Acme Screw & Nut

[Kendall]

Quit a while back, someone [Post]ed that they had purchased an Acme Screw & Nut.  Can we get an update on how that is working out...

- What Size did you purchase (Dia/TPI)?
- What type of material is the screw?
- Was the thread class 'G' or 'C'?
- Was the Nut a Plastic Anti-backlash Nut (or) Standard?
- Did this Mod fix your Banding issue (Better/Same/Worse)?
- How did you couple it to the stepper (Rigid/Flex)?
- How did you attach it to the bed?
- Did you support it at the top?

Pictures would be nice...

[Kendall]

Was this person not my Waco Neighbor?

[RonSII]

Cwalster on soliforum was doing that not sure if he is on here or not???

[Kendall]

Thanks, If you could post which thread, I will try and contact him over there...

I have ordered a bunch of parts to create a new metal bed frame and guidance system, and would like to incorporate these thoughts into my new SS bed concept project...

I looked back on just how long we have been complaining about the Banding issue, and SD's comments, way back then (almost the beginning of SD time) said; Set tight Guys, we're working on a solution, and your really going to be happy with it.  The solution did finally come, but it didn't seem to correct the problem, and (or should I say 'so') all the bad screws were not replaced (some people who did get them, said the banding got worse for them and installed their old screw back in, some it cured).  And the problem still exists today, 4+ months later.  A few days ago I see the same comment from SD again, only this time with an attitude.  Y'all have seen the results that my machine produces (non-sellable parts, b/c of banding), and video's of the screw going around catawompus.  I did request a new screw, but they have made a decision; "Compared to some cases we've seen your wobble doesn't look very bad."  But, they 'are' still thinking about it.  That thinking process of; "Sit & Wait", has not a good track record so...

__________________________________________________________________________

[Ian Johnson]

To be fair, the original solution did work quite well at first, and then it got torpedoed by bad screws.  Now their problem is how to detect bad ones without actually installing and printing with them.  It might be more expensive in costs to go with a higher quality part from a supplier that doesn't need to be watched like a hawk, but not biting the bullet and doing it might cost the company much more in reputation as they keep sending out bad screws while trying to solve the problem.

I hope they do something before they start shipping the S3.  They don't want people showing off S3 prints that look like this -

[Kendall]

Sorting for quality is very expensive manufacturing practice, if it is pinned down to the screw assembly being the culprit; if I were the process engineer, a few questions I would be finding answers for:

- Is the screw straight (bowed to start with)?
- Is it the screws threads off centric (pitch diameter an oval)?
- Is it the coupling attachment alignment (center holes not aligning properly)?
- Could it be the assembly approach of attaching the motor to the screw (maybe it needs fixturing during the gluing or a mechanical insertion approach to ensure alignment (items that can be mechanically adjusted to ensure part to part consistency))?
- Does the quality engineer have an accurate way measuring acceptable components, before each step of the assembly happens?

These items need to be known prior to assembly, or inspected prior, before it's stage of assembly.  Once value has been added during the assembly process, it's a $ killer to correct...

[Jeffery Sanders]

http://www.ebay.com/itm/Haydon-Kerk-CNC-Lead-Screw-40-TPI-Diameter-1-4-6-mm-/130508876979?pt=LH_DefaultDomain_0&hash=item1e62ef6cb3  This is exactly what I bought, and like I said in the other thread; I'm pretty sure I am going to save it for the DLP projector-resin printer I have 90% of the parts for.

[Kendall]

Kool, after you get started, I'd like to see it (I have made the trip to Baylor from my home in 1.5 hours)...  If you need to do any specialized machining, I have some equipment your welcome to use, just by asking (1.5hp CNC Mill, 1.5hp 10" x 26" manual lathe w/DRO, and a 1.5hp small manual mill w/DRO)...

I have purchased (3) 1/4-20 Acme Lead Screw and (2) Acme Anti-backlash Nuts (tried to change my order to (3) nuts, but they shipped to fast) http://www.roton.com/Mating_Components.aspx?family=7060758, and (3) flexible shaft connectors http://www.ebay.com/itm/1PCS-5x6-35mm-DR-Shaft-Coupler-5mm-6-35mm-Flexible-Shaft-Coupling-L25mmxD20mm-/370552725472?pt=LH_DefaultDomain_0&hash=item5646a9dfe0.  I am still thinking about MakerSlide or OpenRail for the guidance system.  Think I am sold on the SS for the printing bed, and considering it for the bed frame also, but would like to use a different type of bed heater that I have been reading about...

______________________________________________

[Kyle K]

Kendall-When do you plan on putting all this together.  I am very interested in your solution and if it looks like a winner I may follow your lead.  It looks like to me that if you install those parts and still see any banding we can rule out bad screws as the culprit.  

[Kendall]

My intent is to come up with a totally different all metal bed design (separate project), self contained, that can just plug and play, without altering anything with the SD2 (this design would not be cost effective for Solidoodle in production vs. trying to keep the cost down), this is a personal project only.  If/When I get this going, the SD2 bed and Z movement will be out of the machine, will still be intact, then I will look for a non-invasive way to modify without having to alter any of the SD2 part from what they presently are...

I do believe that the banding is coming from the screw/nut design concept, and I do believe that the banding can be redesigned out.  Ian, way back in Aug/Sept time frame, hit the ball right on the head, just none of us pick up on it and followed through.  The function is only to lift or retract the bed without backlash, the wobble can still exist without effecting these functions.  This can be redesigned to perform the function as needed, but it would require some add-on modification to the printer...

Results: Better; But you would still be at the mercy of the sloppy bushings, guide pin alignment hole clearances (all of which; with the present spacing, is a magnified amount, out where actual work is being performed).  The ever-changing humility vulnerable wooden frame would still be in the picture.  And you would still be printing on Kapton Tape (or) converting over to Glass (b/c most of the aluminum bed plates shipped are not flat enough to get to fine layers or with some printer, large parts.  Actually, by inspecting my bed and finding areas that were totally flat, fine tuning the Z stop, I was able to create a Slic3r profile for printing .05mm (.002") layers, it looked wonderful, except for the dreaded banding...

[Kendall]

If anyone want to just take off on some theory, here are my thoughts:

- Trap (2) Nuts with a wave spring between that has side clearance to the Nut, so 'sideways' movement doesn't apply side force)
- Crown the top Nut and bottom Nut, so 'catawompus' movement doesn't apply up/down force
- Sandwich the above between (2) stationary surfaces

I believe the area that has to hold the nuts from turning can engineered in such a manner that wouldn't restrict this mechanism from floating (the Nuts do not have to be standard nuts, they could be a tapped thru hole in a piece of material...

And a goal should be; this this mechanism could still bolt up with the same screws and screw hole locations, allowing, the SD2 to return to its factory delivered configuration at any time...

[Ian Johnson]

How do you hold the nuts firm enough against the stationary sufaces without dragging them along during wobble?  Is there a kind of low friction washer that gets used in this kind of application?

[Caswal Parker]

I have been thinking along similar lines of how to decouple X/Y from Z. I talked about it over here: http://www.soliforum.com/topic/994/crazy-z-suspension-carriage-concept-to-reduce-banding/

The other itemI have pictured there, is a 3rd rail attached to the front of the SD, using the hinge mounting points. It does not restrict X/Y Extruder movement, and with a bit of a design change, would not restrict the bed movement either.

With the acme lead screws, how are the ends finished, or is it just cut threaded rod?

[Kendall]

@ Caswal:  What I ordered is just Acme Threaded Rod & Anti-Backlash Nut, I have a small machine shop and plan on making my own alterations.  The anti-backlash nut only has a 5 pound overcome, this may be a problem and also something for me to have to engineer a work-around for.  Also, just going to an Acme Thread does not eliminate the thread pitch side torque, just that it is lifting on a flat surface, instead of the 30° thread surface (that makes it have multipliers in side torque)...

@ Caswal:  I'm not totally convinced that everything will be fixed by floating the Z mechanism, the banding may be also be exaggerated by all the other non stable items, your 3rd guidance suggestion may be required in addition too, and that still doesn't take into considerations the X carriage slop that could also be adding to our problem...

@Ian:  I'm sure that the choice of materials could be crucial, but we have been sliding steel on a dis-similar steel for years with only lubes between.  But yes, there will be some resistance, and with all the other slop in the guidance system, will be hard make one fix, fix all.  The Screws inconsistencies is the driving problem, the others are what is allowing it to happen.  Maybe Caswal's 3rd leg guidance approach (and You suggested this way back in the beginning also), may have to be the included in the solution (this could be a clamp on to the frame top & frame bottom, and clamp on the bed non-evasive prototype for testing results)...

I'm getting embarrassed with how many times I have been saying, "this was suggested way back in the beginning"; 1st rule of Trouble-Shooting; if your approach is not producing results, clear your mind, then revisit your first thoughts, those 1st thoughts, more than likely are the Root Cause...

Conclusion thoughts:
Another thought; would any of this be happening (or that noticeable) if we could control acceleration and de-acceleration (in the CNC world it's called; Ramping), as it makes a changes in direction, ramps down to stop, then ramps up to speed in the new direction (in my Mach3 Software, Called Tuning the Motors).  Eventually, I would like to do an experiment with changing my Controller Software to Mach3 (why, b/c I understand it), but have strayed away from this, b/c also eventually, we would all like to get too dis-buckling the computer, and making our printers, stand alone...

[Ian Johnson]

There is acceleration control, and you can access it through the EEPROM configuration in Repetier.  Here is a brief discussion of it- Jerk controls the start and end speed, which is then followed by acceleration toward the feedrate, which may or may not be reached before the end of the segment - from blog.lincomatic.com

3. Acceleration and Jerk

If your printer vibrates a lot when printing at higher feedrates, causing wiggly prints, you can improve the quality without slowing down your prints too badly by lowering the acceleration and/or XY-jerk values. In Marlin, acceleration is specified in mm/s^2. The default is 3000. On my Printrbot, I slowed it down to 1000 with G-code as follows:

M201 X1000 Y1000; // max accel print
M202 X1000 Y1000; // max accel travel
M204 S1000; // default accel for normal moves

You can also change the value in Configuration.h, if you prefer:

#define DEFAULT_ACCELERATION 1000 // X, Y, Z and E max acceleration in mm/s^2 for printing moves

If you have a slow extruder, you can also slow down the retraction acceleration, but generally, there’s no reason do do this:

#define DEFAULT_RETRACT_ACCELERATION  3000   // X, Y, Z and E max acceleration in mm/s^2 for r retracts

3000 is the default value.

The jerk value is also useful, and affects the speed at which Marlin joins two segments. The default XY-jerk value is 20 mm/s.  Lowering the value will slow down the transitions between segments.  This is especially helpful for smoothing out ringing waviness on 90 degree turns.  I found that 15mm/s is a good balance between speed an quality on my Printrbot.  G-code:

M205 X15; // max xyjerk mm/s

Configuration.h:

#define DEFAULT_XYJERK                20.0    // (mm/sec)

There are also values for the z-axis and extruder, but I didn’t need to change them:

#define DEFAULT_ZJERK                 0.4     // (mm/sec)
#define DEFAULT_EJERK                 5.0    // (mm/sec)

When tuning acceleration and jerk values, a handy way to test the effects of your new settings is the Dry Run feature of Repetier Host.  When Dry Run is enabled, the extruder is disabled while printing, so you can observe its motion without wasting filament.

After tuning the acceleration and jerk values my prints are better quality at higher feedrates, without sacrificing too much speed.  Note that depending on your print, the lower acceleration/jerk values will have more or less effect on the total time it takes to complete it.  Prints that have more short segments will be slowed down a bit more.

Even though my discussion was about slowing down the printer, you can also tune the acceleration/jerk parameters to speed things up, if you have a particularly well-tuned bot.

[Kendall]

Ian, Thanks...

This will give me some place to start, no one ever explained it to me, or if they did, it went right over my head.  I do see (think I see) that you don't know the hill of ramp (in Mach3 it's Ramp to Speed over Time; which lets you build a specific graph hill), but playing around with it should allow me an understanding on how it works, now that I know how to make changes...

One last question, is the board reading the configuration.h file, or do I have to up-date the firmware each time I make a change?

[Ian Johnson]

You can update the values using gcodes, for instance in the start file.  These will be reset when the board resets or powers off.

You can write the values into EEPROM through the config in Repetier, or there is also a gcode command that writes the current settings to EEPROM.  This is "above" firmware.  If you have these acceleration values, or steps/mm etc written in EEPROM, they will remain even if you reflash your firmware.  This is handy because you don't have to redo your custom settings every time you update the firmware.  There is also a gcode command that will reset the EEPROM to "Factory Default" if you want to. 

The EEPROM is great because you can quickly change the settins and see the result without needing to upload the firmware every time.

[Kendall]

Ok, thanks, think I get the jest...  Will give it a try, probably Monday, getting ready for tomorrows Bass Fishing Tourney on Lake Austin, and usually, I'm really pooped the day after and don't feel like doing anything that requires thinking...

[Kendall]

Just received my Acme Screws, set them on a surface plate and rolled them, perfect, they just keep rolling and rolling, no bend.  Chucked them up in my lathe, check them for true with an indicator, within .0005" run-out (that might be my lathe), started the anti-backlash nut on, and started the lathe, they screwed right up, no wiggle feel, no vibration feel, just felt smooth...
I AM A HAPPY CAMPER...

The Flex-Coupling and MakerSlide components have been shipped, can wait...

Would like to purchase a Z stepper motor, so I can leave my SD2 bed intact, does anyone know what the spec's are for a replacement SD2's Z Stepper Motor?  Maybe they would sell me one, but almost afraid to ask...

[Ian Johnson]

I don't still have the link, but I found the motor specs pretty quick by Googling the number on the sticker.  The extruder motor is still a mystery, there's no sticker on that one.

[Kendall]

Motors are everywhere, from below $10 bucks to over $75 bucks...  Tried to match by measurements on physical size, but no luck finding a match.  What I was hoping to find is what Model (??? NEMA 17 ???), Voltage (?), Amp Draw (?), Rating (?), and what ever other items that need to be specified (electronics stuff is out of my field and way over my head)...  The mounting screw pattern and actual physical size does not matter, this can be to suit (as long as it has a 5mm shaft), since this is a new Z axis movement design, what matters is that it will plug into the existing board with matching Voltage/Amp draw spec's, and have the same power as the SD Z Stepper.  Don't have to, but would like to leave my SD2 bed intact, so it can all be put back as original, and use a totally new Stepper on my new assembly.  My new screw is 20 TPI, so it should not need as much HP, and know, will have to mess with the steps per unit of measure, to get back in the correct movement distance (this is mechanical, probably can handle this)...

I am drawing it all in Solidworks, if it works out, will find out what it takes to copyright it into Open Source Technology, and do so, then publish it...

[Ian Johnson]

Here is a page with the Z motor specs- http://www.smartautomation.com.cn/ProductShow.asp?ArticleID=497   It's model sm42ht38-1684a according to the sticker on the motor, and this page was the first hit in Google.  There might be other suppliers, or other motors with similar specs.

[Jeffery Sanders]

http://calculator.josefprusa.cz/ this has a calculator for figuring approx z figures needed (it will give you a ballpark to start from on that new 20tpi LS)

[Jeffery Sanders]

sorry I lied...he does not have a 20tpi formula setup already...

[Josh Smith]

http://store.solidoodle.com/index.php?route=product/product&path=63&product_id=70

They sell the z motor on the site. Their spare parts selection is way bigger now. I stocked up on hardware this week including and extra carriage and rod assembly for some fun modding. 

[Ian Johnson]

The problem is that motor has the rod attached, and you probably won't want the rod. 

[Josh Smith]

I know the picture shows it attached, but I'm not sure if it is since it doesn't say z assembly. I'm hoping the print carriage comes with the x motor since it is shown on the picture there. It would be sort of misleading if it wasn't on there, but who knows. A few of the items aren't perfectly clear about what is included. Anyways, might not be much more than finding the z motor somewhere else and would definitely have mounting screws' holes where they are supposed to be. It would be nice if Solidoodle shared a reliable video of how to remove the rod since some find it easy to get off and some can't pop it off. I haven't been able to remove mine yet either. 

[Support Tech]

Josh,
Which items are unclear? We'll change them in store straight away if that's the case.

Regards,
John

[Josh Smith]

Hey John the motors, the extruder assembly, and the print carriage are unclear to me. Do the motors all come with corresponding fittings?  ( the pulley for the y, the toothed knob for the extruder, and the rod for the z). It looks like the extruder assembly comes with the x motor, does it also have the fan?  Does the print carriage come with just the x motor and just the x rods and extruder mount or the y rods as well?  Or maybe just say it comes as in the picture.  Hate to hijack Kendall's thread without answering his question so can you get the z without the rod?  Maybe one last suggestion for offerings would be to offer replacement wires from the board to the thermistor and hot end as well as the belts.  Thanks for the quick reply and hopefully this helps others. 

[Support Tech]

Josh,
We'll amend the store, but as a general rule it is a "What you see is what you get" scenario. If it is in the picture, you are getting it. I'll answer your questions one by one here to clear things up:

On Sat, Jan 12, 2013 at 11:23 AM, Josh Smith <onejos...@gmail.com> wrote:

Hey John the motors, the extruder assembly, and the print carriage are unclear to me. Do the motors all come with corresponding fittings?

No, just the motor. 

 ( the pulley for the y, the toothed knob for the extruder, and the rod for the z). It looks like the extruder assembly comes with the x motor, does it also have the fan?

No fan, just the extruder. 

 Does the print carriage come with just the x motor and just the x rods and extruder mount or the y rods as well?

The print carriage does come with the X motor. 

 Or maybe just say it comes as in the picture.  Hate to hijack Kendall's thread without answering his question so can you get the z without the rod?

Z comes with the rod attached. It isn't too hard to tease it out if you want a clean one. 

 Maybe one last suggestion for offerings would be to offer replacement wires from the board to the thermistor and hot end as well as the belts.

We'll take it under consideration. Belts are most likely a better possibility than the wires. 

 Thanks for the quick reply and hopefully this helps others. 

No problem! We're always here to help.

Regards,
John

[jon bondy]

On Saturday, January 12, 2013 11:32:26 AM UTC-5, Support Tech wrote:

On Sat, Jan 12, 2013 at 11:23 AM, Josh Smith <onejos...@gmail.com> wrote:

Hey John the motors, the extruder assembly, and the print carriage are unclear to me. Do the motors all come with corresponding fittings?

No, just the motor. 

This response makes no sense to me, John.
 

 Maybe one last suggestion for offerings would be to offer replacement wires from the board to the thermistor and hot end as well as the belts.

We'll take it under consideration. Belts are most likely a better possibility than the wires.

I was sent wire pairs from the controller to the hot end when mine failed.  I know that you are capable of doing this, John.

Jon

[Kendall]

Ian, Whoa; an internet search on this steppers spec's all lead back to Mr. Leo Shen.  It does appear that they will sell in groups of (1) for around $10 bucks, but I have not received an official quote back as of yet.  Just sent support (John) an email asking if SD will sell me one without the screw, guess this will also be a wait and see.  If anyone knows a US source for this stepper (Z) motor, please post it here for me...  Thanks, Kendall

[derelictmybawls]

Do you really want to purchase the exact wires over and over since they seem to be a weak spot? They are probably doing us a favor by not offering them for sale

[Kendall]

Suggested as a possible fix for banding on a different forum, I deiced to order a few.  No way is this going to work, as well as have great flex for alignment, it also has flex in the Z direction, both stretch and compression...

[Kendall]

Finally, all the purchased parts and material are in (still waiting for SD's Z motor & Switch, but it has been shipped), I will get busy and finish the design, and machine up the parts this week coming...  The ACME screw and Anti-Backlash nut run real smooth, and the rod is straight.  Have assembled the Linear Motion Slide Mechanism feel real smooth, so this all make me a happy camper...

[Joey]

Just got a replacement z axis and rod assembly from solidoodle and it is still wobbly. I can see it wobbling when i actuate Z. When you get your new Z axis installed can you post a few pics, some general assembly instructions, and maybe a BOM so we can do the mod? I just want my Z to be nice and straight and I don't really want to mod the firmware just to compensate for banding like some others have done. Thanks Kendall.

[Kendall]

I am building a totally new linear type Z axis.  Have purchased (3) Acme Screws and Anti-backlash nuts, so I will have left-over parts.  After I analyze my new Z axis, and 'if' this cures the problem, then I plan on starting to rework the (old) SD2 assembly; working in an Acme Screw with the Anti-backlash Nut (floating in the X-Y directions, trapped in the Z direction (with adjustable trapped Z direction spacing, so it can move X-Y, and still be able to control the amount of slop)), and a means of floating the motor position to get parallel to the guide pins alignment...

Allot of different theory's floating around about banding, we all know it has something to do with the screw, but no one has pinned it down with an easy fix, that can be applied by just anyone in our community (and that SD can apply to their manufacturing practice without additional cost (or at least a very minimal cost).  I still believe; it is how the rolled threads are manufactured, their tolerances, how straight the screw is, motor alignment positioning (a little non-parallel to the guide pins, and everything gets exaggerated), the amount of slop in the guide bushings, and so on, and on, the possibilities (or combinations, for possibilities of non-adjustable problems causes)...

Basically, for the new linear motion, I am using purchased components (and in-stock materials) that are larger and bulkier than I would like, b/c for ease of this test and $ vested.  If it works, I will concentrate on altering the existing SD2 assembly with fixes that can be easily done by anyone.  My unproven design looks something like this right now:

This design can just slip in with altering any of the SD2 components (including motor mounting hole alignment position), and does not limit the amount of Z movement (travels 6+ inches).  I have some bed assembly proposals made up, but still looking at easily adjustable options, so the existing bed can just drop in without any SD2 alterations.  By the time my present work schedule clears to where I can get back working on it, the last of the components will be in-house (long lead item; 3mm screws for the Z motor assembly (motor was shipped without the screw that hold it together? why???, don't know)...

____________________________________

[Joey]

Yes, I agree. Some folks have done some firmware fixes, but I haven't had time to read up on them.  And I'd rather have a hardware fix in place anyway.

Cool. Looks good. Will be looking forward to see what you come up with.

[Kendall]

They are saying "Totally Gone", in the pictures I have seen, it still shows, but I do agree, the results are better...

[Jeffery Sanders]

I would say the firmware fix was 100% needed.  Afterwards apply hardware fixes for the inadequacy of design.  I took off my AB fixes and went with firmware only and it is almost perfect...I bet it gets adopted by the main Marlin/Sailfish firmware repos as well of submitted, it is that good.

[Support Tech]

Here's the thing:
The nature of this firmware mod is that it accounts for a very specific wobble situation. While we haven't done extensive testing I will say that it is unlikely to fix the issue on all machines. Again, if the banding issue was as one dimensional as this it would be quite an easy fix.

 I should reiterate while I am here that Solidoodle does have an official github page (github.com/solidoodle) and you can submit proposed bugs / bug fixes there.

On Sun, Feb 17, 2013 at 3:07 AM, Jeffery Sanders <jeffery...@gmail.com> wrote:

I would say the firmware fix was 100% needed.  Afterwards apply hardware fixes for the inadequacy of design.  I took off my AB fixes and went with firmware only and it is almost perfect...I bet it gets adopted by the main Marlin/Sailfish firmware repos as well of submitted, it is that good.

--
You received this message because you are subscribed to the Google Groups "Solidoodle" group.
To unsubscribe from this group and stop receiving emails from it, send an email to solidoodle+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[nickythegreek]

The firmware banding firmware fix, coupled with tightening my y-belts and kendalls z-backlash nut have fully resolved my banding issues.  It might not work for everyone, but I bet it will work for at least 50-60% of the banding issues that most users are experiencing.

http://f.cl.ly/items/421q3v451f3p0Y3R1s1f/photo.JPG

you can see in that print where I sent the manual control firmware mod correction gcode line.

[Jeffery Sanders]

"While we haven't done extensive testing I will say that it is unlikely to fix the issue on all machines." Statements like this have me wondering what your R&D guys actually do besides print MLP keychains, bathe in $$$$, and jack off...you continue to ship printer's with a many major faults, lie to customers (I never got what was promised to me, nor has anyone else and that is meant specifically and generally [parts that were supposed to be sent never were included or shipped]),your company maintains a close door policy on product development  when barely any part is even unique and mostly inferior to the OS designs they were derived from especially in light of equal options, expect other's who are not compensated to fix the design inadequacies (dodgy wiring, dangerous thermistor locations, the extruder acrylic is horribly brittle and people stuck with the loltastic heated bed resistor can expect random failures on any print because it is insufficient to generate the heat need to properly affix any large parts)...you can say the problem is, "unlikely to fix all machines" well duh that is common logic, but I have not seen a single print done on a SD branded printer including the parts used to build my printer that were banding free until the kind gentleman came up with a solution...If it does not fix it; then you shipped a rod that was bent or the point were the motor connects to the frame is flexing and causing the wobble (those seem to be the only other problems seen causing z banding (we are not talking about backlash here, just the horrible banding pattern that caused weak dimensionally incorrect parts [circle/hole consistency is also improved with this fix] the surface aesthetics no longer appear sinusoidally banded). It makes the usage of cheap threaded rod which has been cut and rammed onto a shaft indiscriminately (my guess is your worker's or supplier cut each rod in a slightly different spot so the threads are never the exact same length or start in the same location on the thread [that's obvious really].  This fix is good for everyone using cheap threaded rod and tdp not only SD printer users, it gives everyone another option to the somewhat effective mechanical band-aids. I have never seen even layer lines on this printer until I applied this fix, the strength of my prints was also increased since there are no inconsistent layers anymore. Like night and day. I would say my prints were slightly embarrassing before, but now I would gladly sell even lightly sanded parts I create to clients.   I did the opposite of nicky, I started a print with correction on and then issued the command to turn it off and the difference is dramatic to say the least.

[Danny]

I just want to say...
Every part I ever printer was banding free with a stock solidoodle.

Not that I've not had my share of problems else where, (hot end breaking inside of two week, warped bed) cracked jigsaw etc.
I replaced the warped bed with glass, and replaced the resistor for a matrix or wire, I'm lead to believe that SDs are now shipping with heat pads instead of resistors pretending to be heaters.

[Jeffery Sanders]

" I just want to say...Every part I ever printer was banding free with a stock solidoodle. " I don't believe that, sorry.  please post a picture of one of your prints if you don't mind (pref a cali cube) and I will show you your banding with a red arrow from mspaint, (turn the part upside down or to the side) depending on phase.  Print a 20x20 calibration cube and shine a light through it...the layers are not even.

They finally did swap out that resistor for a proper heating method...(which early adopter's have to miss out on or pay for)

[Kendall]

Your very lucky to have minimal banding...  Please [Post] pictures of your parts and tell us what software & settings your using.  Share your experience...

[Francesco Santini]

(I thought I sent this message already but it didn't show up in the list... So I'll try again, I apologize if you receive it twice.)

On Sunday, February 17, 2013 4:17:47 PM UTC+1, Support Tech wrote:

Here's the thing:
The nature of this firmware mod is that it accounts for a very specific wobble situation. While we haven't done extensive testing I will say that it is unlikely to fix the issue on all machines. Again, if the banding issue was as one dimensional as this it would be quite an easy fix.

Hi, I think I would quickly step in the discussion, being the author of said firmware fix. I agree that this solution is 1-dimensional in the sense that it only takes into account the wobbling in the Z direction neglecting XY coupling. However, it is a more general solution than it might seem.

In the original idea (sinusoidal wobbling) the "geometrical" assumption would be that the rod section is elliptical instead of round. This could mean that a rod that is not perfectly aligned with the axis of the nut. However, even if the mechanical problem was more complex than that, any periodic banding would still have a main component that can be described as sinusoidal: thus applying the sinusoidal fix would fix the first harmonics of the banding, leaving only higher spatial frequencies, which would be less evident given the finite layer thickness.

Furthermore, tealvince (in the soliforum) inspired and refined a solution (still based on the same firmware mod) that allows non-sinusoidal (but still periodic) wobbling to be measured and compensated, and he is claiming positive results with his approach.

So to summarize there are two things not taken into account:

1) Non-periodic (ie Z-dependent) wobbling. This would be the case if the rod is actually (badly) bent in the middle of the working zone. In this case, the firmware compensation would work for part of the print and fail in another part. The only viable solution would be to provide straight rods. The firmware could still be modified to take this into account, but finding the right parameters would probably be a hell.

2) XY coupling. This makes sense if there is a significant backlash in the guides of the bed. Personally, I haven't noticed any (and I have a wobbling that is worse than others). If this is the case, it can still be accounted for in the firmware, provided that it can be measured (and tealvince's approach would be a winner in this case).

I would also like to stress that the firmware compensation is completely for free, except for maybe one hour spent with fiddling with the parameters, so IMHO it wouldn't harm to give it a try.
 

 I should reiterate while I am here that Solidoodle does have an official github page (github.com/solidoodle) and you can submit proposed bugs / bug fixes there.

I'm struggling a bit with github, and at the moment I made a pull request to lawsy, since his repo is referenced in different places, including your official firmware update page. I will try to fork the official repo as well, but when I tried last time github did not allow me to fork two "marlin" repos at the same time, or so it seemed. I will try to incorporate this fix in the official repo as soon as I figure out a way with github.

Francesco

 

[Francesco Santini]

Il 17/02/2013 16:17, Support Tech ha scritto:

Here's the thing:
The nature of this firmware mod is that it accounts for a very specific wobble situation. While we haven't done extensive testing I will say that it is unlikely to fix the issue on all machines. Again, if the banding issue was as one dimensional as this it would be quite an easy fix.

Hi, I think I would quickly step in the discussion, being the author of said firmware fix. I agree that this solution is 1-dimensional in the sense that it only takes into account the wobbling in the Z direction neglecting XY coupling. However, it is a more general solution than it might seem.

In the original idea (sinusoidal wobbling) the "geometrical" assumption would be that the rod section is elliptical instead of round. This could mean that a rod that is not perfectly aligned with the axis of the nut. However, even if the mechanical problem was more complex than that, any periodic banding would still have a main component that can be described as sinusoidal: thus applying the sinusoidal fix would fix the first harmonics of the banding, leaving only higher spatial frequencies, which would be less evident given the finite layer thickness.

Furthermore, tealvince (in the soliforum) inspired and refined a solution (still based on the same firmware mod) that allows non-sinusoidal (but still periodic) wobbling to be measured and compensated, and he is claiming positive results with his approach.

So to summarize there are two things not taken into account:

1) Non-periodic (ie Z-dependent) wobbling. This would be the case if the rod is actually (badly) bent in the middle of the working zone. In this case, the firmware compensation would work for part of the print and fail in another part. The only viable solution would be to provide straight rods. The firmware could still be modified to take this into account, but finding the right parameters would probably be a hell.

2) XY coupling. This makes sense if there is a significant backlash in the guides of the bed. Personally, I haven't noticed any (and I have a wobbling that is worse than others). If this is the case, it can still be accounted for in the firmware, provided that it can be measured (and tealvince's approach would be a winner in this case).

I would also like to stress that the firmware compensation is completely for free, except for maybe one hour spent with fiddling with the parameters, so IMHO it wouldn't harm to give it a try.

 I should reiterate while I am here that Solidoodle does have an official github page (github.com/solidoodle) and you can submit proposed bugs / bug fixes there.

I'm struggling a bit with github, and at the moment I made a pull request to lawsy, since his repo is referenced in different places. I will try to fork the official repo as well, but at the moment github is not allowing me to fork two "marlin" repos at the same time, or so it seems. I will try to incorporate this fix in the official repo as soon as I figure out a way with github.

Francesco

[Support Tech]

Great explanation!

We are encouraging the use of the official repo, as there will be more firmware mods / changes from Solidoodles side in the future. If there is something we can do on our side to make this easier, we will do it.

[WheatBread Cat]

On Friday, February 22, 2013 8:37:39 AM UTC-6, Support Tech wrote:

Great explanation!

We are encouraging the use of the official repo, as there will be more firmware mods / changes from Solidoodles side in the future. If there is something we can do on our side to make this easier, we will do it.

Can you please list even 3 things that Solidoodle has actually fixed on their own product since the release that has improved the printer?  The orginal firmware (that shipped with the printer) was horrible and full of bugs, so I doubt anything kept and maintained by SD itself  "under your official repo"  would be of worth the trouble of a pull request...anytime SD is involved it slows down development, puts the ball in "bad supports'" hands to adopt the changes or not and I know there isn't is half the intelligence in the whole SD Co. when compared to each of your brighter customers...we the customer's built your POS company....not you....you took our money.....kept it for months w/o proper delivery of item(s) as promised...I never got half the items I was supposed to upon receipt , or the parts I was told I would get for repairing my z-axis.   Hell, if I just left the printer as I took delivery it would not be working at all...the heated bed, extruder, stepper drivers w/o sinks, improper thermistor application (very dangerous), z-axis, wiring have all been upgraded, replaced, or fixed in less than 6 months of ownership.  The only positive thing I can say is that they shipped me my printer late, but not nearly as late as someone who didn't speak up to move there spot up...still over a month after it was supposed to arrive...if the US still maintained consumer satisfaction based economy where each CO were held responsible vs just being able to let insurances take the hits we might actually get somewhere...there always is the co BBB rating which puts it on par with some of the worst in the US....http://www.bbb.org/new-york-city/Business-Reviews/computer-printers-sales-and-service/solidoodle-in-brooklyn-ny-136978.  

[jon bondy]

I can sense that you are angry and frustrated.  As regards the original issue, both of my printers are using the original firmware, and have never been upgraded.

[Support Tech]

I sense some anger here, but let's not get ahead of ourselves.

On Sat, Feb 23, 2013 at 9:02 PM, WheatBread Cat <jeffery...@gmail.com> wrote:

On Friday, February 22, 2013 8:37:39 AM UTC-6, Support Tech wrote:

Great explanation!

We are encouraging the use of the official repo, as there will be more firmware mods / changes from Solidoodles side in the future. If there is something we can do on our side to make this easier, we will do it.

Can you please list even 3 things that Solidoodle has actually fixed on their own product since the release that has improved the printer?  The orginal firmware (that shipped with the printer) was horrible and full of bugs,

This is the same firmware that is used by the wider 3D printing community. To my knowledge no *bugs* have been found in it - there have just been some features added, and some tweaks made to the parameters in the configuration file. To my knowledge no changes whatsoever have been made to the inner workings of the firmware whatsoever.
 

so I doubt anything kept and maintained by SD itself  "under your official repo"  would be of worth the trouble of a pull request

The utility of it is this: You will be able to list bugs to us. To Solidoodle's knowledge there are no bugs whatsoever in the firmware. We have no reason to change something, that is far as we can tell works fine. We are not privy to any bugs or change requests made on behalf of m-laws. However, if we were made aware, we could divert resources to that sort of thing. Since the addition of the S3, Solidoodle will be maintaining at least two sets of firmware. Please take this time to work with us on the firmware. We'd love to work together as a team.
 

...anytime SD is involved it slows down development,

Here's the thing: Since we are closed source company, you can't see our development. Nor are you necessarily made aware of the iterative development / tweaks to the machine that are made on a daily basis.  Sometimes that makes it seem like we aren't doing much to improve the machine, but the machine has changed a great deal since day 1.

puts the ball in "bad supports'" hands to adopt the changes or not and I know there isn't is half the intelligence in the whole SD Co. when compared to each of your brighter customers...we the customer's built your POS company....not you....you took our money.....kept it for months w/o proper delivery of item(s) as promised...

We do apologize for our earlier customers who experienced a large delay. I think anyone here who was involved from that time period felt a certain frustration with us. However, that era is over, and we are happy to say that lead times for the S2 are starting to approach the 6 week mark. I think a lot of complaints that arise today are related to delays in the process. Look for improvements across the board as 2013 wears on.
 

I never got half the items I was supposed to upon receipt , or the parts I was told I would get for repairing my z-axis.

 

I'm sorry that you slipped through the cracks. If you contact us directly, I'm sure we'd be willing to send you more parts. However, if your requests relates to z-axis banding, I'm not sure a new lead screw will necessarily help.
 

  Hell, if I just left the printer as I took delivery it would not be working at all...the heated bed, extruder, stepper drivers w/o sinks, improper thermistor application (very dangerous), z-axis, wiring have all been upgraded, replaced, or fixed in less than 6 months of ownership.  The only positive thing I can say is that they shipped me my printer late, but not nearly as late as someone who didn't speak up to move there spot up...still over a month after it was supposed to arrive...if the US still maintained consumer satisfaction based economy where each CO were held responsible vs just being able to let insurances take the hits we might actually get somewhere...there always is the co BBB rating which puts it on par with some of the worst in the US....http://www.bbb.org/new-york-city/Business-Reviews/computer-printers-sales-and-service/solidoodle-in-brooklyn-ny-136978.  

That link speaks for itself, but essentially we've only had a single complaint in the history of the company. Again, we are always willing help customers - we don't want to leave anyone in the dark.

A partial list of Solidoodle improvements since May 2012:
Improved electronics
Added electronics cover
Improved extruder mount
Improved printer carriage
Improved all printed parts on printer.
Improved bed heater
Improved shipping technology (foam inserts vs. peanuts)
Complete redesign of hot end
Software tweaks to pronterface/printrun
Development of 1 click installer for pronterface/printrun for Windows
Custom configurations for repetier host
Development of 1 click installer for repetier host for Windows.
Development of the Solidoodle 3

And of course we are always working on further improvements to the machine, and to our process in general. We are always happy to work with the community.

Regards,
John

[Support Tech]

Jeffery,
What can we do to help you? Again, we are always willing to help. Is your machine still having problems?

On Sun, Feb 24, 2013 at 11:19 AM, WheatBread Cat <jeffery...@gmail.com> wrote:

 Now you have 2.

[Ian Johnson]

A thread popped up at the Reprap forum on Z wobble.  Another recommended fix is using a thinner rod for the Z axis, like 5mm.  Since it is smaller than the 8mm guide rods, it won't be able to win a fight against them- it will bend rather than force a misalignment onto the bed. This goes in addition to a flexible coupling on the motor and an unconstrained top end.  People have also recommended putting something soft under motor as well so there is some play there as well.

[Francesco Santini]

This is assuming a coupling between Z and XY, which doesn't seem to be the case in most of our situations, where the wobbling is apparently one-dimensional (otherwise the firmware fix wouldn't be successful)

[RonSII]

The wobbling can be two dimensional in the case of loose parts, also I have seen some great examples of the FW fix helping the banding issue... I have also seen some examples that I thougt it just compounded on some other problems and made it different/worse. My printer had more banding issues with bed wobble and thermal banding than threaded rod issues, once I cleaned up the counterbores for the levelers and tightened them more than stock and also autotuned the extruder I have virtually no banding.

[Kendall]

RonSII, please give us a 1, 2, 3 of exactly what you did with some pictures before and after with the banding...

I, like you, think it is a bunch of different problems; 1st with the nut threads might not being perpendicular to the ends;  2nd, threaded rod is just rolled threads, most of them have an oval spiral pitch diameter, and that isn't even a consistent oval; 3rd, the threaded rods are not straight; 4th we're dependent on 'as close as possible' alignment with the coupling to the motor; 5th, the guide bushings do have slop; 6th, nothing is holding the bed alignment but friction from the spring pressure; 7th, why do you think solidoodle ships the nut loose? B/c when it is tight, other problems raise their ugly heads (and this is why my Floating Jam Nut Fix did not fix everything (it did fix the slop between the nut and screw only, but if the threads or the screw is catawampus, it only exaggerates the side force, leaving you with a different problem)).  All of these items listed are tribal, but all do play a part in the banding, and I believe everyone has just a little different situation with their particular printer, if not, someone would of came up with a real solution by now...