About to start a DIY zoning project.. I have a couple questions...
JimB
close to finishing the renovation of the space over the garage
and now I will really need a zone controller. I have a couple
questions...
1. I see your software has servo motors controls. I can't see to find
an example of how they are used to make a damper?
The ones I'm looking at are from retrozone and they are ~$105 each.
I'd much rather find a cheaper solution since I need about 11 - 14 of
them!
Did I miss a section where it showed how to make one?
2. My initial thought is to use a micro controller instead of a PC to
control it all. Has anyone thought of porting to a PIC micro
controller?
Anyone have any thoughts on how tough it might be? Or am I better off
trying to write it from scratch. (I haven't really looked at the
software yet...)
I'm really excited to get started.. but I really need to finish the
renovation before I start otherwise I'll never finish the room!! :-)
Oh.. and is this the only "forum" for you guys?
Thanks in advance!
Jim
Vadim Tkachenko
> 1. I see your software has servo motors controls. I can't see to find
> an example of how they are used to make a damper?
There's the one I made in 2001:
http://diy-zoning.sourceforge.net/Homeowners/images/motorized_register.jpg
There are more - Tomasz made some (better than I did), but I don't
think I published those pictures anywhere.
Plus this (nice choice of hardware, but I geometry doesn't make much
sense as far as torque is concerned):
http://cyborgworkshop.org/2010/03/28/teaser/
And another one from the same guy:
http://cyborgworkshop.org/wp-content/gallery/misc-shots/register.jpg
These are not dampers, though, these are registers. Dampers are here:
http://cyborgworkshop.org/wp-content/gallery/Projects/WHIA/20090519-185546.jpg
http://cyborgworkshop.org/wp-content/gallery/Projects/WHIA/20090519-185636.jpg
http://cyborgworkshop.org/wp-content/gallery/Projects/WHIA/20090520-204907.jpg
http://cyborgworkshop.org/wp-content/gallery/Projects/WHIA/20090520-205159.jpg
(again, it puzzles me why would one want to reduce torque, but well,
apparently it works).
There are more pictures and articles at his site, some pretty
interesting - for example, this one
(http://cyborgworkshop.org/wp-content/gallery/Projects/WHIA/20090509-183041.jpg),
showing how to make a sensor in a simpler way than I did
(http://diy-zoning.blogspot.com/2010/07/dz-temperature-sensor-howto.html).
The whole directory is open:
http://cyborgworkshop.org/wp-content/gallery/Projects/WHIA/
And this is the label for his HVAC project:
http://cyborgworkshop.org/category/whia/
> The ones I'm looking at are from retrozone and they are ~$105 each.
Which ones?
> I'd much rather find a cheaper solution since I need about 11 - 14 of
> them!
I assume you're considering modulating dampers?
> Did I miss a section where it showed how to make one?
No, I don't think so. Mine is self-explanatory, and those other
pictures show it in enough details, I hope
(http://cyborgworkshop.org/projects/whole-home-information-awareness/building-the-servos/).
> 2. My initial thought is to use a micro controller instead of a PC to
> control it all. Has anyone thought of porting to a PIC micro
> controller?
That guy built something similar on Arduino - I have no idea what,
though, for there's no mention of that project being Open Source
anywhere.
> Anyone have any thoughts on how tough it might be? Or am I better off
> trying to write it from scratch.
I didn't think of a PIC, but an idea to port DZ to Arduino is in the
air for a couple of years now. Reason being, off-the-shelf stackable
platform with quite a few existing add-ons (Ethernet, XBee, PWM,
etc.). I wanted to freeze the feature set before the port begins,
though. Pretty close now.
If/when DZ gets ported to an embedded hardware platform, it will
retain all the features available now, including scheduling and remote
control.
> (I haven't really looked at the software yet...)
You might even like it :)
> I'm really excited to get started.. but I really need to finish the
> renovation before I start otherwise I'll never finish the room!! :-)
Good luck :)
> Oh.. and is this the only "forum" for you guys?
I'm not sure I understand the question. For DZ, this *is* the forum.
There's another family of forums you might want to take a look at
(http://hvac-talk.com/), but the audience there is different - HVAC
contractors, mostly.
> Jim
--vt
Tim Small
> I didn't think of a PIC, but an idea to port DZ to Arduino is in the
> air for a couple of years now. Reason being, off-the-shelf stackable
> platform with quite a few existing add-ons (Ethernet, XBee, PWM,
> etc.). I wanted to freeze the feature set before the port begins,
> though. Pretty close now.
>
> If/when DZ gets ported to an embedded hardware platform, it will
> retain all the features available now, including scheduling and remote
> control.
>
Personally, I'm not sure that porting to an "embedded" platform at this
stage will be a big win - I'm assuming it'll mean ditching Java (and
thus a total re-write), and the hardware to run the code on is such a
low proportion of the cost of a DZ installation at the moment...
It's currently very attractive to be able to prototype and develop on a
dedicated PC (or old laptop or whatever), and then switch to something
like a Beagle Board or an NSLU2 etc. or some common wireless router
hardware at a later date.
It would perhaps be less work (both up front and in the long run) to get
the code running on some such suitable cheapish ARM/Linux compatible
device, and distributing a "plug-and-go" image for it - e.g. DZ
pre-installed on open-embedded/emdebian/debian etc. - with the code
either running under the full OpenJDK JVM, or maybe even under the
Dalvik runtime?
Many such devices have some SPI/I2C/GPIO pins which can be brought out.
Perhaps the problem with these is that they tend to be a rapidly moving
target (relatively rapid board changes etc.) - so perhaps something like
the "dragino" might be better?
http://hackerspaces.org/wiki/Dragino
as used by these guys:
https://www.flukso.net/content/about-flukso
... that having been said - perhaps another route would be some sort of
hybrid solution with an Arduino doing some low-level control stuff, and
the bulk of the DZ code running on a commodity embedded-class Linux
platform (maybe a cheap Linux tablet or phone, tied to the Arduino via
USB, and doubling as a GUI device), or even in the cloud would be a good
way to go?
Hmm...
Tim.
JimB
> Personally, I'm not sure that porting to an "embedded" platform at this
> stage will be a big win - I'm assuming it'll mean ditching Java (and
> thus a total re-write), and the hardware to run the code on is such a
> low proportion of the cost of a DZ installation at the moment...
back end.
The back end will do all the control aspects (measuring, flipping
relays, data reporting.)
The front end will be web based (or some java app) that will
communicate via TCP to the back end to
collect and store data (for logging) and for changing setpoints and
displaying status.
At least that's what I envsion.
>I didn't think of a PIC, but an idea to port DZ to Arduino is in the
>air for a couple of years now. Reason being, off-the-shelf stackable
>platform with quite a few existing add-ons (Ethernet, XBee, PWM,
>etc.). I wanted to freeze the feature set before the port begins,
>though. Pretty close now.
I'm familiar with it!
There are existing add ons as well for ETH, Xbee etc.. (Although I
haven't played with Xbee yet)
>> The ones I'm looking at are from retrozone and they are ~$105 each.
>Which ones?
Currently I have these butterfly damper covers on all my HVAC
outlets. They are adjusted by a multi-turn screw. Maybe I can find a
way to hook to a servo but it seems like a geared apperatus might be
in order... which is beyond what I want to do!
IF I just go with the "off the shelf" dampers and install correctly
then down the road a commercial zoning system can be put in.
Jim
Jerry Zdenek
I was planning on running dZ on a dedicated system that was pretty much
standalone. It's cheaper than buying a new PC and more power efficent
to. There's a lot of ARM based boards out there that run Linux.
I was thinking about one of the andahammer Mini6410's:
http://www.andahammer.com/mini6410-sdk/
However, since I just moved and don't have a house yet, it'll be a while
before I get to it.
Jerry
Vadim Tkachenko
>> Personally, I'm not sure that porting to an "embedded" platform at this
>> stage will be a big win - I'm assuming it'll mean ditching Java (and
>> thus a total re-write), and the hardware to run the code on is such a
>> low proportion of the cost of a DZ installation at the moment...
>
> My thought is splitting this into two parts.. the front end and the
> back end.
> The back end will do all the control aspects (measuring, flipping
> relays, data reporting.)
>
> The front end will be web based (or some java app) that will
> communicate via TCP to the back end to
> collect and store data (for logging) and for changing setpoints and
> displaying status.
>
> At least that's what I envsion.
Not along the same exact lines, but there is currently a webapp you
can use to control DZ from anywhere where Internet is, without a need
to host anything at your home computer. Few weeks ago I posted an
announce and asked if anyone wants it, several people responded, and
there's still space left. Contact me off the list if you want to try
it out.
>>I didn't think of a PIC, but an idea to port DZ to Arduino is in the
>>air for a couple of years now. Reason being, off-the-shelf stackable
>>platform with quite a few existing add-ons (Ethernet, XBee, PWM,
>>etc.). I wanted to freeze the feature set before the port begins,
>>though. Pretty close now.
>
> The reason I'll choose the PIC is that that's what I use at work and
> I'm familiar with it!
:)
>>> The ones I'm looking at are from retrozone and they are ~$105 each.
>>Which ones?
> http://www.retrozone.com/Catalog/mechanical_dampers.htm
One thing I forgot to mention - I'd stay away from high torque servos,
they're simply not necessary with correct design.
> Currently I have these butterfly damper covers on all my HVAC
> outlets. They are adjusted by a multi-turn screw. Maybe I can find a
> way to hook to a servo but it seems like a geared apperatus might be
> in order... which is beyond what I want to do!
Hmm... This sound like a dream application for the stepper. I've been
planning to write a stepper driver for quite a few years now, but
necessity (nor the hardware) never exceeded the pain limit. If you
lurkers tell me which stepper controller is amicable, I might just do
it.
> IF I just go with the "off the shelf" dampers and install correctly
> then down the road a commercial zoning system can be put in.
<evil laugh> Yeah, good luck with financing *that* :)
Seriously, though, look at this:
http://diy-zoning.blogspot.com/search/label/competition
If you know about any others, I'd be happy to look at them as well and
throw in my $0.02 (not necessarily negative).
> Jim
--vt
Brent Hineline
On Oct 27, 2011, at 11:21 PM, Vadim Tkachenko wrote:
> Seriously, though, look at this:
> http://diy-zoning.blogspot.com/search/label/competition
>
> If you know about any others, I'd be happy to look at them as well and
> throw in my $0.02 (not necessarily negative).
This week's future product announcement was the Nest Learning Thermostat (http://www.nest.com/) Although promising, time will tell if it gets any traction for reasons other than it being associated with an ex-Apple VP. It apparently will integrate into existing zoning systems, but certainly doesn't disrupt that market.
Brent
Vadim Tkachenko
>> If you know about any others, I'd be happy to look at them as well and
>> throw in my $0.02 (not necessarily negative).
>
> This week's future product announcement was the Nest Learning Thermostat (http://www.nest.com/)
Thank you, this link made my day.
Very impressive at the first glance. Lots of things to learn from for
me. Some things are outright bizarre - for example, "the blog" without
an RSS or Atom feed link.
> Although promising, time will tell if it gets any traction for reasons other than it being associated with an ex-Apple VP. It
> apparently will integrate into existing zoning systems, but certainly doesn't disrupt that market.
Agree completely. After the dust settles in, it's just a bleeping
thermostat - it won't address most things that DZ and other zoning and
climate control systems address.
I want to grok it completely before I write the long article about it
on DZ blog, I'll post the link to it here.
And I guess the comparison chart between different competition systems
and DZ is long overdue.
And again, tell me what are the comparison criteria that are relevant
to *you*, the users, for my view will be certainly be different.
> Brent
--vt
Michael Spath
The Nest is certainly a slick looking thermostat. I always doubt the
size of the savings claim the manufacture posts.
I have been wondering what it would take to create a DIY/open source
thermostat, one that could support zoning projects like DZ. I have
enough projects to keep me busy and this one is not high in the queue
right now.
This DIY thermostat popped up this week on Hack a day:
http://hackaday.com/2011/10/27/diy-thermostat-keeps-the-harsh-winter-cold-at-bay/
This also popped up this week on Hack a day and made me think of DZ.
Not sure if it really applies:
http://hackaday.com/2011/10/25/wicked-use-of-html5-to-display-sensor-data/
Mike
Vadim Tkachenko
[afterthought, as usual]
> This week's future product announcement was the Nest Learning Thermostat (http://www.nest.com/) Although promising,
> time will tell if it gets any traction for reasons other than it being associated with an ex-Apple VP. It apparently will integrate
> into existing zoning systems,
I doubt that this integration will be efficient or effective. I've
considered making DZ work with conventional thermostats, and then
decided that it's going to be a waste of time and effort because the
only thing they support is on/of state, and DZ needs analog signal to
control the house as a whole - different zones are interrelated. This
thermostat (at least what I saw so far) is a drop-in replacement for a
conventional wired thermostat, and is unlikely to produce analog
signal - which makes it a very expensive microscope to hammer those
nails.
> Brent
--vt
Vadim Tkachenko
> The Nest is certainly a slick looking thermostat.
That's for sure. Though, when I asked my son to watch the video, his
knee-jerk reaction was "One button? Again? I'm sick of *one* button".
And I agree. It would be interesting to compare DZ UI on the
smartphone - when I'm done, I'll post it to Android Market and make it
so anyone can play with it even if they don't have DZ installed.
As I was writing this, two other things came to my mind. One, "If you
create a system that any idiot can use, then only idiots will find it
useful". Nice followup to that is at
http://c2.com/cgi/wiki?IdiotProofProcess
And the other is - "Make things as simple as possible, but not
simpler". With mandatory attribution -
http://en.wikiquote.org/wiki/Albert_Einstein
> I always doubt the size of the savings claim the manufacture posts.
Actually, you'd be surprised at how true this specific claim is. Last
summer, I've installed DZ to control just the ACs in my house (null
dampers, for it was prohibitively expensive to install them wired, and
I'm not done with power efficient wireless setup yet). Electric bill
dropped like a brick, and it lasted till the end of the cooling season
(just about now). This spring, my family rebelled at me trying to be
frugal and demanded the temperature to be lowered 1°C. Well, I'd say
that the penalty certainly was *very* noticeable - can't tell you in
dollar terms, but this year the energy consumption summary per month
showed at least 10%, and quite possibly more, increase.
However, there's a thing that bothers the hell out of me - a product
that is not yet on the market already has testimonials from three
people with just the first name and location. Smells bad to me.
> I have been wondering what it would take to create a DIY/open source
> thermostat, one that could support zoning projects like DZ. I have enough
> projects to keep me busy and this one is not high in the queue right now.
>
> This DIY thermostat popped up this week on Hack a day:
> http://hackaday.com/2011/10/27/diy-thermostat-keeps-the-harsh-winter-cold-at-bay/
Ukraine :D
The question is, though, is it really the thermostat you want on the
wall, or just the display and the controller?
Sensors should be where *you* are, not where the thermostat is. One of
mine is hidden behind a stand up piano in the living room (and works
surprisingly well, I was afraid that it will not pick up temperature
changes fast enough because of air getting stale, but no, it works
perfectly), the other is at the night stand, and yet another is in the
corner behind the speaker next to the home theater (so it would pick
up heat generated by the equipment and react before people notice).
> This also popped up this week on Hack a day and made me think of DZ. Not
> sure if it really applies:
> http://hackaday.com/2011/10/25/wicked-use-of-html5-to-display-sensor-data/
The idea to scan the QR code to start the webapp on the phone is
definitely cool. However, the whole thing is a little bit off because
DZ sensor data stream is not near as dense as what they're dealing
with. I'm currently planning on using Google Visualization API
(http://code.google.com/apis/chart/) for that, and if that turns out
to be unfit, I'll just refactor the DZ2 chart API.
> Mike
--vt
Vadim Tkachenko
>> http://diy-zoning.blogspot.com/search/label/competition
>>
>> If you know about any others, I'd be happy to look at them as well and
>> throw in my $0.02 (not necessarily negative).
>
> This week's future product announcement was the Nest Learning Thermostat (http://www.nest.com/) Although promising, time will tell if it gets any traction for reasons other than it being associated with an ex-Apple VP. It apparently will integrate into existing zoning systems, but certainly doesn't disrupt that market.
As promised, here's a detailed review:
http://diy-zoning.blogspot.com/2011/10/competition-nest-learning-thermostat.html
Wonder what else you guys see in it that I didn't.
> Brent
--vt
JimB
> can use to control DZ from anywhere where Internet is, without a need
> to host anything at your home computer. Few weeks ago I posted an
> announce and asked if anyone wants it, several people responded, and
> there's still space left. Contact me off the list if you want to try
> it out.
dreaming! But I'm not sure I'd
be ready to play with it yet..
> >>> The ones I'm looking at are from retrozone and they are ~$105 each.
> >>Which ones?
> >http://www.retrozone.com/Catalog/mechanical_dampers.htm
>
> One thing I forgot to mention - I'd stay away from high torque servos,
> they're simply not necessary with correct design.
>
> > Currently I have these butterfly damper covers on all my HVAC
> > outlets. They are adjusted by a multi-turn screw. Maybe I can find a
> > way to hook to a servo but it seems like a geared apperatus might be
> > in order... which is beyond what I want to do!
>
> Hmm... This sound like a dream application for the stepper. I've been
> planning to write a stepper driver for quite a few years now, but
> necessity (nor the hardware) never exceeded the pain limit. If you
> lurkers tell me which stepper controller is amicable, I might just do
> it.
there is no access to the spindle from the
inside of the damper! :( So the motor would have to go on the outside
or I'd have to epoxy an extension to the spindle.. And then I'd
have to come up with a way to mount the stepper on the inside.. As
fun as it would be... It seems like a lot of work that might be prone
to breaking!
I'd rather invest my time getting all the sensors in place and
designing the PCB to hold my PIC and relays and spend the money on the
dampers..
> > IF I just go with the "off the shelf" dampers and install correctly
> > then down the road a commercial zoning system can be put in.
>
> <evil laugh> Yeah, good luck with financing *that* :)
something I hacked in my butterfly registers!
Jim
JimB
http://airtec.rectorseal.com/mvclassic_main.asp
Except I have the MVAW "retro" air diffuser.
The link above show a better picture of the guts.
Jim