When I brought the assembled rack inside, I was pleased to see that it was quite stable when the racks were folded closed. That made sense, because the racks were over the base. When I unfolded them with the outer racks opened about 45 degrees, the stability was okay. If I unfolded them out more (or opened the inner racks more than 90 degrees), the rack leaned back at the slightest touch. It didn’t fall, but it was not stable.

I played around with preventing the rear casters from rotating (which would keep the wheels as far back as possible) to see if that would help. It didn’t, and it made the cart harder to maneuver. I had to do something else. I wondered about adding some sort of fold-out support in the back, but that would be fussy to use, and it would look like it was tacked on after the fact. My only real option was adding weight.

There was room under the cart for a 5 lb. dumbbell plate, but I didn’t think that would be enough. (I used canned vegetables to try to determine how much weight I’d need, and it was about 7 lbs.) I also wanted to be able to put as much weight as possible near the front of the cart, which meant I needed multiple smaller weights. I considered ankle weights (which I think use some sort of heavy metal “sand”), inserts for weighted vests, rectangular weight machine plates, etc. Everything I found was either too long, too thick, or too expensive. Then I thought… rebar. I looked it up, and a 10-foot length of #4 rebar (1/2” diameter) weighs almost 7 lbs. And it’s cheap.

I did some rough sketches and calculations and determined I could fit over 10 feet of rebar inside the base. Three four-foot lengths cost about $15. I thought I could cut it with a hacksaw, so I bought an extra set of blades for another $5.
 
The rebar I brought home was dirty and slightly rusty, so I cleaned it up with a wire wheel on a hand drill. Then I tried cutting it with my hacksaw. It took about two minutes to cut through. That wasn’t too bad, and I can always use more exercise.

A few hours later, I had fourteen 8” lengths and ten 3 1/4” lengths. I used my Dremel and grinding wheel to dull the cut edges to protect my fingers.

The pieces stacked neatly inside the base, two layers thick (see the photo at the top of this post). Now I needed to find a way to secure them. My first thought was to cut a cross-shaped 1/2“ plywood panel and attach it to the inner skirt with pocket holes. The rebar would rest on that panel.

I didn’t go that route, though. I knew I’d be installing the weights with the cart on its back (and the base vertical), and I wanted something easier to handle, so I decided to build some sort of cradle that I could move into place easily. I started by wrapping the rebar with duct tape. (This is the first time I’ve used it in a woodworking project.)

Then I started playing around with some 1/4“ plywood panels and 1/2” plywood scraps.

After an hour or so of seat-of-the-pants engineering, I ended up with this cradle. It won’t win any beauty contests, but it did the trick. I was worried about the glue joints on the thin sides at the front, so I reinforced them with brad nails (glued into place) and small gussets (anything that would reinforce the joints without impacting the range of caster rotation.)

I wound up using all the rebar except one of the 8” lengths. I attached the cradle to the cart’s base with eight 3/4" lath screws. The cradle plus rebar weighs nearly 8 lbs. The cart, racks, base, and skirt have a combined weight of around 16 lbs. The cart’s not going to tip over now. 

I still wasn’t done, though. With the dowels installed, it was obvious that some of the spool holders weren’t vertical, but leaned back because their tilt stops were a bit too low (around 1/16”). Of course, I knew about this since I assembled the racks, but I decided to soldier on and worry about it later.

Fortunately, I had some spare magnets, and they’re 1/16” tall. I was able to use them as spacers to level the racks. I didn’t even have to glue them in. If I put a loose magnet anywhere near the one in the center of a block, it flew into place, and it stayed there when I tilted the block. Here, I’ve just corrected the block in the middle and haven’t gotten to the one on the bottom yet.

The spacer magnet is barely visible. Thank goodness for an easy fix.

The last problem is one I have not fixed yet, and possibly will leave as-is. When my wife started filling the rack with thread, she noticed how far out/down the blocks tilted. The wider angle makes it harder for her to place spools in the rack. I had designed them to tilt out about 45 degrees, but they tilted much farther (somewhere between 70-80 degrees). (She told me she'll get used to it, but I'll describe the problem.)

I had noticed this problem when I was constructing the racks, and I even spent an afternoon trying to correct it. Here’s what was happening… The top of this rendering shows the intended design. The tilt stop is supposed to contact the corner of the block’s beveled notch. The bottom of the rendering shows what happens to the spool block if the tilt stop is just 1/32” shorter (horizontally). (The small white rectangles represent the difference.) The stop doesn’t contact the notch at all. Instead, it doesn’t stop the block’s rotation until it contacts the block’s lower front flat.

When I discovered the problem during construction, I experimented with adding thin spacers to the front edges of the stops. I think my spacers were slightly too thick, so they stopped the rotation too early (which is far worse than letting them tilt out too far). At the time, I sanded off the spacers I had just glued on and decided to live with it. Now that I’m finished with the rack, I may revisit the approach (using tape to add spacers of various thicknesses) to see if I can make it work better this time.

As much as I wanted this rack to work like a Swiss watch, it just wasn’t meant to be. If I were to make another rack (something which I do not ever plan to do), I’d come up with a more forgiving design for the stops.

This marks the end of this blog series. Like the project, it dragged on longer than I expected.

But, being an old software engineer, I can’t help but present an enhancement I think about every time I look at my model—a grab handle. Ladies and gentlemen, I present to you… Pteranadon Rackasaurus.

I haven’t run it by my wife yet; she’ll see it when she proofreads this post. (I wouldn’t hold my breath waiting for photos of an implementation. 😊)

Thank you for reading this far. I truly appreciate it. In the next day or two, I’ll take real photos of the finished rack and include them in a project post.