Ok so im working on the next phase of my HVAC project.. the first phase is going really well and the ZX-328n's do a wonderful Job!..
i want to measure the operating speed of the A/C unit's Compressor...
im looking for the RPM of the compressor.. sounds easy if it was an external fan or such but these are sealed 480 volt 3 phase compressors..
the compressor is operated by an inverter drive that takes the sdtandard 120 volt single phase input and turns it into the 3 phase needed for the compressor to operate..
being that the compressor is a scroll type compressor and not a piston type, it makes a distinct "pulsation" that can be heard (and felt) by the human senses each revolution it makes...
my first thought was that the controller sends the frequency data over the communications Link to the outdoor unit.. however this is not the case.. the outdoor unit simply takes the temperature readings and system setpoints from the indoor unit and then acts on them accordingly..
the onboard ATMEGA-32 handles the drive...
so im looking for a device I can connect to the ZX-328 and attach to the casing of the compressor itself that will output a pulse each time a compressor "pulsation" occurs...
then I can do Pulsin commands and measure how many pulses I get in a certain time frame etc to determine what speed the compressor is running at...
any ideas on the hardware I might use for such a purpose?
-Christopher
device to Measure physical Pulsations to determine frequency
-
cadillackid
- Posts: 35
- Joined: 22 August 2009, 16:34 PM
Hmmm, interesting problem. I have never done this. Here are some thoughts off the top of my head. I toss them out for discussion purposes.
I suspect that accelerometers might not have the resolution to feel the vibration, but that might be the first avenue to consider.
Maybe a strain gauge between the compressor and an immobile object?
Maybe a mirror on the compressor housing, and a laser diode shooting toward the mirror and reflecting into a phototransistor. You could detect when the vibration deflects the beam away from the phototransistor and back again.
A simple device would be a pendulum swinging inside a metal container. when it hits the edge it will make an electrical contact. The problem is that you might have issues with resonant frequencies. You could use some variant of this approach, however.
-Tony
I suspect that accelerometers might not have the resolution to feel the vibration, but that might be the first avenue to consider.
Maybe a strain gauge between the compressor and an immobile object?
Maybe a mirror on the compressor housing, and a laser diode shooting toward the mirror and reflecting into a phototransistor. You could detect when the vibration deflects the beam away from the phototransistor and back again.
A simple device would be a pendulum swinging inside a metal container. when it hits the edge it will make an electrical contact. The problem is that you might have issues with resonant frequencies. You could use some variant of this approach, however.
-Tony
device to Measure physical Pulsations to determine frequency
> ... attach to the casing of the compressor itself that will output a
pulse...
Three thoughts come to mind.
- ask the manufacturer if the motor speed control they use provides an
output to sample;
- monitor one of the three phases with a clamp-on coil to determine the
drive frequency;
- mount a piezo disk to the case or, perhaps better, to the coolant
high-pressure line, to sense the mechanical or liquid pulse.
Each of these (except, perhaps, the first) will probably require a
custom interface to make the signals useful to your processor.
pulse...
Three thoughts come to mind.
- ask the manufacturer if the motor speed control they use provides an
output to sample;
- monitor one of the three phases with a clamp-on coil to determine the
drive frequency;
- mount a piezo disk to the case or, perhaps better, to the coolant
high-pressure line, to sense the mechanical or liquid pulse.
Each of these (except, perhaps, the first) will probably require a
custom interface to make the signals useful to your processor.
Tom
I've never used a modern vibration sensing tachometer, but it might make sense to see how one of these operates.
I have used the old fashioned style (with vibrating reeds). They are/were very simple, required no maintenance, and accurate. I doubt one would work for this project though.
I do like the laser and mirror concept, but it would most certainly have to be a sealed unit itself, as it will have to be able to cope with the outdoor environment. Perhaps forgo the mirror, and mount the laser itself to the compressor housing, with the photo-transistor a short distance away. Mount one or the other on a flexible mount, and any movement of the compressor housing would surely put the two out of alignment, allowing for a measurement.
Tapping into the motor speed control would certainly be the most ideal solution.
-Don
I have used the old fashioned style (with vibrating reeds). They are/were very simple, required no maintenance, and accurate. I doubt one would work for this project though.
I do like the laser and mirror concept, but it would most certainly have to be a sealed unit itself, as it will have to be able to cope with the outdoor environment. Perhaps forgo the mirror, and mount the laser itself to the compressor housing, with the photo-transistor a short distance away. Mount one or the other on a flexible mount, and any movement of the compressor housing would surely put the two out of alignment, allowing for a measurement.
Tapping into the motor speed control would certainly be the most ideal solution.
-Don
There are sufficiently sensitive accelerometers. For example, this one would be able to sense the vibration if you go that route:
http://www.sparkfun.com/commerce/produc ... ts_id=9723
http://www.sparkfun.com/commerce/produc ... ts_id=9723
Paul
I took a quick look at the specifications of that accelerometer. It looks nice, but I am not sure how easy it would be to use. It seems as if it has the resolution to read compressor vibration, and the noise seems low enough. But I am not sure how to get the readings with low CPU overhead. It might be possible to get streaming readings, and determine when the sign of the acceleration changes. I did not see an interrupt mode that would do this for you, but maybe there is.pjc30943 wrote:There are sufficiently sensitive accelerometers. For example, this one would be able to sense the vibration if you go that route:
http://www.sparkfun.com/commerce/produc ... ts_id=9723
It seems to me that one would prefer NOT to stream the data, but use interrupts instead. With the correct interrupts, you could just measure the time between interrupts without reading the actual data!
-Tony
-
cadillackid
- Posts: 35
- Joined: 22 August 2009, 16:34 PM
a friend and I were talking last night and the compressor has an aluminum case on it.. so im thinking I might be able to see the stator magnets with a sensitive hall effect sensor.. its a standard sanyo 3 phase mini-compressor and motor information is available as well as a cut-away so I may be able to go that route also.
I also did find an rs-485 communication between the main outdoor board and the motor control system.. since 485 is multi-drop compatible im going to see if I can tag comma off of that..
I would of course have to crack the protocol but I think I have a shot at doing that.. not sure what info it will give me but im sure part of the data stream will have to be drive speed, frequency, or feedback of some sort on it...
tough part is these units are in service so I may order another unit to have to play around with in this regard...
-Christopher
I also did find an rs-485 communication between the main outdoor board and the motor control system.. since 485 is multi-drop compatible im going to see if I can tag comma off of that..
I would of course have to crack the protocol but I think I have a shot at doing that.. not sure what info it will give me but im sure part of the data stream will have to be drive speed, frequency, or feedback of some sort on it...
tough part is these units are in service so I may order another unit to have to play around with in this regard...
-Christopher
-
cadillackid
- Posts: 35
- Joined: 22 August 2009, 16:34 PM
a friend and I were talking last night and the compressor has an aluminum case on it.. so im thinking I might be able to see the stator magnets with a sensitive hall effect sensor.. its a standard sanyo 3 phase mini-compressor and motor information is available as well as a cut-away so I may be able to go that route also.
I also did find an rs-485 communication between the main outdoor board and the motor control system.. since 485 is multi-drop compatible im going to see if I can tag comma off of that..
I would of course have to crack the protocol but I think I have a shot at doing that.. not sure what info it will give me but im sure part of the data stream will have to be drive speed, frequency, or feedback of some sort on it...
tough part is these units are in service so I may order another unit to have to play around with in this regard...
-Christopher
I also did find an rs-485 communication between the main outdoor board and the motor control system.. since 485 is multi-drop compatible im going to see if I can tag comma off of that..
I would of course have to crack the protocol but I think I have a shot at doing that.. not sure what info it will give me but im sure part of the data stream will have to be drive speed, frequency, or feedback of some sort on it...
tough part is these units are in service so I may order another unit to have to play around with in this regard...
-Christopher