Y-carriage design questions

Hi,

I'm currently rebuilding my printer after six weeks of debugging and trying to iron out issues mostly mentioned in this forum.

I disassembled the y-carriage first to look into the problem with my sagging front on the heatbed (about 1.5mm lower than the back edge without adjusting it).
That's why I measured angles around the carriage. What I observed was that the y-axis-rib has an angle of 0.5° at the front end compared to the bottom. If I take the extrusion bottom, z-axis and the squareness of the whole construction as given (I tested on a 30mm steel straightening plate) then there is a 0.5° angle to the whole construction at the front end of the y-axis-rib.
I did some math and got the following:
If I take the angle alpha as 0.5° and the distance (even if the slant distance might be a bit more) as d = 16cm (about length of the rib/ heatbed from the point where I'm measuring) and want to get the height difference at the end of the distance I put it in the mathematical equation for calculating height having an angle and a distance:

sin(alpha) = h / d

sin(0.5°) = h / 160mm

sin(0.5°) * 160mm = h

h = 1.39mm

Hm. Coincidence?

I think one reason might be the point where the y-axis-rib rests on the bearing. If there is any production difference (which is very natural having tolerances producing the lasercut parts) this could worsen any angle in the whole heatbed assembly. Maybe a hint for people who are currently building it and observe notable height differences in the heatbed construction.

Just an observation and mathematical approximation.

The usual photo:



Markus

[EDIT] To solve the problem a part/ cutout of an old bicycle tube might be put between bearing and cutout of the y-axis-rib. This helps also with any slippage.

[EDIT 2] In my case a 0.4mm tiny piece of rubber patch put between hole cutout of the rib and the bearing was sufficient to get all straight and level.

Edited 4 time(s). Last edit at 01/17/2014 08:14AM by markbee.

---

XBee & electronics blog: [lookmanowire.blogspot.com]

Hi Markbee,
My table was much worse than this, and varied with load, humidity and day of the week!
I gave up on it and threw the wood away, rotated the back bearing carriers to sit with the flat on the top,
I replaced the wood with a 2mm piece of aluminium. The front bearing is now connected with a spring Terry clip and spacer.
The spacer was chosen to give a dead level initial setup. All of the hole mounts for the heater bed were transferred to the ali plate.
The limit switch was serviced with an M3 bolt fixed to the orange dearing carrier. The belt was clamped with a split nylon block.

2mm is thick enough to countersink fixing screws for the blocks so they are flush.
Use stainless steel screws if possible to space off the heater pad. This will reduce thermal losses.

Result - my bed is now +/- 0.1mm over it entire surface and no compensation is needed. It is stable.
I have also bolted the entire thing to a rigid base to prevent the entire thing flexing.

A lot of work but worth it.
(there are a lot of design options being explored currently - but I cant wait that long)
Regards
Dennis

I didn't perceive a problem with the plywood ribs or with the fit between the front rib and the bearing, but the MDF bed support was sagging at both back corners. So I kept the ribs but replaced the MDF with a 3mm aluminium triangular bed support. I had already fitted bed leveling adjustment screws as described in the manual bed-leveling instructions. It's level to within about 0.1mm now, except in the left front corner where the glass appears to be not quite flat.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Hi DC42,

I'm still on building stage, and MDF seems a little soft at corners and it began to sack already ! I know it will happen because I'm in Asia country, very hot and humid. Setting the bed level compensation at multipoint are time consuming, I completed the Ormerod hardware and electronics within 12 hours, but spend another 12 hour to solve pronterface problem, ( not solved yet, still missing GET POS etc button ) network still can't operate, and SD card not working as mentioned before in this forum . can't print anything yet.
Markbee : some design / parts are not rigid enough for a precision machine like 3D printer, I already written down those parts and I need to think a way out to modified it, can't print anything yet , stuck at commissioning stage now :-(

If it's sagging at the front corners, don't worry about it - the mdf isn't doing much there, just supporting the cardboard insulator. Just use three screws to mount the bed; centre front, and the two at the back. These are better supported by the ribs, and should sag. The glass will keep the bed flat.

If the bed drops away from the nozzle a lot (more than 1mm or so), drill a small hole above the front Y bearing, pack between the bearing and the y-rib with something, and use a cable tie through the hole and around the bearing to hold it in place. I think this is what markbee has done in another thread.

The drop in the bed may be to do with inaccurately laser cut parts, but could also be due to slight misalignment between the y-axis-end-plates and the y-axis aluminium extrusion, or between the y-axis and z-axis aluminium extrusions.

Ian
RepRapPro tech support