PDA

View Full Version : Frugal data recorder


Don K
July 17th 03, 03:55 AM
In the past, there has been some interest expressed here in
determining energy usage of appliances that cycle on an off,
like refrigerators, air conditioners, etc.

One way to do that is to rig up a data recorder that
would record things like voltage, current, temperatures,
etc. as a function of time.

You can buy a complete 4-channel data recorder for $25
that simulates a chart recorder on your PC. It plugs into
your computer's serial port and comes with software
for Windows. The 4 input channels take inputs between
-10Volts and +10Volts and digitizes them with 10 bit
resolution (~1000 steps). You would have to scale
whatever parameter you're measuring to fit within
the +-10 volt range.

http://www.dataq.com/194.htm

For $25, I think this is a pretty neat toy to play around with.

You could do things like plot the temperatures inside and
outside your house, and how often the furnace or AC kicks in.

How fast does your oven get up to temperature?

How much does the temperature change in your
refrigerator, after you open and close the door?
Etc. etc.

Don

Halcitron
July 17th 03, 05:07 AM
>From: "Don K"
>Newsgroups: misc.consumers.frugal-living
>Date: Wed, 16 Jul 2003 21:55:21 -0400

>
>In the past, there has been some interest expressed here in
>determining energy usage of appliances that cycle on an off,
>like refrigerators, air conditioners, etc.
>
>One way to do that is to rig up a data recorder that
>would record things like voltage, current, temperatures,
>etc. as a function of time.
>
>You can buy a complete 4-channel data recorder for $25
>that simulates a chart recorder on your PC. It plugs into
>your computer's serial port and comes with software
>for Windows. The 4 input channels take inputs between
>-10Volts and +10Volts and digitizes them with 10 bit
>resolution (~1000 steps). You would have to scale
>whatever parameter you're measuring to fit within
>the +-10 volt range.

You do that with a "Voltage Divider"

<http://hyperphysics.phy-astr.gsu.edu/hbase/electric/voldiv.html>

>
>http://www.dataq.com/194.htm
>
>For $25, I think this is a pretty neat toy to play around with.
>
>You could do things like plot the temperatures inside and
>outside your house, and how often the furnace or AC kicks in.
>
>How fast does your oven get up to temperature?
>
>How much does the temperature change in your
>refrigerator, after you open and close the door?
>Etc. etc.
>
>Don
>
>
>
>
>
>
>
>


caveat lector

Halcitron misc.survivalism
Check your six and know when to duck.
NRA Member since 2002
The Law of the Land, is the weapon in your hand.

Smith & Wesson starts where the Bill of Rights stop.

Halcitron
July 17th 03, 05:07 AM
>From: "Don K"
>Newsgroups: misc.consumers.frugal-living
>Date: Wed, 16 Jul 2003 21:55:21 -0400

>
>In the past, there has been some interest expressed here in
>determining energy usage of appliances that cycle on an off,
>like refrigerators, air conditioners, etc.
>
>One way to do that is to rig up a data recorder that
>would record things like voltage, current, temperatures,
>etc. as a function of time.
>
>You can buy a complete 4-channel data recorder for $25
>that simulates a chart recorder on your PC. It plugs into
>your computer's serial port and comes with software
>for Windows. The 4 input channels take inputs between
>-10Volts and +10Volts and digitizes them with 10 bit
>resolution (~1000 steps). You would have to scale
>whatever parameter you're measuring to fit within
>the +-10 volt range.

You do that with a "Voltage Divider"

<http://hyperphysics.phy-astr.gsu.edu/hbase/electric/voldiv.html>

>
>http://www.dataq.com/194.htm
>
>For $25, I think this is a pretty neat toy to play around with.
>
>You could do things like plot the temperatures inside and
>outside your house, and how often the furnace or AC kicks in.
>
>How fast does your oven get up to temperature?
>
>How much does the temperature change in your
>refrigerator, after you open and close the door?
>Etc. etc.
>
>Don
>
>
>
>
>
>
>
>


caveat lector

Halcitron misc.survivalism
Check your six and know when to duck.
NRA Member since 2002
The Law of the Land, is the weapon in your hand.

Smith & Wesson starts where the Bill of Rights stop.

Bill
July 17th 03, 03:35 PM
Don K wrote:

>In the past, there has been some interest expressed here in
>determining energy usage of appliances that cycle on an off,
>like refrigerators, air conditioners, etc.
>
>One way to do that is to rig up a data recorder that
>would record things like voltage, current, temperatures,
>etc. as a function of time.
>
>You can buy a complete 4-channel data recorder for $25
>that simulates a chart recorder on your PC. It plugs into
>your computer's serial port and comes with software
>for Windows. The 4 input channels take inputs between
>-10Volts and +10Volts and digitizes them with 10 bit
>resolution (~1000 steps). You would have to scale
>whatever parameter you're measuring to fit within
>the +-10 volt range.
>
>http://www.dataq.com/194.htm
>
>For $25, I think this is a pretty neat toy to play around with.
>
>You could do things like plot the temperatures inside and
>outside your house, and how often the furnace or AC kicks in.
>
>How fast does your oven get up to temperature?
>
>How much does the temperature change in your
>refrigerator, after you open and close the door?
>Etc. etc.
>
>Don
>
>
>
>
The kicker here is that all the converters you have to have to convert
temperature, AC current, AC voltage, etc. to +/- 10 V cost extra. So in
the long run it costs quite a bit more than $25.

Bill Gill

Bill
July 17th 03, 03:35 PM
Don K wrote:

>In the past, there has been some interest expressed here in
>determining energy usage of appliances that cycle on an off,
>like refrigerators, air conditioners, etc.
>
>One way to do that is to rig up a data recorder that
>would record things like voltage, current, temperatures,
>etc. as a function of time.
>
>You can buy a complete 4-channel data recorder for $25
>that simulates a chart recorder on your PC. It plugs into
>your computer's serial port and comes with software
>for Windows. The 4 input channels take inputs between
>-10Volts and +10Volts and digitizes them with 10 bit
>resolution (~1000 steps). You would have to scale
>whatever parameter you're measuring to fit within
>the +-10 volt range.
>
>http://www.dataq.com/194.htm
>
>For $25, I think this is a pretty neat toy to play around with.
>
>You could do things like plot the temperatures inside and
>outside your house, and how often the furnace or AC kicks in.
>
>How fast does your oven get up to temperature?
>
>How much does the temperature change in your
>refrigerator, after you open and close the door?
>Etc. etc.
>
>Don
>
>
>
>
The kicker here is that all the converters you have to have to convert
temperature, AC current, AC voltage, etc. to +/- 10 V cost extra. So in
the long run it costs quite a bit more than $25.

Bill Gill

Don K
July 18th 03, 04:53 AM
"Halcitron" > wrote in message
...
> >From: "Don K"
> >Newsgroups: misc.consumers.frugal-living
> >Date: Wed, 16 Jul 2003 21:55:21 -0400
>
> >
> >In the past, there has been some interest expressed here in
> >determining energy usage of appliances that cycle on an off,
> >like refrigerators, air conditioners, etc.
> >
> >One way to do that is to rig up a data recorder that
> >would record things like voltage, current, temperatures,
> >etc. as a function of time.
> >
> >You can buy a complete 4-channel data recorder for $25
> >that simulates a chart recorder on your PC. It plugs into
> >your computer's serial port and comes with software
> >for Windows. The 4 input channels take inputs between
> >-10Volts and +10Volts and digitizes them with 10 bit
> >resolution (~1000 steps). You would have to scale
> >whatever parameter you're measuring to fit within
> >the +-10 volt range.
>
> You do that with a "Voltage Divider"


Maybe, but you will have to be aware that the channel inputs
are not totally isolated from each other, and that you cannot
allow a floating voltage difference of more than 20V across
any ports.

To get around this limitation, you could also use a step down
transformer to lower AC voltage and a current-sensing relay
across a small value series resistor to determine when current
is being drawn. This would at least let you plot the duty-cycle
of an appliance.

If you're familiar with op amps, you can make a more
sophisticated front end.

Don

>
> >
> >http://www.dataq.com/194.htm
> >
> >For $25, I think this is a pretty neat toy to play around with.

Don K
July 18th 03, 04:53 AM
"Halcitron" > wrote in message
...
> >From: "Don K"
> >Newsgroups: misc.consumers.frugal-living
> >Date: Wed, 16 Jul 2003 21:55:21 -0400
>
> >
> >In the past, there has been some interest expressed here in
> >determining energy usage of appliances that cycle on an off,
> >like refrigerators, air conditioners, etc.
> >
> >One way to do that is to rig up a data recorder that
> >would record things like voltage, current, temperatures,
> >etc. as a function of time.
> >
> >You can buy a complete 4-channel data recorder for $25
> >that simulates a chart recorder on your PC. It plugs into
> >your computer's serial port and comes with software
> >for Windows. The 4 input channels take inputs between
> >-10Volts and +10Volts and digitizes them with 10 bit
> >resolution (~1000 steps). You would have to scale
> >whatever parameter you're measuring to fit within
> >the +-10 volt range.
>
> You do that with a "Voltage Divider"


Maybe, but you will have to be aware that the channel inputs
are not totally isolated from each other, and that you cannot
allow a floating voltage difference of more than 20V across
any ports.

To get around this limitation, you could also use a step down
transformer to lower AC voltage and a current-sensing relay
across a small value series resistor to determine when current
is being drawn. This would at least let you plot the duty-cycle
of an appliance.

If you're familiar with op amps, you can make a more
sophisticated front end.

Don

>
> >
> >http://www.dataq.com/194.htm
> >
> >For $25, I think this is a pretty neat toy to play around with.

Don K
July 18th 03, 05:08 AM
"Bill" > wrote in message
.com...
> Don K wrote:
>
> >One way to do that is to rig up a data recorder that
> >would record things like voltage, current, temperatures,
> >etc. as a function of time.
> >
> >http://www.dataq.com/194.htm
> >
> >For $25, I think this is a pretty neat toy to play around with.
> >
> The kicker here is that all the converters you have to have to convert
> temperature, AC current, AC voltage, etc. to +/- 10 V cost extra. So in
> the long run it costs quite a bit more than $25.
>
> Bill Gill

If you're clever, you could probably do this on the cheap.

A resistor plus a ($2.50) thermistor or a diode would give you
a temperature indicator.

A $6 transformer plus resistor divider gives an AC voltage indicator.

A $6 transformer across a small series resistance (a length of wire)
would give you an AC current indicator.

(Radio Shack prices)

Don

Don K
July 18th 03, 05:08 AM
"Bill" > wrote in message
.com...
> Don K wrote:
>
> >One way to do that is to rig up a data recorder that
> >would record things like voltage, current, temperatures,
> >etc. as a function of time.
> >
> >http://www.dataq.com/194.htm
> >
> >For $25, I think this is a pretty neat toy to play around with.
> >
> The kicker here is that all the converters you have to have to convert
> temperature, AC current, AC voltage, etc. to +/- 10 V cost extra. So in
> the long run it costs quite a bit more than $25.
>
> Bill Gill

If you're clever, you could probably do this on the cheap.

A resistor plus a ($2.50) thermistor or a diode would give you
a temperature indicator.

A $6 transformer plus resistor divider gives an AC voltage indicator.

A $6 transformer across a small series resistance (a length of wire)
would give you an AC current indicator.

(Radio Shack prices)

Don

Bill
July 18th 03, 03:45 PM
Don K wrote:

>"Bill" > wrote in message
.com...
>
>
>>Don K wrote:
>>
>>
>>
>>>One way to do that is to rig up a data recorder that
>>>would record things like voltage, current, temperatures,
>>>etc. as a function of time.
>>>
>>>http://www.dataq.com/194.htm
>>>
>>>For $25, I think this is a pretty neat toy to play around with.
>>>
>>>
>>>
>>The kicker here is that all the converters you have to have to convert
>>temperature, AC current, AC voltage, etc. to +/- 10 V cost extra. So in
>>the long run it costs quite a bit more than $25.
>>
>>Bill Gill
>>
>>
>
>If you're clever, you could probably do this on the cheap.
>
>A resistor plus a ($2.50) thermistor or a diode would give you
>a temperature indicator.
>
>A $6 transformer plus resistor divider gives an AC voltage indicator.
>
>A $6 transformer across a small series resistance (a length of wire)
> would give you an AC current indicator.
>
>(Radio Shack prices)
>
>Don
>
>
>
>
One thing is that all the devices you mention have to be calibrated. I
was a calibration technician for many years, then I was an engineer in
the Aerospace industry, working with sensor. Calibration, and stability
are rather difficult to achieved with a handful of components.

The simplest would be a voltage sensor. You can do that with a
transformer, a few resistors and a diode. Then you could calibrate it
against a digital voltmeter. Be sure to use the transformer because you
must isolate the input to the A/D converter (that's the computer card)
from the power line.

Temperature is always difficult, but this one you might be able to buy a
sensor for.

Current is also difficult. If you use a transformer across a length of
wire, as you suggest, the voltage output will be too small to detect
with a +/- 10 volt A/D converter. If the input is down in the millivolt
range it will be lost in the noise.

All these things can be done, but not too many people can just pick up a
handful of parts and put together a good sensor.

Bill Gill

Bill
July 18th 03, 03:45 PM
Don K wrote:

>"Bill" > wrote in message
.com...
>
>
>>Don K wrote:
>>
>>
>>
>>>One way to do that is to rig up a data recorder that
>>>would record things like voltage, current, temperatures,
>>>etc. as a function of time.
>>>
>>>http://www.dataq.com/194.htm
>>>
>>>For $25, I think this is a pretty neat toy to play around with.
>>>
>>>
>>>
>>The kicker here is that all the converters you have to have to convert
>>temperature, AC current, AC voltage, etc. to +/- 10 V cost extra. So in
>>the long run it costs quite a bit more than $25.
>>
>>Bill Gill
>>
>>
>
>If you're clever, you could probably do this on the cheap.
>
>A resistor plus a ($2.50) thermistor or a diode would give you
>a temperature indicator.
>
>A $6 transformer plus resistor divider gives an AC voltage indicator.
>
>A $6 transformer across a small series resistance (a length of wire)
> would give you an AC current indicator.
>
>(Radio Shack prices)
>
>Don
>
>
>
>
One thing is that all the devices you mention have to be calibrated. I
was a calibration technician for many years, then I was an engineer in
the Aerospace industry, working with sensor. Calibration, and stability
are rather difficult to achieved with a handful of components.

The simplest would be a voltage sensor. You can do that with a
transformer, a few resistors and a diode. Then you could calibrate it
against a digital voltmeter. Be sure to use the transformer because you
must isolate the input to the A/D converter (that's the computer card)
from the power line.

Temperature is always difficult, but this one you might be able to buy a
sensor for.

Current is also difficult. If you use a transformer across a length of
wire, as you suggest, the voltage output will be too small to detect
with a +/- 10 volt A/D converter. If the input is down in the millivolt
range it will be lost in the noise.

All these things can be done, but not too many people can just pick up a
handful of parts and put together a good sensor.

Bill Gill

Don K
July 18th 03, 11:59 PM
"Bill" > wrote in message
gy.com...


>Current is also difficult. If you use a transformer across a length of
wire, as
> you suggest, the voltage output will be too small to detect with a +/- 10
volt
>A/D converter. If the input is down in the millivolt range it will be lost
in the noise.

For a 10 Amp appliance, if you put a 120V:6V transformer (backwards so that
it
steps up the voltage drop) across 5 feet of #14 wire, you should get an
output of
about +/-1volt peak to peak. About 1.3 watts will dissipate in the wire.

If all you want to do is look at the duty cycle, you don't really care about
calibration.
But you can always calibrate the sensor by comparing as set of measurements
to a known instrument and apply the corrections to your data in the
computer.

> All these things can be done, but not too many people can just pick up a
> handful of parts and put together a good sensor.

I don't recommend that anyone fool around with 120VAC unless they
understand what they're doing. But I think there are a lot of interesting
applications for someone who wants to play around with an inexpensive
data logger.

Don

Don K
July 18th 03, 11:59 PM
"Bill" > wrote in message
gy.com...


>Current is also difficult. If you use a transformer across a length of
wire, as
> you suggest, the voltage output will be too small to detect with a +/- 10
volt
>A/D converter. If the input is down in the millivolt range it will be lost
in the noise.

For a 10 Amp appliance, if you put a 120V:6V transformer (backwards so that
it
steps up the voltage drop) across 5 feet of #14 wire, you should get an
output of
about +/-1volt peak to peak. About 1.3 watts will dissipate in the wire.

If all you want to do is look at the duty cycle, you don't really care about
calibration.
But you can always calibrate the sensor by comparing as set of measurements
to a known instrument and apply the corrections to your data in the
computer.

> All these things can be done, but not too many people can just pick up a
> handful of parts and put together a good sensor.

I don't recommend that anyone fool around with 120VAC unless they
understand what they're doing. But I think there are a lot of interesting
applications for someone who wants to play around with an inexpensive
data logger.

Don

Google