues (though not fully proven yet)

Course of action taken and results:

1.) Got LCD working
- Seemed to work well, managed to print for around 30 mins a few times, but forced a stop to restart each time to cycle the 'stop / start'. Seemed to do what I wanted....fairly happy
- Prints still started in the wrong place a lot!! Still something wrong
(quality seemed much better than being driven by the PC)
- Seemed repeatable but didn't see any prints 100% through as I had other things to test.

2.) Switced back to PC
- Same as usual
- BUT BIG SPOT!! Noticed X endstop was reporting it had been hit sometimes - though printing was continuing ...strange, is that normal!!

Corrective action....

3.) Decided to tidy up wires to mitigate the random endstop reporting (and bond heatsinks in place)
- Full rewire
- Bonded heatsinks

Super tidy now!!

Result - WORSE!

:(

- Now none of the axis will do what I want!
- They all home - but not properly!! only the Y homes properly...ie. it hits home then jumps back. X and Z just go to home and stop and the motors grind.
- No axis responds properly to the controls.
- LCD not working in this state either and steppers not responding to manual control.
- ALL ENDSTOPS SEEM TO WORK if the LED status is the judgement. I have tried all Endstops manually by placing a piece of paper in the slot while the axis is moving. The axis stops.

4.) Reuploaded stock Felix Firmware (as LCD can come later)
- Tried Mac - same as PC.....axis do not home properly.
- No printer functionality.


Oh dear! it's broken.....

2nd rewire.....

Same!! :( I should take a photo to show how super tidy the wiring is....all wires are tinned, cutback and have heat-shrink strain relief. All endstops have the same treatment, nicely tinned ends and neatly placed and crimped by connectors.

Regarding optosensor vs translucent plastic - mine is black and no light passes....have tried wrapping tape around the 'fins' no change.

Oh dear! it's still broken.....


My theory......

RAMPS / Arduino broken (maybe endstops are suspect too?)


Also as the LCD / SD controller performance was much better a less problemmatic than the Dell M4600, I feel the Dell is contributing to the failures i've had. Perhaps though poor USB implementation by Dell - something not right with....dam I hate PC's...

Have decided to implement the LCD / SD controller longer term but sadly my printer is now worse....

Any tips on getting it back to how it was...even though that wasn't 100%.....because i'm stumped now!! it's just simple wiring....no rocket science....

Any prescriptive tests I can do on the Endstops / RAMPS / Arduino with a multimeter?

Thanks!! :D

satoer
Posts: 436
Joined: Mon Oct 01, 2012 8:00 pm
Location: Netherlands

Post by satoer »

If you rewire the whole thing, and the result of that is that things change, it sounds like there is something wrong with the wiring. But there's quite a lot wrong with youre unit, to much for just one faulty wire. And you have to remove and refit the ramps board to rewire so there may be a mecanical electric problem there. Have you inspect all the connectors on the ramp? Are the pins from the ramps correctly in the Aduino? Is the reset button on the arduino exacty alligned with the hole?

Maybe try to switch places between the steppermotor controllers to see if things change, to exclude the controllers?

Is it posible to make a video with all the problems and post it on youtube?
My Felix 1.5 printed designs on Thingiverse
E3D V6 Hot end upgrade for your Felix
Tweezer bracket for your Felix
Led bar upgrade for your Felix

stratogavster
Posts: 120
Joined: Tue Jul 17, 2012 9:04 pm
Location: United Kingdom

Post by stratogavster »

Exactly my thoughts, obviously some underlying issue at play here, and what is causing the issues...the wiring is far too simple to be a major contributor....it's super simple, I think all i've done is disturbed an unsettled dry-joint or cracked trace on the PCB

Jumped ahead a bit and actually reflowed some of the suspect soldering on the base of the RAMPS and Arduino. Must admit these electronics are real low quality! the standard of soldering is very low!

Next rewire....sadly the same..... :(

Next step....swap drivers and optosensors....

Another fault finding question....are there any key voltages etc to be found at various pins and terminals? Perhaps a more strategic test method is need to got to the root of the issues.

Thanks again to everybody!! :)

Post Reply

Return to “General”