AlexanderK176

Almost certainly, and for rectangular windows, diagonal placement is most likely best. I should probably mention that in the text.For my windows, the bottom was simpler and less invasive (top would require new unsightly holes)

Thanks for the excellent points, something to add to the Instructable.I had wondered about the atmospheric pressure effect on IGUs. In fact, the big window (5'x6') that I cleaned was "vacuum locked" for years - the panes were touching in the middle (now that does affect the U-value!). It didn't recover on its own when I made the holes, but it recovered when I pulled on the glass with a suction cup for half a day.

I didn't work work on any windows with metal seals, so I didn't have photos to include to describe this. I have one I may experiment on in the future.My suggestion for metal seals (in the "Overview" Step) was to drill from the side or bottom of the seal. In most cases, the IGU will need to be removed, or at least shifted. On the upside, you can position it conveniently on a workbench, and even rotate it to drain.In glass doors it may be possible to drill right through the door edge (though it would have to be very precise). Then again, usually it is easier to remove the IGU.

Thanks for the feedback. It's true that I didn't focus on the audio enough. When I have time, I may re-do the video. It may have been better off with after-the-fact narration than with the original voice track. I meant to switch to a shotgun mic for outside and forgot, but I may be able to crank up the volume a bit. My video editor seems to have screwed up audio sync already, so we'll see what else happens.I'll include more of the zip-tie video. They are just zip-tied through the nearest holes. If there are no suitable holes, I make new ones.

Our solution was to adopt the neighbor's cat. Well, that's not quite right... The cat adopted us first. A year later, it was made official (neighbor gave him up).

My cat is the same way, I had to build her a ramp (see the photo in the last step).

Much appreciated!

I was rather tempted to do the same (and possibly even IoT it or at least wire it up to a server), but there are a few things that I consider mission critical (like not locking my cat out) and I prefer not to debug such projects using a living being. That's just my opinion, obviously, and is not meant to discourage experimentation.

I may also need to solve this problem soon. My current plan is to replace the simple flap with a microchip-controlled flat. This will require a microchipped cat or a collar tag.

Thanks for the suggestion.

I found a fairly comprehensive design write-up here: http://www.postoilsolutions.org/documents/Earthbox.pdfBetween the ideas in this Instructable and the PDF, the options seem pretty clear now.

I'm not sure how much the advertised electric frame goes for or how much you can afford, but I build something similar (see photo) using this $350 frame: https://www.amazon.com/gp/product/B016Z1U34K and a $40 piece of oak plywood that I simply cut down to size and varnished. The shape is up to you, mine is about 6'x2.5' with a center indent for a wraparound experience. The 3/4" oak plywood turned out to be strong enough without any bracing.

Depending on the type of plastic used in the golf cart top, a "Plastic Welder Epoxy" type product may be a safer option. I've had very good success with them, and you can always undo the repair by sanding the stuff off. A backing layer may still be appropriate. Proper preparation and following instructions (especially correct mixing) are important.I would not go with spot welding in such an application due to the aesthetic deficiencies of the end result, even if it works.

I think I figured it out - my pack became unbalanced and the BMS is not able to re-balance it. Possibly the BMS isn't able to discharge fast enough, or it's faulty, or I'm not waiting long enough (weeks?). Maybe the BMS unbalanced it in the first place, stealing power from a few cells, I'm not sure. The only thing I do know is that its safeties work - both upper and lower voltage limits.I plan to add two 6S JST connectors to use a cheap external balancer should this ever happen again. I only use (and thus charge) the e-bike occasionally and that probably gives the pack enough time to go out of balance.My other option was to replace the BMS, but all the options I found were in China and they listed very slow balancing discharge rates in the specs (70 mA at best)