Searching \ for '[EE] Three-Phase Transformer set for monitoring.' 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/power.htm?key=transformer
Search entire site for: 'Three-Phase Transformer set for monitoring.'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Three-Phase Transformer set for monitoring.'
2010\01\21@165447 by Forrest W Christian

flavicon
face
I'm needing to monitor voltage and the like for some sites with three
phase 120/208V (and perhaps whatever the 240V/???? version is - if
someone would like to enlighten me, that would be great).

I was thinking of doing a  PIC essentially directly-on-the-mains to do
the measurement and then serial the data out via an optocoupler - that
is, connect the pic 'dangerously' to the line , but I'm not particularly
eager to have to deal with all of the connect-a-pic-to-the-line issues -
still an option but want to explore others first.

The simplest thing I can come up with is to stick 3 120V transformers on
a circuit board connected such that you have 3 'line' and 1 'neutral'
inputs, and each of 3 transformers have their primary attached to a line
input and  the netural, and then read the (lower) secondary voltages.  
I've been told on here that this works, but I'm a bit skeptical of the
linearity and/or other issues.  I would guess this varies based on the
type of transformers.

Has anyone actually done this?   Or does anyone else have an idea on how
to measure AC line voltage safely in a way that can be read by a PIC?

-forrest

2010\01\21@170233 by Herbert Graf

picon face
On Thu, 2010-01-21 at 14:56 -0700, Forrest W Christian wrote:
> Or does anyone else have an idea on how
> to measure AC line voltage safely in a way that can be read by a PIC?

What about a hall effect sensor?

TTYL

2010\01\21@171622 by Dwayne Reid

flavicon
face
At 02:56 PM 1/21/2010, Forrest W Christian wrote:
{Quote hidden}

Yep - we do it both ways.

For most of the instrumentation that we sell to Power Generating
Utilities, we use 3- small Hammond 229A24 power transformers.  These
have 120 / 240 primary voltage options and we connect the primary
windings in series.  Also worth noting is that the voltage they are
measuring is typically 70Vac Line to Neutral (120Vac line-to-line).

Because we are operating the transformer with such reduced input
voltage (less than a third of rated input), linearity is absolutely
NOT a problem.

Our other stuff uses a PIC connected directly to the power line (and
powered from that line) with the data sent out serially via an
optical link.  Also works well.

Hope this helps.

dwayne

--
Dwayne Reid   <spam_OUTdwaynerTakeThisOuTspamplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax
http://www.trinity-electronics.com
Custom Electronics Design and Manufacturing

2010\01\21@194951 by Adam Field

flavicon
face
>>I'm needing to monitor voltage and the like for some sites with three
>>phase 120/208V (and perhaps whatever the 240V/???? version is - if
>>someone would like to enlighten me, that would be great).
>>

I'm working on a simple phase meter to measure PF. I've been following
Microchips App note AN521. Basically you connect to AC with a resistor
large enough to limit current below 50uA. Also, protection diodes to
the rails are nice, and necessary on some pins like TOCKI.

Admittedly, I'm only interested in zero crossings, so my input
waveform is very clipped. You would have to size the impedance so you
could measure the peak with the ADC for voltage measurement.

I wasn't sure this was a smart thing to do, but Microchip says it is.

2010\01\22@013138 by Forrest Christian

flavicon
face
On 1/21/10 3:15 PM, Dwayne Reid wrote:
> Because we are operating the transformer with such reduced input
> voltage (less than a third of rated input), linearity is absolutely
> NOT a problem.
>    
I take it the linearity issue really only occurs when the magnetics
saturate... which of course would be around the rated voltage.

I like the idea of running a transformer at half it's rated voltage (or
thereabouts) to keep the saturation from occuring.  One of my 'blind
areas' - I don't even think about running things outside their expected
voltage range.
> Our other stuff uses a PIC connected directly to the power line (and
> powered from that line) with the data sent out serially via an
> optical link.  Also works well.
>    
Figured it would, but my main issue were the mechanicals.   With the
transformer solution, I can put a terminal block on a PCB, and select a
transformer with the leads underneath, and use a DIN-rail 'track' to
mount the whole thing, and pretty much have the line voltage not very
easy to touch.   With a more complex pic circuit, it gets more difficult
since you are also having to hide the components.

Gave some thought to dipping the whole thing in that rubber dip stuff or
potting it in epoxy but that has it's own challeges.

Did you come up with some interesting way to handle the whole
'enclosure' issue?  Or is this used in an application where it doesn't
matter (I.E. in a switch panel or whatever the correct term for what we
used to call 'motor control panels' when I used to do automation in an
industrial environment).

-forrest

2010\01\22@042251 by Alan B. Pearce

face picon face
>Has anyone actually done this?   Or does anyone else
>have an idea on how to measure AC line voltage safely
>in a way that can be read by a PIC?

Microchip have some reference designs using an 18F with their new energy
monitoring chips. The 3 phase one is DS51643B, the single phase one is
http://ww1.microchip.com/downloads/en/DeviceDoc/51884a.pdf

2010\01\22@115636 by Dwayne Reid

flavicon
face
At 11:30 PM 1/21/2010, Forrest Christian wrote:
>On 1/21/10 3:15 PM, Dwayne Reid wrote:
> > Because we are operating the transformer with such reduced input
> > voltage (less than a third of rated input), linearity is absolutely
> > NOT a problem.
> >
>I take it the linearity issue really only occurs when the magnetics
>saturate... which of course would be around the rated voltage.

I didn't do the initial design on the PT (Potential Transformer) & CT
(Current Transformer) modules but I did check them out when I got
responsibility for supporting them.  The original designer seems have
done it right.


>Did you come up with some interesting way to handle the whole
>'enclosure' issue?  Or is this used in an application where it doesn't
>matter (I.E. in a switch panel or whatever the correct term for what we
>used to call 'motor control panels' when I used to do automation in an
>industrial environment).

The stuff we sell is primarily for diagnostic and trouble-shooting
purposes.  We mount the PT's & CT's in two different cases, depending
upon which version has been ordered.  The stand-alone units are
mounted in fiberglass cases made by Rolec.  Very heavy duty cases
indeed.  We use those same cases for our Field Voltage and Field
Current transducers.

The PT's & CT's that are part of the power analysis stuff are mounted
in a metal Hammond enclosure that contains the interface to the NI
data acquisition card that we use.  The enclosure is primarily
intended for interconnect purposes (lots and lots of binding-posts)
but is large enough to house the two (PT & CT) boards and the CT
card's power supply.

FWIW - the CT boards use very nice LEM closed-loop (servo) current
sensors.  The CT loops that Power Generating Utilities use run at 5A max.

We used to use our own data acquisition card, based on an old TI DSP
controller with code specifically written to capture very accurate
power-line-type data.  Eventually, the NI cards became fast enough
and powerful-enough that we could use them instead.  It meant a
complete re-write of the software (not by me, thank goodness) but the
end users seem really happy with it.

dwayne

--
Dwayne Reid   <.....dwaynerKILLspamspam@spam@planet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax
http://www.trinity-electronics.com
Custom Electronics Design and Manufacturing

2010\01\22@120639 by Marcel Duchamp

picon face
On 1/21/2010 10:30 PM, Forrest Christian wrote:
> On 1/21/10 3:15 PM, Dwayne Reid wrote:
>> Because we are operating the transformer with such reduced input
>> voltage (less than a third of rated input), linearity is absolutely
>> NOT a problem.
>>
> I take it the linearity issue really only occurs when the magnetics
> saturate... which of course would be around the rated voltage.
>
> I like the idea of running a transformer at half it's rated voltage (or
> thereabouts) to keep the saturation from occuring.  One of my 'blind
> areas' - I don't even think about running things outside their expected
> voltage range.

Power transformers are not designed to saturate at rated voltage.  One
that does would get very hot very quick.  When saturation occurs, the
impedance that the primary presents to the incoming line drops towards
zero - all you get is Rdc of the primary winding.  Input current then is
Vapplied/Rdc; usually a very big number of amps.

Although that is the case, as Dwayne points out, running at half the
rated value will promote longevity of the transformer as it will run
cooler as well.  Linearity should be excellent.

2010\01\22@132244 by Dwayne Reid

flavicon
face
At 10:05 AM 1/22/2010, Marcel Duchamp wrote:
>On 1/21/2010 10:30 PM, Forrest Christian wrote:
> > On 1/21/10 3:15 PM, Dwayne Reid wrote:
> >> Because we are operating the transformer with such reduced input
> >> voltage (less than a third of rated input), linearity is absolutely
> >> NOT a problem.
> >>
> > I take it the linearity issue really only occurs when the magnetics
> > saturate... which of course would be around the rated voltage.
> >
> > I like the idea of running a transformer at half it's rated voltage (or
> > thereabouts) to keep the saturation from occuring.  One of my 'blind
> > areas' - I don't even think about running things outside their expected
> > voltage range.
>
>Power transformers are not designed to saturate at rated voltage.  One
>that does would get very hot very quick.  When saturation occurs, the
>impedance that the primary presents to the incoming line drops towards
>zero - all you get is Rdc of the primary winding.  Input current then is
>Vapplied/Rdc; usually a very big number of amps.

I find that the linearity of modern power transformers is not perfect
when you are running the transformer at its rated input voltage.  Low
cost transformers are MUCH worse than good transformers - they seem
to be partly into saturation when lightly loaded.  In other words,
the transformer runs hot to the touch while running idle and its
temperature drops when the transformer is running at rated load.

The Hammond part number I mentioned doesn't get very warm while
running idle at rated input voltage.  However, we actually run it at
about 1/3 of its rated voltage when we use for instrumentation purposes.

dwayne

--
Dwayne Reid   <dwaynerspamKILLspamplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax
http://www.trinity-electronics.com
Custom Electronics Design and Manufacturing

2010\01\22@135139 by Forrest Christian

flavicon
face
On 1/22/10 10:05 AM, Marcel Duchamp wrote:
> Power transformers are not designed to saturate at rated voltage.
Thanks for pointing this out.  I somehow had mixed up a 'normal' power
transformer and the ferroresonant CVT's that I used to work with a lot
(aka sola voltage regulators) in my head.   The CVT's are designed to
'saturate' (overly simplistic) to produce a regulated output.  Yes, the
saturation isn't exactly the same as what I was thinking, but explains
my confusion.

I've got some PCB mounted 240V primary transformers on their way.  I'll
be spending some quality time with them and a variac to see how well
this works.

-forrest

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...