Analog dynamic range, accuracy and number of bits

[Habib Bouaziz-Viallet]

Hi all,

Anyone could tell me what are the relations between Analog dynamic range
(figure with no dimension), accuracy over this range and number of bits ?
In other words, how to compute the number of bits of an ADC knowing the
analog range and the accuracy wanted over this range.

Ex :
Analog range = 1:500
accuracy over this range : +-0.1%
number of bits = ?

Best Regards, Habib

[upsid...@downunder.com]

On 02 Nov 2012 08:10:18 GMT, Habib Bouaziz-Viallet <h...@elsewhere.eu>
wrote:

I do not know how you define dynamic range, but the old rule of thumb
is that the signal to noise ratio SNR (signal power to quantization
noise power) expressed in decibels, can be calculated simply as

SNR = 6 x number_of_bits

[Fred Bartoli]

Habib Bouaziz-Viallet a �crit :

Depending on what you want/have to do there are a lot of parameters that
can enter into the equation.

Better to tell us what you want to do.

--
Thanks,
Fred.

[Robert Macy]

Sounds more like a 'test' question, but here goes:
if accuracy is +/- 1/1000 and the smallest range to largest implies
500X, 1/500000; which is a little more than 19 bits, so therefore you
need AT LEAST 20 bits.

[John Larkin]

On 02 Nov 2012 08:10:18 GMT, Habib Bouaziz-Viallet <h...@elsewhere.eu>
wrote:

12 bits!


--

John Larkin Highland Technology Inc
www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom timing and laser controllers
Photonics and fiberoptic TTL data links
VME analog, thermocouple, LVDT, synchro, tachometer
Multichannel arbitrary waveform generators

[Spehro Pefhany]

Or maybe 19, but if it's a test question, show your work.

[Phil Hobbs]

For AC, where you have to fit the p-p signal into the ADC and the DC
offset doesn't count as signal, it's

SNR_max = 6.02*N + 10*log(1.5).

The odd-looking 10 log 1.5 (1.76 dB) comes in because you lose a factor
of 8 in power (2*sqrt(2) in voltage) from accommodating the peak-to-peak
sine wave signal, but you gain a factor of 12 in power because the RMS
noise due to sampling is sqrt(1/12)* (1 LSB).

If it's DC, where the offset counts as signal, you get back the factor
of 8, so it's

SNR_max = 6.02*N + 10*log(12), or roughly 6N+10.8 dB.

This only applies where Widrow's theorem is valid, i.e. the signal is at
least a few LSBs in size, so that it's a good approximation to treat
quantization as additive noise of sqrt(1/12) LSB.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

[Habib Bouaziz-Viallet]

Le 02/11/12 15:36, Robert Macy a �crit :

In fact i'm far from a test question ... it's real life and i've got the
same approach.

Energy Watt-Meter 0,1% accuracy, 230Volts/50Hz. I range (0,02In .. In,
10 In)), U range (0,8Un .. Un .. 1,5Un)

--> 24bits on ADC for currents measurements.

Any objection from gurus of the analog ?

Habib

[John Larkin]

On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
<h.bouazi...@free.fr> wrote:

>Le 02/11/12 15:36, Robert Macy a écrit :

I've done revenue-quality energy meters with as few as 7 bits of ADC.
10 or 12 is plenty. We did one nice meter using an HC05 microprocessor
and its on-chip ADC. The trick is to add some noise to the current
signals to smear the bits, and do the downstream algorithms properly.

But nowadays, several people make metering chips that do all the work
for you.

[upsid...@downunder.com]

On Fri, 02 Nov 2012 08:46:53 -0700, John Larkin
<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
><h.bouazi...@free.fr> wrote:
>

>>Le 02/11/12 15:36, Robert Macy a �crit :

While adding dithering noise, oversampling and postprocessing low pass
filtering will certainly give you some extra bits, at least if the
actual ADC is monotonous, I am a bit suspicious of the accuracy of
that meter (I am not claiming that it would be worse than traditional
electromechanical kWh meters).

The situation would be easy if only resistive loads existed, but in
the real world, the voltage waveform contains a lot of distortion,
current drawn by a non-PFC electronics loads (such as "energy saving"
lamps) complicate the situation further.

However, if it can be assumed that the load remains constant for
several cycles (e.g. 1 s) averaging will help and less bits are needed
in the ADC.

[Tim Wescott]

Except that the accuracy is in percent.

If that's percent of full range, then it's 0.2% overall -- ah ha! This
is probably where the 1:500 comes in.

If what the OP really means is that he needs to know the input to one
part in 500, then he needs an ADC with an effective number of bits (ENOB)
of 500, or a 9-bit ADC.

In truth, the OP needs to look at his candidate ADCs various error
specifications to figure out which one will work. All he knows from the
above estimate that it needs to be nine or more.

(And note: there are plenty of ADCs out there that have eight good bits
and four more for marketing: bit count Does Not Mean that the accuracy
will be sufficient).

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com

[rickman]

To know how many bits you need to use in the converter, you have to
define if you want to calculate the power to 0.1% or if you want to
measure the voltage and the current to 0.1% and you need to know if the
0.1% accuracy is applied to the full range of measurement or if it is
relative to the largest value measured.

If you need to calculate the power to 0.1%, then you need each
measurement to 0.05% since the percentage errors will add as you
multiply to get power. If you want to know the total energy consumed to
0.1% you also need to be accurate over the full range since a given user
may only consume at lower levels.

So the lsb of your converter must be no larger than 0.05% of your lowest
current measurement in addition to the 500:1 current range or 1 part in
1,000,000. (excuse my use of periods and commas being different from
yours.)

This would imply a 20 bit converter for the current, but don't assume a
converter with 24 bits is accurate to 24 bits. There are mostly two
types of converters with 24 bit results, slower ones for scales and
other measurements at rather low sample rates which I don't think you
can use. Then there are audio type converters which will sample at
higher rates, but are specified for AC performance mostly. The useful
number of bits (ENOB - effective number of bits) is sometimes given, but
usually they spec SNR which is a similar number in dB. You will get 1
bit for each 6.02 dB of SNR minus a small factor to account for
quantization noise (~1.5 dB IIRC).

So 24 bits may be good enough, but make sure you have at least 122 dB of
SNR ((20 * 6.02) + 1.5). This is not so easy in a small, low power chip.

How fast do you plan to sample? You will get some improvement in SNR by
averaging multiple measurements, either over a cycle or over multiple
cycles of the power line. Random errors will average out (or more
"accurately" they average down since they never go away). "Accuracy" is
also impacted by systematic errors which won't average out.

One thing that bugs me is that they measure the current of the meter so
you pay to power the durn measuring device! I would expect the power
company to treat that as part of it's own operating costs...

Rick

[John Larkin]

On Fri, 02 Nov 2012 21:45:08 +0200, upsid...@downunder.com wrote:

>On Fri, 02 Nov 2012 08:46:53 -0700, John Larkin
><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>
>>On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
>><h.bouazi...@free.fr> wrote:
>>

>>>Le 02/11/12 15:36, Robert Macy a écrit :

The mechanical disk meters are actually amazingly good, not easy to
match with electronics.

>
>The situation would be easy if only resistive loads existed, but in
>the real world, the voltage waveform contains a lot of distortion,
>current drawn by a non-PFC electronics loads (such as "energy saving"
>lamps) complicate the situation further.

If you simultaneous-sample (or nearly ss) the voltage and current
waveforms, and multiply-integrate, you get true power regardless of
waveform, as long as nothing clips.

Voltage waveforms aren't bad and don't need dithering. Current is
nasty, ugly waveforms and a huge, like 20,000:1, dynamic range.

ADC channel crosstalk is one gotcha that has to be managed. Magnetic
coupling, too.

>
>However, if it can be assumed that the load remains constant for
>several cycles (e.g. 1 s) averaging will help and less bits are needed
>in the ADC.

Energy is a long-term integral, millions or billions of samples in a
billing period, so that works out. Even damand metering seldom
averages over less than about 5 minutes, still a lot of samples.




--

John Larkin Highland Technology, Inc

jlarkin at highlandtechnology dot com

http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom laser drivers and controllers

Photonics and fiberoptic TTL data links

VME thermocouple, LVDT, synchro acquisition and simulation

[Habib Bouaziz-Viallet]

Le 02/11/12 21:12, rickman a écrit :

> This would imply a 20 bit converter for the current, but don't assume a
> converter with 24 bits is accurate to 24 bits. There are mostly two
> types of converters with 24 bit results, slower ones for scales and
> other measurements at rather low sample rates which I don't think you
> can use. Then there are audio type converters which will sample at
> higher rates, but are specified for AC performance mostly. The useful
> number of bits (ENOB - effective number of bits) is sometimes given, but
> usually they spec SNR which is a similar number in dB. You will get 1
> bit for each 6.02 dB of SNR minus a small factor to account for
> quantization noise (~1.5 dB IIRC).

Yes most of ADC 24bits are nearly 18bits ENOB but they (AD, TI, Cirrus
...etc) claims that Gaussian noise gives the major part in the SINAD.
Gaussian noise on I and U would canceled themselves on the power
calculation (mean (discrete sum (U, I)) on 1s) because the two
acquisitions sequences are not correlated each other ... we will see
even if with GNU-Octave (awgn() ....) the results are something like
spectacular !

>
> So 24 bits may be good enough, but make sure you have at least 122 dB of
> SNR ((20 * 6.02) + 1.5). This is not so easy in a small, low power chip.
>
> How fast do you plan to sample? You will get some improvement in SNR by
> averaging multiple measurements, either over a cycle or over multiple
> cycles of the power line. Random errors will average out (or more
> "accurately" they average down since they never go away). "Accuracy" is
> also impacted by systematic errors which won't average out.
>

We plan to get 8Ksamples/s (out of the FIR filter after the Delta-Sigma
modulator of the ADC)

> One thing that bugs me is that they measure the current of the meter so
> you pay to power the durn measuring device! I would expect the power
> company to treat that as part of it's own operating costs...

i do the hacks they pay for ... like everyone else ! the most important
is the next holidays in Greece or elsewhere with my wife !

Habib

[Habib Bouaziz-Viallet]

Le 02/11/12 16:46, John Larkin a �crit :

> On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
> <h.bouazi...@free.fr> wrote:
>

>> Le 02/11/12 15:36, Robert Macy a �crit :

Pardon me John but i really doubt that you could achieve to implement
any power meter with even a real 12 bits ADC over the 1:500 range of
current, no offense John.

> But nowadays, several people make metering chips that do all the work
> for you.
>
>

Hey John, some guys are paying me to implement a good IEC62053-21 class
0.2S power meter ... they will whip me out if i repeat you words ...
Please John don't try to ruin my next Holidays !

Habib

[rickman]

On 11/2/2012 7:30 PM, Habib Bouaziz-Viallet wrote:
> Le 02/11/12 21:12, rickman a �crit :

>
>
>> This would imply a 20 bit converter for the current, but don't assume a
>> converter with 24 bits is accurate to 24 bits. There are mostly two
>> types of converters with 24 bit results, slower ones for scales and
>> other measurements at rather low sample rates which I don't think you
>> can use. Then there are audio type converters which will sample at
>> higher rates, but are specified for AC performance mostly. The useful
>> number of bits (ENOB - effective number of bits) is sometimes given, but
>> usually they spec SNR which is a similar number in dB. You will get 1
>> bit for each 6.02 dB of SNR minus a small factor to account for
>> quantization noise (~1.5 dB IIRC).
>
> Yes most of ADC 24bits are nearly 18bits ENOB but they (AD, TI, Cirrus

> ....etc) claims that Gaussian noise gives the major part in the SINAD.

> Gaussian noise on I and U would canceled themselves on the power
> calculation (mean (discrete sum (U, I)) on 1s) because the two
> acquisitions sequences are not correlated each other ... we will see
> even if with GNU-Octave (awgn() ....) the results are something like
> spectacular !

To get good accuracy you also need the two converters to be very well
synchronized. You can do the calculations to see just *how* synchronized.

Where are you from that you use U for voltage? I haven't seen that
before. I thought this was something standardized by SI? I also saw
you use In and Un. What does that mean?

>> So 24 bits may be good enough, but make sure you have at least 122 dB of
>> SNR ((20 * 6.02) + 1.5). This is not so easy in a small, low power chip.
>>
>> How fast do you plan to sample? You will get some improvement in SNR by
>> averaging multiple measurements, either over a cycle or over multiple
>> cycles of the power line. Random errors will average out (or more
>> "accurately" they average down since they never go away). "Accuracy" is
>> also impacted by systematic errors which won't average out.
>>
> We plan to get 8Ksamples/s (out of the FIR filter after the Delta-Sigma
> modulator of the ADC)

How will you verify the accuracy of the meter? Using a simple resistive
load is pretty easy, but how can you verify its accuracy in a real world
situation with spikes and inductive loads?

Rick

[John Larkin]

On Sat, 03 Nov 2012 00:40:56 +0100, Habib Bouaziz-Viallet
<h.bouazi...@free.fr> wrote:

>Le 02/11/12 16:46, John Larkin a écrit :

>> On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
>> <h.bouazi...@free.fr> wrote:
>>

>>> Le 02/11/12 15:36, Robert Macy a écrit :

But I've done it many times. With ADCs of 7 bits, 8 bits, 12 bits, and
once 16 bits.

This works:

http://www.highlandtechnology.com/DSS/V180DS.shtml

It uses a 16 bit ADC, because it can do DC measurements too. You don't
need so many bits for AC, because you can dither and average.

I've done meters that passed the ANSI C12 meter requirements using a 7
bit single-slope ADC.

>> But nowadays, several people make metering chips that do all the work
>> for you.
>>
>>
>Hey John, some guys are paying me to implement a good IEC62053-21 class
>0.2S power meter ... they will whip me out if i repeat you words ...
>Please John don't try to ruin my next Holidays !
>
>Habib

Oh. Sorry.

What sort of current sensor do you plan to use? That's the nasty part.

--

John Larkin Highland Technology, Inc

jlarkin at highlandtechnology dot com

http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom laser drivers and controllers

Photonics and fiberoptic TTL data links

[John Larkin]

You don't need a high sample rate for power metering. The Nyquist
criterion isn't relevant. I designed the C180, a nice 16-channel meter
that sampled at 27 Hz per voltage:current pair; sold thousands of them
for utility end-use studies.

Hey, they are on ebay now!

http://www.ebay.com/itm/SYNERGISTIC-E180E-CONTROL-PORTABLE-SURVEY-METER-C180-XP-/250636684548

http://btus.lbl.gov/HPCBS/Element_5/Deliverables/E5P2_2T3a.pdf

[John Larkin]

On Fri, 02 Nov 2012 20:28:04 -0400, rickman <gnu...@gmail.com> wrote:

>On 11/2/2012 7:30 PM, Habib Bouaziz-Viallet wrote:

>> Le 02/11/12 21:12, rickman a écrit :

>>
>>
>>> This would imply a 20 bit converter for the current, but don't assume a
>>> converter with 24 bits is accurate to 24 bits. There are mostly two
>>> types of converters with 24 bit results, slower ones for scales and
>>> other measurements at rather low sample rates which I don't think you
>>> can use. Then there are audio type converters which will sample at
>>> higher rates, but are specified for AC performance mostly. The useful
>>> number of bits (ENOB - effective number of bits) is sometimes given, but
>>> usually they spec SNR which is a similar number in dB. You will get 1
>>> bit for each 6.02 dB of SNR minus a small factor to account for
>>> quantization noise (~1.5 dB IIRC).
>>
>> Yes most of ADC 24bits are nearly 18bits ENOB but they (AD, TI, Cirrus
>> ....etc) claims that Gaussian noise gives the major part in the SINAD.
>> Gaussian noise on I and U would canceled themselves on the power
>> calculation (mean (discrete sum (U, I)) on 1s) because the two
>> acquisitions sequences are not correlated each other ... we will see
>> even if with GNU-Octave (awgn() ....) the results are something like
>> spectacular !
>
>To get good accuracy you also need the two converters to be very well
>synchronized. You can do the calculations to see just *how* synchronized.

A reasonable ADC could sample current then voltage (in that order!)
maybe 10 usec apart. That's only about 0.2 degrees. Since the
front-end analog stuff probably includes some lowpass filtering, it's
easy to skew the filter time constants to take out that 0.2 degree
error.

[Robert Macy]

On Nov 2, 7:58 am, Habib Bouaziz-Viallet <h.bouazizvial...@free.fr>
wrote:
> Le 02/11/12 15:36, Robert Macy a écrit :

I don't recommend counting on dithering and such, better to just do
straight forward ADC. else certain waveforms will cause strange power
measurements. I highly recommend TI's ADC used in the soundcards. I
get 22.5+ bits out of them and sometimes better.

There is a company in South Africa I use to compare my power meter
design to, both cost and performance. They were good, low cost AND
accurate, really tough competitor, but memory is not so good and lost
the name with those pesky 8 hard drive crashes ALL in a two year
period.

From memory, we did the power meter esign with an analog multiplying
chip AND my current transformer was a different design so didn't take
any metal. I used air core, almost, just a small bit of core whose
permeability could be initially over 10 to 1 and when operating change
30 to 1 and still keep in spec.pretty forgiving current sensor.

Don't forget to put in 'exterior' field sensors to cause the meter's
reading to skyrocket if some one tries to use a magnet to disable your
meter. That's what we did to punish those who attempted to saturate
the current sensor core to lower the power reading. I'll bet they were
surprised the first time they tried that and had to pay more!

[Les Cargill]

The disk meters are really good. Very clever transducer, IMO.

> The situation would be easy if only resistive loads existed, but in
> the real world, the voltage waveform contains a lot of distortion,
> current drawn by a non-PFC electronics loads (such as "energy saving"
> lamps) complicate the situation further.
>

I believe the disk meters are more or less integrating current meters
( the movement is actually proportional to power, but voltage will
converge on a constant fairly quickly for a good network ).

That's not quite the same thing as "digital"/TOD meters, where you
basically pull a data stream from them. I don't recall what those
actually measure. I'd want to basically do a histogram
of the phase angle & amplitude myself, but I don't think
that's what they really do.

> However, if it can be assumed that the load remains constant for
> several cycles (e.g. 1 s) averaging will help and less bits are needed
> in the ADC.
>

Right - the ... momentum of the meter helps.

--
Les Cargill

[Jan Panteltje]

On a sunny day (Fri, 2 Nov 2012 18:57:15 -0700 (PDT)) it happened Robert Macy
<robert...@gmail.com> wrote in
<68d0718d-7dd8-4677...@l12g2000vbj.googlegroups.com>:

>Don't forget to put in 'exterior' field sensors to cause the meter's
>reading to skyrocket if some one tries to use a magnet to disable your
>meter. That's what we did to punish those who attempted to saturate
>the current sensor core to lower the power reading. I'll bet they were
>surprised the first time they tried that and had to pay more!

That could be illegal, as somebody may put something magnetic on the meter
without any bad intent.

[Habib Bouaziz-Viallet]

Le 03/11/2012 01:33, John Larkin a écrit :
> This works:
>
> http://www.highlandtechnology.com/DSS/V180DS.shtml
I like VMEBus, i remember i have been designing VME IO's cards interface
with VME a controller named VIC (Cypress) ... it was about 20 years ago !

>
> It uses a 16 bit ADC, because it can do DC measurements too. You don't
> need so many bits for AC, because you can dither and average.
>

Oh yes with a re&l 16bits we can expect the IEC tests benches.

> I've done meters that passed the ANSI C12 meter requirements using a 7
> bit single-slope ADC.
>

> What sort of current sensor do you plan to use? That's the nasty part.

For currents it will be CT with a ratio 1:100 and furthermore we have to
design the interface to support Rogowski Transformers.
VT for voltages.

Accuracy is specified the interface only. A procedure to calirate and
zeroing the interface *with* the sensor (all over the range) will be
done by software.

Best Regards, Habib.

[upsid...@downunder.com]

On Fri, 02 Nov 2012 20:28:04 -0400, rickman <gnu...@gmail.com> wrote:

>Where are you from that you use U for voltage? I haven't seen that
>before. I thought this was something standardized by SI? I also saw
>you use In and Un. What does that mean?

I have only seen V used for voltage in anglo-saxon literature, in many
other cultures (notably German) , U has been used.

n = nominal

[Habib Bouaziz-Viallet]

Le 03/11/2012 01:28, rickman a écrit :
> To get good accuracy you also need the two converters to be very well
> synchronized. You can do the calculations to see just *how* synchronized.

Right. U and must be synchronized of course (each sync signal every 1s)
in order to re-synchronize the two ADC's)! Remember i have to compute
Power/Energy. each U, I sampled *synchronously* and make the
sum/mean/square/product ... over 1s.

>
> Where are you from that you use U for voltage? I haven't seen that
> before. I thought this was something standardized by SI? I also saw
> you use In and Un. What does that mean?

I live in France. U (or sometimes V) is for Voltages. Un stands for V
rated and In stands for I rated. Excuse my English ...

> How will you verify the accuracy of the meter? Using a simple resistive
> load is pretty easy, but how can you verify its accuracy in a real world
> situation with spikes and inductive loads?

Good point. The bench for IEC6.... -21 conformance will be done by an
independent agency. Some sort of test equipement (PPS400.3 and PRS ...)
will be used. I don't know exactly how and what these guys will handle
any test protocol :-(

Best Regards, Habib.

[upsid...@downunder.com]

On Sat, 03 Nov 2012 00:30:39 +0100, Habib Bouaziz-Viallet
<h.bouazi...@free.fr> wrote:

>> So 24 bits may be good enough, but make sure you have at least 122 dB of
>> SNR ((20 * 6.02) + 1.5). This is not so easy in a small, low power chip.
>>
>> How fast do you plan to sample? You will get some improvement in SNR by
>> averaging multiple measurements, either over a cycle or over multiple
>> cycles of the power line. Random errors will average out (or more
>> "accurately" they average down since they never go away). "Accuracy" is
>> also impacted by systematic errors which won't average out.
>>
>We plan to get 8Ksamples/s (out of the FIR filter after the Delta-Sigma
>modulator of the ADC)

That would make 160 samples/cycle at 50 Hz, ie. a sample every 2.25
degrees of the mains cycle. The conduction angles can be quite short
in power supplies, so one might ask, does it make sense to use
sampling rates that are exact multiple of the mains frequency.

There has been a similar discussion about energy metering here in
sci.electronics.design a few months ago, so check groops.google.com

As a general plea to all posting information request to this group,
provide as much as possible information about the problem (limited by
any NDAs) and you will get much better responses immediately.

In the worst case, some requests look like "do my home works,
please":-) and the responses will reflect that.

[Andrzej Ekiert]

Dnia 03-11-2012 o 11:29:19 Habib Bouaziz-Viallet
<h.bouazi...@free.fr> napisał(a):

> Right. U and must be synchronized of course (each sync signal every 1s)
> in order to re-synchronize the two ADC's)! Remember i have to compute
> Power/Energy. each U, I sampled *synchronously* and make the
> sum/mean/square/product ... over 1s.

In practice, you will need to clock the two ADCs from the same source. You
can, however, get away using a single ADC for current and voltage. When
designing a 3 phase energy meter I used one ADC with a single
sample-and-hold and multiple inputs. The phase shift was corrected quite
easily with a linear phase FIR filter. Reached Class 0.5 (MID Class C)
accuracy on a 12-bit ADC (internal, on a micro) with 2 external gain
ranges on current inputs. For the old Class 0.1 (0.1% accuracy) just use a
pair of good dedicated 16-bit Sigma-Deltas and you'll be fine.

>> Where are you from that you use U for voltage? I haven't seen that
>> before. I thought this was something standardized by SI? I also saw
>> you use In and Un. What does that mean?
> I live in France. U (or sometimes V) is for Voltages.

... and it makes perfect sense, since V is the SI unit for voltage. V = 5V
looks weird, U = 5V is better.

> I don't know exactly how and what these guys will handle any test
> protocol :-(

The metering guys don't care about any digital ways of reading the
measurement data for calibration. All they want is a LED pulse output that
behaves exactly the same as the rotating plate in electromechanical
meters. All their test and calibration benches are configured for a pulse
output.

Regards,
Andrzej Ekiert
--
http://www.protronik.pl

[Robert Macy]

On Nov 3, 1:26 am, Jan Panteltje <pNaonStpealm...@yahoo.com> wrote:
> On a sunny day (Fri, 2 Nov 2012 18:57:15 -0700 (PDT)) it happened Robert Macy

> <robert.a.m...@gmail.com> wrote in
> <68d0718d-7dd8-4677-8f8f-86d7b81be...@l12g2000vbj.googlegroups.com>:

>
> >Don't forget to put in 'exterior' field sensors to cause the meter's
> >reading to skyrocket if some one tries to use a magnet to disable your
> >meter. That's what we did to punish those who attempted to saturate
> >the current sensor core to lower the power reading. I'll bet they were
> >surprised the first time they tried that and had to pay more!
>
> That could be illegal, as somebody may put something magnetic on the meter
> without any bad intent.

not in the countries they deploy these meters. The govt and industry
are all pwerful there.

[Robert Macy]

On Nov 3, 3:29 am, Habib Bouaziz-Viallet <h.bouazizvial...@free.fr>
wrote:
> ,,,snip...

> will be used. I don't know exactly how and what these guys will handle
> any test protocol :-(
>
> Best Regards, Habib.

Don't they test up to PF of 6? or is that 10? Digital has a it of
trouble there since we're starting to subtract to large numbers to
find the power. Those 'disk' meters have an incredibly good accuracy
curve thatconverges down to 'almost' zero.

[Jan Panteltje]

On a sunny day (Sat, 3 Nov 2012 06:04:32 -0700 (PDT)) it happened Robert Macy
<robert...@gmail.com> wrote in
<1d7c9479-95ee-4b0a...@y8g2000yqy.googlegroups.com>:

>On Nov 3, 1:26 am, Jan Panteltje <pNaonStpealm...@yahoo.com> wrote:

>> On a sunny day (Fri, 2 Nov 2012 18:57:15 -0700 (PDT)) it happened Robert =

>Macy
>> <robert.a.m...@gmail.com> wrote in
>> <68d0718d-7dd8-4677-8f8f-86d7b81be...@l12g2000vbj.googlegroups.com>:
>>
>> >Don't forget to put in 'exterior' field sensors to cause the meter's
>> >reading to skyrocket if some one tries to use a magnet to disable your
>> >meter. That's what we did to punish those who attempted to saturate
>> >the current sensor core to lower the power reading. I'll bet they were
>> >surprised the first time they tried that and had to pay more!
>>

>> That could be illegal, as somebody may put something magnetic on the mete=

>r
>> without any bad intent.
>
>not in the countries they deploy these meters. The govt and industry
>are all pwerful there.

Strange, I have one, not sure about that gov thing and industry,
that is the same here, but if it was to show wrong values I could sue.
Would be easy to demonstrate, hook up a 1kW space heater,
readout, put magnetic object on meter, if it speeds up,
you win the case always.
Could be a magnet as placeholder for one of those 'safety guides'
you are supposed to hang next to the meter.

[John Larkin]

Most CTs, especially cheap ones, have more phase shift, up to a few
degrees, at low currents. That can be fudged in software, by blending
in a bit of saved quadrature current samples at lower average
currents. *IF* the CT is known and characterized.

--

John Larkin Highland Technology Inc

www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom timing and laser controllers

Photonics and fiberoptic TTL data links

[John Larkin]

On Sat, 03 Nov 2012 11:29:19 +0100, Habib Bouaziz-Viallet
<h.bouazi...@free.fr> wrote:

>Le 03/11/2012 01:28, rickman a écrit :
>> To get good accuracy you also need the two converters to be very well
>> synchronized. You can do the calculations to see just *how* synchronized.
>Right. U and must be synchronized of course (each sync signal every 1s)
>in order to re-synchronize the two ADC's)! Remember i have to compute
>Power/Energy. each U, I sampled *synchronously* and make the
>sum/mean/square/product ... over 1s.

Why not just trigger two ADCs simultaneously?

There's no need to synchronize the ADC trigger times to the AC line.
That falls out of the math.

[John Larkin]

On Sat, 3 Nov 2012 06:07:39 -0700 (PDT), Robert Macy
<robert...@gmail.com> wrote:

>On Nov 3, 3:29 am, Habib Bouaziz-Viallet <h.bouazizvial...@free.fr>
>wrote:
>> ,,,snip...
>> will be used. I don't know exactly how and what these guys will handle
>> any test protocol :-(
>>
>> Best Regards, Habib.
>
>Don't they test up to PF of 6? or is that 10?

PF is always in the range of -1 to +1.

Digital has a it of
>trouble there since we're starting to subtract to large numbers to
>find the power. Those 'disk' meters have an incredibly good accuracy
>curve thatconverges down to 'almost' zero.

A digital power meter just simultaneous-samples the voltage and
current signals, multiplies, and integrates. That works for all
waveforms and all power factors. No trig, no subtraction required.

[John Larkin]

On Sat, 03 Nov 2012 12:43:44 +0200, upsid...@downunder.com wrote:

>On Sat, 03 Nov 2012 00:30:39 +0100, Habib Bouaziz-Viallet
><h.bouazi...@free.fr> wrote:
>
>>> So 24 bits may be good enough, but make sure you have at least 122 dB of
>>> SNR ((20 * 6.02) + 1.5). This is not so easy in a small, low power chip.
>>>
>>> How fast do you plan to sample? You will get some improvement in SNR by
>>> averaging multiple measurements, either over a cycle or over multiple
>>> cycles of the power line. Random errors will average out (or more
>>> "accurately" they average down since they never go away). "Accuracy" is
>>> also impacted by systematic errors which won't average out.
>>>
>>We plan to get 8Ksamples/s (out of the FIR filter after the Delta-Sigma
>>modulator of the ADC)
>
>That would make 160 samples/cycle at 50 Hz, ie. a sample every 2.25
>degrees of the mains cycle. The conduction angles can be quite short
>in power supplies, so one might ask, does it make sense to use
>sampling rates that are exact multiple of the mains frequency.

It makes sense to "walk" the waveform with N samples uniformly spaced
per cycle. But that doesn't require a high sample rate; one can
equivalent-time sample, like a sampling oscilloscope. For example, in
a 50 Hz system, the line period is 20 milliseconds. You could sample
every 20.1 milliseconds, which is 361.8 degrees, which is equivalent
to sampling every 1.8 degrees. Every 200 samples, every 4.2 seconds,
you will have tested all 1.8 degree spots on the line waveform.

If the voltage waveform is sinusoidal, there is no power in the
current harmonics, so a high sample rate isn't needed to measure power
and energy.

[John Larkin]

With a 16 bit or better ADC, there will be enough noise around that
dithering isn't necessary.

else certain waveforms will cause strange power
>measurements.

I don't understand that. Dithering the current signal just smears out
the ADC quantization. It does show up in the RMS current value, if you
compute that off to the side, but it will be a few ADC LSBs, not a big
deal.

[Habib Bouaziz-Viallet]

Le 03/11/2012 17:13, John Larkin a �crit :

> On Sat, 03 Nov 2012 11:29:19 +0100, Habib Bouaziz-Viallet
> <h.bouazi...@free.fr> wrote:
>

>> Le 03/11/2012 01:28, rickman a �crit :

>>> To get good accuracy you also need the two converters to be very well
>>> synchronized. You can do the calculations to see just *how* synchronized.
>> Right. U and must be synchronized of course (each sync signal every 1s)
>> in order to re-synchronize the two ADC's)! Remember i have to compute
>> Power/Energy. each U, I sampled *synchronously* and make the
>> sum/mean/square/product ... over 1s.
> ;
> Why not just trigger two ADCs simultaneously?

In fact there are two type of daughter board, one for currents (CT,
Rogowski) and the other for voltage. I have to synchronize the
delta-sigma modulators of each Delta-Sigma ADC's, 1 time per second to
avoid false measurements based on each local quartz drifting ... That's
the meaning of synchronizing.

H

[John Larkin]

On Sat, 03 Nov 2012 17:48:23 +0100, Habib Bouaziz-Viallet
<h.bouazi...@free.fr> wrote:

>Le 03/11/2012 17:13, John Larkin a écrit :
>> On Sat, 03 Nov 2012 11:29:19 +0100, Habib Bouaziz-Viallet
>> <h.bouazi...@free.fr> wrote:
>>

>>> Le 03/11/2012 01:28, rickman a écrit :
>>>> To get good accuracy you also need the two converters to be very well
>>>> synchronized. You can do the calculations to see just *how* synchronized.
>>> Right. U and must be synchronized of course (each sync signal every 1s)
>>> in order to re-synchronize the two ADC's)! Remember i have to compute
>>> Power/Energy. each U, I sampled *synchronously* and make the
>>> sum/mean/square/product ... over 1s.
>> ;
>> Why not just trigger two ADCs simultaneously?
>In fact there are two type of daughter board, one for currents (CT,
>Rogowski) and the other for voltage. I have to synchronize the
>delta-sigma modulators of each Delta-Sigma ADC's, 1 time per second to
>avoid false measurements based on each local quartz drifting ... That's
>the meaning of synchronizing.
>
>H

Does each ADC have its own oscillator? You'd need to have your two
oscillators match to something like 5 PPM!

It would be simpler to use, say, two 16 bit SAR ADCs and just trigger
them simultaneously. There are nice SAR ADCs with SPI interfaces that
sample and digitize as you read them out, with no explicit trigger
needed. ADS7866 sorts of things. No oscillator needed, no phase drift
problem.

Why separate the ADCs? Just run the analog signals from the acq boards
to one location, and digitize there.

[rickman]

What the hell does "nominal" mean? When I last looked it up it meant
"in name only", or in other words, AC power line might be 120V *nominal*
since the actual voltage can vary hugely over time.

How would you make use of "nominal" voltage and current in a measuring
device? I can already tell you what that power level is, 120V * your
current!

As to the issue of V being anglo-saxon, I thought V for voltage was SI,
an *international* system of units which define not only the units and
how they are measured, but also the nomenclature which includes the
abbreviations. Don't you use SI?

Rick

[rickman]

On 11/3/2012 6:29 AM, Habib Bouaziz-Viallet wrote:
> Le 03/11/2012 01:28, rickman a �crit :

>> To get good accuracy you also need the two converters to be very well
>> synchronized. You can do the calculations to see just *how* synchronized.
> Right. U and must be synchronized of course (each sync signal every 1s)
> in order to re-synchronize the two ADC's)! Remember i have to compute
> Power/Energy. each U, I sampled *synchronously* and make the
> sum/mean/square/product ... over 1s.
>>
>> Where are you from that you use U for voltage? I haven't seen that
>> before. I thought this was something standardized by SI? I also saw
>> you use In and Un. What does that mean?
> I live in France. U (or sometimes V) is for Voltages. Un stands for V
> rated and In stands for I rated. Excuse my English ...

Interesting. The name, "SI" is actually the abbreviation in French for
international system. I believe the meetings and some of the actual
international standards (like the old meter bar) were kept in France, so
much of the language for the system came from French.

So how do the French come up with U for voltage when the SI unit is V?

What exactly does "rated" mean? Is that what you call your "measured"
value, the sample? Oh, is the 'n' the index, like V1, V2, V3,... Vn?

You don't need to apologize for your English. It is tons better than my
French. That's one bad thing about living in the US. Other than
Spanish which is becoming more common here, we don't get exposed to many
languages other than English.

>> How will you verify the accuracy of the meter? Using a simple resistive
>> load is pretty easy, but how can you verify its accuracy in a real world
>> situation with spikes and inductive loads?
> Good point. The bench for IEC6.... -21 conformance will be done by an
> independent agency. Some sort of test equipement (PPS400.3 and PRS ...)
> will be used. I don't know exactly how and what these guys will handle
> any test protocol :-(

Yeah, I'm sure this is a very complex area with all sorts of good and
expensive test equipment. But at some point they have to verify that it
all works correctly under the crazy conditions found on real circuits.

Rick

[rickman]

On 11/3/2012 7:42 AM, Andrzej Ekiert wrote:
> Dnia 03-11-2012 o 11:29:19 Habib Bouaziz-Viallet
> <h.bouazi...@free.fr> napisał(a):
>
>> Right. U and must be synchronized of course (each sync signal every
>> 1s) in order to re-synchronize the two ADC's)! Remember i have to
>> compute Power/Energy. each U, I sampled *synchronously* and make the
>> sum/mean/square/product ... over 1s.
>
> In practice, you will need to clock the two ADCs from the same source.
> You can, however, get away using a single ADC for current and voltage.
> When designing a 3 phase energy meter I used one ADC with a single
> sample-and-hold and multiple inputs. The phase shift was corrected quite
> easily with a linear phase FIR filter. Reached Class 0.5 (MID Class C)
> accuracy on a 12-bit ADC (internal, on a micro) with 2 external gain
> ranges on current inputs. For the old Class 0.1 (0.1% accuracy) just use
> a pair of good dedicated 16-bit Sigma-Deltas and you'll be fine.
>
>>> Where are you from that you use U for voltage? I haven't seen that
>>> before. I thought this was something standardized by SI? I also saw
>>> you use In and Un. What does that mean?
>> I live in France. U (or sometimes V) is for Voltages.
>
> ... and it makes perfect sense, since V is the SI unit for voltage. V =
> 5V looks weird, U = 5V is better.

I didn't think of it that way. The SI V is the abbreviation for the
unit "volts", but the symbol V or U is a math symbol and not necessarily
the same and the units abbreviation. Duh!

Rick

[rickman]

On 11/3/2012 1:03 PM, John Larkin wrote:
> On Sat, 03 Nov 2012 17:48:23 +0100, Habib Bouaziz-Viallet
> <h.bouazi...@free.fr> wrote:
>

>> Le 03/11/2012 17:13, John Larkin a �crit :

>>> On Sat, 03 Nov 2012 11:29:19 +0100, Habib Bouaziz-Viallet
>>> <h.bouazi...@free.fr> wrote:
>>>

>>>> Le 03/11/2012 01:28, rickman a �crit :

>>>>> To get good accuracy you also need the two converters to be very well
>>>>> synchronized. You can do the calculations to see just *how* synchronized.
>>>> Right. U and must be synchronized of course (each sync signal every 1s)
>>>> in order to re-synchronize the two ADC's)! Remember i have to compute
>>>> Power/Energy. each U, I sampled *synchronously* and make the
>>>> sum/mean/square/product ... over 1s.
>>> ;
>>> Why not just trigger two ADCs simultaneously?
>> In fact there are two type of daughter board, one for currents (CT,
>> Rogowski) and the other for voltage. I have to synchronize the
>> delta-sigma modulators of each Delta-Sigma ADC's, 1 time per second to
>> avoid false measurements based on each local quartz drifting ... That's
>> the meaning of synchronizing.
>>
>> H
>
> Does each ADC have its own oscillator? You'd need to have your two
> oscillators match to something like 5 PPM!
>
> It would be simpler to use, say, two 16 bit SAR ADCs and just trigger
> them simultaneously. There are nice SAR ADCs with SPI interfaces that
> sample and digitize as you read them out, with no explicit trigger
> needed. ADS7866 sorts of things. No oscillator needed, no phase drift
> problem.
>
> Why separate the ADCs? Just run the analog signals from the acq boards
> to one location, and digitize there.

That is what he is saying, he synchronizes the sampling of the two
converters. He is not talking about synchronizing them to the power
signal.

Rick

[Habib Bouaziz-Viallet]

Le 03/11/2012 18:03, John Larkin a �crit :

> Does each ADC have its own oscillator? You'd need to have your two
> oscillators match to something like 5 PPM!

Yes correct, each ADC has it's own oscillator. No need such a 5ppm as
long as i resync the internal DS modulator, one time each second.

>
> It would be simpler to use, say, two 16 bit SAR ADCs and just trigger
> them simultaneously. There are nice SAR ADCs with SPI interfaces that
> sample and digitize as you read them out, with no explicit trigger
> needed. ADS7866 sorts of things. No oscillator needed, no phase drift
> problem.
>
> Why separate the ADCs? Just run the analog signals from the acq boards
> to one location, and digitize there.

We have specified some insulation groups and among them Currents and
Voltages are two independent goups with 2KV of dieclectric strenght.
That is the reason of we could not run analog signals on one and only
one ADC for Current and voltage simultaneously. I admit it's an weird
thing ... but guys are insisted heavily on the point ...

Remember my holidays are a non negligible part of the project ... ;-)

H

[Habib Bouaziz-Viallet]

Le 03/11/2012 18:32, rickman a écrit :
> But at some point they have to verify that it all works correctly under
> the crazy conditions found on real circuits.

As usual,

H.

[Andrzej Ekiert]

On Sat, 03 Nov 2012 18:23:25 +0100, rickman <gnu...@gmail.com> wrote:

> What the hell does "nominal" mean? When I last looked it up it meant
> "in name only", or in other words, AC power line might be 120V *nominal*
> since the actual voltage can vary hugely over time.
>
> How would you make use of "nominal" voltage and current in a measuring
> device? I can already tell you what that power level is, 120V * your
> current!

Actually, that 'n' does not stand for "nominal" any more. Energy meter IEC
standards define all these terms:
In - rated current / courant assignée (for meters connected via current
transformers)
Ib - basic current / courant de base (for directly connected meters, can
also be CT based!)
Un - reference voltage / tension de référence

These are just "nominal" values used when specifying various parameters of
the meter, such as ranges of acceptable errors. For example, a Class 1
meter must have an active energy error below 1.5% at PF = 1 in the current
range from 0.05*Ib to 0.1*Ib, etc.

Currently in Europe we're using the EU MID directive instead of old IEC
standards, but the terminology stays mostly the same.

ae
--
http://www.protronik.pl

[John Larkin]

On Sat, 03 Nov 2012 13:38:37 -0400, rickman <gnu...@gmail.com> wrote:

>On 11/3/2012 1:03 PM, John Larkin wrote:
>> On Sat, 03 Nov 2012 17:48:23 +0100, Habib Bouaziz-Viallet
>> <h.bouazi...@free.fr> wrote:
>>

I sounds like they are clocked delta-sigmas with independent crystal
oscillator clocks, and they need to be resynchronized once a second to
correct for clock drift. If that's so, it seems like a weird
architecture.

[rickman]

So what did the OP really mean when he wrote "I range (0,02In .. In,
10 In)), U range (0,8Un .. Un .. 1,5Un)"?

Actually I think this is mixing things a bit. U may be fine for the
math symbol, but when he is saying "0,8Un" shouldn't that be Un = 0,8V
or something like that? But clearly he is not talking about a voltage
range of of 0.8 to 1.5 volts. At this point I have no idea what he is
describing here. Is he talking about a range relative to the full scale
of the unit?

That makes more sense and I understand why "nominal" is used. That is
just what it should mean and I didn't get that.

Rick

[rickman]

On 11/3/2012 3:08 PM, John Larkin wrote:
> On Sat, 03 Nov 2012 13:38:37 -0400, rickman<gnu...@gmail.com> wrote:
>
>> On 11/3/2012 1:03 PM, John Larkin wrote:
>>> On Sat, 03 Nov 2012 17:48:23 +0100, Habib Bouaziz-Viallet
>>> <h.bouazi...@free.fr> wrote:
>>>

>>>> Le 03/11/2012 17:13, John Larkin a �crit :

>>>>> On Sat, 03 Nov 2012 11:29:19 +0100, Habib Bouaziz-Viallet
>>>>> <h.bouazi...@free.fr> wrote:
>>>>>

>>>>>> Le 03/11/2012 01:28, rickman a �crit :

I don't see how you came to that conclusion. I doesn't really matter.

Rick

[John S]

On 11/2/2012 3:12 PM, rickman wrote:

> One thing that bugs me is that they measure the current of the meter so
> you pay to power the durn measuring device! I would expect the power
> company to treat that as part of it's own operating costs...
>
> Rick

Either way, you pay for it.

[Andrzej Ekiert]

Dnia 03-11-2012 o 20:26:38 rickman <gnu...@gmail.com> napisał(a):

>
> So what did the OP really mean when he wrote "I range (0,02In .. In,
> 10 In)), U range (0,8Un .. Un .. 1,5Un)"?
>
> Actually I think this is mixing things a bit. U may be fine for the
> math symbol, but when he is saying "0,8Un" shouldn't that be Un = 0,8V
> or something like that? But clearly he is not talking about a voltage
> range of of 0.8 to 1.5 volts. At this point I have no idea what he is
> describing here. Is he talking about a range relative to the full scale
> of the unit?

Range relative to the "nominal" values of the unit. e.g.:
Un = 230V, 0.8Un = 184V, 1.5Un = 345V
In = 5A, 0.02In = 100mA, 10In = 50A

ae

[Andrzej Ekiert]

Dnia 03-11-2012 o 17:10:54 John Larkin
<jjla...@highnotlandthistechnologypart.com> napisał(a):

> Most CTs, especially cheap ones, have more phase shift, up to a few
> degrees, at low currents. That can be fudged in software, by blending
> in a bit of saved quadrature current samples at lower average
> currents. *IF* the CT is known and characterized.

Mine were pretty linear, but for high accuracy meters one needs to take
into account the temperature-phase characteristic. Self-heating of the
wire that goes through the CT can heat up the CT and change its phase
shift.

ae

[Tim Williams]

"rickman" <gnu...@gmail.com> wrote in message
news:k73kln$594$1...@dont-email.me...

> I didn't think of it that way. The SI V is the abbreviation for the
> unit "volts", but the symbol V or U is a math symbol and not necessarily
> the same and the units abbreviation. Duh!

'U' is traditionally used by physicists to denote potential.

Using it also for electrical potential, I believe, is more of a cultural
thing. E, V and U have been variously used to represent voltage.

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

[Robert Macy]

On Nov 3, 9:18 am, John Larkin

<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sat, 3 Nov 2012 06:07:39 -0700 (PDT), Robert Macy
>

> <robert.a.m...@gmail.com> wrote:
> >On Nov 3, 3:29 am, Habib Bouaziz-Viallet <h.bouazizvial...@free.fr>
> >wrote:
> >> ,,,snip...
> >> will be used. I don't know exactly how and what these guys will handle
> >> any test protocol :-(
>
> >> Best Regards, Habib.
>
> >Don't they test up to PF of 6? or is that 10?
>
> PF is always in the range of -1 to +1.
>
>  Digital has a it of
>
> >trouble there since we're starting to subtract to large numbers to
> >find the power. Those 'disk' meters have an incredibly good accuracy
> >curve thatconverges down to 'almost' zero.
>
> A digital power meter just simultaneous-samples the voltage and
> current signals, multiplies, and integrates. That works for all
> waveforms and all power factors. No trig, no subtraction required.
>
> --
>

> John Larkin                  Highland Technology Incwww.highlandtechnology.com  jlarkin at highlandtechnology dot com

>
> Precision electronic instrumentation
> Picosecond-resolution Digital Delay and Pulse generators
> Custom timing and laser controllers
> Photonics and fiberoptic TTL data links
> VME  analog, thermocouple, LVDT, synchro, tachometer
> Multichannel arbitrary waveform generators

maybe I'm thinkging 'crest factor'

[rickman]

On 11/3/2012 7:53 PM, Tim Williams wrote:
> "rickman"<gnu...@gmail.com> wrote in message
> news:k73kln$594$1...@dont-email.me...
>> I didn't think of it that way. The SI V is the abbreviation for the
>> unit "volts", but the symbol V or U is a math symbol and not necessarily
>> the same and the units abbreviation. Duh!
>
> 'U' is traditionally used by physicists to denote potential.
>
> Using it also for electrical potential, I believe, is more of a cultural
> thing. E, V and U have been variously used to represent voltage.
>
> Tim
>

Yes, I remember now using E a lot in college. I don't recall U so much,
but then I did undergraduate as a chemist where we were looking at work
function and chemical potentials. I mostly have forgotten more of that
than I remember.

I do still remember the difference between H2O and H2SO4...

Rick

[rickman]

So why do they want to replace the disk meters with digital ones? You
could add a VERY simple optical monitor to the disk meter to read the
number of revolutions and provide all the other benefits of digital
without the mess of the ADCs, etc. I suppose the old meters can be a
bit more expensive than the fancy, dancy digital ones?

Rick

[lang...@fonz.dk]

On 4 Nov., 00:53, "Tim Williams" <tmoran...@charter.net> wrote:
> "rickman" <gnu...@gmail.com> wrote in message
>
> news:k73kln$594$1...@dont-email.me...
>
> > I didn't think of it that way.  The SI V is the abbreviation for the
> > unit "volts", but the symbol V or U is a math symbol and not necessarily
> > the same and the units abbreviation.  Duh!
>
> 'U' is traditionally used by physicists to denote potential.
>
> Using it also for electrical potential, I believe, is more of a cultural
> thing.  E, V and U have been variously used to represent voltage.
>

don't remember seeing ohm's law written other U=I*R in all my school
years

current isn't written as A(mpere) or C(urrent) either but I

-Lasse

[George Herold]

On Nov 2, 9:00 pm, John Larkin <jlar...@highlandtechnology.com> wrote:
> On Fri, 02 Nov 2012 20:28:04 -0400, rickman <gnu...@gmail.com> wrote:
> >On 11/2/2012 7:30 PM, Habib Bouaziz-Viallet wrote:
> >> Le 02/11/12 21:12, rickman a crit :
>
> >>> This would imply a 20 bit converter for the current, but don't assume a
> >>> converter with 24 bits is accurate to 24 bits. There are mostly two
> >>> types of converters with 24 bit results, slower ones for scales and
> >>> other measurements at rather low sample rates which I don't think you
> >>> can use. Then there are audio type converters which will sample at
> >>> higher rates, but are specified for AC performance mostly. The useful
> >>> number of bits (ENOB - effective number of bits) is sometimes given, but
> >>> usually they spec SNR which is a similar number in dB. You will get 1
> >>> bit for each 6.02 dB of SNR minus a small factor to account for
> >>> quantization noise (~1.5 dB IIRC).
>
> >> Yes most of ADC 24bits are nearly 18bits ENOB but they (AD, TI, Cirrus
> >> ....etc) claims that Gaussian noise gives the major part in the SINAD.
> >> Gaussian noise on I and U would canceled themselves on the power
> >> calculation (mean (discrete sum (U, I)) on 1s) because the two
> >> acquisitions sequences are not correlated each other ... we will see
> >> even if with GNU-Octave (awgn() ....) the results are something like
> >> spectacular !

>
> >To get good accuracy you also need the two converters to be very well
> >synchronized.  You can do the calculations to see just *how* synchronized.
>

> A reasonable ADC could sample current then voltage (in that order!)
> maybe 10 usec apart. That's only about 0.2 degrees. Since the
> front-end analog stuff probably includes some lowpass filtering, it's
> easy to skew the filter time constants to take out that 0.2 degree
> error.

Cool, but how do you match the low passes?
(do you measure the C's?)

George H.

>
> --
>
> John Larkin         Highland Technology, Inc
>
> jlarkin at highlandtechnology dot comhttp://www.highlandtechnology.com

>
> Precision electronic instrumentation
> Picosecond-resolution Digital Delay and Pulse generators

> Custom laser drivers and controllers

> Photonics and fiberoptic TTL data links

> VME thermocouple, LVDT, synchro   acquisition and simulation- Hide quoted text -
>
> - Show quoted text -

[k...@att.bizzz]

On Fri, 02 Nov 2012 23:07:03 -0500, Les Cargill
<lcarg...@comcast.com> wrote:

>upsid...@downunder.com wrote:
>> On Fri, 02 Nov 2012 08:46:53 -0700, John Larkin
>> <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>
>>> On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
>>> <h.bouazi...@free.fr> wrote:
>>>
>>>> Le 02/11/12 15:36, Robert Macy a �crit :

>>>>> On Nov 2, 1:10 am, Habib Bouaziz-Viallet<h...@elsewhere.eu> wrote:
>>>>>> Hi all,
>>>>>>
>>>>>> Anyone could tell me what are the relations between Analog dynamic range
>>>>>> (figure with no dimension), accuracy over this range and number of bits ?
>>>>>> In other words, how to compute the number of bits of an ADC knowing the
>>>>>> analog range and the accuracy wanted over this range.
>>>>>>
>>>>>> Ex :
>>>>>> Analog range = 1:500
>>>>>> accuracy over this range : +-0.1%
>>>>>> number of bits = ?
>>>>>>
>>>>>> Best Regards, Habib
>>>>>
>>>>> Sounds more like a 'test' question, but here goes:
>>>>> if accuracy is +/- 1/1000 and the smallest range to largest implies
>>>>> 500X, 1/500000; which is a little more than 19 bits, so therefore you
>>>>> need AT LEAST 20 bits.
>>>>
>>>> In fact i'm far from a test question ... it's real life and i've got the
>>>> same approach.
>>>>

>>>> Energy Watt-Meter 0,1% accuracy, 230Volts/50Hz. I range (0,02In .. In,

>>>> 10 In)), U range (0,8Un .. Un .. 1,5Un)
>>>>

>>>> --> 24bits on ADC for currents measurements.
>>>>
>>>> Any objection from gurus of the analog ?
>>>>
>>>> Habib
>>>

>>> I've done revenue-quality energy meters with as few as 7 bits of ADC.
>>> 10 or 12 is plenty. We did one nice meter using an HC05 microprocessor
>>> and its on-chip ADC. The trick is to add some noise to the current
>>> signals to smear the bits, and do the downstream algorithms properly.
>>
>> While adding dithering noise, oversampling and postprocessing low pass
>> filtering will certainly give you some extra bits, at least if the
>> actual ADC is monotonous, I am a bit suspicious of the accuracy of
>> that meter (I am not claiming that it would be worse than traditional
>> electromechanical kWh meters).
>>
>
>The disk meters are really good. Very clever transducer, IMO.
>
>> The situation would be easy if only resistive loads existed, but in
>> the real world, the voltage waveform contains a lot of distortion,
>> current drawn by a non-PFC electronics loads (such as "energy saving"
>> lamps) complicate the situation further.
>>
>
>I believe the disk meters are more or less integrating current meters
>( the movement is actually proportional to power, but voltage will
>converge on a constant fairly quickly for a good network ).

No, you're ignoring PF. It is an amazing widget but it *is* an
integrating power meter, *not* a current meter.

>That's not quite the same thing as "digital"/TOD meters, where you
>basically pull a data stream from them. I don't recall what those
>actually measure. I'd want to basically do a histogram
>of the phase angle & amplitude myself, but I don't think
>that's what they really do.

Measure voltage and current and multiply. Why do you care about the
phase angle? It's meaningless if there are any uncorrected electronic
power supplies downstream.

>> However, if it can be assumed that the load remains constant for
>> several cycles (e.g. 1 s) averaging will help and less bits are needed
>> in the ADC.
>>
>
>Right - the ... momentum of the meter helps.

It comes out in the wash. There is very little mass there.

[Les Cargill]

Oops! Right!

>> That's not quite the same thing as "digital"/TOD meters, where you
>> basically pull a data stream from them. I don't recall what those
>> actually measure. I'd want to basically do a histogram
>> of the phase angle & amplitude myself, but I don't think
>> that's what they really do.
>
> Measure voltage and current and multiply. Why do you care about the
> phase angle? It's meaningless if there are any uncorrected electronic
> power supplies downstream.
>

Wouldn't phase angle give you power factor? Or is deviation of PF more
like distortion?

>>> However, if it can be assumed that the load remains constant for
>>> several cycles (e.g. 1 s) averaging will help and less bits are needed
>>> in the ADC.
>>>
>>
>> Right - the ... momentum of the meter helps.
>
> It comes out in the wash. There is very little mass there.
>

--
Les Cargill

[k...@att.bizzz]

Think harmonics (or ugly waveforms from a SMPS).

[John Larkin]

On Sun, 04 Nov 2012 10:11:39 -0600, Les Cargill
<lcarg...@comcast.com> wrote:

>k...@att.bizzz wrote:
>> On Fri, 02 Nov 2012 23:07:03 -0500, Les Cargill
>> <lcarg...@comcast.com> wrote:
>>
>>> upsid...@downunder.com wrote:
>>>> On Fri, 02 Nov 2012 08:46:53 -0700, John Larkin
>>>> <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>>>
>>>>> On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
>>>>> <h.bouazi...@free.fr> wrote:
>>>>>

>>>>>> Le 02/11/12 15:36, Robert Macy a écrit :

If you take simultaneous E and I samples at some sensible rate,

Auto-zero and then multiply the E and I samples. Lowpass filter to get
power, integrate to get energy.

Compute RMS current and voltage by squaring/filtering/rooting.

Multiply RMS current by RMS voltage to get VAs.

PF = power/VAs.

The only glitch here is that we lost the leading/lagging sign of the
PF. That can be fixed with some more signal processing.

--

John Larkin Highland Technology Inc

www.highlandtechnology.com jlarkin at highlandtechnology dot com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom timing and laser controllers

Photonics and fiberoptic TTL data links

[Les Cargill]

Ah - okay. For some reason, I thought power factor correction was mainly
about phase.

>>>>> However, if it can be assumed that the load remains constant for
>>>>> several cycles (e.g. 1 s) averaging will help and less bits are needed
>>>>> in the ADC.
>>>>>
>>>>
>>>> Right - the ... momentum of the meter helps.
>>>
>>> It comes out in the wash. There is very little mass there.
>>>

--
Les Cargill

[Les Cargill]

John Larkin wrote:
> On Sun, 04 Nov 2012 10:11:39 -0600, Les Cargill

<snip>

>> Wouldn't phase angle give you power factor? Or is deviation of PF more
>> like distortion?
>
>
> If you take simultaneous E and I samples at some sensible rate,
>
> Auto-zero and then multiply the E and I samples. Lowpass filter to get
> power, integrate to get energy.
>
> Compute RMS current and voltage by squaring/filtering/rooting.
>
> Multiply RMS current by RMS voltage to get VAs.
>
> PF = power/VAs.
>
> The only glitch here is that we lost the leading/lagging sign of the
> PF. That can be fixed with some more signal processing.
>
>

Interesting! Thanks, John. Sorry that was so basic. FWIW, I
defer to people who really know when power distro comes up - it's
not my field at all. I keep thinking it's mainly
impedance mismatch, when it's not really...

--
Les Cargill

[John S]

No, no, no! Think back about the definition of Power (which is the the
integral of V(t)*I(t) averaged over the period of interest. Then think
of apparent power as Irms*Vres (taken over the same period of interest).
Power factor is Power/(Irms*Vrms). These definitions can be used
anywhere at any time under any circumstances and are always correct.
Knowing phase is a special case of using the above on sine waves to make
it easier to calculate under those circumstances.

[John Larkin]

On Sun, 04 Nov 2012 14:06:56 -0600, John S <Sop...@invalid.org>
wrote:

PF is signed. Power/(Irms*Vrms) isn't.

[Jamie]

John S wrote:

> On 11/4/2012 12:26 PM, Les Cargill wrote:
>
>> k...@att.bizzz wrote:
>>
>>> On Sun, 04 Nov 2012 10:11:39 -0600, Les Cargill
>>> <lcarg...@comcast.com> wrote:
>>>
>>>> k...@att.bizzz wrote:
>>>>
>>>>> On Fri, 02 Nov 2012 23:07:03 -0500, Les Cargill
>>>>> <lcarg...@comcast.com> wrote:
>>>>>
>>>>>> upsid...@downunder.com wrote:
>>>>>>
>>>>>>> On Fri, 02 Nov 2012 08:46:53 -0700, John Larkin
>>>>>>> <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>>>>>>
>>>>>>>> On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
>>>>>>>> <h.bouazi...@free.fr> wrote:
>>>>>>>>

>>>>>>>>> Le 02/11/12 15:36, Robert Macy a écrit :

Oh my fucking god, you can't even get that correct...

Jamie

[rickman]

>>>>>>> of the phase angle& amplitude myself, but I don't think

>>>>>>> that's what they really do.
>>>>>>
>>>>>> Measure voltage and current and multiply. Why do you care about the
>>>>>> phase angle? It's meaningless if there are any uncorrected electronic
>>>>>> power supplies downstream.
>>>>>>
>>>>>
>>>>> Wouldn't phase angle give you power factor? Or is deviation of PF more
>>>>> like distortion?
>>>>
>>>> Think harmonics (or ugly waveforms from a SMPS).
>>>>
>>>
>>> Ah - okay. For some reason, I thought power factor correction was mainly
>>> about phase.
>>>
>>
>> No, no, no! Think back about the definition of Power (which is the the
>> integral of V(t)*I(t) averaged over the period of interest. Then think
>> of apparent power as Irms*Vres (taken over the same period of interest).
>> Power factor is Power/(Irms*Vrms). These definitions can be used
>> anywhere at any time under any circumstances and are always correct.
>> Knowing phase is a special case of using the above on sine waves to make
>> it easier to calculate under those circumstances.
>
> PF is signed. Power/(Irms*Vrms) isn't.

Why isn't Power/(Irms*Vrms) signed? What happens if the current is
flowing in the opposite direction to the voltage? The RMS value will be
negative and the above equation will be negative. That is the meaning
of a negative power factor, no?

Rick

[John Larkin]

On Sun, 04 Nov 2012 16:26:09 -0500, rickman <gnu...@gmail.com> wrote:

>On 11/4/2012 3:29 PM, John Larkin wrote:
>> On Sun, 04 Nov 2012 14:06:56 -0600, John S<Sop...@invalid.org>
>> wrote:
>>
>>> On 11/4/2012 12:26 PM, Les Cargill wrote:
>>>> k...@att.bizzz wrote:
>>>>> On Sun, 04 Nov 2012 10:11:39 -0600, Les Cargill
>>>>> <lcarg...@comcast.com> wrote:
>>>>>
>>>>>> k...@att.bizzz wrote:
>>>>>>> On Fri, 02 Nov 2012 23:07:03 -0500, Les Cargill
>>>>>>> <lcarg...@comcast.com> wrote:
>>>>>>>
>>>>>>>> upsid...@downunder.com wrote:
>>>>>>>>> On Fri, 02 Nov 2012 08:46:53 -0700, John Larkin
>>>>>>>>> <jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>>>>>>>>
>>>>>>>>>> On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
>>>>>>>>>> <h.bouazi...@free.fr> wrote:
>>>>>>>>>>

>>>>>>>>>>> Le 02/11/12 15:36, Robert Macy a écrit :

No. Wikipedia defines PF as a dimensionless value from 0 to 1, but
hangs "leading" and "lagging" off to the side.

If voltage and current are in phase, PF is 1 and the sign is
indeterminate. If current phase leads voltage (capacitive load), by
even one degree, PF is called "leading" and if current waveform peaks
later, it's called "lagging." Capacitive (leading) loads are usually
assigned a negative PF if the word "leading" isn't used instead.
Whichever terminology is used, something often needs to be said about
whether the current leads or lags the voltage, and the simple
Power/(Irms*Vrms) equation loses that.

For nonlinear and time-variant loads (like burst-mode temperature
controllers) the exact meaning of PF is debatable.

[John Fields]

On Sun, 04 Nov 2012 12:29:36 -0800, John Larkin
<jjla...@highNOTlandTHIStechnologyPART.com> wrote:

>On Sun, 04 Nov 2012 14:06:56 -0600, John S <Sop...@invalid.org>
>wrote:

>>

>>No, no, no! Think back about the definition of Power (which is the the
>>integral of V(t)*I(t) averaged over the period of interest. Then think
>>of apparent power as Irms*Vres (taken over the same period of interest).
>>Power factor is Power/(Irms*Vrms). These definitions can be used
>>anywhere at any time under any circumstances and are always correct.
>>Knowing phase is a special case of using the above on sine waves to make
>>it easier to calculate under those circumstances.
>
>PF is signed. Power/(Irms*Vrms) isn't.

---
What's your point?

--
JF

[k...@att.bizzz]

On Sun, 04 Nov 2012 12:26:44 -0600, Les Cargill

Just think of it as PF=W/VA It doesn't matter whether the reason is
phase offset of two sine waves or harmonic distortion.

[Jim Thompson]

Wikipedia seems to think PF is only 0 to 1...

http://en.wikipedia.org/wiki/Power_factor

...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

[rickman]

This is getting into the twilight zone territory. Near the top of this
post you are quoted saying, "PF is signed". Now you are saying it is
NOT signed, no?

The point is that power factor is a result of reactance in the load, or
non-linearities. When reactive, the current will be out of phase with
the voltage. Blah, blah... The power factor with a reactive load
tells you how much of the current is reactive and how much is
dissipative. But what if the load is actually shoving power back into
the line?

BTW, Wikipedia is not a primary source. In other words, I don't rely on
it for significant facts unless the references are consulted. Wikipedia
often gets facts *wrong*. Who was the reference they cited for the
power factor info?

Rick

[Jamie]

k...@att.bizzz wrote:

> On Sun, 04 Nov 2012 12:26:44 -0600, Les Cargill
> <lcarg...@comcast.com> wrote:
>
>
>>k...@att.bizzz wrote:
>>
>>>On Sun, 04 Nov 2012 10:11:39 -0600, Les Cargill
>>><lcarg...@comcast.com> wrote:
>>>
>>>
>>>>k...@att.bizzz wrote:
>>>>
>>>>>On Fri, 02 Nov 2012 23:07:03 -0500, Les Cargill
>>>>><lcarg...@comcast.com> wrote:
>>>>>
>>>>>
>>>>>>upsid...@downunder.com wrote:
>>>>>>
>>>>>>>On Fri, 02 Nov 2012 08:46:53 -0700, John Larkin
>>>>>>><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>>>>>>>
>>>>>>>
>>>>>>>>On Fri, 02 Nov 2012 15:58:18 +0100, Habib Bouaziz-Viallet
>>>>>>>><h.bouazi...@free.fr> wrote:
>>>>>>>>
>>>>>>>>

>>>>>>>>>Le 02/11/12 15:36, Robert Macy a écrit :

Some one finally hits the nail.....!

Jamie

[John Larkin]

On Sun, 04 Nov 2012 19:43:34 -0500, rickman <gnu...@gmail.com> wrote:

>On 11/4/2012 5:28 PM, John Larkin wrote:
>> On Sun, 04 Nov 2012 16:26:09 -0500, rickman<gnu...@gmail.com> wrote:
>>
>>> On 11/4/2012 3:29 PM, John Larkin wrote:
>>>>
>
>
>>>> PF is signed. Power/(Irms*Vrms) isn't.
>
>
>>>
>>> Why isn't Power/(Irms*Vrms) signed? What happens if the current is
>>> flowing in the opposite direction to the voltage? The RMS value will be
>>> negative and the above equation will be negative. That is the meaning
>>> of a negative power factor, no?
>>>
>>> Rick
>>
>> No. Wikipedia defines PF as a dimensionless value from 0 to 1, but
>> hangs "leading" and "lagging" off to the side.
>>
>> If voltage and current are in phase, PF is 1 and the sign is
>> indeterminate. If current phase leads voltage (capacitive load), by
>> even one degree, PF is called "leading" and if current waveform peaks
>> later, it's called "lagging." Capacitive (leading) loads are usually
>> assigned a negative PF if the word "leading" isn't used instead.
>> Whichever terminology is used, something often needs to be said about
>> whether the current leads or lags the voltage, and the simple
>> Power/(Irms*Vrms) equation loses that.
>
>This is getting into the twilight zone territory. Near the top of this
>post you are quoted saying, "PF is signed". Now you are saying it is
>NOT signed, no?

It has a phase. Some people call it "leading" and "lagging". Some
prople use - and + to express the same thing. That's just conventions.

If utilities want to switch in PF correction devices, they need to
know which direction they are trying to correct. It matters. They are
usually trying to correct for lag, so they usually switch in real or
simulated (rotating) capacitors.

>
>The point is that power factor is a result of reactance in the load, or
>non-linearities. When reactive, the current will be out of phase with
>the voltage. Blah, blah... The power factor with a reactive load
>tells you how much of the current is reactive and how much is
>dissipative. But what if the load is actually shoving power back into
>the line?

That's a different issue, but the same ideas apply.

>
>BTW, Wikipedia is not a primary source. In other words, I don't rely on
>it for significant facts unless the references are consulted. Wikipedia
>often gets facts *wrong*. Who was the reference they cited for the
>power factor info?

I took two semisters of Electrical Machinery in college, and I've
designed a dozen or so electronic AC power meters, many of which have
been tested by utilities, universities, and test labs. I wouldn't cite
Wiki if I thought they were wrong.

There's debate about what PF really means in
non-constant/unbalanced/non-sinusoidal situations.

[Jamie]

Please read this. I won't bother to explain it myself because everyone
thinks I am an idiot, they maybe correct in that insertion, I am an
idiot in trying to lead people in the correct direction.

True power (Resistive), Reactive Power(Xl,XC etc..) and Apparent, the
difference between the two = PF...

http://www.allaboutcircuits.com/vol_2/chpt_11/2.html

Jamie

[rickman]

On 11/4/2012 7:57 PM, John Larkin wrote:
> On Sun, 04 Nov 2012 19:43:34 -0500, rickman<gnu...@gmail.com> wrote:
>>
>> BTW, Wikipedia is not a primary source. In other words, I don't rely on
>> it for significant facts unless the references are consulted. Wikipedia
>> often gets facts *wrong*. Who was the reference they cited for the
>> power factor info?
>
> I took two semisters of Electrical Machinery in college, and I've
> designed a dozen or so electronic AC power meters, many of which have
> been tested by utilities, universities, and test labs. I wouldn't cite
> Wiki if I thought they were wrong.
>
> There's debate about what PF really means in
> non-constant/unbalanced/non-sinusoidal situations.

You miss the point about Wikipedia. If I cited my cousin on this
problem what would you say about that? He isn't wrong just because he
isn't a primary reference. But he still *isn't* a primary reference, so
I don't cite him. Wikipedia is full of errors whether this one is right
or wrong.

The point is that citing Wikipedia without citing the real source of the
info is just like citing my cousin. It lends nothing to your case.

Rick

[rickman]

On 11/4/2012 7:57 PM, John Larkin wrote:
> On Sun, 04 Nov 2012 19:43:34 -0500, rickman<gnu...@gmail.com> wrote:
>>
>> BTW, Wikipedia is not a primary source. In other words, I don't rely on
>> it for significant facts unless the references are consulted. Wikipedia
>> often gets facts *wrong*. Who was the reference they cited for the
>> power factor info?
>
> I took two semisters of Electrical Machinery in college, and I've
> designed a dozen or so electronic AC power meters, many of which have
> been tested by utilities, universities, and test labs. I wouldn't cite
> Wiki if I thought they were wrong.
>
> There's debate about what PF really means in
> non-constant/unbalanced/non-sinusoidal situations.

[rickman]

On 11/4/2012 8:20 PM, Jamie wrote:
> Please read this. I won't bother to explain it myself because everyone
> thinks I am an idiot, they maybe correct in that insertion, I am an
> idiot in trying to lead people in the correct direction.
>
> True power (Resistive), Reactive Power(Xl,XC etc..) and Apparent, the
> difference between the two = PF...
>
> http://www.allaboutcircuits.com/vol_2/chpt_11/2.html

Thank you.

Rick

[John Larkin]

Case? I'm telling you how people define and measure PF in the real
world. If you disagree on anything substantive, say so.

[John Larkin]

The ratio, actually. PF = real_power/apparent_power

where real power is in watts and apparent power is in volt-amps.

[rickman]

On 11/4/2012 9:18 PM, John Larkin wrote:
> On Sun, 04 Nov 2012 20:32:29 -0500, rickman<gnu...@gmail.com> wrote:
>
>> On 11/4/2012 7:57 PM, John Larkin wrote:
>>> On Sun, 04 Nov 2012 19:43:34 -0500, rickman<gnu...@gmail.com> wrote:
>>>>
>>>> BTW, Wikipedia is not a primary source. In other words, I don't rely on
>>>> it for significant facts unless the references are consulted. Wikipedia
>>>> often gets facts *wrong*. Who was the reference they cited for the
>>>> power factor info?
>>>
>>> I took two semisters of Electrical Machinery in college, and I've
>>> designed a dozen or so electronic AC power meters, many of which have
>>> been tested by utilities, universities, and test labs. I wouldn't cite
>>> Wiki if I thought they were wrong.
>>>
>>> There's debate about what PF really means in
>>> non-constant/unbalanced/non-sinusoidal situations.
>>
>> You miss the point about Wikipedia. If I cited my cousin on this
>> problem what would you say about that? He isn't wrong just because he
>> isn't a primary reference. But he still *isn't* a primary reference, so
>> I don't cite him. Wikipedia is full of errors whether this one is right
>> or wrong.
>>
>> The point is that citing Wikipedia without citing the real source of the
>> info is just like citing my cousin. It lends nothing to your case.
>>
>> Rick
>
> Case? I'm telling you how people define and measure PF in the real
> world. If you disagree on anything substantive, say so.

You don't get what I'm saying at all, do you?

Rick

[John Larkin]

Actually, no. Something about your cousin?

[Phil Hobbs]

On 11/03/2012 07:53 PM, Tim Williams wrote:
> "rickman"<gnu...@gmail.com> wrote in message
> news:k73kln$594$1...@dont-email.me...
>> I didn't think of it that way. The SI V is the abbreviation for the
>> unit "volts", but the symbol V or U is a math symbol and not necessarily
>> the same and the units abbreviation. Duh!
>
> 'U' is traditionally used by physicists to denote potential.
>
> Using it also for electrical potential, I believe, is more of a cultural
> thing. E, V and U have been variously used to represent voltage.
>
> Tim
>

E gets confused with electric field--having a quantity and its line
integral called by the same symbol is, *ahem*, suboptimal.

Technical vocabulary is funny, especially in EE and other engineering
fields. Normally it's sort of folksy but very descriptive, e.g.
"soakage" or "walkoff". Sometimes it's quite irrational,
though--everybody says "voltage" for "potential", but if you say
"amperage" for "current", everybody laughs at you.

The Brits used to call it "electric tension". Do you UK folks really
say "the tension at this point is fifteen millivolts?"

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

[Michael A. Terrell]

Phil Hobbs wrote:
>
> The Brits used to call it "electric tension". Do you UK folks really
> say "the tension at this point is fifteen millivolts?"

They say lots of weird things. I blame Lucas for the poor lighting
and all that warm beer they drink. It really messes with their minds!
;-)

[Martin Brown]

On 05/11/2012 14:43, Phil Hobbs wrote:
> On 11/03/2012 07:53 PM, Tim Williams wrote:
>> "rickman"<gnu...@gmail.com> wrote in message
>> news:k73kln$594$1...@dont-email.me...
>>> I didn't think of it that way. The SI V is the abbreviation for the
>>> unit "volts", but the symbol V or U is a math symbol and not necessarily
>>> the same and the units abbreviation. Duh!
>>
>> 'U' is traditionally used by physicists to denote potential.

U and V both are and u and v for that matter after that you end up with
subscripts - though I have seen S,s & T,t used in hand based analysis.

>> Using it also for electrical potential, I believe, is more of a cultural
>> thing. E, V and U have been variously used to represent voltage.
>>
>> Tim
>>
>
> E gets confused with electric field--having a quantity and its line
> integral called by the same symbol is, *ahem*, suboptimal.
>
> Technical vocabulary is funny, especially in EE and other engineering
> fields. Normally it's sort of folksy but very descriptive, e.g.
> "soakage" or "walkoff". Sometimes it's quite irrational,
> though--everybody says "voltage" for "potential", but if you say
> "amperage" for "current", everybody laughs at you.
>
> The Brits used to call it "electric tension". Do you UK folks really
> say "the tension at this point is fifteen millivolts?"

No, but we still talk of HT and EHT even today.

--
Regards,
Martin Brown

[Jim Thompson]

Lucas, Prince of Darkness ?:-)

[tm]

"Phil Hobbs" <pcdhSpamM...@electrooptical.net> wrote in message
news:RLGdnZQMJbMUTQrN...@supernews.com...

> On 11/03/2012 07:53 PM, Tim Williams wrote:
>> "rickman"<gnu...@gmail.com> wrote in message
>> news:k73kln$594$1...@dont-email.me...
>>> I didn't think of it that way. The SI V is the abbreviation for the
>>> unit "volts", but the symbol V or U is a math symbol and not necessarily
>>> the same and the units abbreviation. Duh!
>>
>> 'U' is traditionally used by physicists to denote potential.
>>
>> Using it also for electrical potential, I believe, is more of a cultural
>> thing. E, V and U have been variously used to represent voltage.
>>
>> Tim
>>
>
> E gets confused with electric field--having a quantity and its line
> integral called by the same symbol is, *ahem*, suboptimal.
>
> Technical vocabulary is funny, especially in EE and other engineering
> fields. Normally it's sort of folksy but very descriptive, e.g. "soakage"
> or "walkoff". Sometimes it's quite irrational, though--everybody says
> "voltage" for "potential", but if you say "amperage" for "current",
> everybody laughs at you.
>
> The Brits used to call it "electric tension". Do you UK folks really say
> "the tension at this point is fifteen millivolts?"
>
> Cheers
>
> Phil Hobbs
>
> --

Seen on a T shirt:

E/C^2 -1^-2 PV/nR

JT should know this one.

[rickman]

M 1 T?

Rick

[Michael A. Terrell]

Jim Thompson wrote:
>
> On Mon, 05 Nov 2012 10:02:31 -0500, "Michael A. Terrell"
> <mike.t...@earthlink.net> wrote:
>
> >
> >Phil Hobbs wrote:
> >>
> >> The Brits used to call it "electric tension". Do you UK folks really
> >> say "the tension at this point is fifteen millivolts?"
> >
> >
> > They say lots of weird things. I blame Lucas for the poor lighting
> >and all that warm beer they drink. It really messes with their minds!
> >;-)
>
> Lucas, Prince of Darkness ?:-)

Except tommorow, when he's the 'Prince of Dorkieness'.

[John Larkin]

On Sat, 3 Nov 2012 19:46:09 -0700 (PDT), George Herold
<ghe...@teachspin.com> wrote:

>On Nov 2, 9:00 pm, John Larkin <jlar...@highlandtechnology.com> wrote:
>> On Fri, 02 Nov 2012 20:28:04 -0400, rickman <gnu...@gmail.com> wrote:
>> >On 11/2/2012 7:30 PM, Habib Bouaziz-Viallet wrote:
>> >> Le 02/11/12 21:12, rickman a crit :
>>
>> >>> This would imply a 20 bit converter for the current, but don't assume a
>> >>> converter with 24 bits is accurate to 24 bits. There are mostly two
>> >>> types of converters with 24 bit results, slower ones for scales and
>> >>> other measurements at rather low sample rates which I don't think you
>> >>> can use. Then there are audio type converters which will sample at
>> >>> higher rates, but are specified for AC performance mostly. The useful
>> >>> number of bits (ENOB - effective number of bits) is sometimes given, but
>> >>> usually they spec SNR which is a similar number in dB. You will get 1
>> >>> bit for each 6.02 dB of SNR minus a small factor to account for
>> >>> quantization noise (~1.5 dB IIRC).
>>
>> >> Yes most of ADC 24bits are nearly 18bits ENOB but they (AD, TI, Cirrus
>> >> ....etc) claims that Gaussian noise gives the major part in the SINAD.
>> >> Gaussian noise on I and U would canceled themselves on the power
>> >> calculation (mean (discrete sum (U, I)) on 1s) because the two
>> >> acquisitions sequences are not correlated each other ... we will see
>> >> even if with GNU-Octave (awgn() ....) the results are something like
>> >> spectacular !
>>
>> >To get good accuracy you also need the two converters to be very well
>> >synchronized.  You can do the calculations to see just *how* synchronized.
>>
>> A reasonable ADC could sample current then voltage (in that order!)
>> maybe 10 usec apart. That's only about 0.2 degrees. Since the
>> front-end analog stuff probably includes some lowpass filtering, it's
>> easy to skew the filter time constants to take out that 0.2 degree
>> error.
>
>Cool, but how do you match the low passes?
>(do you measure the C's?)
>
>George H.
>

Just design the amplifiers to have 0.2 degrees relative phase shift.
If that's right to 10%, it's plenty good enough.


--

John Larkin Highland Technology, Inc

jlarkin at highlandtechnology dot com

http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators

Custom laser drivers and controllers

Photonics and fiberoptic TTL data links

VME thermocouple, LVDT, synchro acquisition and simulation

[John Larkin]

On Sat, 3 Nov 2012 18:34:05 -0700 (PDT), Robert Macy
<robert...@gmail.com> wrote:

>On Nov 3, 9:18 am, John Larkin
><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> On Sat, 3 Nov 2012 06:07:39 -0700 (PDT), Robert Macy
>>
>> <robert.a.m...@gmail.com> wrote:
>> >On Nov 3, 3:29 am, Habib Bouaziz-Viallet <h.bouazizvial...@free.fr>
>> >wrote:
>> >> ,,,snip...
>> >> will be used. I don't know exactly how and what these guys will handle
>> >> any test protocol :-(
>>
>> >> Best Regards, Habib.
>>
>> >Don't they test up to PF of 6? or is that 10?
>>
>> PF is always in the range of -1 to +1.
>>
>>  Digital has a it of
>>
>> >trouble there since we're starting to subtract to large numbers to
>> >find the power. Those 'disk' meters have an incredibly good accuracy
>> >curve thatconverges down to 'almost' zero.
>>
>> A digital power meter just simultaneous-samples the voltage and
>> current signals, multiplies, and integrates. That works for all
>> waveforms and all power factors. No trig, no subtraction required.
>>
>> --
>>
>> John Larkin                  Highland Technology Incwww.highlandtechnology.com  jlarkin at highlandtechnology dot com

>>
>> Precision electronic instrumentation
>> Picosecond-resolution Digital Delay and Pulse generators

>> Custom timing and laser controllers

>> Photonics and fiberoptic TTL data links

>> VME  analog, thermocouple, LVDT, synchro, tachometer
>> Multichannel arbitrary waveform generators
>

>maybe I'm thinkging 'crest factor'

A 100 amp class meter might be designed to clip at 200 amps peak or
so. The occasional monster current spike, like a short or a nasty
motor start, will have so little total energy that it won't matter.

[k...@att.bizzz]

You fail (no surprises).

[Spehro Pefhany]

I think it should be E/C^2 (-1)^1/2 PV/nR

For mjT, of course, or miT if you're not an EE.

[Phil Hobbs]

Perhaps it was actually (-1)**(1/2), rather than (-1)**(-2) ?

Which for engineers is of course mjT. ;)

[Jim Thompson]

Naaaah! The math classes used "i". I was already out of MIT before I
encountered the use of "j"

[Phil Hobbs]

I use "i" for physics problems, where the forward Fourier transform in
time is +i*2*pi*f*t, and "j" for EE, where the forward transform is
-j*2*pi*f*t, because it reminds me what the sign convention is.

Physicists care about spatial transforms more than EEs do, and a
positive-going plane wave is exp(+-i*(k*x- omega*t)), so you can't have
both frequencies positive and deal with positive-going waves--you have
to pick your favourite two out of three.

[Jim Thompson]

On Tue, 06 Nov 2012 12:21:30 -0500, Phil Hobbs

I use "j" exclusively now for complex expressions, and have no trouble
with +/- for direction of the wave >:-}

[Phil Hobbs]

I doubt you have to worry much about wave propagation on-chip. ;)

[Jim Thompson]

On Tue, 06 Nov 2012 13:11:35 -0500, Phil Hobbs

Getting close to that... latest stuff is 40nm feature size.

[Phil Hobbs]

Nah, you're all RC transmission lines, and the clock frequencies stopped
increasing years ago.

An interesting bit of work just came in--building SPICE models of an
ion-trap mass spectrometer. The current product works OK, but
apparently was designed by a chemist, so there are a few production
problems and lots of mysteries.

[John Larkin]

http://www.weeklystandard.com/blogs/report-woman-wearing-mit-shirt-barred-voting-florida_661853.html