New and improved x-axis carriage

Hello all,
I have been working on an improvement of the x axis carriage, hoping to overcome the following issues:

X-axis sag: I bought a 10mm bearing, but it was a pain to find and still not as tight fitting as I would have liked.
IR sensor placement: My sensor fitted perfectly on the original carriage, but I was unable to connect the cable as the recess wasn't deep enough. This could just be that my sensor/loom connector was deeper than others, but still a pain. My bodge over the last month was to place 2 washers between the PCB and the carriage to keep the sensor straight while allowing enough space for the cable to be plugged in. This had the undesired effect that only the very tips of the short bolts were doing anything and it was very easy to overtighten and strip the thread from the carriage.

Solution:
(See attached file)
This is only a slight modification of the RRP carriage (I hope it's allowed to publish modifications of official versions!), that I hope address all the issues above.

The changes aren't obvious at first glance.
The bolt slot and nut recess now extend 2mm closer to the x-axis arm. The nut recess has also been very slightly enlarged to try and avoid all file work. This has the disadvantage of leaving the wall between the nut recess and bearing cylinder recess very thin indeed, however a printed test of this section showed that it still works and the original 9mm bearing can now be used to hold the carriage directly against the x-axis arm, completely eradicating x axis sag!
I have also increased the recess for the IR sensor plug from 2.05mm to 4.5mm, which should be plenty for anyone with the problem I was experiencing above.

I previously posted here a version that I had hoped would also increase the stability of the carriage to stay on the bearing (as some people had been reporting problems of it slipping off. Unfortunately this modification didn't work (it fouled on the Z axis support when homing the X axis. I have therefore redacted this modification and re-attached a (hopefully) working carriage below.

Please note this modification of the RRP file is still only at the testing stage and hasn't yet completed rigorous sea trials! - use at your own peril!

Kind Regards,
Squags

PS
The modified solidworks file is now very slightly too big to be allowed to be uploaded to the forum. Please send me a message with your email address if you'd like it and I'll happily email it.

Edited 2 time(s). Last edit at 01/31/2014 10:45AM by Squags.

Please note, since first posting this morning I have experienced problems with the new carriage and have therefore modified it further - please delete any attachment previously downloaded and download afresh (post and attachment above now corrected)

Nice work, and I finally know clearly what X-axis sag is now.
So when I spent about 90 minutes modifying the placement for the bearing and adding in an extra homemade part to eliminate this it wasn't because of a mistake I made? Damn. Ah well, knowing now was worth the frustration. I will post pictures of my "I can't print a fix because I don't have a 3D printer as I am currently building it" fix when I am back on land.
Brilliant. Heheh.
Any updates on the success of your version 2? It will probably be the next thing I print.

First thing to note is that having printed the part in black, the (re-calibrated) IR sensor is now far more accurate at homing the Z axis - Where as before I had a value of 687 at 0.6mm, with 897 at 0.1mm; it is now 698 at 1.7mm and 974 at 0.1mm (with the same lighting conditions as before).
It's a complicated part and as such if your printing quality isn't spot on, it's important to fettle and fit it adequately. Mine needed a bit of persuasion to get the belt (both ends) properly tucked in as there's now no room for them to stick out and rub against the x axis arm. I also had to file the edge that tucks behind the Z arm (when homing X axis) as it was catching a little at first.

I'll update this again once I've got round to printing something else (I'm in the middle of 2 weeks of MSc exams at the moment, so the printer is more of a distraction than a help!

Many thanks Squags,

Printed, installed and running droopless - excellent solution. Also found that the bearing fit is tighter than the RS original part, which is a good thing. I had already removed the tags and filed down the sensor plug so didn't actually need the extra space for that, but a good answer too.

I'm still suffering from blobs and strings which meant a lot of fettling for the nut traps, but I've found the thread for corrective suggestions so hope to print an even smarter one (or two?) in PLA before I crank up the wick and go for ABS.

Thanks again.

---

Steve - RS Ormerod - 7th Feb 2014
"... the clouds looked as if they had heard about snow and were considering the idea." - (Terry Prachett, The Light Fantastic)