Productively Destructive

I started work yesterday on getting Fragwell Lab in order. I had originally planned to insulate and drywall the building before putting it to use. That's not a good idea. In my experience, better planning for large projects like that is to be had from a variety of dress rehearsal where you're actually doing what you set out to do, but without the finish work.

Essentially this amounts to having no insulation in the lab while I'm still using it. This is not ideal by any stretch of the imagination, but it's a good way to get a feel for things. Right now, the lab is sitting at 47.6F with myself, the overhead fluorescent lamp, my laptop and the small stereo as the only heat sources. The door is shut and the only major heat escape points are the vents at either end up top. With hot air rising, I'm sure I'm losing a lot of heat.

I've assembled the desk I picked up at the Logan DI store back on October 4 (almost two months ago? Jeez.) and I have to say I'm pleased with the outcome. The desk cost me $30 and it's a 250lb 30x72" Herman Miller (as in Herman Miller Aeron chairs) beast with a 1" top and a base bent and welded from 3mm sheet steel. $30 wouldn't have bought me an X-mart crapdesk new.

I've added Rule #2 to my Rules of Heavy-ass Gear:

1. If the wall-plug doesn't look weird, it's not a real computer.
2. If I can assemble a desk with a screwdriver and Allen wrench, it's crap.

I have to admit I'm glad the spiders I've run into so far in my clean-up project are dead. There have been some real keepers. I'm not scared of spiders or anything, I just hate them with a fury. I've found a few dozen spider skeletons and trashed a few egg sacs. Hopefully the population will be reduced next spring. The last thing I need is actual bugs in my systems.

The workshop had some shelves installed at some point. With the plan of insulation and drywall, they were no longer necessary so I removed them. I gained quite a few screws during the process. I'll be putting those to use, for sure.

One of the first things I added to get the shop put to work was a stereo. I need music while I work. I bought a pair of Mission 734 loudspeakers from the Logan DI about a month ago, paying $16 for the pair. They weren't in good shape, with one woofer cone missing the center cap and another cone cracked all the way out from center to the surround. Having had no luck tracking replacements down via eBay, I decided to check with Mission across the pond to see what they had to say... and their spares email address was dead. Great.

So I was stuck having to find a way to repair polypropylene, which is notoriously difficult. Time was on my side, however. The oils and such that polypropylene products emit over time are pretty much gone after more than a decade so standard super glue, applied in several coats over the course of three days managed to fix the problem. With the woofers being same for both Bass and Mid sections, I put the repaired driver in the Mid position in one tower and it has turned out to be fine.

The missing center cap on the remaining damaged driver will need replacing, but it's not a huge issue since I'm not running the speakers at high wattage and the air is quite cold. Based on what I see from the design, the voice coil is cooled by the moving air. There's also the airtight thing, which isn't as important since the driver in question is mounted in the Bass segment, which is ported.

Upon first listen, I noticed the highs were missing from the same tower the formerly broken driver was in. Pulling the tweeter and checking resistance revealed that the thing was blown. The cool thing was the coil and dome were separate from the rest of it so I could pull it apart and see what was wrong. Forensice analysis showed that the tweeter had been pushed inward at some point and one of the small wires connecting to voice coil to the outside world had suffered a break. The center of the dome was still pushed in. Couldn't see this without disassembly because there's a wire mesh grating covering the dome. Since the wire was coated with the usual enamel and the dome was metal-coated plastic, I just globbed solder and flux over the broken ends and left it to burn away the enamel. It worked and the joint was smooth. I figure at such high frequencies it won't be seeing all that high of stress anyway so it works and that's what matters. So that's a $900 pair of speakers for $16 and a couple hours of labor. The cabinets are nice and they sound fine. I will count that as a win.

I'm sure there are audiosnobs who scoff at my solutions, but they're the same guys who swear on their mother's grave that a CD rewinder improves their listening experience.

I'm driving the Mission 734s with a small Hitachi bookshelf stereo from the mid-80s. I bought it back in like 2002 because it was one hell of a performer for $8 and the small size was crazy. When I opened it one day to give it a cleaning, I saw that the transformer in the amplifier was toroidal. For those who don't know, this basically means donut-shaped. Winding these transformers is difficult and they are really only seen in high-grade industrial gear or professional music equipment. Often, they were the only way to get anything AC-powered into a 1U rackmount case without using a switch-mode power supply, something you just can't use in certain situations--even today.

So I basically ended up with a treasure. It's the only example I have of a small receiver kit that sounds truly great. The AM/FM module uses the older-style tuning capacitor, something I swear by.

But I don't think it will be enough. I'm going for 5.1 channel surround sound so I need to go further. Likely, I'll be using a separate decoder and feeding the outputs to separate amplifiers for that. The Hitachi unit would be running the rear channels while a very nice Philips 7851 from 1977 is perfect for the front channel and has switchable speaker banks to allow the use of smaller speakers when I don't want to run the Mission units.

Can't do it Captain. I don't have the power!

I look at the standard American 115VAC 15A circuit and laugh. Not because I'm always maxing them out, but because when I do exceed their limits, I REALLY exceed those limits.

Fragwell Lab, I just discovered last night, has a dedicated 20A circuit on Fragwell Manor's 100A load center. I had thought I would need to upgrade the load center to be able to handle a stronger circuit for the lab, but after doing the math, no. Even with everything running, including the electric dryer, we're still not touching half of capacity. So the logical thing, then, is to run a bigger line to the lab.

I'd like to have the ability to run an EV charger one day if the opportunity arises, but that would be a pretty big load itself so I need to think big. A 230VAC circuit with a 30A rating it basically the minimum for me. I'm looking more toward 50A. That's like 11KW at full load. I'll certainly be running my own load center in the lab to break that out.

The loads:

Climate control:
15A 1 phase dedicated circuit

This will be for the A/C unit and, in the event the computers aren't providing enough heat during the winter, a ceramic heater. The cooler will be using the whole circuit when in cool mode. I doubt the heater will be needed.

Legacy Supercomputer:
20A 2-phase dedicated circuit
Legacy is my planned eight-node render farm. The systems I have in mind don't come close to maxing even one phase line, but that's only with eight nodes. I could scale up or use systems that use a lot more energy at some point. The average stripped system will draw around 2A at full load. That's including the loss from the power supply, a basic video card, full memory slots, high speed LAN card and a hard drive. I mentioned heat output previously. You can see why I don't think the heater will ever turn on.

Legacy front-end:
15A 1 phase dedicated circuit
This is just where the main computer where I do my work will be set up. It won't come close to maxing the circuit.

Lights:
15A 1phase circuit
All lighting will be connected to this one.

DC rails:
20A 1 phase circuit
The DC center will be connected to this one. The plan is to have a 115VAC to 48VDC converter which will branch out to several sockets for use with DC-DC converters to drop the voltage to lower voltages such as 12VDC and 5VDC.

History Lesson

I ran across a page that was supposed to be about 100 battery types, but had a lot more information that that on it. Over the course of a week at work, I still didn't finish reading it during my downtime. At the moment, I'm only up to World War II. It's kind of Euro-centric, but that's a desired change of pace. There are even humorous side-notes about inventors.

Here's a link: http://www.mpoweruk.com/history.htm

It's a lot of information in a small space. I'll be saving a copy for my daughter to read later on down the line when she's ready. Maybe I'll make storytime out of it in a while.