so far so good...

[WayOutWest]

My liteplacer arrived yesterday!

Stayed up until the wee hours of the morning and made it as far as step "X.17: Finished Frame" -- wow, it actually looks like a pick-and-place now!

The instructions are absolutely fantastic. I think any company considering manufacturing any sort of product that requires customer assembly ought to get Juha to give a lecture to their documentation-writers. The world would be a much happier place.

The only major obstacle so far is the laser-cut holes for the 608ZZ flanged bearing, in both the "Y-bearing plate" and "Y-pulley plate". They are too small! I filed and filed and filed them, and they're simply cut too small. In the end I had to hammer the bearing into the hole, which warped the plate! Juha, please consider throwing out your stock of these plates and having a new set cut with a larger hole. Some sort of rubber gasket to hold the bearing in the hole is better than relying on the laser-cutting tolerance to be just small enough to grip the bearing but just large enough to allow insertion -- I don't think the lasercutters are accurate enough for this. Also the holes for the M5-35s in the "Y-bearing plate" are also too small: the non-threaded part of the M5-35 is actually wider than the threaded part, and it does not fit into the hole. I had to re-drill the holes a tiny bit larger to get the screw to go all the way in.

Another minor comment: it would be helpful for those of us in the US if you could make sure to include at least one extra of every metric screw/nut/washer. Unfortunately the hardware stores here carry only a very very very limited supply of metric parts, so if I lose one of the screws I will have to order the replacement online -- at least a two-day delay and quite expensive unless I wait a week. So far I haven't lost any parts yet, thankfully!

Today I'll try to finish the remaining mechanical steps.

[WayOutWest]

A few other suggestions:

- It might be best to attach the luer lock AFTER completely assembling the pnp head. I accidentally ripped it off while putting the rod into the shaft, but fortunately I didn't damage the threads.

- In many places the instructions say to put the extrusion nuts into the aluminum bars before putting the relevant screw (and whatever is attached to it) into the extrusion nut. Often I found it was easier to do this the other way around: attach the nut to the screw (and whatever it holds) first, then insert the nut-screw-assembly complex into the aluminum rod, then tighten the screws. Obviously it isn't worth re-doing all the instructions and graphics, but it might help to mention this at the first step where the extrusion nuts are encountered: "we will be using a lot of these extrusion nuts, and you have the option of attaching the nut to the screw before inserting the screw instead of afterward if you like; for short screws this can be easier than trying to land the screw in the hole of a nut that is already inside the aluminum rod".

- The customer photo showing the fully assembled pnp head was IMMENSELY helpful. I would have gotten stuck on that step if I hadn't had the photo. Thank you to whoever sent it in!

- It was a very good idea to have the most complicated and tedious part of the job (the pnp head) go first. That way by the time I was tired out I was doing steps that had a lot more "positive feedback" in terms of seeing the machine come together. If the pnp head were the last step I wouldn't have gotten as much done in the first day.

- I suggest adding a preparatory step telling the user to sort the bags into four piles: washers, screws, nuts, and everything else. Then sort the washer/screw/nuts piles by diameter, then by length. It sounds silly and tedious, but it saves a lot of time later. Almost every step of the instructions requires selecting a washer+screw+nut, often a different size from the step before, so not having to dig through the whole pile of bags every time makes things go a lot faster. The two minutes of sorting saves twenty minutes of searching.

- The laser-cut holes for the eccentric spacers are VERY VERY tight. Maybe consider increasing the hole size a tiny bit? It shouldn't affect the functioning of the spacer -- as long as the hole is smaller than the octagonal part of the spacer it won't matter, right? So no need for those holes to be so tight.

- Everything about this machine seems exceptionally well-designed, except the high-Z limit switch. That does not look terribly reliable, having it triggered by a stray bolt sticking out of the machine... But maybe it will turn out to work better than I thought.

- The text of the instructions mentions that the HBLSS5 bracket is not symmetrical, but I didn't notice that... unfortunately the first one that is used is part of the pnp head, and if you install it backwards you won't notice until many many steps later when you install the Z-axis motor. For stupid people like me it might help to make the warning about the HBLSS5 orientation bold, or red, or even blinking Or maybe an arrow on the diagram; I find that I pay much more attention to the diagrams than the text.

- In "Frame, Step 1" you mark the front side of the machine with a green line, and then in the next step you use the green line to mark the *back* of the machine. This was a bit confusing at first. I suggest using a different color (blue, red, etc) for the back. I know you've already put a ton of effort into the graphics, but it shouldn't be hard to do a color-for-color replacement in the "Frame, Step 2" image so people don't assume that the green marking in the two steps means the same thing.

- It would really help to see "Frame, Step 14: Y axle" fully exploded for the entire rod, with all five shaft collars and the entire length of the device. I know that's a really huge diagram, but it would make it clear which "end" of the machine to start from.

Okay, more tonight or tomorrow!

[WayOutWest]

Ah yes, one last hint.

For those in the US with our demented system of measurement, it turns out that a 4'x2' work surface is exactly the right size for the liteplacer and a laptop. So for a tabletop you can just drive to the hardware store and pick from what they have.

Home Depot carries a large variety of surface materials in that particular size. I got a thick 3/4" piece of MDF for the support and also thin "whiteboard" and "chalkboard" sheets to conceal the texture of the MDF. I'm not sure if a white background or black background will work best yet; most of the chips I use are black, but they arrive on a white tape. So I'll try both backgrounds.

[mrandt]

Hello Adam,

welcome to the forum and greetings to the "Evergreen State" from another happy LitePlacer user in Europe

Thank you for sharing your experience so far.

I also thought Juha's instructions and drawings were great - and he is constantly improving them based on user feedback. So I bet he will appreciate your comments.

As I did not like the Z-high-limit switch, I designed a small 3D-printed part to mount it underneath the gantry - see photos below:

Malte's LitePlacer Z High Limit Switch

high_limit_z_01.jpg (18.8 KiB) Viewed 6913 times

Malte's LitePlacer Z High Limit Switch

high_limit_z_02.jpg (20.84 KiB) Viewed 6913 times

If you have access to a 3D-printer, I will happily share the STL file - let me know.

Best regards
Malte

[mawa]

as long as the hole is smaller than the octagonal part of the spacer it won't matter, right? So no need for those holes to be so tight.

The idea of the eccentric spacers is to adjust the distance between the V role and the maker slide. If the beveled part fits snuggly in the hole you can very precisely do this adjustment while - and that is of some importance - the screw is a bit loosened.

By adding additional clearance having a hole larger than the beveled diameter would let this part to be able to diametrically move inside the hole.

Therefore I would suggest to take the time and hone down the hole with a file until you can insert the beveled part and it is able to rotate 360°. I used a round diamond coated file and it took me only a few minutes.

[WayOutWest]

As I did not like the Z-high-limit switch, I designed a small 3D-printed part to mount it underneath the gantry - see photos below:

Oh neat! Yes, that looks like a much better solution. I like it!

If you have access to a 3D-printer, I will happily share the STL file - let me know.

Funny that you should mention it, that is my next project after the liteplacer. In fact I was in the middle of setting up the 3D printer when the liteplacer arrived and "bumped" it down the priority list! Yes I'd definitely be interested in the STL file, but it may be a few weeks before I can go back to working on 3D printing... the 3D printer is strictly hobby whereas the liteplacer will hopefully be used for a few lightweight work tasks.

[WayOutWest]

as long as the hole is smaller than the octagonal part of the spacer it won't matter, right? So no need for those holes to be so tight.

By adding additional clearance having a hole larger than the beveled diameter would let this part to be able to diametrically move inside the hole.

But wouldn't the pressure of the makerslide on the v-groove wheels keep it pressed against one side of the hole?

I'm probably completely wrong about this; I'm a total amateur when it comes to all things mechanical (that's why I was so glad to come across the liteplacer... Juha did all of the mechanical engineering far better than I ever could and yet hasn't tried to hide the programmable parts from people who want to tweak or improve them).

[WayOutWest]

Update, start of day #3: all the mechanical assembly tasks are finished.

A few more minor comments:

- Drat! I wound up at the last step with two extra M5-8 screws and lacking two M5-12 screws. Curses! The up-looking camera will have to wait.

- I couldn't find instructions for mounting the drag chains, but fortunately one of the photos on liteplacer.com was taken of a model with them installed. If it weren't for this I probably would have mounted them backwards!

- Unfortunately I didn't realize that the 20awg-4conductor shielded wiring is not exactly a commonplace part in hardware stores, so I didn't order it beforehand. Bummer! It looks like I may be stuck until Monday waiting for parts to arrive. At first I wondered why the wire wasn't included with the liteplacer, but I realized that much cable would add a lot to the shipping weight, and coming from Finland that's an unnecessary expense. Is this the reason for not including it? Might be nice to mention that fact and warn the customer to order the cable at the same time they order the liteplacer... or at least not assume that it can be found at a local hardware store.

I have noticed, however, that the two power conductor pairs of many USB cable do in fact meet the specifications (the cable has four conductors, but two of them are thin data-only 24awg). Since I have a veritable ocean of USB cables lying around I might cannibalize a dozen of them instead of waiting until Monday Is it a bad idea to use the 24awg pair that shares a shield for the limit switches? I assume this is a bad idea since the noise from the drive wires will couple to the limit switch wires, which is exactly the problem that the shielding is supposed to guard against.

[WayOutWest]

Day four: motion on all four axes, all six limit switches checked and working.

More notes:

- It would be useful to mention the phenomenon of "motor stalling" earlier on. What is this, anyways, and why does it cause the *software* running on the TinyG's microcontroller to crash? This caused me a great deal of frustration and I still think there's no excuse for it... when the host kept losing its connection to the TinyG I thought clearly the problem must have been some sort of software thing, since the job of the TinyG's software is to report physical malfunctions, not crash itself in response to them!

- Also, why is it necessary to reboot the TinyG when a limit switch is triggered? This makes no sense. The "Max X" limit switch should simply forbid any further motion in the positive X direction, not crash the microcontroller. If I trigger the Max-X limit switch, I should be able to send a "move negative X" command immediately, without having to walk over to the machine to hit a physical button.

Very disappointed with the TinyG, it is not of the same quality as the rest of this machine. Its default reaction to things seems to be crashing.

[mawa]

Well the TinyG keeps track of all positions and when you hit any limit these positions are invalid and any continuation is impossible.

Hitting a limit switch is an exception and therefore a clean reset with appropriate notification of the host is IMHO the best and safest reaction.

If you have your machine running clean without stalling and your homing XYZ works the chance of hitting a switch is very small.