Art School: Semester 2, Week 7

Some things never quite go the way you want, or expect.

This past week was, as previously mentioned, “reading week”, or (effectively) Spring break.

I will detail the things I did for school below, as usual. But what needs discussion before then is heating. As in keeping warm.

You see, we live in Canada. And it’s Winter. Now, admittedly, we live in British Columbia, and it’s nothing like Alberta for weather. But it does get a bit cold. Down towards 0 degrees Celsius (that’s 32 degrees F, for my fellow Americans still living with stone age measurement units).

Now, before we go on, there is another part to this story: the local gas company got in touch a few weeks back and told us they had to swap out our gas meter. Apparently there is a program in which they randomly select homes and replace the older gas meters there with new ones. The old ones get taken off and checked for accuracy. Unstated is what happens if they aren’t accurate, but my guess is if a particular flock of meters — model, production run, or whatever — turns out to be inaccurate over time, they are replaced ASAP with new ones.

Anyway, Wednesday morning was our day to have the meter replaced. The guy arrived on time and was gone in less than an hour. New meter installed, pilot lights re-lit on the water heater and hot water boiler for the heating system. But we had him just turn off the two gas fireplaces, since we never use them. All seemed well.

So, back to the part about this being Canada and a bit nippy. I got up on Thursday morning to find it colder than usual downstairs. I went looking around and discovered that the pilot light was out in the boiler. The water heater was fine — we could shower — but there was no heat in the house. Uh oh.

I spent some time at it. I re-lit the pilot light a few times and it would just go out again. I could get the boiler to run for a while, but after a few cycles it would shut down and the pilot light would shut down with it, so it would not restart. Also, the on/off cycles got shorter and shorter, and there was a swishing noise in the water pipes. (It’s hot water heat: if a thermostat asks for it, a boiler heats water in a closed system and a pump pushes it around to radiators along the baseboards in the area around the thermostat.) I think the swishing noises are new — air in the lines, somehow.

I also discovered a hose bib in there was dripping and there was water on the floor. Not a lot, but some, and that was new.

Debugging this lead me to at least one interesting issue: the guy from the gas company didn’t relight the pilot lights in the fireplaces, which is fine, but he also didn’t bleed the air out of those gas lines. It took some time to get gas all the way down to those fireplaces, get them re-lit, and working again. Which we needed to get the house back up to temperature. I think that was an oversight on the part of the gas guy. The air left in those lines could work its way around and pilot lights could go out. I thought that getting the fireplaces working might solve this issue, but no such luck.

And I called the gas company. They told me they would not send anyone out because their records said everything worked after their visit, and because the hot water heater was still working. In their minds that meant the issue was with the boiler, and not the system, despite the fact that the system had been just fine until they diddled with it.

So I called the landlord. The first pass at getting help came back “How about 10pm three nights from now?” which seemed a bit odd, since there would be no way to get parts for the system at that point if any were needed. I pushed back, and we were supposed to get a call.

We put our lives on semi-hold while we waited for someone to get back to us, but no one did on Thursday. We turned the fireplaces off on Thursday night and the house was OK all night.

Friday was Chinese New Year. Guess what town shuts down in large measure on that day? The guy wasn’t available, but it was still reasonably warm (above 0 C at night) and the house is pretty well insulated, so we were fine.

Saturday afternoon the repair guy arrives. After an hour, he determines that the heat exchanger in the boiler is clogged. It overheats and the entire system shuts down, including the pilot light, as a safety measure. The fact that this came up after the meter exchange is a coincidence. This boiler is ancient and dying.

And the fix is too expensive to contemplate: a minimum of $1000 to tear open the system and clean the heat exchanger. And we know the house is going to be torn down in a few months, so not worth the expense. I am talking with the landlord about the options here.

And of course, next week is supposed to be cold at night. Down to -8 C or so. Entertainingly cold. So the entire situation is amusingly complicated and weird. Just like the rest of our lives.

Also in the “that’s not what I expected” category: Medium tells me that I am now a “top writer in Art.” Really? Somehow I suspect that has less to do with my writing and more to do with how few people tag things as being about Art on this site. Maybe they need a guerrilla event where lots of people create Medium accounts and write posts about — literally — anything and give them the Art tag. Given how few posts and how few reads it took me to get this “honour,” I suspect anyone can have it with a few hours of effort.

Art History



From last week, we had the “before” picture of the sculpture:

The sculpture in its raw state

After two days in the shop and a bit I have the following mess in the locker (sorry the picture is a bit blurry):

Most of the sculpture, now in smaller bits

There are some things in there that need more explanation, but first, here’s the actual out put of the second day in the shop:

Day 2 in the shop… the results

And here’s what I did to that cylinder thing in the back at home as I layed out the spiral on it:

Spiral layout

Now, the details. Day one in the shop consisted of figuring out how to cut 10 equilateral triangles from my maple stock. Each one is 4.5" on a side. (Yes, inches. For various reasons having to do with being too close to the USA, we’re stuck with stone age units when measuring wood, pipe, and a fair number of other things. Go figure.) The challenge with those was the simple act of making them without cutting off my fingers. The shop has a really nice compound mitre saw, and with an extra fence added to support them while cutting, I managed it.

I also created 10 more triangles but these are only 4.5" on one side. The other sides are longer, and the total “height” if you will is 6". The actual lengths of the other sizes is the square root of ((2.25*2.25) + (6*6)), or 6.40", give or take. (Isn’t math fun? Everyone remember the Pythagorean theorem now?)

I also cut the dowels down to about leg segment lengths, so they would fit in my locker. Then I went off to do other (non-sculpture) things.

The second day I turned that rolling pin looking thing show above (which is the central spindle of the virus) from another chunk of maple. It’s 2.25" in diameter and 9" long, plus some bits on the ends to help glue it in place in the final piece. (The wide flange on one end will go away… it’s scrap.)

I also worked on figuring out how to bevel the edges on all of my triangles to make assembling them work. That, it turns out, wasn’t simple. I’d hoped to be able to cut them with a saw, but there is no safe way to do that. So I then hoped to sand them down to the right angles on the disk sander, which has an adjustable table on it, but that can’t be done directly because the adjustment won’t go far enough in the direction I need it to go.

Just now, as I write this, I realize that’s not actually true. Argh.

In the interests of honesty and story telling, I’ll tell you the whole story so you can see how I messed up, and why I did what I did.

First off, the angles on a regular icosahedron are all 138.19 degrees. But, the head of the virus isn’t actually a regular icosahedron. It’s 20 sided, but it’s elongated in a particular way. Thankfully, after careful consideration, I think all the angles continue to be the same.

Also, that 138 degree number means the bevel between pieces is 42 degrees:

180–138 = 42

And thus I need to set the sander’s bench at 42 degrees up or down from the 90 it is at by default. That will let me sand the bevels I need.

But it’s not that simple. I can’t get it to 42 degrees up because that would pinch the wood between the table and the disk. Very dangerous, as well as leaving no room for my hands to hold the wood.

And I can’t set it 42 degrees down because it doesn’t go that far. It stops a few degrees short of the angle I need.

(Those of you who are really alert have already spotted my error, but I missed it, so I’ll keep going and come back to it.)

The fix, I decided, was to make a carrier to hold my wood at 42 degrees to the sanding disk. I went off and did that — it took some time — and I cut five test triangles to sand with it and then assemble, just to test it all out.

And when that was all done I discovered they were all wrong.

I was sanding way too much off, and the resulting assembly was trying to come out nearly (but not quite) square. I stared at that for a few minutes and realized that I’d failed to divide the angle in half. That is, the total bevel angle is 42 degrees, but I only need to sand of 21 degrees from each piece, not 42 degrees from each piece. Doh!

So I went off and built a second carrier at 21 degrees, and cut more test triangles, and assembled them. That’s the Chrysler logo you see on the left of the photo above. It’s right, and it works, though the scrap would I used was all warped, it definitely proved out the angle theory, and the carrier.

In that same photo, the 3 bits in the middle are the carrier. Effectively it is a sled and two different top pieces, each cut to hold the two shapes of triangles I have to work with. (And error two is in that sentence… clearly I was not thinking clearly! More on that in a bit.) The thing on the right in that photo is one of the actual non-equilateral triangles I will need to sand down for this.

OK… so… does everyone see my errors? Yes? No? Well, for those that don’t:

  1. The first error was that I didn’t halve the bevel angle, as discussed above. 42 instead of 21 degrees. Had I gotten that right at the start, things would have gone very differently.
  2. I failed to take into account the fact that when I halved the bevel angle I need to sand off, I can go back to just tilting the sander table. It’s true I cannot get it to 42 degrees, but I can get it to 21 degrees. So I don’t need the carrier at all. I can just set the angle on the sander’s table and sand directly. All the time I spent building the carriers was wasted.
  3. Even worse, if I actually needed the carrier, the one I built for sanding the non-equilateral triangles only lets me sand the short side. It can’t hold the wood so I can sand the two long sides. I’d need a third widget to hold it on the carrier in that orientation, and I didn’t build it. Thankfully, however, I can skip all of this and just use the sander’s table directly.

I am a total idiot. I spent hours making that silly carrier on a blunder, and then failed to rethink fully when I discovered said blunder, so I actually spent more time remaking it. Wow.

On the plus side, sanding will be easier now. Sigh.

I’ll be able to confirm all of this on Monday, but I am pretty sure I wasted at least four hours as a result of those mistakes.



I spent two full days at it, and then some. Premier crashed a lot, and lost things. It really didn’t like what I wanted to do, probably because it was complicated.

The work is two minutes long, and I am making something a bit musical, semi inspired by Bolero and a couple of other works that build to a crescendo. I’ve got a main beat going once a second, and minor beats going 3 additional times per second. In my first pass at this, that was 120 main beats and another 360 minor beats, each created as a separate clip from a source sound. (They are just two different sounds with 120 copies of one and 360 copies of the other, but from Premier’s perspective they were 480 separate clips, if I understand the terminology properly.)

There were a fair number of other things in there as well, probably getting my sound piece to something close to 550 clips total. When I had it all tweaked out and it sounded good, I exported it to a MP3 and AIFF files, and both of those sounded awful. Premier lost a lot of stuff, and mangled what it left.

I struggled along trying to rebuild it and make it work, but nothing would export and sound right. And the net informs me that for Premier to export correctly, the inputs should have been WAV files, not MP3 files. (Of course that advice is not a sure thing; it’s just what I found searching for why this was such an ugly issue.)

After the second day I went to bed trying to figure out what to do, and I made up an alternative. I exported each track from the separately to a WAV file. That converted each track in the original project into a single, two minute track, combining all the clips in it. For example, one track originally contained 120 main beats as separate clips. Another had the 360 sub beats.

Once the exports were done, I created a new project from scratch and imported the 16 WAV files. Now there were only 16 clips — not over 500 — and that meant Premier wasn’t struggling as much.

Once again I edited in the changes I needed — volume adjustments, a couple of effects, etc. — and then exported the file. It took about 15 seconds to export the MP3 file, and (surprise!) another hour to re-index the project. (While the indexing happened, I copied the MP3 file off to another machine and listened to it. It sounded good.) When it eventually finished the indexing, I exported to AIFF, which took about 30 seconds, followed by another hour of re-indexing.

Why it thought it had to re-index, I have no idea. I edited nothing, and the exported files were done quickly. I did figure out that the reindexing takes forever because (a) Premier uses exactly one CPU thread — no multitasking — and (b) the WAV files I created are uncompressed, and so each is about 22 MB in size. 22 MB times 16 is a lot of data to index, apparently. It shouldn’t be, but it is.

As you can tell, I have no love of Premier. I think they need a serious set of testers to find the bugs, and programmers willing to fix them.

And of course the third project in the media class is a video art work done in — you guessed it — Premier. Yay.

Anyway, it’s done. It’s not due for another week, but it’s done. For the curious, here’s a picture of the PC’s screen showing the final layout of the tracks:

Adobe Premier Pro CC, driving me crazy

Here is a link to the MP3 file that I produced and will turn in. It is possible you’re listening to this before I’ve turned it in. It’s a preview!

Jeff’s Construction Sounds Audio Project

I don’t claim it is anything fantastic, but it’s something. For me it plays directly in Chrome when I click on the link. I hope it works for others as well.

That should end everything I need to do in Premier for two weeks or so. What the video project will entail, I have no idea. Yet.


  • I came across a monster of a sculpture project that I thought my instructor and shop manager might find interesting. They did. You might too. Below is an article and a YouTube link to a short video in Japanese about it. The first article links to an shorter, silent video as well.
  • You know those irritating password security questions that we’re all asked to create answers for, and that nearly everyone does wrong? (Hint: if they ask your mother’s maiden name, and you answer with her actual maiden name, you might as well stand on a street corner handing your account password out to random strangers on pink post-it notes. Answer with “RedPonyIsntIt” instead, or whatever works for you. Something that is not easy to search in a generic database. And yes, most of those questions can be answered with a few searches. Mother’s maiden name? What street did you grow up on? Make of first car? And so on. All easy data to get.) Well, I found a set of much more interesting security questions that really ought be used. And so I give you Nihilistic Security Questions, by Soheil Rezayazdi:
  • Here’s a very interesting article about Facebook and how it is killing online comedy. Generalize this, however. FB is killing a lot of things, including — to some degree — democracy. I grow less and less happy with FB every day, and use it for the bare minimum as a result. But it does have a stranglehold on some kinds of communication. I hate that. Warning, some profanity in this piece. The person being interviewed is very upset with the state of the comedy world and says so very clearly.
  • The usual index of art school posts and other things here on Medium, because Medium’s display isn’t chronological, as far as I can tell.

Dog Pictures

Sculptor/Artist. Former programmer. Former volunteer firefighter. Former fencer. Weirdest resume on the planet, I suspect.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store