Searching \ for '[EE]: optically isolated mains voltage acquisition' 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=voltage
Search entire site for: 'optically isolated mains voltage acquisition'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: optically isolated mains voltage acquisition'
2002\09\21@150130 by rusque (Listas)

face
flavicon
face
Hello,

   I'm trying to make a circuit to do mains voltage acquisition using the
ADC of a PIC and a pair of optos, but I'm having a hard time trying to make
it work correctly. The circuit is:


                             4N25    VDD    VDD
                           ,------,   |      |
    Phase--------100K-+----|A    B|   |     1K
100-240VAC             |    |     C|---'      |
  Neutral---------------+--|K    E|---,      |
                      | |  '------'   |      |
                      | |             +------+---to PIC ADC input
                      | |  ,------,   |      |
                      | '--|A    B|   |      |
                      |    |     C|---'     1K
                      '----|K    E|---,      |
                           '------'   |      |
                                     GND    GND

   I'm getting a very distorted and noisy sine on the circuit output. Also,
I have a problem that the LEDs on the optos aren't conducting in the
zero-crossings. Someone have made a circuit like this?

   Any sugestions are very welcome.

   Thank you,

   Brusque

-----------------------------------------------------------------
Edson Brusque                 C.I.Tronics Lighting Designers Ltda
Research and Development               Blumenau  -  SC  -  Brazil
Say NO to HTML mail                          http://www.citronics.com.br
-----------------------------------------------------------------

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


2002\09\21@174450 by Ray Gallant

flavicon
face
source=I'm sure you'll get lot of better responses so I will just monitor for now.  Would you tell me what results you are looking for?  i.e. just voltage level, or do want peak, KW, PF, KVA, KWH or current also (all single phase?).  In my voltage front end, I have a circuit that includes a PT, a dual op-amp (hi common mode ratio), one pot for cal and a few discretes, then connected to the pic's 8 bit A/D.  I use internal Vref = 5V.  The PT reduces the AC and gives isolation.  My PT has additional secondary windings for my power supplies.  My results equal true RMS!  I have the fastest RMS routine (assembler) around!  Problem with this design is that the op-amp requires + and - 12VDC as rails.  
--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.


2002\09\21@183316 by rusque (Listas)

face
flavicon
face
Hello Ray,

>I'm sure you'll get lot of better responses so I will just monitor for now.
>Would you tell me what results you are looking for?  i.e. just voltage
level,
> or do want peak, KW, PF, KVA, KWH or current also (all single phase?).

   the first "incarnation" of this device just need to know RMS voltage
level. But I'll expand it in the future for another readings (peak, KW etc).

>In my voltage front end, I have a circuit that includes a PT, a dual op-amp
> (hi common mode ratio), one pot for cal and a few discretes, then
connected
> to the pic's 8 bit A/D.  I use internal Vref = 5V.  The PT reduces the AC
> and gives isolation.  My PT has additional secondary windings for my power
> supplies.  My results equal true RMS!  I have the fastest RMS routine
> (assembler) around!  Problem with this design is that the op-amp requires
> + and - 12VDC as rails.

   What is PT? Power transformer?

   I have already done this with a current transformer wired as:

       VDD
        |
       10K
        |
        +---PICADC
        |
       10K
        |
       Xformer-secondary-with a 100R resistor in parallel
        |
       GND

   And it's working very well besides it's simplicity. I'm thinking about
doing this with transformers for voltage measuring, but optos would be more
elegant/cheap and would waste less space.

   Best regards,

   Brusque

-----------------------------------------------------------------
Edson Brusque                 C.I.Tronics Lighting Designers Ltda
Research and Development               Blumenau  -  SC  -  Brazil
Say NO to HTML mail                          http://www.citronics.com.br
-----------------------------------------------------------------

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


2002\09\21@194902 by Jinx

face picon face
That circuit is more useful for zero-rossing detection because of
the high gain The 4N25 data sheet will have linear applications

http://www.fairchildsemi.com/ds/4N/4N25.pdf

http://www.vishay.com/docs/PTgeneric.pdf

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


2002\09\21@224238 by Jinx

face picon face
After looking thrrough the previously posted optocoupler pdfs
I relaise they probably don't give you enough information. Sorry,
assumed they would

As noted here, the problem is that the LEDs are non-linear

(linear optos info)

http://www.ssousa.com/pdf/an060.pdf

http://www.taosinc.com/pdf/til300.pdf

You'd need to match the non-linearity in the opto's transistor if
you intend to use them on the AC for diagnostic purposes

Possibly you could use a 555 in voltage-controlled PWM mode. I
think that's how I've seen IR headphone transmitters do it. Thought
there was a magazine constructional article lying around here in
easy reach but can't seem to locate it just now. Possibly it could be
done by FM of an IR signal

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


2002\09\22@014600 by Russell McMahon

face
flavicon
face
> I'm trying to make a circuit to do mains voltage acquisition using the
> ADC of a PIC and a pair of optos, but I'm having a hard time trying to
> make it work correctly. The circuit is:


The "trick" in transferring analog signals via optos is either to linearise
the opto OR know its response curve OR (by far the easiest) use specially
designed optos which have a single transmitter and two matched receivers
which are both illuminated by the same transmit LED. One receiver is placed
in a  feedback loop on the sending side and a matching circuit is
used by the other receiver on the receiving side. The "receiver-receiver"
tracks the behaviour of the "transmit-receiver" and thereby replicates the
input condition.

To save explaining this and all its variants at length I'll cut to the chase
and suggest you look at the Siemens / Infineon IL300 as one example of a
purpose built part. Infineon's app note 50 is a 17 page PDF that shows a
number of ways of using the IL300. (I can send a copy offlist if you can't
find it). While it is normal to use an opamp in the input circuit their
simplest circuit requires two transistors on the input side and a single
resistor on the output.



       Russell McMahon

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


2002\09\22@073741 by rusque (Listas)

face
flavicon
face
Hello Russell,

> The "trick" in transferring analog signals via optos is either to
linearise
> the opto OR know its response curve OR (by far the easiest) use specially
> designed optos which have a single transmitter and two matched receivers
> which are both illuminated by the same transmit LED. One receiver is
placed
> in a  feedback loop on the sending side and a matching circuit is
> used by the other receiver on the receiving side. The "receiver-receiver"
> tracks the behaviour of the "transmit-receiver" and thereby replicates the
> input condition.

   maybe a linearizing table could be created on the microcontroller.
Probably the less expensive solution.

> To save explaining this and all its variants at length I'll cut to the
chase
> and suggest you look at the Siemens / Infineon IL300 as one example of a
> purpose built part. Infineon's app note 50 is a 17 page PDF that shows a
> number of ways of using the IL300. (I can send a copy offlist if you can't
> find it). While it is normal to use an opamp in the input circuit their
> simplest circuit requires two transistors on the input side and a single
> resistor on the output.

   Ok, Jinx already mentioned the TIL300. And I have no problems
undertanding it, but I'm thinking a transformer would be simpler.

   I'm trying to find the Infineon's AN50, but can't find where are the ANs
http://www.infineon.com. Please, send it to me off-list.

   Thank you very much,

   Brusque

-----------------------------------------------------------------
Edson Brusque                 C.I.Tronics Lighting Designers Ltda
Research and Development               Blumenau  -  SC  -  Brazil
Say NO to HTML mail                          http://www.citronics.com.br
-----------------------------------------------------------------

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


2002\09\22@095948 by Chris Loiacono

flavicon
face
Ed, my experience concurs with Russel's thoughts on this.

If you want to go the easy wa out, a few companies, such as CR Magnetics
http://www.crmagnetics.com produce true RMS voltage transducers for single and
three phase lines.

While you can try to linearize the curves of the emitters in various optos,
you will next find that you can't  purchase a pair that are exactly alike,
and I promise you will not be pleased with the resulting frustration. (Been
there, done that, bought the tee-shirt and wore it out.)

What Russel calls purpose-built parts is the only way to do this without
magnetics.
If you must continue with the circuit sheme below, you can try MCT62's, but
they have a pair of LED's in inverse parallel. While I have found the pairs
to be *fairly* consistent, they are high gain devices (high CTR) and you
will have the difficulty that Jinx describes because of the shape of the
resulting output curve.

I have used a few methods of sensing AC load current on the fly, but have
never set out to detect the line voltage with any degree of accuracy. Two
ideas come to mind, and I may be off, but you asked for "any ideas", right?

1. With a known Vp range you can measure current with a current sense
resistor and by a simple application of Ohm's law derive the RMS V.

2. Do the same with a CT and rectify the output before input to the ADC

3. Some variation or combination of the above....

{Quote hidden}

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


2002\09\22@153212 by Ray Gallant

flavicon
face
source=I really like Russell's IL300.  I have used it for that specific application as well as on a 3 phase power monitor.  I did the following all on the high side of the opto:
Step A:  create + & - DC rails voltage suplies for op-Amps
Step B:  create Vref for opto of about -2.5 VDC to be calibrated, includes op-Amp & pot.
Step C:  create voltage divider & pot for each phase A, B, C
Step D:  create voltage divider/op-Amp circuit for delta of phase A & phase B, includes and pot for cal.
Step E:  create Opto circuit with op-amps and a pot for gain.
Step F:  create multiplexer (mux) circuit for 4 channels (phA, phB, phC & Delta phA->phC.  (make note of the MUX's slew rate!)
Did the following on low side.
step G:  Included two 4n35 (basic opto's) where their inputs are from the uP and the outputs go to the MUX's channels select.  The final result of the unit was 0.75% accurate KVA each phase.  It supports 600, 480, 240, 115, delta or 4 wire or single or dual phase with max CT's of 5000:5.  It was challenging for me and I finally developed a constant 5Amp DC current supply to cal the section I did not talk about, the CT inputs.  The best part of this circuit is that is doesn't need PT's (potential transformer) and needs only one IL300 so variances among the same opto batches are calibrated out.  Need a HV constant AC voltage source for spec checking your unit?, just ask!  It's a great chip!  Hope this wasn't too long a message.  Best Regards, Ray

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


2002\09\23@070316 by Alan B. Pearce

face picon face
>The "trick" in transferring analog signals via optos is either to linearise
>the opto OR know its response curve OR (by far the easiest) use specially
>designed optos which have a single transmitter and two matched receivers
>which are both illuminated by the same transmit LED. One receiver is placed
>in a  feedback loop on the sending side and a matching circuit is

There is another way - pass the linear input signal through a voltage to
frequency converter. Then the only signal going across the opto barrier is a
square wave, which can easily have the frequency to voltage recovered from
it. If very good linearity is required this is probably the most linear way
of transferring across an isolation barrier.

--
http://www.piclist.com hint: To leave the PICList
EraseMEpiclist-unsubscribe-requestspam_OUTspamTakeThisOuTmitvma.mit.edu


2002\09\23@231214 by Mike Singer

picon face
May be I'm missing something, but guys, why not to stick
PIC or AVR into the environment directly without opto,
feeding it through divider or some 300v DC/ 5v DC converter,
STM Viper for example. Then you could use IR or RF to
transfer data.

Regards.
Mike.

Ray Gallant wrote:
{Quote hidden}

--
http://www.piclist.com hint: To leave the PICList
piclist-unsubscribe-requestspamspam_OUTmitvma.mit.edu


2002\09\24@075021 by Chris Loiacono

flavicon
face
It takes thousands of volts worth of isolation to protect the micro's input
from the line transients that occur, which, BTW are amazaingly quick and
sneak right past MOV's. While a DC/DC converter could conceivably filter
these, it would be a very costly method.

Optical isolators are common, fail-safe if applied properly while being
inexpensive and available in a wide variety of output types to fit any
application. So cost, availability and reliability are good reasons for
using them.

{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\24@133102 by Mike Singer

picon face
OK. How about massive metal bar with the cave. Place
there some low power microcontroller with battery. Connect
PCB ground to the metal bar only in one point. Place
fast Zener diode between this point and input and so on.
Let this microcontroller communicate to outer world by IR.
Jinx said LEDs start emitting at micro amperes, if I'm not
mistaken.
This way you could get live analog feelings of what is going
there. Much better then diminished ones by optical isolators.
(some association with rubber isolation not to be mentioned
here: women in PicList)

Mike.

Chris Loiacono wrote:
{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\24@135447 by Ray Gallant

flavicon
face
source=Typical isolation factor is twice the rated voltage plus 1000Volts (i.e. for 240V = 1480Volts).  I would recommend 8 or at least at the very least 4 samples per quadrant for 50/60Hz especially if you ever consider KVA results someday.

Could you offer more specific details on?
> PIC or AVR into the environment directly without opto,
> feeding it through divider or some 300v DC/ 5v DC converter,
> STM Viper for example. Then you could use IR or RF to
> transfer data.
>
I don't recommend RF in that environment unless you add a few orders of filtering.  Also could you please converse on the incoming peak-peak value, instantaneous spikes as well phase shift handling method in your example.  I'd like to try your concept if feasible.
Tx.  Sounds like a nice challenge.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\24@141459 by Peter L. Peres

picon face
part 1 2 bytes content-type:TEXT/PLAIN; charset=US-ASCII


part 2 2555 bytes content-type:APPLICATION/PDF; name="linear_opto.pdf" (decode)

part 3 154 bytes
--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\24@141851 by Peter L. Peres

picon face
*>Jinx said LEDs start emitting at micro amperes, if I'm not
*>mistaken.

I ran a standard red led as voltage reference today using 50k from 3V.
This is 30uA. The light is visible in normal light (~400lux).

There are several easy ways to linearise optoisolators, but they all
involve using a second opto. A scheme that needs no linearisation is a diy
led+CdS photoresistor one (can't buy there ready made afaik). It has the
usual problems CdS has but it may be accurate enough. I have done this and
it works fine, the output resistance is very linear with LED current. Used
it as 4-20mA converter too.

Peter

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\24@143941 by Chris Loiacono

flavicon
face
> Subject: Re: [EE]: optically isolated mains voltage acquisition

Peter,
This looks too simple to work well. But then again, the same applies to me.
:)
Are you in a good enough mood to offer a little more explanation of how this
and the CdS idea work???

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\24@155516 by Peter L. Peres

picon face
On Tue, 24 Sep 2002, Chris Loiacono wrote:

*>> Subject: Re: [EE]: optically isolated mains voltage acquisition
*>
*>Peter,
*>This looks too simple to work well. But then again, the same applies to me.
*>:)
*>Are you in a good enough mood to offer a little more explanation of how this
*>and the CdS idea work???

The transistor circuit is an adaptation from a book schematic that uses an
opamp and two ocs. The transistor circuit can be considered a inverting
'Norton' opamp for this analysys. Its gain reduces any nonlinearity in the
transfer characteristic of the OC, by this gain (beta). The second OC
should be run at a comparable voltage differential for the output to be
accurate but it should be close under all conditions.

If ocs with simple photodiodes would be available and inexpensive then
this gymnastic would be unnecessary.

The CdS + LED circuit relies on the respective curves of CdS resistance vs
light and LED emittance vs current. The device I built used a green led
and a 12.7mm dia. CdS photoresistor in a black tube with black epoxy
sealing (Devcon I think). It was better than 2% over one order of
magnitude.

Peter

PS: The o.p. could use a capacitive divider too. This gives voltage
reduction + galvanic separation in one shot. A standard 2nF/1kV X2 ceramic
cap has Zc ~= 1.5Megs at 50Hz. Just what the doctor ordered to get 2Vpp
out using an inverting opamp with Rfb ~= 7.5k, or use a proper
differential divider with 2nF/200nF from each circuit (L and N) to circuit
gnd and amplify the difference voltage in an opamp.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\24@165012 by Chris Loiacono

flavicon
face
So, simple IS good. Very cool! Thanks - I will enjoy playing with this one
and I promise to share anything new that comes out of it.\
c

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2002\09\25@070729 by Jinx

face picon face
----- Original Message -----
From: Peter L. Peres <KILLspamplpKILLspamspamACTCOM.CO.IL>
To: <RemoveMEPICLISTTakeThisOuTspamMITVMA.MIT.EDU>
Sent: Wednesday, September 25, 2002 6:11 AM
Subject: Re: [PICLIST] [EE]: optically isolated mains voltage acquisition


Peter, what is mains in your pdf ? 120V or 220V ?

--
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


2002\09\25@110410 by Peter L. Peres

picon face
On Wed, 25 Sep 2002, Jinx wrote:

*>----- Original Message -----
*>From: Peter L. Peres <spamBeGoneplpspamBeGonespamACTCOM.CO.IL>
*>To: <TakeThisOuTPICLISTEraseMEspamspam_OUTMITVMA.MIT.EDU>
*>Sent: Wednesday, September 25, 2002 6:11 AM
*>Subject: Re: [PICLIST] [EE]: optically isolated mains voltage acquisition
*>
*>
*>Peter, what is mains in your pdf ? 120V or 220V ?

220V since peak voltage is 310V

Peter

--
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


More... (looser matching)
- Last day of these posts
- In 2002 , 2003 only
- Today
- New search...