Tuesday, July 31, 2007

SketchUp + HVAC-Calc = ?

Loosely quoted, "Good companies have many great ideas. Great companies know which great ideas to discard".

Alas, I will not have time to implement this, nor ways to approach the Big Guys to make this happen, but I wonder - am I the only one that sees an obvious symbiosis of SketchUp and HVAC-Calc?

Granted, it's not a five minute project, but all the necessary components are already there - and it won't be that difficult (especially for big guys like Google) to implement a real-time heat loss and gain recalculation as one works on the model of the house...

One of the most important features of SketchUp is that it, being a Google application, evolves fast. Plugins are there, too. It would be really interesting to see where it goes - SketckUp Blog is a nice place to keep an eye on it.

More on SketchUp later - it might have many obvious and not so obvious ramifications on the ways home automation, HVAC and general improvements are done. For example, the most obvious side effect of one creating a house model is that all the hidden spaces you've never thought about become obvious, and the task of routing your cable system becomes significantly less intimidating.

to be continued...

Sunday, July 29, 2007

Short take: two HVAC units vs. Zoning System

Talk about outsmarting oneself.

The whole point of The Dual Nightmare article was to convince the reader that there's no way two HVAC units can serve the house as good as one zoning system. Alas, the subject didn't click together with what people were looking for. Let's see if this one is better.

You will never buy a good zoning system, either

About a month ago I was trying to convince you that you will never buy a good HVAC unit. Today I'll try to hammer the last nail into the coffin.

Assumptions are a bit different in this case - if you are even investigating zoning systems, it means that the situation is pretty bad. Buckle up, it's about to get worse.

A slightly elaborated version of how DZ got actually started looks like this:

  • Existing systems were extremely dumb;
  • They were priced astronomically;
  • People who were selling them didn't have a clue;
  • People who were installing them knew just about as much, and, to add insult to injury, were actually afraid of zoning systems.
Situation hasn't changed much since -

Zoning systems are still dumb. Even the smartest ones are - the manufacturers haven't realized yet that they absolutely must play nice with other kids in the sandbox, and they keep producing proprietary components - yeah, I'd like to see the Carrier Infinity (a.k.a. Bryant Evolution) zoning system work with a pretty damn fine Lennox split unit that I have absolutely no motivation to replace or upgrade, it works just fine. My favorite pet peeve - none of them allow any real time diagnostics without special tools - what's the big deal, Ethernet and SNMP's been there forever. You'd think that with the price of chips going down so much, they'd have it in each and every unit, but no - there is just one, Carrier Infinity System Access Module - and even with that, the details on the version that allegedly supports Ethernet and wireless are very, very sketchy. Can't wait for it to come out.

The prices are still where they were.

Salespeople are still the same. Sometimes it is simply hilarious - see HVAC-Talk for the story of a Carrier dealer, out of all people, trying to talk the customer out of buying Infinity Zoning.

HVAC contractors have changed a bit - some of them did. Smart people hang out at HVAC-Talk, others are proudly proclaiming that they haven't heard of Internet and for sure don't need it to teach them how to do their job.

So, basically, it all boils down to the following:

  • If you don't do your homework, you're screwed - you will pay too much for something that is not going to ease your pain;
  • If you do your homework, you're screwed - what is out there is too dumb to ease your pain (up to 8 zones? forget it, I need twice as much, and why is it such a huge problem to begin with?);
  • If you compromise and buy what they sell you, you're screwed (read: Vendor lock-in).
So, what's the best way out of this? Don't buy a zoning system...

Or, you know, you could just make one...

Sunday, July 22, 2007

A/C Can't Keep Up?

Been there...

First of all, if that's the case (and not the phenomenons described in Thermostat Wars), it means that something is pretty much busted in your HVAC system.

Not necessarily the unit itself - ducts may leak, filter may be dirty (how many years ago did you change it last time?), you may have cut the trees on the southern side that provided you with shade and so on and so forth. Best bet would be to assess the overall system status - but that's going to cost you.

Next best assumption is - you don't have money not only to fix something if it is broken (1), but even to assess the status. Or, for some reason, this is not practical (say, the assessment, fix and/or upgrade is planned and budgeted, but the right time hasn't come yet). Well, this is not a nice situation to find yourself in, but not all hope is lost.

All the things below are pretty trivial in retrospect, but sometimes not obvious upfront, so I'm going to lay them out anyway.

First of all, set the thermostat to a lower temperature early in the morning and leave it there. Your house will store some energy and will help your A/C fight with the sun. You must realize this will somewhat increase your electric bill - but not much.

You probably heard that you can't run evaporative (a.k.a. swamp) cooler together with your other (known as refrigerating) A/C - the swamp cooler will be busy injecting humidity into the air, whereas refrigerating A/C will be busy removing it from there - thus adding $$$ to your electric bills and doing little else.

Also, be ready that the house will be significantly more difficult to cool some time after you stop using evaporative cooler - all that humidity it pumped in is still there, contained in the air, clothes, carpets, stuffed furniture and building frame.

Close all the blinds and shutters you have. If some of your windows don't have them, go buy some - they're relatively cheap and will significantly improve the insulation - not only they block the sun radiation, but also introduce insulating air pocket when closed.

Go get some insulation sheets from your favorite home supply store and stuff them between the shutters and windows - bad windows may contribute as high as a quarter of total heat load to the house.

Along the same lines, if you're planning to improve the house, don't go cheap on windows - get dual or triple pane tinted windows - and do your homework on selection. Makes the world of a difference - our upstairs bedroom on the south side of the house with good windows installed was significantly cooler than north facing rooms with old, single pane aluminum frame windows.

Make sure your ceiling fans work for you, not against you.

Well, that's just about it for starters...

In addition, there's a nice article about Tricks to Keep Your House Cool This Summer (except for the section about fans, which I disagree with) - and make sure you read the comments. Have to apply a good sanity filter to those, though...


(1) Having home warranty is relatively painless way of making sure this never happens; will cost you anywhere up from $250 a year, depending on the features - whether you have a pool, etc. - but it opens another can of worms altogether.

Saturday, July 21, 2007

Trivia: Not Web Surfing, Blog Hopping

Not that I just realized it, but it's just I have to conduct a lot of research recently, and it is now painfully obvious - the Web as it was before is dead. Static web sites are mostly either brochureware and therefore useless, or product, service or design specifications, or simply abandoned. All current information has shifted to blogs.

Search engines seem to have heed the trend - information contained in blogs is indexed significantly faster than static pages - I won't be surprised to discover that blog hosting sites have feeds to search engines that allow indexing in almost real time.

A corollary: good blog browsing tool is of paramount importance. Web browser is simply inadequate. RSS reader is better, but not much.

Hmm...

Friday, July 20, 2007

Short Take: Fractal Design

For years (I think, starting somewhere in 2000), I've been fighting my delusions of grandeur when explaining the concept of fractal design to people - I thought that obviously somebody must've thought of something this simple way before I did, and, hence, it is a common concept that I heard of somewhere and subliminally decided that I coined the term.

Apparently, not quite. Despite the fact that Google search for fractal design and fractal architecture returns more than enough matches, my patience was exhausted well before I could find anything similar to what I had in mind.

So let me put the stake in the ground, then, and introduce a definition:

Fractal Design: design pattern characterized by approximately same level of complexity at any level of abstraction.

It is a counterpoint to God Object (a.k.a. Blob), Interface Bloat, Spaghetti Code, abused Ravioli Code , Big Ball Of Mud and possibly other Antipatterns.

Starts making sense in context of Complexity Management, which, in turn, is not very well defined in the software engineering domain (this nice essay is just about the only article connecting the two), though, apparently, is widely accepted and practiced in business.

Short Take: Hysteresis in RTH7500

Frequently asked question with a very vague answer - what is the hysteresis (a.k.a. dead band) of my thermostat?

For Honeywell RTH7500, the measured (therefore not exact) typical value is about 1°C (about 1.8°F) - for the sensor located in the room behind the wall where the thermostat is.

Worst observed value is 2.5°C (about 6.3°F) - for the sensor located on the first floor (the thermostat is at the second) - this is not actually a property of the thermostat, but rather an indication of how badly the airflow was designed.

Another question - can hysteresis be controlled in RTH7500? The answer is: probably not.

If it can, there's no way to do it that is described in the documentation that is sold with the thermostat (yes, I found the manual, feel free to ask me what those numbers mean).

Wednesday, July 18, 2007

Honey, there are gremlins in the house!

Like the saying goes, "Flying is the Second Greatest Thrill Known To Man, the First is Landing".

The most thrilling part of getting into a new house is the first night. Or any night, for that matter, until the dust settles. It's the time when you find out with excitement that the people that sold you the house forgot to mention that the neighbor behind the southern fence likes to play his electric guitar off key at 5AM, or that the guys across the street get their truck that is older than me out of the garage and start fiddling with the engine and revving it at about 11PM... You get the picture.

Then, deeper into the night, you start hearing noises. The one I'm talking about drove me nuts for several months until I figured out what it was. It's quite rare (so far), because the only HVAC units affected by it are the ones that are slowly ramping up and down.

I'm taking about loud bangs in more or less rapid succession, seemingly all across the house. Gremlins.

Hindsight is always 20/20, and now that all the facts are laid out before you it is quite simple to see what it is: it's the flat tin panels that comprise the ducts flipping back and forth when the pressure changes, all of them at different times. What makes it fun is that the sound pattern changes as the building ages - panels that used to be well insulated and thus somehow dampened are not anymore because of the insulation breakdown, and the more time passes, the more pronounced the phenomenon becomes.

It is also an indication of the fact that the pressure in the ducts is quite high, probably way above what it should be.

The best part? There's probably nothing you can do without significant amount of money or elbow grease injections, so might as well get used to it.

Proper fix, though, assuming the pressure is within reason, is to apply sound dampening materials to the outside surface of the ducts. I guess the trick would be to find a material that sticks to the surface and doesn't let it bang, and at the same time won't break down for a long time.

There are also products called "acoustic duct liners", but knowing how widespread the problem of undersized ducts is, I would be extremely cautious stuffing anything alien into the ducts, unless you're absolutely sure the duct itself is grossly oversized.

Which returns us all the way back to the heat loss and gain (a.k.a. heat load) calculation, but that's a different story for a different time...

Short Take: Where's My RTH7500 Manual?

Exactly where you put it, dummy.

Correction: I am the dummy.

So here I am, trying to figure out the way to configure the hysteresis and, at the same time, save myself a long walk to the shelf at the second floor where I *think* I put the plastic bag with the manual and spare parts. Also, it's middle of the night, it's dark, the light switch is out of the way, everyone in the house is asleep and I don't want to step on the cat strategically placed somewhere down my path.

So I'm thinking: it's 2007, let's google it up.

Yeah, right.

I guess my patience is thinned by previous encounters with HVAC machinery and service, unlike one of this gentleman.

Mind you, all configuration settings in this thermostat are not described by words. They are described by numbers. There's no way in hell I can remember numbers I have to use once in a few months, if not years. And the paper manual is nowhere to be found. And what other companies usually put up (a printable PDF document) is nowhere to be found, either.

Nice.

Another score for Rite Temp.

And don't lose your manual, otherwise you end up with a quite expensive brick on your wall.

Update: I've found the manual...

Tuesday, July 17, 2007

Rite Temp GPMG8085C: Informal Review

Rite Temp GPMG8085C

Who would've thought that this blog would be one of the most often referred to sources of information about GPMG8085C... All right, if that's what y'all want...

Ever heard of "The Golden Triangle"? "Good, Cheap, Fast - Pick Two". So, this thermostat may just as well be the closest approximation. $20 cheaper than the next guy (Honeywell RTH7500, I'll come back to it another time), does the same job, if not better.

Comparable quality and length of DIY installation.

Installation complexity: no-brainer (YMMV). A few advices, though:

  • Make sure you RTFM;
  • Don't drop the wires into the wall;
  • Programming is best done when the thermostat is laying on the desk before you, not when you're standing next to the wall with the thermostat on it;
  • At least for me, the preinstalled battery still works after a few months of operation, may want to ignore that paranoid advice to remove it.
Significantly bigger and nicer backlit touchscreen than the competition has (to be completely fair, they do have a big touchscreen on VisionPro, but that one is almost twice as expensive, so no cigar), and it has a nice textured button that you can grope (English experts, give me a non-ambiguous word and have my thanks in return) in the dark without being afraid to push a wrong button and reset something.

You have to really hate your customers not to provide an option like that (example: RTH7500 needs you to press any button, and, invokes the action corresponding to that button, even in the dark. So you have to be really careful where you tread, and the fact that the screen is backlit is useless - you can't comfortably use it anyway).

There's a stylus and a holder, but feel free to ignore it - fingers and/or fingernails do the job just fine. No visible scratch marks after few months of use.

7-day programmability is certainly nice, but the schedule selection is really weird - you can't set the schedule boundaries at arbitrary time, but only at one of two settings (forgot what times exactly, will update later). Kind of inconvenient, and this is probably the most annoying feature, or, rather, lack thereof. Competing device shines here - you can have your heat and cold any time.

Side note: I'm wondering why nobody figured yet that anticipation is something you're quite willing to have for a period starting, say, at 6PM, when you return from work, but not really want for the period starting, say, at 9PM - when you are on a time of day electric plan, which makes energy after 9PM three times cheaper than before. Remember, you read it here first.

If you don't need 7-day programming, there are other, cheaper Rite Temp models that offer compatible feature set, but 5+1+1 or 5+2 day programming.

I'm quite willing to forgive lack of schedule configurability for including an extremely important feature - you can manually change the hysteresis. Setting it to the maximum value makes the temperature spread quite noticeable (up to 5°F), but makes the unit cycle about three times less often than it would otherwise do. Don't have equipment to measure the actual gain, but would certainly recommend this thermostat to people who are either energy conscious and don't mind little discomfort, and to people who have grossly oversized units - you guys may suffer from temperature swings even worse, but at least your electric bill will drop like a rock because the unit won't breathe like a dog outdoors in Arizona summer anymore.

Can't say anything about humidistat - not much use for it in the middle of Arizona summer.

Another useful feature - filter usage counter. Depending on the climate, your unit may cease operating for a couple of months in mid-season, so you can afford to buy a higher quality, more expensive filter without a fear that you're wasting money (just think of how much it'll cost you to clean the inner coil when the time comes, and reconsider buying that inexpensive $3 filter).

You definitely need to read the manual provided with it in order to learn to operate - but then again, this is true about any more or less complicated home appliance - it takes significantly higher IQ to figure out how to connect the 5.1 home theater and the rest of components together without electrocuting oneself, so no big deal here.

At least all the settings have icons assigned to them, and you can figure out what is what, unlike some other designs, where features are indicated with numbers (come on, it's not 1960 anymore...)

Bottomline

Well worth the money I paid for it.

Disclaimer

God, what have we come to... Anyway - use your common sense, don't do stupid things, and read this first.

Friday, July 13, 2007

Friday, the 13th: Good time to start looking for a new job

Effective immediately, I'm fair game. Job offers are accepted. Resume: plaintext, brief, detailed.

Wednesday, July 11, 2007

DZ Live: July 11 2007 Presentation Draft

Below are links for the presentation happening later today:

Questions from the audience and answers to those questions may or may not be available later, depending on whether anybody is going to be transcribing the session (hint: if you want your question answered, this is the right place to post it).

Tuesday, July 10, 2007

Hardware Architecture: Centralized vs. Distributed

Hardware and software are the ying and yang of what modern appliances strive to be. Evolution and oscillations of the software part of DZ is tightly coupled to choices hardware offers.

At all times, cost was the predominant factor. $99 temperature sensor is not a valid choice for someone who is most likely using a mechanical thermostat, but $3 chip probably is.

So what that it doesn't have a nice box - rich and poor are affected by extreme temperatures in strikingly similar ways, and if you can't afford to cover the chip, and you had to staple the cables to the walls instead of paying couple of thousand dollars to the crew that hides them into the walls, so be it. At least, you're comfortable now.

Same goes for professionally installed dampers vs. DIY registers.

So all right, it is clear enough that $3 chip doesn't have a human interface. All the control was initially being provided through the debug console - sure, it was limited to whoever had the password and the inclination to fiddle with it, the rest of the family had it by the way of voicing opinions and complaining (which, surprisingly, ceased pretty soon after the system was activated).

But it is crystal clear that this solution doesn't stand a chance in the long run - in order to pass the SWMBO Compliance Certification, the system must have at least rudimentary controls in each zone (provided the zone is used often enough; it is unlikely that a server room is going to require much intervention often).

Which presents quite a few interesting options.

Rudimentary controls may or may not be terminally dumb.

The dumbest possible control will include a display (two 7-segment LED is just fine) and two buttons (which actually give you more than two possible inputs: you can use "both buttons depressed" as a signal to change the control context). Plain 7-segment LEDs, though, cross the border between "cheap as in inexpensive" and "cheap as in obnoxious" - a graphical LCD won't be that expensive today, especially given the fact that it will constitute just a fraction of the total cost.

Now, what difference does it make if you have two buttons or eight? Granted, two buttons is 4 times cheaper (by themselves), but again, much less so, given the infrastructure considerations: first of all, having 8 soft keys gives you unlimited flexibility for all practical considerations, second, they won't be that much more expensive compared to the total cost.

Does the sensor have to be in the same box as the LCD and the buttons? It depends. First of all, the box has to be located where it is convenient to access it - most probably, next to the light switch. But the sensor has to be located where the actual temperature has to be measured, and that is where the complaining person is - in the bed, at the coach, next to the oven, you name it. On the other hand, the complaining persons lived with a single thermostat for all their life, so even the sensor that is in the box is already going to be an improvement. Another consideration is that this case requires less wiring (assuming it's wired; wireless solutions are a whole different can of worms).

Yet another component is the servo controller. Historically, servo controllers were always centralized - one controller, bunch of wires, servo boosters and capacitors at each servo, good enough.

Later, Nic van der Walt provided schematics for distributed 1-Wire servo controller, which would support bang/bang dampers with no problems at all, and modulating dampers that will be quite noisy - latency of 1-Wire protocol with the bunch of sensors on it is insufficient to provide software controlled quiet transition.

Today, there are chips with PWM supported in hardware, and if they have enough brains to support even the simplest quiet transition possible, then you have the distributed servo controller right there and then.

Considerations above assumed we're ignoring existing 24V dampers (which are mostly bang-bang, and regardless of whether they are or not, they're well beyond the approved spending limit for the audience) and using R/C servos to control the dampers, as shown here.

This just about concludes the story. There are three major topics left uncovered - the HVAC actuators, hardware interface selection, and making all the gadgets wireless - but the article is getting too long, so stay tuned.

Monday, July 9, 2007

Thank you for using @#$%S...

Someone who's been through one major and one minor flood, heat pump compressor, four or five pool pump motors, couple of pool filter tanks, few dead refrigerators is not likely to treat warranty issues lightly. Thus, after being somewhat dissatisfied with service contractors taking a bit too long (up to three-four days) to show up, I decided to abandon my warranty company in favor of @#$%S, a big, domestically owned, reputable company. There is safety in numbers, I thought.

How naive of me.

Jumping ahead, I did go to extreme depths to explain to them that I have not one, not even two, but three units and did have to pay extra premium for the service.

Being diligent, I decided to use the service that's been advertised among other warranty clauses - preventive maintenance. So I call them up, they give me a window a couple of weeks later, it's still February, I'm thinking, what could go wrong...

Right.

Guys show up, take a look at the unit, say it's not on their record, spend half an hour on the phone with headquarters, say this unit is not on their records. I say fine, there's another one, which is on your records, go take care of that. They do, profusely apologize, and leave. I call up again and explain that the unit in question is not quite up to snuff and has to be looked up, they say sure, and schedule another appointment two weeks later.

No call, no show. With a window like 9 to 12 or 1 to 5, it's quite aggravating - half a day is wasted.

Another call, another two weeks. A crew arrives with gusto, steps up to the unit and says "Dude! This unit is not on our records!" as I stare at them in disbelief. So I make them write down the model and serial numbers of all HVAC equipment, upon completion of which task they notify me that I have not three, but five units, namely, one rooftop, two condensers and two air handlers. Fine, as long as the unit gets taken care of. They say they can't do anything now without headquarters approving, and leave.

Another hour on hold, another conversation, another two weeks. Another crew arrives. Guess what? The unit is still not on records.

Another hour on hold, another conversation, another two weeks. Another no call, no show.

I'm thinking, it's costing me too much aggravation already, the hell with it - if it breaks, they'll fix it. So I leave it alone.

So the summer comes. In June, one of my HVAC units goes dead (the one @#$%S "maintained"). I call @#$%S, they say it'll take them 9 (nine) days for them to dispatch the serviceman. I tell them this is unacceptable, keep biting their ankles every day - and 8 days later, the guy shows up. Replaces the fuse and is on his way. Not that I would believe that the fuse is the root cause, but there's nothing I can do with it - all in all, he's a professional...

Within two weeks, the unit is dead again, with the same symptoms. To make the long story short, it took them over a week to send someone over (mind you, we're talking about August in Phoenix), and the unit did get fixed. Ground short.

This is not a kind of service one pays premium for, so BBB complaint gets filed, processed, and somewhere in the middle of the winter a symbolic compensation comes. End of story.

Or so I thought.

What made me dig up this stinking corpse is that an appliance I bought a while ago, it turns out, has its warranty served by @#$%S. Damn, I say, let's hope it works this time.

Yeah, right.

Scheduled visit number one, 1 to 5 window. No call, no show. At 4:30 service center is called, and they are telling me that the contractor is actually parked in front of my house. Right. At 5:30 another call to service center, and they're told about the fact that contractor never made it in from the driveway, about pretty contemptible track record, of preexisting BBB complaint. They politely apologize and say they will have to call back reschedule.

A day passes. No call. Calling them. Another hold, another conversation, another 1 to 5 service window. Another no call, no show.

I wonder, are they brain dead? Okay, it may be fine to screw things up once. Not so okay, but happens twice in a row. But six??? And a complaint??? And, after that, two times in a row again?

Now, for those curious about the actual identity of the company... Pretty easy to figure out - I just want to leave myself some maneuvering space in case they ever get to read this and decide to send me a cease and desist. That would be fun to watch, though, and something tells me that I'm not the only one having this kind of experience. And have no doubts, another BBB complaint is on the way.

The most saddening thought? They're not alone. This winter, the unit I was trying to take care of in this story has finally died (pretty non-trivial failure, as it eventually turned out), and the contractor that was trying to fix it tried everything he could think of (took couple of weeks to order the parts, have them delivered, and another appointment scheduled), then just "recharged" the unit and said "Let's pray that it keeps working".

I guess he wasn't praying hard enough. But that's a different story for a different time.

DZ Live: Confirmed

As previously announced, a presentation tentatively named "DIY Zoning: 7 years down the road" will take place at Phoenix Java User's Group meeting on July 11 2007.

Look for schedule and site directions on PHXJUG site.

Does Open Source Pay Off?

Depends on who you're asking.

As you can see, big corporations are simply ecstatic with it. Why, you don't have to go through expensive interview and trial and error interview process, and all you have to do is to ask a candidate for the list of Open Source projects they are or have been involved in and then spend fifteen minutes googling the projects to figure out if the candidate is any good. Or, even simpler, just steal the idea and/or the implementation, if the project is obscure enough and your subject matter area is guaranteed to never see the public scrutiny. Within reasonable timeframe, that is - all you've gotta do is to generate profit that will allow to hire the best team of lawyers before the poor sod even realizes he's been ripped off. If ever.

As for the other side... It depends, again.

In almost 7 years since its inception, DZ Donations page at SourceForge generated sixteen dollars and spare change. Yes, you've read that right, sixteen dollars in seven years. Not to mean any disregard to people who made the donations (I will always remember each three of you by name), this is just about how much someone in my profession makes in time that it takes to smoke a cigarette without haste.

So, again, does it pay off? As you can see, not really. At least when you're observing the direct benefits.

As for indirect...

Companies hate to hire. They only hire if there's no other way to ease their pain. Every time they have to hire, they have to spend significant resources on recruiters in- or out-of-house, tear people away from work to conduct interviews, face uncertainty since there are so many clueless people familiar with software engineering shop talk that they penetrate the entry barriers fortified by recruiters with such ease, and given the fact the software engineering itself is not an exact science, but rather art - it may not even become obvious that someone flashing buzz words and certifications is a complete fool until much later...

But situation is becoming very different when there are cold, hard facts that you can present to your prospective employer. Because by giving them facts you are removing doubts and uncertainty and becoming a credible, verifiable and predictable asset (just think of the fact that more and more employers are starting to pay attention to your credit score).

Achievements in their immediate subject matter expertise area are the best for them (but not the best for you: you will be forced to sign the waiver basically making them own everything you and your parents have ever created, and your firstborn too - so you may want to talk to a lawyer and craft the "previous inventions clause" before you even start looking for a job).

Achievements in other subject areas, no matter how bizarre, do count, though you'd never know whether for or against :) Which brings another note to mind - you'll have to insist that recruiters include your extracurricular activities in their version of your resume that they send out to their contacts - because they may not possess enough expertise to see what is relevant and what is not.

Coming back to the original question, does Open Source pay off? Let me just ask you to visualize the idea of Linus Torvalds going unemployed and watch you laugh.

As for donations... Well, will someone tag a black Squeezebox or, even better, Transporter for me, please? Don't expect me to hold my breath, though...

Friday, July 6, 2007

Short Take: Modulating Dampers

A damper, in HVAC terminology, is an air valve used to regulate the flow of air through the duct.

There are two general kinds of dampers (definitions are probably not in line with HVAC terminology, but hopefully understandable enough):

  • Simple, also known as bang-bang, dampers - these can only be in fully open or fully closed position;
  • Modulating dampers are able to stay in any position between fully open and fully closed.

First kind is, of course, cheaper, but has serious drawbacks - first, the noise they make is quite noticeable (one can get used to it, though), second, they provide only limited control and are not suitable to advanced applications.

Modulating dampers, on the other hand, are silent by design and provide adequate control, but require more sophisticated control systems in order to utilize the capabilities.

Zoning For The Poor

Poor people pay more, it is often said. Let's see what would be an average cost of a DIY zoning system assembled out of components specifically intended for DIY applications. Let's make following assumptions:

  • We're dealing with an average family with two kids, geek parents who spend some time working home therefore four bedroom house (fourth bedroom being the office), two bathroom, dining room, family room, laundry room - total of 9 zones (that'll be a pessimistic estimate);
  • All the components will be bought through Smarthome;
  • Functionality will try to mimic what DZ offers today.
Let's see how much is this going to cost us at today's prices....

HVAC 4 Zone Controller - Rs485 - $349.99, power supply sold separately - that's just four zones, out of nine required;

Alternatively, HVAC 4-Zone Controller - $229.99, plus power supply - $11.99;

Thermostat Display Unit - $199.99, one per zone;

Or, as an alternative, wall mounted temperature sensor - $99.99 per zone, plus LCD Wall Display Unit - on sale for $107.99, for all four zones;

6-In Normally Open 24V AC Duct Damper or 6-In Normally Closed 24-VAC Damper- $59.99 (or, if you're feeling generous, 6 Inch Diameter Damper Normally Closed for low, low cost of $98.99 - I don't see much difference);

X10 Universal Module for $22.99 or Appliance Module for $34.99, depending on your configuration, one per damper;

Now, all the stuff above is just the beginning, but it'll be enough for a ballpark: the cheap version of the above for 4 zones will cost you, depending on components, anywhere between 973.86 and 1685.87 plus tax, shipping and handling.

For 9 zones (ignoring the fact that a 9 zone controller is not being sold, and counting only extra cost for dampers and sensors), the prices will be 1888.71 and 3355.72, respectively.

That is, not including cabling and insallation.

Oh, and that gives you only a bang-bang version using unreliable by design X10 protocol. Forget about remote control and monitoring.

For comparizon, DZ installation for the same physical configuration will cost you:

  • Anywhere from $50 for a beaten up 10 year old computer to $400 for a shiny new one to run the control system on (with web access, remote control and monitoring and unlimited number of zones);
  • Roughly $50 to $100 for 1-Wire adapter and servo controller, depending on source and capabilities;
  • $10 to $15 per zone plus some manual labor for modulating dampers;
  • About $3 a zone for a 1-Wire temperature sensor.
All of the above amounts to $152 to $552 for 4 zone system, and $217 to $617 for a 9 zone system (difference being primarily the price of the computer and servo controller).

Take your pick.

Update: Forgot to include cost of 1-Wire adapter and servo controller into DZ installation cost. Fixed.

Software Architecture: Centralized vs. Distributed

Evolution of software moves in mysterious ways. Especially if you're not the one contemplating it, but the one making it happen. Hindsight is always 20/20, and now that I'm looking back at the way DZ evolved, I'm seeing that I was merely trodding the same path as countless others - namely, coming all the way from monolithic single host software design to modular distributed.

Just as dinosaurs became extinct, DZ ceased to be a monolith even before it become public - the first GPL'd version was already split in several parts able to talk to each other via TCP/IP, and the parts were linked together in a modular container driven architecture.

Further down the road, the project started shedding more pieces as better versions of them arrived elsewhere (scheduler would be one example). It also became clear that as computers become smaller and cheaper, the day will come when it would be not only possible, but beneficial to utilize multiple computers for a task that could be perfectly done by just one - just for the very reason of having a particular computer in a convenient place for connecting hardware.

Most prominent examples would be -

  • An HTPC which has to reside in a rigidly defined place, next to the rest of AV equipment;
  • A wireless router that is a focal point of wireless and Ethernet connectivity;
  • A workstation, which by itself provides several relevant data streams (CPU temperature, motherboard temperature, hard drive temperature).
Which makes distributed communications difficult to underestimate. Initially, it was just about protocols, now, it's getting all about application interoperability and plugins (Cacti, Munin, Nagios, Mister House to name a few).

Stay tuned.

Wednesday, July 4, 2007

The Dual Nightmare

There is an ongoing heated discussion among homeowners and HVAC professionals about what solution to the problem of properly balancing the house is better - zoning the house, or having multiple units installed. There are all kinds of arguments on both sides.

The proponents of zoning are usually saying that it would be silly to use one switch to turn on or off all the lights in the whole house, whereas the opponents are pointing out that it is silly to pump the amount of air intended for all ducts together through one.

The opponents of zoning are usually saying that it is more reliable to have two simple HVAC units than it is to have one complicated HVAC unit and one complicated zoning system, whereas the proponents are pointing out that maybe HVAC professionals may be simply out of their depth and are trying to cover that up by silly arguments.

The opponents of zoning are saying that having two units is an adequate solution to the balancing problem and point out to the fact that existing commercial zoning systems are rarely extending "affordable" options beyond four zones, to which I say they are completely nuts and give the room (read: zone) count for three quite average houses that I've lived in: roughly 11 zones in Iowa farmhouse (2 levels, 4 bedrooms, full basement), 14 zones in one house in Arizona (3 levels, 4 bedrooms, game room, library) and 13 zones in the house I live in now (2 levels, 3 bedrooms). Oh, I do include bathrooms, too.

But I digress.

Sometimes you just have two units - for example, when you simply buy the house with them installed - it would be silly to rip them out, wouldn't it? So one just has to deal with it...

Whereas having two units is certainly nice, it is by no means a guarantee of any improvement whatsoever. Like it was pointed out before, all your luck depends on having the job done right.

Sometimes your luck is improved by the fact that the house was designed with heat exchange physics in mind (little heat transfer between floors, narrow staircases).

Sometimes it is degraded by the fact that aesthetic considerations took priority over common sense - wide open floor plans with walls being mostly glass, rooms spanning several floors vertically, and different floor count in different parts of the house.

Which is exactly the situation I have to deal with right now. The north side of the house (1 floor, 2 rooms + 2 bathrooms) is served by a nice powerful variable speed high SEER Lennox split unit. The south side (2 floors, 4 rooms + bathroom + kitchen + laundry room) is served by a puny rooftop Trane and a set of drinking straws they called ducts.

So let me skip the rant about how unthoughtful that was, and get directly to describing various interactions between two units that may happen. Let's assume for simplicity that we're talking about cooling season behavior.

Unhappy Load Meter

If both units kick in at the same time, and you are on a load controlled electrical billing plan, you're not gonna be happy. Neither is your electrical company, but that's a completely different story for a different time.

Cold Spots

It may happen that both units serve the same room[s], directly or indirectly, and the room has enough or more than enough supply and return capacity, and neither of the thermostats is in it. If both units kick in at the same time, the room will get uncomfortably cold.

Hot Spots

It may happen that the cold air supplied by unit A drags down the temperature at the thermostat controlling unit B. In this case, it will not get unhappy and cause unit B to switch on until unit A is off. Which will cause the rooms normally served by unit B to become uncomfortably hot.

The Skew

To add insult to injury, it is quite possible that the layout described above also makes unit A serve, at least partially, the zone that is supposed to be served by unit B - and in this case, unit A will keep running and running, possibly not even catching up with demand - and it will not get any help from unit B because the effect described above robbed unit B of necessary control input and is preventing it from switching on. You end up with one unit trying to do the work of two, for which it is clearly undersized. To further aggravate the situation, it is possible that the overworked unit will go beyond its normal design range, and then you have a dead unit (ask me how I know).

For more than one floor, The Skew is rather a rule than an exception.

Ask yourself a question, don't you always have to set the thermostat downstairs a couple of degrees lower than upstairs, otherwise it feels hot and stuffy? The scenario above explains it vividly. All there is is a leak of cold air from upstairs through the staircase (or a tall room) that causes the thermostat downstairs to make the unit stay off.

Catch-22

To prevent The Skew (further defined as one of the units working significantly longer than the other), you need to carefully balance the thermostats in such a way that the race never starts. This will usually mean that you will have one of the thermostat set below the comfort level, and the other above. Also, this is going to cost you, because the difference between "comfortable" and "balanced" positions of thermostats is going to make a significant contribution to your electric bill. Also, it robs you of comfort. Also, it will make you fiddle with controls all the time, unless you have monitoring set up, in which case you'll just spend half of the time fiddling.

It would be interesting to know what side of argument described at the beginning of the article you are now...

Tuesday, July 3, 2007

SWMBO Compliance Certification

Weighting the priorities is the most delicate task there is.

On one side, you have a bunch of unhappy people, some of them complaining of being hot, some of being cold.

On the other hand, you have The Other Decision Maker which notifies you in no ambiguous terms that those blue wires on that ivory wall (beautiful combination, if you ask me) won't be tolerated. At least no longer than a week. Or a month. Or a year. Or two.

Depending on how well and how fast you can silence the unhappy gang. You do it fast, you're forgiven.

Until the time when you're not home, but Someone Very Important feels a need to change the setpoint. Right there and then.

Which brings you to the point of explaining that you have to take that remote, press this button five times, that button three times, then... wait, you've just run out of their attention span - one of the kids fell into the pool.

So much for advanced user interfaces.

Squeezebox Black

I'm still toying with the idea to use my Squeezebox, along with a SlimServer plugin, say, xAP for SlimDevices, but by now it is painfully clear that the only person who would be able to ever use that interface, and/or willing to use it, would be me. Though scratch that, I'd probably go directly to web interface. More about that later.

Honeywell T87 Round®

So apparently the Honeywell Round® is still the best thing since sliced bread. Granted, the mechanical one had a drawback that caused the worst kind of thermostat wars: one could swing it all the way up or down in one swift motion, and people on both extremities of the age span could have trouble applying controls in moderation - but it was cheap and simple.

Honeywell CT8775 Round®

And, no matter how badly does a digital thermostat in a round body freaks you out, there are decades of usability research behind it. It solves the task at hand perfectly.

Until you run into something as simple as a heating/cooling schedule. Which shatters your interaction model completely. Not even talking about zoning.

Carrier Infinity control panelHoneywell TH8320

Now, you either have to add more buttons (left), or use soft keys (right). The former solution makes the design more expensive and less durable, the latter makes ergonomics more complicated (but at the same time, more flexible).

In other words, you probably can't get the best of both worlds (simplicity and flexibility) in one package, a compromise is required. Most sensible compromise seems to be in a form of arbitrarily complicated master panel that one has to spend a week programming, and a very few critical functions and overrides exposed to the end user.

To be continued...

Monday, July 2, 2007

Thermostat Wars

The predominant kind of thermostat war is the one between those who understand the concept of the thermostat and those who don't. It is difficult to believe, but there are still people who don't get it. They would crank the thermostat all the way down when they feel hot and all the way up when they feel cold. I kid you not, several years ago there was a guy who came to the Residential HVAC forum with a question, "how to educate the significant other not to do that". After few days and several dozen messages, the community agreed that the only way to deal with his particular problem was to install a fake thermostat. This case is beyond repair, let's concentrate on the other.

Out of remaining, two kinds can be distilled: those that require marriage counseling and cause HR violations , and those that don't. The latter is the kind I would like to cover in this article.

The final division is the wars between the occupants of different rooms, and, those apparently happening just in the mind of a single person waging the war.

The first kind is actually pretty easy - this is what the zoning systems are for. I mean, the real ones that do allow you to set different temperature for different zones.

As for the last kind... Imagine a situation when there's a person who claims they are absolutely comfortable with a certain temperature - let's say, for the sake of argument, it's 78°F. And indeed, when put into a scientifically controlled environment, they feel just fine. But in real life...

They fiddle with the thermostat all the time. Sometimes they push it up. Sometimes they push it down. Sometimes, they claim the thermostat is not calibrated (it is). When cornered, they are claiming the temperature is not right, and demand a precision thermometer. And you know what?

Most of the time they are right.

The cause for it is trivial from a control systems engineer standpoint, but is carefully hidden in a plain view for the rest of the world.

Almost all thermostats, except the very advanced ones, are based on a very simple concept of a hysteresis. Temperature goes X degrees above the setpoint, cooling on. Temperature goes X degrees below the setpoint, cooling off. Simple?

Not quite.

First of all, not all thermostats (as a matter of fact, very few) will let you change X. According to some claims (google it up if you want), X = 2.5 (I suppose Fahrenheit) for mechanical thermostats, somewhat less for digital. 5°F spread is a lot.

Side note: Rite Temp GPMG8085C (and possibly other Rite Temp thermostats) will let you control hysteresis. Doesn't mean I endorse it, though - it's got its quirks. Doesn't mean I condemn it, either - it's doing a good job comparable to Honeywell RTH7500 for $20 (20%) less.

Then, the situation gets very interesting as temperatures get milder. The closer is the ambient temperature to the setpoint temperature, the more suffering there is (that'll be mornings, afternoons, and the whole mid-seasons). The reason?

Deviations of ambient temperature are sufficient to cause severe discomfort, but are not sufficient to kick the temperature at the thermostat beyond the dead band and thus to force the HVAC unit to actually do something.

Things can become really bad if the house is big enough, and the part of it where the thermostat is is insulated better than other parts - all in all, "mild" is a relative term. A typical case would be a tri-level house somewhere in Arizona.

Which is where and how it all started. Is there an answer? Yes, and it's been already implemented several years ago, but I'm not telling. It would suffice to say that if you want to find out, by this time you know where to look for it.

Sunday, July 1, 2007

HVAC Interoperability? Forget about it...

"The Home Plug and Play Specification represents a strong first step towards unifying the industry behind a common open industry specification and accelerating the deployment of smart interoperable devices in the home," stated Kevin Kahn, Director of Intel's Communications Architecture Lab and Intel Fellow. "This will enable new applications for the PC bringing increased convenience and security to the homeowner."

This inspiring quotation comes from a press release dated... January 10 1997. Over 10 years ago.

So, what do we have today?

The Home Plug and Play Specification never materialized.
Same is the fate of CEBus, for all practical purposes.
BACnet, Modbus, and LonTalk never made it down to residential applications.

Up to this day, there is no sign of existence of a unified diagnostic interface across residential HVAC devices.

Only the very advanced and recent systems such as Carrier Infinity have remote access, but even for that you have to pay a hefty $100 annual fee, and it does NOT have a simplest possible thing - the Ethernet interface. Yes, instead of providing an ability to plug a CAT-5 cable into it, it goes across the SkyTel pager network, quote, "because it didn't require a computer network to operate".

Side note: forgive me for being blunt, but I don't really understand why two pieces of equipment, both of which reside in my house, keep nagging for money and require a roundtrip across the world, rely on complicated and expensive infrastructure, absolutely refuse to talk to anything that hasn't been paid for and/or thought of by the designers? Two answers come to my mind:

  1. The manufacturers want to milk the customers and make them pay for most number of upgrade modules they can (market segmentation in action, also, see Carrier Infinity vs. Bryant Evolution);
  2. The designers didn't know about the fact that an Ethernet connector hardware costs pennies to implement and the computer network in a strict sense is not even necessary (hint: static IP, DHCP, zeroconf, Ethernet hub).
You pick the answer you like better.


Let me skip the rest of the long list of arguments that lie just beneath the surface, and state the fact as it appears from a consumer standpoint: Interoperability between commercially available hardware appliances is effectively nonexistent.

One of possible reasons for this is that the residential HVAC industry hasn't reached maturity - technically speaking, we're barely at introduction stage yet. None of the manufacturers seem to have realized that with drastically shortened product development and obsolescence it is much more profitable to play nice together than to rip the sandbox apart trying to get the biggest share of toys.

Related is the fact that the explosive growth of IT, achievements in hardware and software interoperability and the benefits they can yield are mostly being ignored.

Yet another fact is that most probably the manufacturers are stuck in the middle of the hardware development pipeline - and they still can't wake up and realize that unless they manage to shake off the illusion that they can pull it off, cut the losses, drop the pipeline and start working as the present day demands, they will be overrun by smaller, more agile companies that are able to utilize all these nice things referred and alleged to above.

Mark my words.