Testing Update

A quick update: I did more testing last night after adjusting a plank and managed to get the wall up to 10ft/min; a very reasonable climbing speed. Ultimately, i would like to get it running up to 15 or 20ft/min.

I suspect that the stalling issue I had recently was due to a change in temperature as much as issues with the planks. Basically, the small, low torque motor I am using only has to overcome the friction in the gearbox. It doesn’t need to really drive anything, but when the temperature drops and the oil becomes more viscous that drag in the gearbox increases.

At 10ft/min the motor is turning at 600rpm, and having a torque curve like the one below, I am guessing it is putting out, at most, 200 oz*in of torque at that speed. So very little torque.


I think my options for dealing with this are: lower oil viscosity, use a lower ratio gear box, or a higher torque motor. The use of a lower ratio gearbox might seem counter intuitive but if we go from a 60:1 gear ratio to a 15:1 ratio the motor will be at the same operating point and the wall will be moving much faster. Since the motor isn’t really driving the wall, it just needs to overcome the friction in the gearbox, which will be comparable, this should work. In the end I will probably end up trying all three options.

The 60:1 gearbox wasn’t my preferred choice when I was designing the thing i was hoping for 20 to 30:1 but it was at princess auto and it was cheap. I guess this is what happens when you compromise on your design.

More Testing

This week I pulled out and trimmed a couple of the planks that rubbed badly and things seem to be running smoother. I also did a test run with the wall set to 30° overhang. Aside from the climbing being really hard (I seriously need to buy some holds with texture) it ran pretty well.

Me climbing on the wall, set to overhanging. Thanks to my shoe for taking the picture.

On the back side of the wall the chain was dragging on the cross members (due to the previously discussed lack of a track on the back side) but it wasn’t noticeable. I still think I will try to fix this in the near future, just to prevent undo wear on any components.

I am very impressed with how solid the thing feels with the wood in place. I had added more bracing to the design after playing with just the frame, but now I think I will take it out.

BUT when I put the wall back to vertical the damn thing seems to stall before ramping up to full speed. I’m not sure what the problem is but I’m guessing it is just a tighter fitting plank binding. I guess we’ll see.

Planks on Testing

Planks are finally on. It took a lot of planing and sanding, but they are on.

After getting the planks on I was anxious to start climbing. I looked around my garage and realized: I don’t own any climbing holds. I could have mooched and borrowed holds but I thought it would be faster to just start screwing whatever crap I had laying around my garage to the wall. Morgan also volunteered some scrap wood to the effort.

The completed wall and very professional looking holds. I am pretty excited about this!

Here is a gif of my first, low speed test. And yes I realize my choice of climbing pants is around 15 years behind the times. Or you can see the full video in all its glory. Most of the noise in the video is from a crane across the street, not the wall.


My initial thoughts are that the whole thing runs pretty well and feels solid. It squeaks a bit but a little more wax in the tracks will fix that. (I need to find some bee’s wax. It, apparently, is a better lubricant for wood.) The chain isn’t spaced perfectly on some of the planks which leads to thumping noises as they go over the drive sprockets. So I will be pulling off those planks to adjust their fit.

I do wish the whole thing was about 6″ taller just to lend a little more buffer when you are climbing, 12″ would be even better. I don’t think you would be able to make a wall any shorter than my 9′-2″ ceilings usable, unless you made it permanently overhanging. If I were to start from scratch I would knock a hole in the ceiling and have the top poke into the attic.

Running at ~30° overhang would let you fit the track portion of this wall under an 8′ ceiling, which would still be usable. So there are options for this to work with a lower ceiling.

After all the trouble it took to make the planks I would consider laminating 2 layers of 3/4″ plywood together to make the planks or trying 1″ plywood. This would eliminate the issues I had with twisted boards, bowed boards, and varying thickness. So it would remove labour from modifying boards while adding some by requiring gluing them together. I suspect it would lead to a better end product though. You would be able to use cheap plywood for the back layer and better stuff for the outside and it will make it easy to countersink the t-nuts so you can used the standard bolt for a given climbing hold. AND you could ensure that all the planks would fit snugly in the guide channels so they wouldn’t feel wobbly without any sanding.

Additionally, in the future, I would try including a track on the return side of the wall as well as the climbing side. This would keep the planks running vertically, and prevent the chain from rubbing on the crossmembers in the wall regardless of chain tension and wall angle. It could be a looser fitting track to keep friction down. I think this addition would make it easier to control chain tension and make the unit easier to use. I might try to add this feature to this wall just to try it out.

Planks on return side aren’t in a track and sit crooked. This isn’t a huge issue but it does make chain tensions a harder to adjust, and it means when you have the wall set to overhang the chain will sag in the middle and might rub unless the chain is tight.

Next step will be electrical design. First, I think I will just purchase a prefab stepper driver, power supply, arduino etc and get this wall running as a stand-alone unit. Later I think I will design a custom board with everything on a PCB, maybe add some whizzy features like wifi so you can control it from your phone.

Lastly, I want to mention how impressed I am with my Bosch cordless rotary hammer (this one if your curious). I had only intended to use it for putting up routes, but my corded drill wasn’t suitable and I ended up using my hammer drill for everything on the planks. So it drilled around 150 t-nut holes (1/2″ x 1.5″ deep) and sunk around 150 1.5″ decking screws on about half a charge.

If you are shopping for a drill for route development this one is light enough for drilling on lead at 6.25 lbs and will punch more than a dozen 3/8″ x 3″ holes in moderately hard limestone per charge. However, if you only plan to bolt on rappel I would consider finding a heavier drill with a bigger battery (more Ah) to get a few more holes per charge. I am considering buying a bigger battery for this one; bosch makes compatible batteries with almost 3x the capacity.

Topo and Update

Hi all. This week I drew up a topo for the new route (Family Jewels 5.10b/5.8) described in last weeks blog post. It can be downloaded here. Also, I realised that neither of the topos for emerald wall include direction to find the cliff: it’s here.

I have been putting some work into the wall. Adding the planks now. This has required a bit of debugging. It has proved a little finicky. It needs everything aligned properly and a good amount of tension in the chain which, I suppose, isn’t surprising. And it is taking a little more torque than I expected to run. So I will either need to use a bigger motor, a lower ratio gear box, or just run at a lower top speed for now. Or maybe just more lube it is really only a little short on torque.

(edited 14Oct2016: topo link now goes to mini guide for Emerald Wall)

Adventure Time: Bolting in the Crowsnest Pass

You know what’s more fun than building climbing equipment? Climbing! Especially putting up new routes or problems.

Morgan and I set out early Saturday for the emerald wall in the Crowsnest pass. We were hoping to put up a new line but didn’t have a clear objective in mind. We weren’t sure if we would go ground up or top down, but we knew it would be mulitpitch and bolted (this wall tends to offer few gear placements and most of them are in absolute choss).

The view from the emerald lake parking lot.

After considerable deliberation in the parking lot we settled on a line we had eyed a few years ago on a previous excursion. It starts left and up the scree slope from the line, Rags to Riches, that Morgan put up last year. It aims for a, more or less, direct line from the scree cone to the ledge with a big lone tree, starting maybe 40 feet left of plumb.

Since we were bolting on lead, with all the neccesary gear tethered to the leaders harness, we decided to lead in blocks. Just to save time at the belays, moving gear. I took the first block and the crux of the day ended up being a bulge only about 15feet off the deck at around 5.9 or maybe 5.10a. (the grade will likely change as following parties tear off more of the holds) Though, the real crux was getting the first bolt in about 8 feet off the deck, the stance got super pumpy while drilling. Morgan may have given me a bit of an boost while drilling.

Morgan at the belay on top of pitch 1.

I got up about 2.5 pitches before the battery in our drill ran dead and we swapped leads. Aside from the early crux the climbing was all nice, cruisy 5.8ish slab, with decent quality rock. From there Morgan lead the next pitch and a half to the big ledge with the tree.

Morgan torquing up the first bolt on his block. The tree we ended at can be seen above him.

There was some pretty poor rock quality through some of the block Morgan led. We hucked off a lot of choss and there is a lot more to go. But there is some steeper climbing with nice movement that will be awesome once it has been cleaned up. Here are some pictures from the top of pitch four. Including me with a sandwich.

There is another pitch or two above that we would like to add to this route. (they will be awesome, definitely great rock) But our hardware supply was dwindling so we elected to head down and add a couple bolts to the earlier pitches to make them a little better protected. While rappelling and adding these bolts, however, we managed to bust the tip off our drill bit (no spare packed) and run the second battery dead. Soooo, the first pitch is presently pretty run out and the crux of the route. Morgan hopes to return very soon and rectify this.

*Update July 7, 2016* Morgan went out yesterday and added bolts to the first and second pitches to protect the previously run out sections. So it should be reasonably well protected at this point. He also added an alternate first pitch (5.5) to make the route more accessible and consistent after getting on the first pitch again and deciding it was closer to 5.10b.TOPO HERE!

Preliminary Test

I have been away on vacation, so not much has gotten done and I am heading out climbing this weekend so my progress will not be accelerating any time soon. But I am still plugging away at this and last night I did a little testing: I ran the wall without planks on.

Second chain assembled and mounted. Test motor in place.

The motor and driver are borrowed off another project, so it won’t be quite the same on the final assembly. I am hoping to get away with a slightly smaller motor, which should be both cheaper and a little quieter. This one did produce a bit of a whine when running.

Success…. So far.

Sorry the video is a little shaky, I am holding down the “go” button way off to one side while holding my phone in the other.

I only tried a couple, slower speeds (up to ~8 fpm) but so far everything seems to be running smoothly. Top speed will, hopefully, be around three times faster. The motor bracket provided great alignment without any tweaking.

Final Assembly: Part 1

With the last of my parts picked up I could get down to business again.

Next up was assembling the chain. To attach wooden planks (2x8s) to the chain we need to swap out every 14th link with an attachment link. This was a lengthy process since it requires pushing out 116 pins to replace 58 links so we can mount 29 planks. It’s a lot of time with a chain breaker.

2 down 56 more to go. An attachment link is shown in the top left. They have 2 bent flanges that we can use to screw the link to our moving planks.

Additionally, I needed to make a spacer since I am using a bearing on the drive shaft that I hadn’t accommodated in the design. So I cut one out of some scrap plywood, I think it will be stiff enough to support the bearing.

bearing spacer

I had to go buy some metric bolts for the gearbox (which was a surprise) then we could get down to business and mount the drive shaft and chains. This ended up being a two man job: one to support the shaft and the other to ram it into the gearbox. It might have been easier, if heavier to lift into place, to get the shaft in the gearbox on the ground and then mount the whole assembly to the wall.

The shaft, gearbox, and bearing in place. The chain isn’t really done yet it was just up to check the length. Everything is still working as it did on my computer.

With all the attachment links put in the chain we put it on the sprockets. It looks like there will be 28 7″ wide planks and one small 2″ plank to fill the remainder. ie. the chain length was not evenly divisible by 7″.

chain in place


Next up will be a bunch of woodworking to rip our planks to 7″ width from standard 2x8s.