Solved Z Rods not parallel

Hi everyone!
New to 3D printing and recently received a ROBO3D R1.
First print was the "Calibration Box". The first 4mm or so was sloppy, but as it continued to print, things smoothed out a little with issues at one corner.
I did a quick search of the forum and couldn't find anything concerning the Z Rods not being parallel.
With the extruder at it's lowest, I took a 14" set of linear calipers and measured the distance between the smooth Z Rods. The distance was 11.892 inches.
Then I ran the Z axis to the top limit. The distance was now over 12.005 inches. More than 1/8th inch.
Knowing a little about precision linear bearings, this isn't the way it's done.
The rods should be precisely parallel and I imagine the X carriage can be binding at the bottom.

What I did to correct the problem was to remove the X axis carriage and drill the 8mm holes in the right motor side deeper to allow the distance between the brackets to be less and installed adjusting screws in the ends.
Now when the extruder is run up to the top, the distance between the Z Rods is the same as when at the bottom.
Has anyone else had this problem and does ROBO3D know about it?

 

I forgot to add that the second printing of the "Calibration Box" had improved, but there were still issues at the corners and I'm suspecting it's from jiggle in the Z Rods.
Next upgrade Z Rod Stabilizers.

 

I'd suggest printing these rod caps and installing them on the top to get a better seat on the top mounts

https://www.dropbox.com/s/p2qplbikfsjs1vm/rodcap.STL?dl=0

 

Thanks, these will only stop the jig

So, how do I know if your rod caps will position the Z Rods at the proper distance apart.
I think I'd rather use the adjustable type.
I would also like to know what the distance is supposed to be, center to center of the Z Rods.
Anyone got the blueprints?

 

You could have one z at a length of 350mm and the other at 300mm. The height doesn't really matter. You need to do the paper level test and level things that way. Also make sure you have the correct firmware installed for right threaded rods you have. Another thing is to make sure that the couplers are not loose.

 

Not sure you're catching the problem.
There was nothing wrong with the leveling of the extruder. Did leveling, couplers are not loose and correct firmware. All that is good.
What was wrong, was the Z axis rods were not parallel with each other. The X axis rods were either too long or the holes were not injection moulded deep enough. A difference of over .125 from bottom to top is too much even for the sloppy bearings ROBO uses in these printers.
You can't have 2 parallel rods on a linear system that aren't precisely parallel and it's possible that the problem I was having with the first 5mm or so layering had to do with the bearings binding at the very bottom.
As I said, to correct the problem, I had to bore the X Rod holes in the right motor bracket about .075 inch deeper.
Second printing of the Calibration Box was much improved.

 

I think you're over analyzing the issue. The hole spacing on the top matches the hole spacing on the bottom. This will square up the system.

 

I'm not over analyzing anything.
This is mechanically wrong for a precision linear bearing system. I worked at GM for 31 years and know a little about machinery that uses linear bearings.
Why do you think I even own a set of 14" vernier calipers? (They didn't let us take the expensive digital stuff home when we retired!)

If the hole spacing of the bearings in the X brackets is wrong, the extruder carriage will bind at the bottom and at the top, if the top and bottom of the Z Rods are constricted and can't move.
If the center to center distance of the Z Rods is 310mm at the top and 310 at the bottom of the rods, but the X carriage bearing distance center to center is larger (or smaller), the extruder will bind as is moves toward the bottom and toward the top.
This will make the stepper motors work harder and cause irregularities while printing.

I think that the extruder carriage spreading the Z Rods apart 4mm is excessive and will cause problems.
Don't you find it odd that my printing improved after I fixed the problem?
Aren't there any engineers here in the forum?

 

I can understand what you're saying, but many owners are reporting problems and questions and aren't getting any answers.
Since I couldn't find anyone talking about this problem and knowing that it might be a solution for some, it should be considered and not tossed aside.

Plus having worked at GM and met many "Engineers", I became aware that there are two types of engineers. The real engineers that understand how the physical world works and the ones that their high school "Guidance" councilor told them, "Oh, I see you're good at math. You should become an mechanical engineer".
Once met an engineer working for GM that didn't know what a car fender was!??

 

For sure MatterControl is buggy. It wouldn't run on my Windows 7 laptop, but oddly enough, I did get it to print on my iMac both in OS X and Windows 7 running in Parallels.
I think the graphics chip in the laptop couldn't handle all the info being sent through it.
Also tried to run it in a laptop with Linux Mint and couldn't get it to recognize the printer.
Now it seems to be running well on a first year Intel Mac Pro with Mountain Lion.
Anyway, my suspicion is that the X carriage rods were cut too long on my printer. It would be nice to know the proper length.
This is what happens when you hire 14 year old girls to work in a factory...

 

Also, if I were an engineer or CEO for that matter working for ROBO 3D. I would watch these forums like a hawk!
If you want to stay in business, what better place to get feedback.

 

No, this fault should exhibit a printing problem at the bottom and the top.
At the bottom, if the X Rods were too long, the stepper motors would be struggling to raise the extruder due to the Z axis bearings binding. I don't know how much it would take or how strong the stepper motors are for this to be a factor.
This was why my first layers weren't coming out smoothly. As the extruder made it's way up, the binding would lesson as the Z Rods would be allowed to bend out.
This is also why my printing of the Calibration Box improved as it got taller and since the Z Rods aren't constricted at the top, they are still allowed to flex some and if I kept printing up to the top limit it would have been fine.
With the one exception of the jiggle seen in the extruder as it goes higher. (To bad "Inertial Dampening" is science fiction!)
My point is, that some people are tearing their hair out trying to get their printer to print properly and this binding could be one cause.
Not everyone has the equipment and tools that I have to measure accurately the spread of the Z Rods.
Maybe they could print an adjustable ruler to check the spread. At least they would know if they had the 1/8th inch slop that I had and know that something isn't right.

 

This is why I think the main problem I had, is that the X axis rods are cut too long. This would be the easiest explanation. The holes in the X brackets wouldn't suddenly change and the jig cutting the Z Rod support holes in the bottom mount wouldn't suddenly change, but Lin Xiuo Mi might have gotten the length cut wrong on the X Rods. Too long and they would try to spread everything apart. (I made that name up by the way.)
Maybe only a few were made. Maybe mine's the only one, but in any case my printing improved by deepening the holes in the X bracket and adding the adjusting screws so there wasn't any binding at the bottom.

 

There's a little expression 'making a mountain out of a molehill' that comes to mind

And you're definitely over analyzing the issue.

These are not a pick and place or a packaging system that requires nanometer precision.

 

LOL! An error of .130 or 3.3mm is hardly nanometer or mµ! I would be happy with 1mm, but I made it better than that.
That's okay if you don't understand what I'm talking about.
Bottom line no more binding and better printing, end of discussion.

 

Oh I get what you're saying but you're exagerating the issue.

Bracing at the top would have resolved the issue just the same.

A good engineer knows over engineering a system can be just as costly as an under engineered. there's also no need for the unjustified arrogance and holier than thou attitude.