Here's what the components look like installed in the box, prior to adding the cabling for the connectors. I will be making a new lid for the box to give more room for cables coming up from the Shield.

The carriages on both towers sit slightly crooked. One roller on each carriage does not make contact with the rail. This is probably from my installing the rollers while the carriage was already in place on the rails. In principle, three points of contact are the most stable, but I'd rather all four be in contact, since they are supposed to be an interference fit.

Wiring is done. Whether or not it's done correctly remains to be seen.

I've run into an issue setting the reference voltage. the video says MaxAmps = Vref * 2. According to the Amazon listing, the stepper motors have a rated current of 0.4 amps. That implies Vrev = 0.2 V. But when I try to set Vref that low, the voltage goes crazy. I've set Vrev=0.4V, which is about as low as I can reliably set it.

What Vref setting did you use for the stepper motors?

On most installations some of the guide rollers are on fixed centres and some are on eccentric mounts. This allows any clearance to be taken up. I suggest you take off the guide rollers and see if you have the two types.

Originally Posted by tspeer

Wiring is done. Whether or not it's done correctly remains to be seen....

Well, that didn't go well. I found the female pins to be harder to crimp properly, and they were far harder to insert into the shells. The male pins could be pulled until they clicked, but the female pins had to be pushed in.

The male plugs from the cables would not stay plugged into the female cables, and would fall out easily.

I've removed all of the female-to-female cables from the box. I ordered some 4-pin and 1-pin female-to-female cables that are intended for use as jumper cables on Arduino breadboards, and I'll see if they work better.

My backup plan is to cut all of the connectors off the cables I've made, cut the wires coming from the stepper motors and switches (gulp), and use soldered splices to use the cabling to extend the wires. That will allow the factory-made connectors to be connected directly to the Shield inside the box. I suspect this is how the Windy Hills prototype was done.

Originally Posted by Bartplus4

On most installations some of the guide rollers are on fixed centres and some are on eccentric mounts. This allows any clearance to be taken up. I suggest you take off the guide rollers and see if you have the two types.

AFAIK, all of the supplied rollers were identical. I couldn't tell any difference between them.

On this system, I think it's the compliance of the rubber rollers that is intended to avoid any mechanical slop and compensate for variation along the rails.

All four rollers should be touching and that is also the case on my and Nathan's machine. I'm very certain this issue is due to the assembly issue. I will send you two new carriages and then it should fit perfectly. Again I apologize for this issue. I will also fix the manual to make the assembly go smoother.

I set Vref to 0.5V since my steppers are rated at 2A. And now I just realized that something was mixed up with the Amazon links. I'm confident that your steppers will still work fine on the machine if you get the Vref setting correct. Another option would be you return the current steppers and get the 2A steppers. Sorry again for this inconvenience.
https://www.amazon.com/gp/product/B0...?ie=UTF8&psc=1

Originally Posted by Ma4din

All four rollers should be touching and that is also the case on my and Nathan's machine. I'm very certain this issue is due to the assembly issue. I will send you two new carriages and then it should fit perfectly. Again I apologize for this issue. I will also fix the manual to make the assembly go smoother.

I set Vref to 0.5V since my steppers are rated at 2A. And now I just realized that something was mixed up with the Amazon links. I'm confident that your steppers will still work fine on the machine if you get the Vref setting correct. Another option would be you return the current steppers and get the 2A steppers. Sorry again for this inconvenience.
https://www.amazon.com/gp/product/B0...?ie=UTF8&psc=1

The steppers specified in your instructions are listed as "4pcs Nema 17 Stepper Motor 26Ncm (36.8oz.in) 0.4A 34mm Length for 3D Printer/CNC Machine/Robotics". I don't think they are going to work. I can't set Vref low enough to limit the current to their rating, and I'm not confident they will have enough torque. I'll order the 2A version.

Here are photos of the tower carriages taken from above.

I found the problem. And old design of the nut blocks snuck it's way into the production line. They are to tall which doesn't allow the lead nut to be centered on the lead screw. Which then causes your carriage to be pushed of the rails and probably also is the reason why it was so difficult to assemble the carriage.

I will send you new nut blocks and carriages.

Did you already glue the nut blocks in? Also on the x axis? Because that issue will persist on the x axis as well even though it does have a lesser affect. If you glued them in I have to send you new X carriages as well.

Yes, they are glued in, all axes.

Instead of using medium CA, per the instructions, I assembled the carriages without gluing, then applied thin CA to allow the CA to wick into the assembled joints. I figured that was the way for them to be aligned with the screws and carriages and I didn't have to worry about the CA going off before I got everything in place.

OK. I will send you new components over the weekend.

Originally Posted by Ma4din

...
I will send you new nut blocks and carriages. ...

Thanks. I received the new stepper motors and I've started disassembling the rig.

When you send the replacements, please include two replacement spacers. The thin CA I used to secure the nut holder to the carriage ran down and glued a spacer to the carriage in each case.

I also need a replacement R1 and M5x25 screw. One of the nuts securing it to the base was cross-threaded on the screw, and wouldn't turn. I ended up pulling the nut completely through the plastic to get it off. The thread on the end of the screw is damaged, so it needs to be replaced as well.

You are lucky. Originally I was planning to ship yesterday but couldn't make it. So I will add the additional missing parts. No problem.

I had to wait for the new motors to start the disassembly, and I discovered the issues then. I'm still waiting for the new cables for the box to come in, too, but that won't hold up getting the rest of the system together.

The new motors, by the way, are noticeably longer than the previous ones. I'm a lot more confident these will have the oomph to do the job.

I have a 60" bow made in the Martin Brungard style, and it will be interesting to see how it fits. It has a spring scale on the back side, so it's heavier than a typical bow, but it also has a turnbuckle that allows the tension to be adjusted to a specific value.

I've been using stainless steel safety wire, but I also bought the nichrome wire recommended in the instructions. It'll be interesting to see which one works better. The safety wire may be a larger diameter, but it can take a lot of tension to minimize bow in the wire. The nichrome may result in a smaller kerf.

Replacement parts arrived today - thanks!

The instructions should be more explicit with regard to wire routing. It was only the on the second time around that I noticed and appreciated the purpose of the slot in base of R3/L3. I see now it is for routing of the stepper wires out of the tower. The wires have to be routed before installing the Z-2020 rails in R5/L5 and R3/L3 because once the rails are in place, there is no room for the connectors to pass through. The instructions make no mention of this, nor do they show the wires in the illustrations.

I found it odd that the wires came directly off the connectors of the limit switches, with no provision for strain relief. Also, the wires exiting the bottom of R3/L3 will be rubbing against the sharp edges of the printed part. So I used a hot knife to cut a slot in the side of R3/L3 to accept a cable tie. The cable tie secures the wires to R3/L3. I suggest this slot be incorporated into the production parts.

FWIW, I covered the outside of the Z-2020 with a strip of Scotch tape to seal in the wires. The tape matched the width of the rail perfectly. Although the stepper wires are on the outside, and shouldn't interfere with the carriage, now they are completely inclosed.

Assembling all four rollers to the carriage and then clipping it onto the rails works SO much better than trying to put the screws in with the carriage in place, as shown by the illustration in the directions. This time I'm waiting to glue in the nut holders until I know everything is properly in place.

I have the mechanical assembly finished. Unfortunately, my order of female cables from Amazon for the box appears to be lost in the mail, so it may be another week until I can test it out.

There's a lot of freeplay in the left tower. I've place a framing square next to it to show the movement as I move it back and forth:

Left Tower Freeplay (0 min 19 sec)

Is there any way to shim or correct for this?

I haven't glued the nut holders in, yet, and that may help. But it seems to be related to only three rollers contacting the rails at a time.