Dibloff's Build Log

Topics Related to the ORD Bot Printer

Re: Dibloff's Build Log

Postby dibloff » Sun Nov 04, 2012 9:47 pm

Following Orcinus recommendations I made a bracket for the fan side of the extruder. This will constrain all the mass to the hadron bracket and add some stiffness to the system. I think it turned out to be nice. Now I need to get some 45 mm long M3 screws. I think this extruder will stay put now.
2012_1104_Hadron 009.JPG
dibloff
 
Posts: 144
Joined: Sat Sep 29, 2012 2:04 am

Re: Dibloff's Build Log

Postby orcinus » Sun Nov 04, 2012 9:58 pm

Looking good!
For some reason, i prefer connectors to screw terminals too.

Except i never invested in a crimping tool and a stock of connectors, so i always do makeshift ones from headers (male headers soldered on the board, female headers with heat shrink insulation as plugs - should really be the other way around to avoid the possibility of short-circuiting something, but meh...)
orcinus
 
Posts: 720
Joined: Tue Feb 14, 2012 4:03 am

Re: Dibloff's Build Log

Postby dibloff » Mon Nov 05, 2012 2:03 am

I had some doubt about the end stop switch wiring. If you look back my post here:
viewtopic.php?f=33&t=1567&start=20#p14509
you can see that I used C and NO terminals on all end switches. Honestly I have not even looked the Azteeg X3 schematic back then. It was obvious for me that if the switch arm will be activated it’ll close a circuit. (It's funny, noone here said a word about it, so in hindsight it must have been right). Later I consulted the schematic on my wall and it uses the C and NC terminals. So I was about to change all the solder connections. Then I found this:
http://www.soliforum.com/topic/377/upgr ... azteeg-x3/
He recommends to use C and NO (Which I did initially). I also looked up the schematic directly from the panucatt webpage, and guess what, they recommend to use C and NO too…
Here is the panucatt schematic. The red lines are the recommended connection a month ago, the solid black lines are as of today.
panucatt_new_schematic.jpg
dibloff
 
Posts: 144
Joined: Sat Sep 29, 2012 2:04 am

Re: Dibloff's Build Log

Postby regressedEE » Mon Nov 05, 2012 2:42 am

Dibloff, really appreciate the detailed build log as it's already saved me from a couple of missteps. In regards to the limit switches, what I'm thinking is that I will install 2 Y switches connected in a NC loop, one at the front and one at the back (ie, X3 signal -> limit switch 1 C and NC -> second switch C and NC -> X3 negative. You'll have to enable the pull up resistor so that the signal goes high when either switch is triggered. I can then put the stop on the idler side of the Y axis OR on the motor side and will not have to change any wiring since it's all one continuous loop and either switch can break it. Same applies for X; I don't think there's really a need for Z since the Z bottom will always set the lower limit. Of course, this assumes you will be setting your Xmax and Ymax values in your Merlin config files otherwise you would have to wire each limit switch separately so you can distinguish which switch has been triggered (max or min). The fact that most people don't use this technique tells me that I'm missing something pretty obvious, but I can't think of any reason this wouldn't work.

PS -- it would seem to me that wiring the Z axis motors in series would ensure a more even distribution of current between the two motors and thus keep them from getting out of sync. Yes, you loose a little torque and speed, but how fast are you going in the Z direction anyway?
regressedEE
 
Posts: 25
Joined: Sat Oct 13, 2012 2:33 pm

Re: Dibloff's Build Log

Postby fma » Mon Nov 05, 2012 7:27 am

It is always better to use C and NC contacts, so if a wire is disconnected, you immediatly see it (it will activate the end-stop)...

About the Z motor wiring, as long as the tork is OK for the motors, it dos not matter if one motor get more current than the other. 1 step will be 1 step.
Frédéric
fma
 
Posts: 213
Joined: Mon Aug 20, 2012 7:53 am

Re: Dibloff's Build Log

Postby dibloff » Sun Nov 11, 2012 1:28 am

OK Folks. Here comes chapter 7. I’m basically done with all the mechanical setup, and wiring. The following will include some harsh engineering language so please proceed with caution.
I was doing some adjustments to make sure my Y gantry is parallel to the bottom transverse beam. While doing so I realized my tray is tilting to the left (viewing from the build side). I started to measure the distance from the edges of the tray to the bottom transverse beam and found out the difference between left and right sides is ~3 mm. After some thinking and further measurements I found out that the nylon spacer that holds the V-Bearings (it’s on the left side) is compressed. I removed it and measured the height. It was 2.91 mm. This is with the load removed... As you remember these are ¼” height, so they should be 6.35 mm. I measured one unused spacer and it’s indeed 6.35 mm high. So I did some calculations below. I used 5 and 10 Nm of torque in the calculation, because I do not exactly know how much I have exerted. I know I used a short handle ratchet, and I know it must have been 5 to 10 Nm. Also the nut has a nylon insert which will have an additional 1-1.5 Nm prevailing torque (this will decrease the torque). So the real number must be somewhere between 5 and 10 Nm.
calculation.jpg

So As you can see @ 5 Nm of torque the M5 thread will generate 5269 N of clamp force, while @ 10 Nm it’ll double - 10538 N.
The original Hadron Bot BOM recommends using 94639A352 from McMaster. This is a 6/6 Nylon spacer. This is an unreinforced Nylon. http://www.engineeringtoolbox.com/engin ... _1225.html recommends 2-3.6 GPa for Elastic Modulus. I used 2.5 GPa in my calculation. The following calculation was made for a hollow cylinder in simple uniaxial tension. With all these plugged in the compression (u) of the nylon washer will be 0.42 mm @ 5 Nm and 0.839 mm @ 10 Nm of tightening torque. So this is @ the location where the nylon washer sits. To find out the deflection @ the end includes some additional math. Here is a drawing that includes all the dimensions and notation:
displacement_layout.jpg

As you can see the deflection @ the left end of the tray (x1) can be as high as 1.8 mm (@10 Nm torque) and on the right end (x2) – 1 mm (@10 Nm torque). The total misalignment (X) can be 2.8 mm – close to what I measured.
Considering all these I made two more calculations with aluminum and steel for the spacer. As you can see in worse case scenario the same max misalignment (X) will be 0.1 mm for Aluminum and 35 micrometers for Steel (both @ 10 Nm of tightening torque).
OK. OK. I know it’s too much math for someone, the conclusion from all these are that using a nylon spacer will put the tray into a misalignment by default. Even if you have a torque wrench and you can control the tightening torque it’ll compress the spacer. Some can say that by suspending the hot bed on springs one can level it by adjusting screws that mount it. Sure you can do it, but I just simply hate to make something with a design flaw like that. Besides, plastic creeps. With compression on it creeps even more. With heat cycles (the darn thing is under the hot bed!!!) it even creeps more, so over time you’ll lose clamp force and your tray will start to wobble like a Hawaiian belly dancer, while you're sittng there and thinking why the darn thing is not printing as it should.
You can use the famous 92510A562 aluminum spacer for 83 cents or the 92415A742 steel spacer for 84 cents. The BOM calls for 8 pieces.
dibloff
 
Posts: 144
Joined: Sat Sep 29, 2012 2:04 am

Re: Dibloff's Build Log

Postby fma » Sun Nov 11, 2012 9:58 am

I also noticed that. If this is not a problem for the Y axis, it becomes a real problem with the gantry. The problem is that the 6 V-wheels which slide on the Z rails can't be in the same plan, leading to the excentric wheels not to be totally push on the rail. Then, with flexion, this lead to Z wobbling because of the M8 screws!

I can't order thing to McMaster from France, but I will try to make steel spacers with the correct height.
Frédéric
fma
 
Posts: 213
Joined: Mon Aug 20, 2012 7:53 am

Re: Dibloff's Build Log

Postby dibloff » Mon Nov 12, 2012 2:43 am

I did a close shot which compares the nylon washer and excentric one. Hope you can see the 0.4 mm difference :)
2012_1111_eBay 001.JPG
dibloff
 
Posts: 144
Joined: Sat Sep 29, 2012 2:04 am

Re: Dibloff's Build Log

Postby dibloff » Mon Nov 12, 2012 2:52 am

I decided to use 16 gage wire for the hot end - Panucatt recommends using 16-20 gage. It sucks to buy 25 foot of wire whe you need less than 3 feet. Fortunately these reels were ~$7 each.
2012_1111_eBay 023.JPG

For the hot bed I used 14 gage wire. Green in color :) I used shrink tubing to hold the wire bundle together, and I made some brackets to mount the cable assembly. I tapped the holes on the Hadron Bracket and used M4 screws to secure the cable assembly.
2012_1111_eBay 003.JPG

2012_1111_eBay 004.JPG

2012_1111_eBay 005.JPG
dibloff
 
Posts: 144
Joined: Sat Sep 29, 2012 2:04 am

Re: Dibloff's Build Log

Postby dibloff » Thu Nov 15, 2012 4:07 am

I replaced all the spacers. I used steel spacers, this way the compression will be the same as on the excentric nut side. I think everything runs smoother now.
I also got a piece of glass from an old scanner I took apart a while ago. I cut the glass to size and grinded the edges to remove sharp corners. It is 3.2 mm thick. I read people are using 3-4 mm thick glass so this should be fine. This will be on the top of the hot bed.
I need to figure out where is the center of the coordinate system on the Hadron. I have not seen any marking for it on any of the drawings. I've got 6 end switches wired up but for the time being I'll only use 3 of them. I need to figure out which ones I have to use (on the bottom row of the Azteeg X3).
Also I have not seen any notes for the axes. I beleive the hot bed is the X axis. Then the gantry left right must be the Y axis, and the gantry up-down the Z axis. Again, I have not seen a drawing pointing this out.
dibloff
 
Posts: 144
Joined: Sat Sep 29, 2012 2:04 am

PreviousNext

Return to ORD Bot

Who is online

Users browsing this forum: No registered users and 4 guests