On Thu, 3 Jul 2008 00:50:24 -0700 (PDT), kazkotx@gmail.com wrote:
>I tend to forget that the AC system includes a heater. It's not
>something that we use much if ever here in Dallas. So I understand
>what a fan limit switch is for now.
>http://toad.net/~jsmeenen/blower.html
>Until I understand the wiring better, it almost looks as if the AC
>uses high speed on the fan and the heater uses MED speed on the fan.
That's the way almost all systems work. Hot air is less dense than cold air
and thus needs less fan to move it. Even most heat pumps, emitting the tepid
air that they do, use dual speed fans.
John
--
John De Armond
See my website for my current email address
http://www.neon-john.com
http://www.johndearmond.com <-- best little blog on the net!
Tellico Plains, Occupied TN
Unable to locate Coffee -- Operator Halted!
On Wed, 9 Jul 2008 09:50:41 -0700 (PDT), kazkotx@gmail.com wrote:
>Neon John, I am trying to see if one could devise a method of using a
>Lennox Harmony2 zone controller to control a three speed blower. If I
>were a bit more of a geek, I could probably devise a relay control
>circuit that would accept the variable fan speed voltage from the
>Harmony2 and convert it to three stages. I can follow directions, but
>my circuit designing days have ended. I wonder what creative ways one
>could convert the variable voltage control signal to the blower fan
>into three speeds.
In the good old days I'd grab an LM3907 quad op-amp (Is that the right number?
Brain fade) and whip up a triple comparator circuit. there would be a pot on
each one so that I could trim each stage to the right voltage. And then trim
it again in a few years when it aged.
Today I'd order a Basic Stamp or competitor, three relays, a couple of
capacitors and resistors and do it in software.
The very most basic Basic Stamp doesn't have an A/D converter but one can be
whipped out using 1 I/O pin, a capacitor and one or two resistors. You hang a
capacitor on one pin and program it as an output driven low. A resistor is
connected between the capacitor/pin and the incoming signal.
To do a conversion, the pin is programmed as an input and a timer started.
When the voltage on the cap reaches the zero-to-one transition point (the
input is a schmitt trigger and so the transition voltage is stable), stop the
timer and program the pin as an output and drive it low again to discharge the
cap. Do a look up in a previously prepared and programmed lookup table that
relates the time interval to voltage on the input.
This is a crude "single slope A/D converter" that uses the RC time constant of
the input resistor and the capacitor as the measuring element. With the use
of one more part and one more I/O pin, it can be converted to a "dual slope
converter" that can be quite accurate if quality parts are used.
the one part is a 2 terminal constant current regulator. I don't recall the
part number but they're common. Connect it between the first I/O pin and the
second one.
The Dual Slope conversion works like this. Initially, both pins are
programmed as output lows and the cap is discharged. At the start of the
conversion, program the cap pin as an input and the other pin tri-state or
high. Doesn't really matter. When the voltage on the cap exceeds the pin's
trip voltage, start a timer that will run long enough to let the capacitor
charge well past the trigger voltage.
When the timer expires, program the second pin output low and start another
timer while watching the first pin. The constant current device will
discharge the capacitor at a linear rate. When the voltage drops to the first
pin's reset point, stop the timer. Use the timer value to look up the
corresponding voltage value in the table.
This dual slope technique is more complicated but it neutralizes out several
pesky sources of errors, not the least of which is dielectric absorption on
the capacitor. And, of course, temperature drift.
That degree of accuracy probably isn't necessary for this application. I'd
try the single slope method first.
On the output side, the PIC processor that is the basis of the Basic Stamp can
drive a low coil current relay directly so no extra parts other than snubber
diodes needed.
Here's the Basic Stamp info
http://www.parallax.com/ProductInfo/Microcontrollers/tabid/121/Default.aspx
Seems like the low end version is only about $16 from Digikey. You don't need
the prototyping board or any of the other stuff that they offer. Only a
source of regulated voltage which you maybe can pull off the zone controller
board, and a serial cable to hook to your PC. It's a simple terminal
interface so you can communicate with it using windows terminal program or
much better, the free PuTTY.
They have lots of applications code available for downloading, including what
it takes to make a simple A/D like I've described.
One other thought. Before you go messing in the analog world, take a close
look at the circuitry that makes up the zone controller's analog output. It's
most likely that they made a simple 2 or 3 bit D/A converter out of some
resistors and 2 or 3 output pins on the processors. It might even be a PIC
processor.
This is a very common technique when the voltage accuracy requirement isn't
very high. If that's the case then you can simply tap the digital bits
directly off the processor's pins and buffer the signals to drive the relays
directly. At most you might have to do some diode or CMOS TTL logic.
Yet another thought. The unit might depend on maintaining a constant
temperature and pressure in the evaporator for part of its efficiency. That
may be why it originally used a continuously variable fan.
I suggest doing some investigating, including trying to find someone at the
factory to talk to or perhaps a technical paper in some journal that details
the unit's underlying engineering.
If a variable speed fan is integral to the design, that can be done for little
more than the cost of the Basic stamp. Fractional HP variable frequency
drives (VFDs) are dirt-cheap these days. I bet that I could find one for
under $50. VFDs typically take single phase line current in and generate the
proper voltage and frequency to drive a three phase motor. A three phase fan
motor might therefore be necessary. I've seen some VFDs that can drive a
split-phase motor but I can't remember where. I'd have to go looking.
If this all seems too complicated then drop me an email (address is on my
website in a spam-bot-proof form) and we'll talk about it. I work cheap,
especially when I'm having fun and this sounds like a fun project.
John
--
John De Armond
See my website for my current email address
http://www.neon-john.com
http://www.johndearmond.com <-- best little blog on the net!
Tellico Plains, Occupied TN
I like you ... you remind me of me when I was young and stupid.
XML site map
>something that we use much if ever here in Dallas. So I understand
>what a fan limit switch is for now.
>http://toad.net/~jsmeenen/blower.html
>Until I understand the wiring better, it almost looks as if the AC
>uses high speed on the fan and the heater uses MED speed on the fan.