Searching \ for '[EE]: Accurate mains frequency measurement' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=accurate+mains+frequency
Search entire site for: 'Accurate mains frequency measurement'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Accurate mains frequency measurement'
2001\06\27@045834 by D Lloyd

flavicon
face
Hi,

Anyone knocked up an accurate (at least 0.1%) mains frequency measurement
device with a PIC? The basic requirement is to datalog the mains frequency
over a period of time as it can often be unstable.

Reports of it being "speeded up" at the end of months to get the right
number of cycles in are often suggested.....(!)

I imagine some kind of fast zero cross detect or fast comparator to detect
the same point on the mains cycle with a pulse output to the PIC would be
the way to go (rather than A-D sampling) but if anyone has any better
ideas....or links....

Best Regards,
Dan

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@055327 by Jinx

face picon face
> Reports of it being "speeded up" at the end of months to
> get the right number of cycles in are often suggested.....(!)

I believe that within any 24 hour period the cycle count is very
accurate and will average 50 or 60Hz. The frequency will drop
during peak and then rise during off-peak

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@063928 by D Lloyd

flavicon
face
part 1 1406 bytes content-type:text/plain; charset=us-ascii
Hi,

Thanks for that but this is part of the problem. Anyone you ask will say,
"I believe that the mains does this....", but no one can tell you for sure
(as, like you say, it is not precise unless averaged over a relatively long
time) which is why I wanted to monitor it. We often drive metering clocks
from the mains, and they are to the required accuracy, but I wished to see
what the mains *really* (i.e. more accurately) got up to while no one was
looking....both short and long term.

Cheers,
Dan




(Embedded     Jinx <spam_OUTjoecolquittTakeThisOuTspamCLEAR.NET.NZ>TakeThisOuTspamMITVMA.MIT.EDU>> image moved   27/06/2001 10:54
to file:
pic21419.pcx)





Please respond to pic microcontroller discussion list
     <
.....PICLISTKILLspamspam@spam@MITVMA.MIT.EDU>
Sent by:  pic microcontroller discussion list <PICLISTspamKILLspamMITVMA.MIT.EDU>


To:   .....PICLISTKILLspamspam.....MITVMA.MIT.EDU
cc:
Subject:  Re: [EE]: Accurate mains frequency measurement

Security Level:?         Internal


> Reports of it being "speeded up" at the end of months to
> get the right number of cycles in are often suggested.....(!)

I believe that within any 24 hour period the cycle count is very
accurate and will average 50 or 60Hz. The frequency will drop
during peak and then rise during off-peak

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics






part 2 165 bytes content-type:application/octet-stream; (decode)

part 3 144 bytes
--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@071506 by Jinx

face picon face
> Thanks for that but this is part of the problem. Anyone you ask
> will say, "I believe that the mains does this....", but no one can tell
> you for sure (as, like you say, it is not precise unless averaged
> over a relatively long time) which is why I wanted to monitor it. We
> often drive metering clocks from the mains, and they are to the
> required accuracy, but I wished to see what the mains *really* (i.e.
> more accurately) got up to while no one was looking....both short
> and long term.

A rough indicator would be to monitor the reading of a mains-
synchronised clock (analogue or digital) like a factory clock
or VCR/radio display. I'm reasonably fussy about having my VCRs
set correctly as I often tape by the timer. I set them by the BBC
"pips" on the hour at some point, and find it's a few seconds out
(slow) by the next time it needs seasonally putting forward or
back an hour

I also have a professional interest in horology, as I build public and
commercial timepieces and have made some medium-term
measurements as I build them, and also keep tabs on them after
installation. I find the same result as with the VCR and, assuming
the BBC has access to a world-class chronometer, conclude that
over a long period (6-12 months), mains frequency on average is
very very slightly less than 50Hz

I would be surprised if there are not government regulations or
standards to be upheld by the power generating companies. The
degree of accuracy they achieve may well be within these (if any)
guidelines, and suitable for general purpose usage. Inaccuracies
may be accepted as operating practicalities, and that applications
requiring more accurate timing have access to a more appropriate
source such as beacons / satellites

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@071523 by Bob Ammerman

picon face
> Hi,
>
> Anyone knocked up an accurate (at least 0.1%) mains frequency measurement
> device with a PIC? The basic requirement is to datalog the mains frequency
> over a period of time as it can often be unstable.

I have done this. (Unfortunately as a work-for-hire so I can't publish the
code:-() It is relatively easy. You can just use the PICs crystal as a
timebase, a zero cross detector with some hysteresis, and low-pass filter
the result.

> Reports of it being "speeded up" at the end of months to get the right
> number of cycles in are often suggested.....(!)

Actually, it is adjusted much more often than once a month.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@071928 by Bob Ammerman

picon face
From: "D Lloyd" <EraseMEdan.lloydspam_OUTspamTakeThisOuTGB.ABB.COM>
>
> Hi,
>
> Thanks for that but this is part of the problem. Anyone you ask will say,
> "I believe that the mains does this....", but no one can tell you for sure
> (as, like you say, it is not precise unless averaged over a relatively
long
> time) which is why I wanted to monitor it. We often drive metering clocks
> from the mains, and they are to the required accuracy, but I wished to see
> what the mains *really* (i.e. more accurately) got up to while no one was
> looking....both short and long term.
>
> Cheers,
> Dan

Dan,

For this you are going to need quite a bit better than 0.1%.

In the US at least: System frequency will vary by small amounts under heavy
load swings. It will be averaged out 'as quickly as possible', often within
minutes, nearly always within hours, certainly within days.

Accuracy over intervals of days is almost certainly going to be better than
any crystal-based timing reference (including, probably temperature
compensated ones (ie: ovens)).

btw: In one of my alternate lives I develop control systems for power
plants, so...

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@072805 by D Lloyd

flavicon
face
part 1 2437 bytes content-type:text/plain; charset=us-ascii
Hi,





(Embedded     Bob Ammerman <RAMMERMANspamspam_OUTPRODIGY.NET>spamspam_OUTMITVMA.MIT.EDU>> image moved   27/06/2001 12:13
to file:
pic22678.pcx)





Please respond to pic microcontroller discussion list
     <
@spam@PICLISTKILLspamspamMITVMA.MIT.EDU>
Sent by:  pic microcontroller discussion list <KILLspamPICLISTKILLspamspamMITVMA.MIT.EDU>


To:   RemoveMEPICLISTTakeThisOuTspamMITVMA.MIT.EDU
cc:
Subject:  Re: [EE]: Accurate mains frequency measurement

Security Level:?         Internal


> Hi,
>
> Anyone knocked up an accurate (at least 0.1%) mains frequency measurement
> device with a PIC? The basic requirement is to datalog the mains
frequency
> over a period of time as it can often be unstable.

I have done this. (Unfortunately as a work-for-hire so I can't publish the
code:-() It is relatively easy. You can just use the PICs crystal as a
timebase, a zero cross detector with some hysteresis, and low-pass filter
the result.

* I had a go at this last year (with a 68HC11) and found that the zero
cross detect was very slow and the cumulative time delay between detection
of the pulse and being able to see it in the micro was not very repeatable.
I think it was maybe my bog-standard opto zero cross detect (slow) that
scuppered it. It know it's pretty straight forward (especially with
capture/compare) but getting that pulse in fast and with a known low delay
seemed to be the problem....


> Reports of it being "speeded up" at the end of months to get the right
> number of cycles in are often suggested.....(!)

Actually, it is adjusted much more often than once a month.

* I know - this was just one of the things "mentioned" to me when I quizzed
several people about the subject.

Best regards,
Dan


Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

> I imagine some kind of fast zero cross detect or fast comparator to
detect
{Quote hidden}

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics






part 2 165 bytes content-type:application/octet-stream; (decode)

part 3 144 bytes
--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@072816 by Laszlo Kohegyi

flavicon
face
Hi,

>what the mains *really* (i.e. more accurately) got up to while no one was
>looking....both short and long term.
>
>Cheers,
>Dan

I'm afraid none of the simple sw methods (zero crossing etc) will work
because of the high level of noise, periodic and transient. You will
need some analog circuitry:

1. a good bandpass filter around 50-60Hz
or
2. a pll, locked to the mains input.

After these a simple counting could be enough.
If you want a fully digital solution, I think you should use an A/D
and some sophisticated algorithm such as

3. FFT
or
4. FIR or IIR filter (again bandpass around 50-60Hz).


Best regards
Les

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@073140 by D Lloyd

flavicon
face
part 1 2117 bytes content-type:text/plain; charset=us-ascii
(Embedded     Bob Ammerman <spamBeGoneRAMMERMANspamBeGonespamPRODIGY.NET>spamBeGonespamMITVMA.MIT.EDU>> image moved   27/06/2001 12:18
to file:
pic17337.pcx)









Please respond to pic microcontroller discussion list
     <
TakeThisOuTPICLISTEraseMEspamspam_OUTMITVMA.MIT.EDU>
Sent by:  pic microcontroller discussion list <RemoveMEPICLISTspamTakeThisOuTMITVMA.MIT.EDU>


To:   PICLISTEraseMEspam.....MITVMA.MIT.EDU
cc:
Subject:  Re: [EE]: Accurate mains frequency measurement

Security Level:?         Internal


From: "D Lloyd" <EraseMEdan.lloydspamGB.ABB.COM>
>
> Hi,
>
> Thanks for that but this is part of the problem. Anyone you ask will say,
> "I believe that the mains does this....", but no one can tell you for
sure
> (as, like you say, it is not precise unless averaged over a relatively
long
> time) which is why I wanted to monitor it. We often drive metering clocks
> from the mains, and they are to the required accuracy, but I wished to
see
> what the mains *really* (i.e. more accurately) got up to while no one was
> looking....both short and long term.
>
> Cheers,
> Dan

Dan,

For this you are going to need quite a bit better than 0.1%.

* You are right - I'd like something a lot more accurate but I think 0.1%
would be a reasonable starting point.

In the US at least: System frequency will vary by small amounts under heavy
load swings. It will be averaged out 'as quickly as possible', often within
minutes, nearly always within hours, certainly within days.

Accuracy over intervals of days is almost certainly going to be better than
any crystal-based timing reference (including, probably temperature
compensated ones (ie: ovens)).

btw: In one of my alternate lives I develop control systems for power
plants, so...

* It's not a major requirement....I just wanted to see if anyone had got a
fast zero/cycle detect circuit, really ;-)

Thanks again,
Dan

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics






part 2 165 bytes content-type:application/octet-stream; (decode)

part 3 144 bytes
--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@073342 by Bob Ammerman

picon face
> > Hi,
> >
> > Anyone knocked up an accurate (at least 0.1%) mains frequency
measurement
{Quote hidden}

repeatable.
> I think it was maybe my bog-standard opto zero cross detect (slow) that
> scuppered it. It know it's pretty straight forward (especially with
> capture/compare) but getting that pulse in fast and with a known low delay
> seemed to be the problem....

Slow is not an issue. We are dealing with at least seconds here. What does
matter is consistency on the zero cross.

>
> > Reports of it being "speeded up" at the end of months to get the right
> > number of cycles in are often suggested.....(!)
>
> Actually, it is adjusted much more often than once a month.
>
> * I know - this was just one of the things "mentioned" to me when I
quizzed
> several people about the subject.
>
> Best regards,
> Dan

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@073349 by Jinx

face picon face
> btw: In one of my alternate lives I develop control systems for power
> plants, so...
>
> Bob Ammerman

Bob, how are the US power generating companies regulated with
respect to frequency standards ? This isn't something I've looked
into under the Sale Of Goods Act in NZ, but I assume that the mains
electricity supplied must be fit for the purpose for which it was bought.
Cases have been won against the power companies for supplying
over-voltage that damaged equipment - I wonder how the long-term
frequency of the mains comes into the product description, particularly
in the litigious US

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@073601 by Bob Ammerman

picon face
Actually, it will work fine as long as:

1: You do a minimal amount in hardware. A simple RC filter goes a long way!

2: You ignore edges that 'just don't make sense' in software. This is
basically a sort of PLL in software.

3: You perform your computation over quite a few cycles of the signal.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

{Original Message removed}

2001\06\27@074006 by Bob Ammerman

picon face
I'm not sure if the lawyers have gotten into this to protect the unwashed
masses from inaccurate clocks, but government agencies _do_ regulate this. I
don't know the exact standards to which the power suppliers are held.

Bob Ammerman
RAm Systems
(contract development of high performance, high function, low-level
software)

{Original Message removed}

2001\06\27@074628 by Jinx

face picon face
Dan, just out of interest, what are you going to use as an external
timing reference for your datalogging ?

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@081155 by D Lloyd

flavicon
face
part 1 964 bytes content-type:text/plain; charset=us-ascii
Hi,

I had thought about using an Internet atomic clock timeserver, eg US Navy
or Rolex  (and allowing for the propagation trip delay) as they are there,
accurate and free to use. Knowing how bad PCs are for timekeeping, this
seemed a reasonable solution.

Regards,
Dan






(Embedded     Jinx <RemoveMEjoecolquittEraseMEspamEraseMECLEAR.NET.NZ>EraseMEspamEraseMEMITVMA.MIT.EDU>> image moved   27/06/2001 12:48
to file:
pic01727.pcx)





Please respond to pic microcontroller discussion list
     <
RemoveMEPICLISTspam_OUTspamKILLspamMITVMA.MIT.EDU>
Sent by:  pic microcontroller discussion list <RemoveMEPICLISTTakeThisOuTspamspamMITVMA.MIT.EDU>


To:   EraseMEPICLISTspamspamspamBeGoneMITVMA.MIT.EDU
cc:
Subject:  Re: [EE]: Accurate mains frequency measurement

Security Level:?         Internal


Dan, just out of interest, what are you going to use as an external
timing reference for your datalogging ?

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics






part 2 165 bytes content-type:application/octet-stream; (decode)

part 3 144 bytes
--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@082029 by Jinx

face picon face
> Hi,
>
> I had thought about using an Internet atomic clock timeserver, eg
> US Navy or Rolex  (and allowing for the propagation trip delay) as
> they are there, accurate and free to use. Knowing how bad PCs
> are for timekeeping, this seemed a reasonable solution.
>
> Regards,
> Dan

I'd be interested in knowing what you decide, how you implement
it and what the results are

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@082645 by D Lloyd

flavicon
face
part 1 1128 bytes content-type:text/plain; charset=us-ascii
Hi,

See below:




(Embedded     Jinx <RemoveMEjoecolquittKILLspamspamCLEAR.NET.NZ>KILLspamspamMITVMA.MIT.EDU>> image moved   27/06/2001 13:20
to file:
pic07997.pcx)





Please respond to pic microcontroller discussion list
     <
PICLISTSTOPspamspamspam_OUTMITVMA.MIT.EDU>
Sent by:  pic microcontroller discussion list <spamBeGonePICLISTSTOPspamspamEraseMEMITVMA.MIT.EDU>


To:   KILLspamPICLISTspamBeGonespamMITVMA.MIT.EDU
cc:
Subject:  Re: [EE]: Accurate mains frequency measurement

Security Level:?         Internal


> Hi,
>
> I had thought about using an Internet atomic clock timeserver, eg
> US Navy or Rolex  (and allowing for the propagation trip delay) as
> they are there, accurate and free to use. Knowing how bad PCs
> are for timekeeping, this seemed a reasonable solution.
>
> Regards,
> Dan

I'd be interested in knowing what you decide, how you implement
it and what the results are

* If I get "time" (groan) to do something good about all this, I'll
certainly let you know.

Thanks again,
Dan

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics






part 2 165 bytes content-type:application/octet-stream; (decode)

part 3 144 bytes
--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@083908 by Jinx

face picon face
> * If I get "time" (groan) to do something good about all this, I'll
> certainly let you know.
>
> Thanks again,
> Dan

Cheers. I've had a look around the US Code Of Federal
Regulations but didn't see anything about regulations covering
frequency standards for power generators. Possibly looking in
the wrong place. Maybe tomorrow I'll pop off a line to Pacific
Gas & Electric, Hoover Powerplant etc and my local utility co

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@085211 by Andy Jancura

picon face
Hi Dan,

two problems I see.

1) precision time constant
2) random noise on cables (burst, spikes,...)

I don't believe, that this can be done with simple solution.


Andrej

_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@090723 by Olin Lathrop

face picon face
> Anyone knocked up an accurate (at least 0.1%) mains frequency measurement
> device with a PIC? The basic requirement is to datalog the mains frequency
> over a period of time as it can often be unstable.
>
> Reports of it being "speeded up" at the end of months to get the right
> number of cycles in are often suggested.....(!)
>
> I imagine some kind of fast zero cross detect or fast comparator to detect
> the same point on the mains cycle with a pulse output to the PIC would be
> the way to go (rather than A-D sampling) but if anyone has any better
> ideas....or links....

Getting a square wave into the PIC from the mains is the easy part.  I think
you will need considerably better than .1% accuracy to get meaningful
measurements.  The frequency variations tend to be very small.  It's
probably more useful to think of it as a long term phase shift.  I think the
tricky part of measuring this will be to get a good enough reference
frequency.  Perhaps you can rig something that receives one of the various
time signals from WWV, WWVL, WWVH, or equivalents around the world.

Another very accurate reference is the color burst signal in your TV set.
All the major networks somehow lock or derive their color carriers from the
atomic clocks in Boulder.  Yes, you can make legally traceable time
measurements with high accuracy from a TV color subcarrier.  I remember
seeing equipment that did exactly this about 25 years ago.

The GPS clock is also supposed to be extremely accurate.  I think it is also
derived from the atomic clocks, but am not sure.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, EraseMEolinspamEraseMEembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@091745 by Maurizio Viterbini

flavicon
face
Hi Jinx, I found this simple circuit that could be useful for you.


Two-Transistor Atomic Frequency Standard
http://techlib.com/electronics/atomic.html

ciao
--
Maurizio Viterbini
Istituto di Fisica dell'Atmosfera
Via Fosso del Cavaliere 100
00133 Roma
Italy

tel. +39 06 49934200
fax +39 06 20660291
email @spam@viterbin@spam@spamspam_OUTifa.rm.cnr.it

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@092412 by Maurizio Viterbini

flavicon
face
Maurizio Viterbini wrote:
>
> Hi Jinx, I found this simple circuit that could be useful for you.
>
> Two-Transistor Atomic Frequency Standard
> http://techlib.com/electronics/atomic.html

please add www

{Quote hidden}

--
Maurizio Viterbini
Istituto di Fisica dell'Atmosfera
Via Fosso del Cavaliere 100
00133 Roma
Italy

tel. +39 06 49934200
fax +39 06 20660291
email .....viterbinspam_OUTspamifa.rm.cnr.it

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@092418 by Scott Newell

flavicon
face
>Another very accurate reference is the color burst signal in your TV set.
>All the major networks somehow lock or derive their color carriers from the
>atomic clocks in Boulder.  Yes, you can make legally traceable time
>measurements with high accuracy from a TV color subcarrier.  I remember
>seeing equipment that did exactly this about 25 years ago.

Is this still true?  How do you know if the carrier is still clean from the
network, or regenerated out of some fancy digital video manipulation box at
the local affliate?


>The GPS clock is also supposed to be extremely accurate.  I think it is also
>derived from the atomic clocks, but am not sure.

Each GPS sat has several atomic clocks onboard.  Look for a receiver with a
pps output--some receivers have variable internal processing delay from
true tick to serial output, making it very difficult to get at stable
timebase.


newell

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@092746 by J.Feldhaar

flavicon
face
Olin Lathrop schrieb:

>

-snip-

> The GPS clock is also supposed to be extremely accurate.  I think it is also
> derived from the atomic clocks, but am not sure.
>

Yes, Olin, they have two rubidium and two cesium standards per satellite, they
are constantly controlled by the tracking and control stations and there is a
good number of "atomic clock" replacements on the market, I was getting the
first series of these made by HAMEG ready to work. They were as good as 1/10E-11
in stability. (But unfortunately no PIC in there...)

Greets
Jochen Feldhaar DH6FAZ

>
> ********************************************************************
> Olin Lathrop, embedded systems consultant in Littleton Massachusetts
> (978) 742-9014, TakeThisOuTolin.....spamTakeThisOuTembedinc.com, http://www.embedinc.com
>
> --
> http://www.piclist.com hint: The list server can filter out subtopics
> (like ads or off topics) for you. See http://www.piclist.com/#topics

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@094829 by Alan B. Pearce

face picon face
>The GPS clock is also supposed to be extremely accurate.  I think it is
also
>derived from the atomic clocks, but am not sure.

This is correct. each GPS satellite has its own caesium beam clock on board,
and the time is kept closely adjusted by the control network. There was an
article in QST around 18-24 months ago about building a frequency standard
that locked onto the 1pps pulse some GPS receivers provide. The author was
able to get something like 1E-14 accuracy from an ovened crystal locked onto
the GPS. There have also been other articles on doing this, one I have seen
was in a German ham radio magazine, but it did not seem to have been done as
elegantly as the QST one.

It is possible to get a scanned PDF of the QST article, and the hex code to
program a 16C73 as per the article, from the authors web site at
http://www.rt66.com/~shera/ The quality is not the best as it is a scan of
the paper article, but with the extra info on the website this is not a
problem.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@095632 by Dwayne McCoy

flavicon
face
Interesting experiment, pardon my stream of consciousness approach to
answering this.

Two things,

1) Since most power generators are on "the grid" they all have to be pretty
much in sync. I'm sure it is pretty accurate because most clocks I deal with
use line frequency as a standard and are pretty close, I check with
NBS/Navel observatory fairly often (I like an accurate watch).

2) what about running the signal through a band pass (say 50 to 70 cycle) or
low pass (70 cycle) filter to remove the High frequency noise and instead of
measuring frequency measure period of a cycle every so often say 1 second,
or for that matter look for variation in adjacent cycles. I believe a PIC
can .01 ms or better with fair ease.

Another suggestion Transform the mains down to a reasonable voltage this
will remove a good portion of the high frequency noise, isolate you from the
mains and give you a voltage that is much less dangerous to work with. You
might look at the output of the transformer on ano scope an see if hf noise
is an issue.

Just my $/50 worth

DMc


{Original Message removed}

2001\06\27@101132 by Jinx

face picon face
> Maurizio Viterbini wrote:
> >
> > Hi Jinx, I found this simple circuit that could be useful for you.
> >
> > Two-Transistor Atomic Frequency Standard
> > http://www.techlib.com/electronics/atomic.html

Thanks a lot - I'll make one up and see how it goes

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@111847 by D Lloyd

flavicon
face
part 1 1181 bytes content-type:text/plain; charset=us-ascii
Hi,

Thanks to everyone who has replied (so far!)....I'm sort of half glad it
isn't as easy as it might appear as I thought it was just me....

Probably easier to hire something in, that is for sure....but that is not
as much fun.

If I do get a shot at this, I'll be sure to post how it went.

Cheers,
Dan





(Embedded     Jinx <TakeThisOuTjoecolquittKILLspamspamspamCLEAR.NET.NZ>KILLspamspamspamMITVMA.MIT.EDU>> image moved   27/06/2001 15:12
to file:
pic31302.pcx)





Please respond to pic microcontroller discussion list
     <
.....PICLISTspamRemoveMEMITVMA.MIT.EDU>
Sent by:  pic microcontroller discussion list <RemoveMEPICLISTspamspamBeGoneMITVMA.MIT.EDU>


To:   spamBeGonePICLIST@spam@spamspam_OUTMITVMA.MIT.EDU
cc:
Subject:  Re: [EE]: Accurate mains frequency measurement

Security Level:?         Internal


> Maurizio Viterbini wrote:
> >
> > Hi Jinx, I found this simple circuit that could be useful for you.
> >
> > Two-Transistor Atomic Frequency Standard
> > http://www.techlib.com/electronics/atomic.html

Thanks a lot - I'll make one up and see how it goes

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics






part 2 165 bytes content-type:application/octet-stream; (decode)

part 3 144 bytes
--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@122140 by Robert A. LaBudde

flavicon
face
At 11:42 AM 6/27/01 +0100, Dan wrote:
>Thanks for that but this is part of the problem. Anyone you ask will say,
>"I believe that the mains does this....", but no one can tell you for sure
>(as, like you say, it is not precise unless averaged over a relatively long
>time) which is why I wanted to monitor it. We often drive metering clocks
>from the mains, and they are to the required accuracy, but I wished to see
>what the mains *really* (i.e. more accurately) got up to while no one was
>looking....both short and long term.

Average power line frequencies are very accurate over long times.

The answer is simple: It's the law!

When synchronous electric clocks were invented, the power companies were
required to keep the average frequency accurate to keep the clocks
accurate. They even speed up the frequency to compensate for short outages,
so clocks don't have to be reset for cumulative losses. If an outage lasts
a substantial amount of time (minutes or more), they don't make adjustsments.

This dates back, I believe, to the 1930's.

================================================================
Robert A. LaBudde, PhD, PAS, Dpl. ACAFS  e-mail: TakeThisOuTralspamspamlcfltd.com
Least Cost Formulations, Ltd.            URL: http://lcfltd.com/
824 Timberlake Drive                     Tel: 757-467-0954
Virginia Beach, VA 23464-3239            Fax: 757-467-2947

"Vere scire est per causas scire"
================================================================

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@123846 by Douglas Butler

flavicon
face
I remember touring Hoover (Boulder) Dam and they had the original 1930's
era control equipment in a glass case.  It was a very accurate
mechanical clock and an "accumulator" which kept track of how the power
cycles were ahead of, or behind the clock.  Some guy would watch the
accumulator and adjust a needle valve to set the frequency of the dynamo
that excited the main alternators.

During the depression a human was the best & most cost effective PID
controller you could get!

Sherpa Doug

> {Original Message removed}

2001\06\27@180722 by Olin Lathrop

face picon face
> >Another very accurate reference is the color burst signal in your TV set.
> >All the major networks somehow lock or derive their color carriers from
the
> >atomic clocks in Boulder.  Yes, you can make legally traceable time
> >measurements with high accuracy from a TV color subcarrier.  I remember
> >seeing equipment that did exactly this about 25 years ago.
>
> Is this still true?  How do you know if the carrier is still clean from
the
> network, or regenerated out of some fancy digital video manipulation box
at
> the local affliate?

I don't know for sure.  I was told once that TV stations affiliated with a
major network receive sync signals from that network, including the color
subcarrier.  Everything in the TV station is then synched to that.  Maybe
newer equipment doesn't work that way, but I would think switching between
different color subcarrier sources is a bad thing because the receivers
would all see a sudden color phase jump and would take a number of scan
lines to adjust to the new signal.  In the mean time, all the colors would
go screwy.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olinEraseMEspamembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\27@203110 by Barry Gershenfeld

picon face
>Bob, how are the US power generating companies regulated with
>respect to frequency standards ?

It seems to me that if you are connected to the "power grid", you
wouldn't have much of a choice.   Lag behind "the world" slightly
and you will find the power grid is driving your equipment like
a motor in order to keep it "up to speed".  Try to push it
ahead and you will find yourself trying to supply (nearly)
infinite power trying to get "the world" to speed up with
you.   So you "have to" run at the right frequency (and phase!)

My questions:
(1) Where does the authority to change the line
frequency originate?  Meaning, who's at the top that everyone
else has to follow?   Or maybe it's distributed, with computer
controls coordinating each power station's contribution to
the overall effort to go "faster" or "slower" .

(2) How extensive is each power grid?  I know there is some
separation they put in after a few really embarrassing big
blackouts.  Do all the power grids run in sync anyway?  (Note:
your mileage may vary.  This is kind of the U.S. view.  Maybe
the Outback is powered by lots of little independent
generators...)

BTW I'm working on a device with a "60 Hz interrupt" built in.
It will be interesting to see if I can get anything resembling
accuracy from it. I'm already seeing readings like 62 Hz on
occasion, so I'm reading the suggestions (like low-pass
filtering) intently.

Barry

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\06\28@031432 by D Lloyd

flavicon
face
part 1 3389 bytes content-type:text/plain; charset=us-ascii

During the depression a human was the best & most cost effective PID
controller you could get!

* Does he also double as transistor man out of the 'Art of Electronics'
bible? ;-)

Dan





(Embedded     Douglas Butler <RemoveMEdbutlerEraseMEspamspam_OUTIMETRIX.COM>EraseMEspamspam_OUTMITVMA.MIT.EDU>> image moved   27/06/2001 17:30
to file:
pic03145.pcx)





Please respond to pic microcontroller discussion list
     <
@spam@PICLISTRemoveMEspamEraseMEMITVMA.MIT.EDU>
Sent by:  pic microcontroller discussion list <EraseMEPICLISTspam@spam@MITVMA.MIT.EDU>


To:   @spam@PICLISTspam_OUTspam.....MITVMA.MIT.EDU
cc:
Subject:  Re: [EE]: Accurate mains frequency measurement

Security Level:?         Internal


I remember touring Hoover (Boulder) Dam and they had the original 1930's
era control equipment in a glass case.  It was a very accurate
mechanical clock and an "accumulator" which kept track of how the power
cycles were ahead of, or behind the clock.  Some guy would watch the
accumulator and adjust a needle valve to set the frequency of the dynamo
that excited the main alternators.

During the depression a human was the best & most cost effective PID
controller you could get!

Sherpa Doug

> {Original Message removed}
part 2 165 bytes content-type:application/octet-stream; (decode)

part 3 131 bytes
--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2001\06\28@095141 by Jason Harper

picon face
"Robert A. LaBudde" <spamBeGoneralEraseMEspamLCFLTD.COM> wrote:

> They even speed up the frequency to compensate for short outages,
> so clocks don't have to be reset for cumulative losses.

I've heard this claim before, but I don't see how it's possible, given that
outages typically don't involve the entire power grid.  How can they
selectively increase the frequency to only the customers downstream of the
problem?
       Jason Harper

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2001\06\28@100326 by Barry King

flavicon
face
Barry, et al.,

We've talked about this before, so you'll want to dig in the
archives, too, and I'll keep this brief.

I asked a power company engineer about this whole issue a while ago
when I was setting up a system using synchronous motors and needed to
know the limits of error of this "constant" speed.

You said:
> It seems to me that if you are connected to the "power grid", you
> wouldn't have much of a choice.   Lag behind "the world" slightly
> and you will find the power grid is driving your equipment like
> a motor in order to keep it "up to speed".  Try to push it
> ahead and you will find yourself trying to supply (nearly)
> infinite power trying to get "the world" to speed up with
> you.   So you "have to" run at the right frequency (and phase!)

Not really.  All generating equipment on a grid has to be running in
sync of course, but only the smallest generators can run by just
monitoring their own phase lead- that only works if the generator is
small enough so that its unable to affect the phase.

Any large generation facility has to have a reference oscillator and
will actively control frequency and power output to achive the
desired line voltage and frequency.  A Gigiwatt is enough power
output to cause large voltage and or frequency fluctuations.  Even a
Megawatt plant needs to regulate carefully to be sure its help rahter
than hurting power quality.

The power plants use a hierarchy of phase-locked reference
oscillators.  Nowadays they may be switching to GPS based references,
but I suspect that since the system they have is working, it has not
been changed.

There are a few (I think its three in the US) main grid blocks, and
each of them has a reference, usually at one of the main generating
facilities.  (I think one is at Niagra/Horseshoe Falls for the
Northeast US and SE Canada, not sure.)  This reference is fed (used
to be by leased telephone lines, nowadays who knows?) to sub-masters
in each area (approximately state by state, but that depends on
generator locations, not state lines).  Sub-masters feed smaller
local power stations.

Again, these are all phase locked, and they do something to
compensate for propagation delays when they set a power plant's
reference up.

So they have a reference.  But the original question was, what do
they do with it- how good is the frequency regulation?  The engineer
I talked to (and this was a while ago) said that frequency does vary
with load.  The grid managers do try to run fast (and therefore high
line voltage, by the way) when they are lightly loaded at night to
"catch up" to 5184000 cycles per day, they do try to regulate day by
day when possible.  Unless they go offline (blackout) when all bets
are off.

To answer these questions:
> >Bob, how are the US power generating companies regulated with
> >respect to frequency standards?

> (1) Where does the authority to change the line
> frequency originate?  Meaning, who's at the top that everyone
> else has to follow?   Or maybe it's distributed, with computer
> controls coordinating each power station's contribution to
> the overall effort to go "faster" or "slower" .

In the U.S., the power companies cooperate on this through grid
management organizations which are set up by the government
regulatory agencys.  But the power quality standards (frequency or
line voltage regulation, for example) that the companies HAVE to meet
are pretty loose, and have to do more with operation of the grid than
service quality.  The frequency police are NOT going to show up if
your power company runs at 61 Hz for a while unless its messing up
the larger grid.

So to get back to original question- is the power line a useful
timebase?  It is for applications such as wall clocks which need a
long term accurate timebase, but can stand some short term error, and
can be re-set to remove accumulated error or after an interruption.

-the other Barry.
------------
Barry King
NRG Systems "Measuring the Wind's Energy"
http://www.nrgsystems.com
Check out the List Archive!  Start at:
PIC/PICList FAQ: http://www.piclist.com/faq

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2001\06\28@150537 by Peter L. Peres

picon face
If he wants 0.1% then of course he should spend a bit on the analog side
of the problem. I think so.

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2001\06\28@150548 by Peter L. Peres

picon face
> zero crossing detector

Why does nobody use a squaring multiplier followed by a differentiator and
hysterezis ? Or at least a precision rectifier followed by differentiator
and hysterezis ?

The zero crossing is that part of the mains waveform (and of any
symmetrical AC waveform) where the S/N ratio always tends to 0.

In fact with mains the points at about 45 and 135 degrees are the
'quietest' afaik.

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2001\06\28@150601 by Peter L. Peres

picon face
50Hz = 20 msec = 10 + 10 msec = 10000 + 10000 usec

(at 60Hz it is slightly worse)

For 0.1% you need to resolve at least twice as good, i.e. 5 usec. This
must include any timing jitter in the PIC counter algorythm and edge
detection, and the group delay etc in the analog section. Getting an
analog circuit (mains driven) to under 5 usec jitter is an interesting
exercise afaik. Getting the PIC to resolve the edge with less than what is
left in time, and not add any more jitter to it, is even more interesting.

If you apply the idea of many measurements and averaging you gain accuracy
by sqrt(Nmeasurements), but is this what you want, or do you want absolute
instantaneous frequency for logging ?

With a 1/sec readout you should be able to get a 7 times improvement
(almost 11 times if you measure each half period separately). That would
put the edge detect/analog part accuracy at 50 usec which is pretty tight
but can be done imho.

Peter

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2001\06\29@090638 by Olin Lathrop

face picon face
> If you apply the idea of many measurements and averaging you gain accuracy
> by sqrt(Nmeasurements), but is this what you want, or do you want absolute
> instantaneous frequency for logging ?
>
> With a 1/sec readout you should be able to get a 7 times improvement
> (almost 11 times if you measure each half period separately). That would
> put the edge detect/analog part accuracy at 50 usec which is pretty tight
> but can be done imho.

But if you measure 1 second worth (60 cycles) you don't do 60 or 120
independent measurements.  You measure the total time for all 60 cycles,
then divide by 60.  This will reduce any period measurement noise directly
by 60.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olinspamBeGonespamembedinc.com, http://www.embedinc.com

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email RemoveMElistserv@spam@spamspamBeGonemitvma.mit.edu with SET PICList DIGEST in the body


More... (looser matching)
- Last day of these posts
- In 2001 , 2002 only
- Today
- New search...