Another Dial Indicator Bracket

[RetireeJay]

I don't know, maybe there are as many dial indicator bracket designs as there are guitar players in Nashville, but I went ahead and designed my own anyway.

This one is specifically for the (my) Printrbot Plus, but it may fit on other Printrbots and I believe the design principles may be applicable to most.

The idea is to have a way to mount a dial indicator that is (A) extremely repeatable and (B) extremely easy to place and remove. I do this by using a design that "clips" onto the horizontal X-axis rods that mount the extruder.

Some things I have learned by using this bracket:
(1) It becomes very easy to accurately verify bed leveling using a movable dial indicator.
(2) An adjustment accuracy of 0.1mm or better is very desirable for repeatable results of part adhesion to the bed
(3) The weight of the suspended carriage is more than sufficient to overcome all hysteresis in the coupling of the nuts to the threaded rods; it is not necessary to add any springs or fancy mechanisms. On the other hand, the original Printrbot design for capturing the nuts was inadequate; I could see the nuts turn a fraction of a rotation every time the rods reversed direction. I printed parts which capture the nuts and lock their rotation with zero play to the carriage (while still allowing the nuts to overtravel free of the carriage in the downward direction in case of a problem with the Z zero switch)
(4) The stepper control is extremely repeatable. I can pick a certain Z elevation, move up or down from there, and then return to the same elevation. The display in Repetier and the display on the dial indicator will match exactly what they were when I started.

I have attached the STL files for the bracket and the wedge, but if someone has the software to take my *.AD_PRT files and put them into a more universal format, I'd really appreciate it. I'm sure that people trying to duplicate this for their machines will need to make small adjustments due to the actual dimensions of their printers and also the way that their materials print (i.e. the hole size for the indicator mounting tab). So the STL files won't be directly usable by very many people.

I've also attached a PDF file with drawings of the part, critical dimensions annotated.

EDIT: I printed this at 100% infill. I wanted it to be as rigid as possible.

DialIndicatorBracket.JPG

smDial Indicator Side View.jpg

smDial Indicator Rear View.jpg

smDial Indicator Wedge.jpg

smDial Indicator Front View Mounted.jpg

smDial Indicator Side View Mounted.jpg

[Mooselake]

Thanks, Jay! It's exactly what I need.

I think a bit of filing will take care of any printing inaccuracy. I'll try printing it in PLA and see if it's flexible enough - if not, I'll have to dig out the roll of ABS.

Kirk

[Tdeagan]

Merry Christmas! As a gift in return for your many contributions, here's your dial indicator bracket in OpenSCAD. Do with it as you will

Code: Select all

//RetireeJay's Another Dial Indicator Bracket
fudge = .01;
$fn = 30;

x_offset = -30;
base_x = 80;
base_y = 53;
base_z = 7.021;

total_z = 18.07;

dial_cutout_r = 9;
dial_cutout_h = 2.5;
dial_cutout_rim = 4.5;
dial_offset = -17;

dial_passthru_x = 16.1;
dial_passthru_y = 6.8;
passthru_offset = -9;

top_trim_blk_x = 21;
top_fillet_r = top_trim_blk_x/2+fudge;
hedge = 3;

bot_y = 20;

bot_fillet_r = (base_y-bot_y)/2 + fudge;
bot_fillet_offset = 3;
bot_width = 7;

bot_x =base_x + x_offset- bot_fillet_r -bot_fillet_offset+fudge;

rod_r = 4;
rod_dist = 41.830;
rod_top_thick = 7;
rod_top_y = 5;

min_thick = 1.804;

bend_r = 2;
outer_bend_r= 5;
bend_gap = 2;
bend_offset = -3;

inner_cutout_y1 = 13;
inner_cutout_y2 = 6;
inner_cutout_x= 33;
inner_fillet_r = bot_fillet_r + 7;

bumper_h = 4.9;
bumper_offset = -13.2;

bot_bumper_r = 6.64/2;
bot_bumper_h = 3.66;
bot_bumper_angle = 135.03;
bot_bumper_w = 5;
bot_bumper_l = -bend_offset;

//---- MAIN ------------------------------------------------------

difference(){
   //base block
   body();

   //base cuts
   rod_cut();
   dial_cut();
   top_cutaway();
   bottom_cutaway();
   
}
   
//------------------------------------------------------------------   
module body(){
   
   union(){
      translate([x_offset,-base_y/2,0])
      cube([base_x,base_y,base_z]);
      
      hull(){
         translate([bumper_offset,base_y/2,base_z + bumper_h -1])
         rotate([90,0,0])
         cylinder(r=1, h=base_y-fudge);
         
         translate([bumper_offset,base_y/2,base_z -1])
         rotate([90,0,0])
         cylinder(r=1, h=base_y-fudge);
         
         translate([bumper_offset-bumper_h,base_y/2,base_z -1])
         rotate([90,0,0])
         cylinder(r=1, h=base_y-fudge);
      }
      
      translate([-rod_top_thick,-base_y/2,base_z-fudge])
      cube([rod_top_thick, rod_top_y, rod_top_thick+fudge]);
      
      translate([-rod_top_thick,base_y/2-rod_top_y,base_z-fudge])
      cube([rod_top_thick, rod_top_y, rod_top_thick+fudge]);
      
      union(){
         translate([base_x+x_offset-bot_bumper_l, -bot_bumper_w/2, 0])
         #cube([bot_bumper_l,bot_bumper_w,total_z]);
         
         translate([0,-bot_bumper_w/2,base_z+bot_bumper_h-fudge])
         difference(){
            union(){
               translate([base_x+x_offset-(total_z-bot_bumper_h-base_z)*.75, 0,0])
               #cube([(total_z-bot_bumper_h-base_z)*.75,bot_bumper_w,total_z-bot_bumper_h-base_z+fudge]);
               
               translate([base_x+x_offset-bot_bumper_r,bot_bumper_w,bot_bumper_h*1.35])
               rotate([90,0,0])
               # cylinder(r=bot_bumper_r, h= bot_bumper_w);
            }
            
            translate([base_x+x_offset-bot_bumper_l, 0,0])
            rotate([0,-bot_bumper_angle,0])
            #cube([total_z-bot_bumper_h-base_z,bot_bumper_w,total_z-bot_bumper_h-base_z+fudge]);      
            
            translate([base_x+x_offset-(total_z-bot_bumper_h-base_z)*.75,0, total_z-bot_bumper_h-base_z])
            #cube([total_z-bot_bumper_h-base_z,bot_bumper_w+fudge,total_z-bot_bumper_h-base_z+fudge]);      
         }

      }
      
   }

}

module rod_cut(){
   translate([0,base_y/2+ fudge/2,base_z+rod_r])
   rotate([90,0,0])
   cylinder(r=rod_r, h= base_y + fudge*2);
}

module dial_cut(){
   //dial cutout
   translate([dial_offset,0,-fudge/2])
   cylinder(r=dial_cutout_r, h=dial_cutout_h+fudge);
   
   //dial passthru
   translate([passthru_offset-dial_passthru_x,-dial_passthru_y/2,-fudge/2])
   cube([dial_passthru_x, dial_passthru_y, total_z+fudge]);
}

module top_cutaway(){
   translate([x_offset-(fudge/2),-base_y/2-fudge/2,-fudge/2])
   difference(){
      cube([top_trim_blk_x+fudge,base_y+fudge, total_z]);
      
      translate([dial_cutout_r+dial_cutout_rim,base_y/2,-fudge/2])
      cylinder(r=dial_cutout_r+dial_cutout_rim, h = total_z+(fudge*2));
      
      translate([(top_trim_blk_x-fudge)/2,0,0])
      difference(){
         translate([fudge*2,hedge,-fudge/2])
         cube([top_trim_blk_x/2-fudge,top_trim_blk_x+fudge, total_z+(fudge*2)]);
         
         translate([0,hedge,-fudge/2])
         cylinder(r=top_fillet_r,  h = total_z+(fudge*2));
         
      }
      
      mirror([0,1,0])
      translate([(top_trim_blk_x+fudge)/2,-base_y+fudge/2,0])
      difference(){
         translate([0,hedge,-fudge/2])
         cube([top_trim_blk_x/2+fudge,top_trim_blk_x+fudge, total_z+(fudge*2)]);
         
         translate([0,hedge,-fudge/2])
         cylinder(r=top_fillet_r,  h = total_z+(fudge*2));
         
      }
      
   }
}

module bottom_cutaway(){
   //sides
   translate([bot_fillet_r+bot_fillet_offset,bot_y/2,-fudge/2])
   cube([bot_x,bot_fillet_r*2, total_z]);

   translate([bot_fillet_r+bot_fillet_offset, bot_y/2+bot_fillet_r, -fudge/2])
   cylinder(r = bot_fillet_r, h = total_z);   

   translate([bot_fillet_r+bot_fillet_offset, -(bot_y/2)-(bot_fillet_r*2), -fudge/2])
   cube([bot_x,bot_fillet_r*2, total_z]);
   
   translate([bot_fillet_r+bot_fillet_offset, -(bot_y/2+bot_fillet_r), -fudge/2])
   cylinder(r = bot_fillet_r, h = total_z);
   
   //middle
   
   difference(){
      translate([bot_width,0,-fudge/2])
      union(){
         translate([0,-inner_cutout_y1/2,0])
         cube([inner_cutout_y1,inner_cutout_y1,base_z+fudge]);
         translate([0,-inner_cutout_y2/2,0])
         cube([inner_cutout_x,inner_cutout_y2,base_z+fudge]);
      }
      
      translate([bot_fillet_r+bot_fillet_offset, bot_y/2+bot_fillet_r, -fudge/2])
      cylinder(r = inner_fillet_r, h = total_z);   

      translate([bot_fillet_r+bot_fillet_offset, -(bot_y/2+bot_fillet_r), -fudge/2])
      cylinder(r = inner_fillet_r, h = total_z);   
   
   }

   //tail
   translate([base_x+x_offset+bend_offset-bend_r,base_y/2,min_thick+bend_r])
   rotate([90,0,0])
   cylinder(r=bend_r, h= base_y);

   translate([base_x+x_offset+bend_offset-bend_r,-base_y/2,min_thick+bend_r])
   cube([bend_gap, base_y, base_z - min_thick - bend_r+fudge]);
   
   difference(){
      translate([base_x+x_offset-outer_bend_r,-base_y/2,0])
      cube([outer_bend_r, base_y-fudge, base_z/2]);
   
      translate([base_x+x_offset-outer_bend_r,base_y/2,min_thick+bend_r])
      rotate([90,0,0])
      cylinder(r=outer_bend_r, h= base_y);
   }

}



dial_indicator.jpg

[RetireeJay]

Thanks, Tim! That was quick!
I'll get it posted on Thingiverse.

(I spent lots more time developing the part. Even printed several iterations before arriving at this one. Such is the life of a EE who's trying to play ME )

[Mochaboy]

My turn for a question ...

What's that for?

[RetireeJay]

We spend a lot of time making sure our printbeds are level and the spacing from printhead to printbed is exactly correct. This dial indicator bracket is a tool that makes it much easier to perform those adjustments, because now we can see analytically to the nearest 0.001" (0.0254mm) what the spacing is from our X-carriage rods to the bed. At any point on the bed. Thus we can see bed leveling & bed warping.

And with just a bit of experience, we can find out exactly what spacing works well for the bed we are using & the nozzle & filament we are using. Then we will be able to return to that exact spacing easily and reliably. Problems with parts swelling or bending due to heat and / or humidity can now be controlled.

That's what it's for. Maybe you never have those problems, but I always do. I've ruined prints and print beds from bad adjustment of the z height.

I'm also working on an optical endstop for Z which will improve adjustability and repeatability - but even then, a perfect endstop won't compensate for swelling or changes of materials on the bed all by itself. The dial indicator will help me to do that. I've been using it now for a couple of weeks and I really love it.

[Mochaboy]

I find that I have a lot of trouble printing below .2 layers especially on the first layer...It's almost always because of bed height...I'll give it a shot - do you a link to a specific dial indicator you like?

[RetireeJay]

I got the one I'm using at Harbor Freight (Pittsburgh house brand). It's in inches, not mm, but it sure is helpful. I just set it to 0.263" - it's just a number - and my prints on Garolite come out great. Of course your number will be different.

[RetireeJay]

BTW, if you are going out and purchasing a dial indicator for the first time, there are a couple of procedures you should know about that the H-F indicator does not tell you about (it comes with no instructions at all, if I recall correctly).

1) The plunger is gently spring-loaded, but it is not shockproof. So never allow the plunger to be pushed or pulled to a significant displacement and then release it suddenly so that it travels to the stop with nothing to slow it down. The sudden jerk when it hits the end could damage the delicate mechanism inside.

2) When you "clip" the indicator + bracket assembly onto the rods, if the Z position is low enough for the tip of the plunger to rub against the bed while you are getting it into the "clipped" position - then you should gently lift the plunger (from the top) out of contact with the bed and let it back down again. The best precision is obtained when the plunger meets an object on a perpendicular path of motion, not a sliding or diagonal path.

[Mooselake]

Here's one from Tim's openSCAD file. Pic is from Tapatalk and my Nexus 7.



That's the pre-leveling diagonal difference, time to start turning dem screws.

This was printed with PLA, 0.2mm layers, 20% infill every 2 layers, my normal settings. The lock tab on the bottom might have cracked, but it's still working. The HF digital gauge (more $ than I wanted to spend, but pretty handy) just clears the extruder, and the attachment point on the back was a good friction fit with a tiny bit of filing.

Great gadget, Jay. Thanks!

Kirk

Edit: Whoops, display was in inches. Bed is now 0.00 at all 4 corners, -0.01 in center so a tiny dip. This was with the bed off, will heat and redo later.

[plexus]

Here's Jay's adaptor printing at 150-200mm/s on my PB+. The bracket worked with my dial which is a cheap $20 one I got at a local tool shop. the back is black and the printing of the 0.001" scale is in a different place however the dial fit after a little cleaning up of the ABS. thanks jay! great tool. I have been using my feeler gauge but its not exact for leveling purposes.

[RetireeJay]

I have received complaints that the way the bracket hooks onto the rods is too fragile. This is in part due to the use of materials other than nylon, but in an effort to accommodate the majority of users, I've tweaked the design a little.

1) The upper hooks are now about 1.3mm thicker.

2) The flexing part of the lower clip is now an equal thickness for a much longer span, so the flexing is not concentrated in a small hinge zone. To compensate for the thinner cross-section, which weakens the "clipping" force, I've made it wider. However, to keep the principle of "3 points of contact" I've sculpted the shape at the lower clip so that the contact with the rod is limited to a small area.

3) In order to prevent the "small-area" problem with blobbiness, I've extended the depth of the upper hooks to be exactly equal to the depth of the lower clip. When you are printing the part, then, the printhead will shuttle around among those three features until the print is finished.

I've attached both an STL file and an AD_PRT file (Alibre). I have not printed this part myself, so I will appreciate feedback from someone on how this works out in practice before I publish it on Thingiverse. (The AD_PRT file is zipped because the forum won't allow posting the file in its native format.)

[plexus]

I have received complaints ... [snip]

= feedback.

I am KISSlicing now... going to print in polycarbonate @ 300C.

Done. turned out nice... except. the bottom clip is too thin. it basically breaks off with minimal force. you can see from the path that its only wide enough for loops and no infill. this is going to make it quite fragile along the layer interface. some infill would help to strengthen it. remember the bending happens parallel to the layers here so its going to be fragile. it needs reinforcement. nylon will work because it has excellent layer interface properties and flexibility. however for a rigid plastic, for which this part would be best printed, this area needs some attention.

[RetireeJay]

I'll see what I can do. I'm thinking maybe making the "spring" part along the bottom (in that picture) and then making the vertical part thicker so it's stronger. That way we won't be asking layers adhere to each other as strongly.

Interesting challenge.

[RetireeJay]

OK, here's a revision. The "springs" for the bottom clip have been moved to a location where there should be less inter-layer cracking problem. If you look at the design now, you can see that the "spring constant" is easily adjustable just by changing the size of the cutout.

[plexus]

OK, here's a revision. The "springs" for the bottom clip have been moved to a location where there should be less inter-layer cracking problem. If you look at the design now, you can see that the "spring constant" is easily adjustable just by changing the size of the cutout.

Printed V9 in polycarbonate. nice! it works great and clamps on tight. i feel no risk to the "spings" breaking. with PC I find the final print to be a "dirty" so it needed some significant cleaning up: removal of "strings", smoothing of edges and machining of the dial hole to fit the dial. all typical stuff with PC. its strong and rigid. thanks Jay!

[Printbird]

This is great! I can't wait to use this instead of the "business card" style bed leveling. I was about to make a bracket, then came across your part on thingiverse.
Thanks for your efforts!

I'll finish and post my build of Printrbot v2.1 writeup this weekend, hopefully I can pass on some helpful info too.

One thing about the .SCAD, the code above and the text file off Thingiverse both seem to generate an invalid portion for the lower clip.
Is there a newer version of the file or is my OpenSCAD version perhaps too new? (2013.06)

badbracket.PNG

[RetireeJay]

Tdeagan is the one who converted the file to Open SCAD.

Did you try just using the STL file? Or do you need a modification like different spacing between the rods or something? I can't generate SCAD from my software, but I can create a new part with a new STL. I think in one of my versions I tried to make the rod spacing "parametric" but I'm not sure if that's in my "final" version.

If you're printing with a very rigid material, you may want a different design; my design works great with nylon or any plastic that is at least somewhat flexible and also has excellent layer-to-layer adhesion.

[Printbird]

I don't anticipate needing any different spacing (printrbot plus v2.1), and was just going through the motions and taking a look at SCAD. I just wanted to make sure the SCAD effort wasn't wasted by an accidental file truncation.
I'll be printing in PLA which I think should be suitably flexible. (and will report back with my likely successful results) If, for some reason, I need to reinvent the wheel, I'll be sharing it here!

Thanks!

[REPRAP SQUAD]

If I remember correctly the spacing between the rods on the plus v1 and v2 is a little bit different. As for the v2. 1 I would assume it's the same as the v2. Can anyone confirm this to be true as it seems like I checked it quite a while back and I think the plus v1 rods are spaced a little further apart. Not much if I remember correctly.

[Printbird]

You're right, the v2.1 (and I assume the v2) rods are a little closer together than in the V9 version of this adapter. I'll see if I can import into solidworks and modify, otherwise, perhaps take a closer look as to why the .SCAD file is acting weird on the lower hook thingy. (Also, my dial indicator has a larger mounting ring than the slot allows. So, if I have time, I will investigate the SCAD, as that is probably adjustable too.) I think the idea of mounting directly on the X axis rods is a good one, as I don't see a very obvious place to mount it on the head itself.

[RetireeJay]

If you want, I can provide dimensioned drawings so that you can re-design it yourself for your specific situation. It's not really a big project if you're comfortable with a CAD program.

[Printbird]

I appreciate it--I've been using Solidworks for my test projects so far with good results, so your drawings would be helpful (I see a PDF in your Thingiverse object that has some dimensions too.) Because it's fun, I will probably just go ahead and try to make a specific one for the v2.1 and see how that goes.
(and Happy Thanksgiving!)

[RetireeJay]

OK, I have added several more dimensions to the drawing. I also discovered that I could export the drawing in DWG format, but the forum won't let me attach it. If you really want the DWG you could PM me with an e-mail address and I could send it to you directly. (Or maybe Thingiverse would allow it to be posted there???)

Jay

[REPRAP SQUAD]

I appreciate it--I've been using Solidworks for my test projects so far with good results, so your drawings would be helpful (I see a PDF in your Thingiverse object that has some dimensions too.) Because it's fun, I will probably just go ahead and try to make a specific one for the v2.1 and see how that goes.
(and Happy Thanksgiving!)

Keep me updated on the one you modify as it would be nice to print one for my v2.

[Printbird]

I aim to get to it tomorrow morning. Shall keep you posted!

[edit: didn't get to it this weekend. Hopefully I'll have time soon.]

[Printbird]

OK, I have added several more dimensions to the drawing. I also discovered that I could export the drawing in DWG format, but the forum won't let me attach it. If you really want the DWG you could PM me with an e-mail address and I could send it to you directly. (Or maybe Thingiverse would allow it to be posted there???)

Jay

Great, thanks for taking the time to do this!

[Printbird]

Thanks for the inspiration and design for reference. I went ahead and did this this morning. The results are amazing; my initial layer is perfectly flat with great adhesion.
Here's the STL:
viewtopic.php?f=89&t=5712

[RetireeJay]

See my further notes on Dial indicator usage at viewtopic.php?f=89&t=5712&p=37141#p37141

[REPRAP SQUAD]

Thanks for the inspiration and design for reference. I went ahead and did this this morning. The results are amazing; my initial layer is perfectly flat with great adhesion.
Here's the STL:
viewtopic.php?f=89&t=5712

I'm printing out your new bracket right now and it looks great so far. I've had lots of recent issues with my printer as of late. So a great print would let me know it was worth the 2hrs of tare down and rebuild of my print plate setup