[sci.electronics] Need analog opto-isolator or equiv.

elmquist@nachos.SSESCO.com (Chris Elmquist) (09/25/90)

Does anyone have a nifty circuit that will replicate a 0v to 5v DC
signal across an opto-isolated path..??  I need to read a meter
in a hostile environment with an A/D in a PC.  I don't want the
PC to get near this meter... this meter will be read, maybe, 10 or
15 times a second so frequency response is not critical...

I've thought about serial A/D and D/A chips...  but I haven't
found any that use the same data format over the serial link...
maybe I missed something.

I've also though about a voltage->frequency ... frequency->voltage
converter and this seems workable...  maybe someone has a circuit
like this.

Any ideas would be appreciated.

Thanks--  Chris
elmquist@nachos.ssesco.com

larry@kitty.UUCP (Larry Lippman) (09/25/90)

In article <222@nachos.SSESCO.com>, elmquist@nachos.SSESCO.com (Chris Elmquist) writes:
> Does anyone have a nifty circuit that will replicate a 0v to 5v DC
> signal across an opto-isolated path..??  I need to read a meter
> in a hostile environment with an A/D in a PC.  I don't want the
> PC to get near this meter... this meter will be read, maybe, 10 or
> 15 times a second so frequency response is not critical...
> ... 
> Any ideas would be appreciated.

	Use an "isolation amplifier" specifically designed for this
purpose, such as one manufactured by Analog Devices, Burr-Brown, etc.
You can probably obtain an encapsulated device providing 1,000 volt
isolation for as little as $ 25.00 in small quanitites.

	My advice is to not even think of designing your own analog
isolation circuit for a one-of-a-kind application.

Larry Lippman @ Recognition Research Corp.  "Have you hugged your cat today?"
VOICE: 716/688-1231   {boulder, rutgers, watmath}!ub!kitty!larry
FAX:   716/741-9635                  {utzoo, uunet}!/      \aerion!larry

grayt@spock (Tom Gray) (09/25/90)

Newsgroups: sci.electronics
Subject: Re: Need analog opto-isolator or equiv.
Summary: 
Expires: 
References: <222@nachos.SSESCO.com>
Sender: 
Reply-To: grayt@halligan (Tom Gray)
Followup-To: 
Distribution: 
Organization: Mitel. Kanata (Ontario). Canada.
Keywords: 

In article <222@nachos.SSESCO.com> elmquist@nachos.SSESCO.com (Chris Elmquist) writes:
>Does anyone have a nifty circuit that will replicate a 0v to 5v DC
>signal across an opto-isolated path..??  I need to read a meter
>in a hostile environment with an A/D in a PC.  I don't want the
>PC to get near this meter... this meter will be read, maybe, 10 or
>15 times a second so frequency response is not critical...
>
>I've thought about serial A/D and D/A chips...  but I haven't
>found any that use the same data format over the serial link...
>maybe I missed something.
>
>I've also though about a voltage->frequency ... frequency->voltage
>converter and this seems workable...  maybe someone has a circuit
>like this.
>
>Any ideas would be appreciated.
>
>Thanks--  Chris
>elmquist@nachos.ssesco.com


 The most straight forward way to do this would would be to linearize an
opto-coupler. You can do this by using two opto-couplers. One opto-coupler
is used to create a feedback loop around an op amp. The op amp is
used to create a constant current source in the feedback path - a typical
cook book op amp circuit.

The input porion of the opto-coupler is driven by a current which you wish
to repilcate across some isolating barrier. It is connected to the current sourc
source and to ground. The ungounded input of the opto is connected to the
inverting input of an op amp. The output of the op amp is connectd
to ground through the output section of the opto-coupler.
A feedback lop is set up with the optical connection of the opto-coupler
and the circuit is stable.

Now the interesting part occurs. Connect the light emiiting portion
of another opto in series with the output of the op amp and
the light emiitoing portion of the first opto. The feedback of
the op amp is making sure that the current though the light emitting
prtion of the first opto is usfficent to produce the input current
in its light receiving portion. Similarly the same current is
flowing through the light emitting portion of the second opto
as that of the first. Thus the current in the light receiving
prtion of the second opto will be the same as the current in the
 light receivng portion of the first opto. However this is the 
input current which is supposed to be transferred across
the barrier.

The op amp feed back calculates the voltages required to
create a liner current transfer mechanism from two
opto's. As long as the transfer characteristics of
the two opto's are amtched, there will be a linear
tranfer characteristic across the barrier. 

henry@zoo.toronto.edu (Henry Spencer) (09/25/90)

In article <222@nachos.SSESCO.com> elmquist@nachos.SSESCO.com (Chris Elmquist) writes:
>Does anyone have a nifty circuit that will replicate a 0v to 5v DC
>signal across an opto-isolated path..??  ...
>I've thought about serial A/D and D/A chips...  but I haven't
>found any that use the same data format over the serial link...

Of course, if you're willing to throw in a few more optoisolators, you
can just use parallel-i/o converters, which generally agree on the data
format. :-)
-- 
TCP/IP: handling tomorrow's loads today| Henry Spencer at U of Toronto Zoology
OSI: handling yesterday's loads someday|  henry@zoo.toronto.edu   utzoo!henry

paulh@cimage.com (Paul Haas/10000) (09/27/90)

In article <4054@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes:
]In article <222@nachos.SSESCO.com>, elmquist@nachos.SSESCO.com (Chris Elmquist) writes:
]> Does anyone have a nifty circuit that will replicate a 0v to 5v DC
]> signal across an opto-isolated path..??  I need to read a meter
]> in a hostile environment with an A/D in a PC.  I don't want the
]> PC to get near this meter... this meter will be read, maybe, 10 or
]> 15 times a second so frequency response is not critical...
]> ... 
]> Any ideas would be appreciated.
]
]	Use an "isolation amplifier" specifically designed for this
]purpose, such as one manufactured by Analog Devices, Burr-Brown, etc.
]You can probably obtain an encapsulated device providing 1,000 volt
]isolation for as little as $ 25.00 in small quanitites.
]
]	My advice is to not even think of designing your own analog
]isolation circuit for a one-of-a-kind application.
]
You can also put a voltage controlled oscillator in the hazardous area
and use it to drive an opto-isolater.  Or choose your favorite way to
convert an analog signal to a digital one.
---
Paul Haas  paulh@cimage.com
(313) 677-2412

rick@sesi.UUCP (Rick Lotoczky) (09/28/90)

> In article <222@nachos.SSESCO.com> elmquist@nachos.SSESCO.com (Chris Elmquist) writes:
> >Does anyone have a nifty circuit that will replicate a 0v to 5v DC
> >signal across an opto-isolated path..??  ...
> >I've thought about serial A/D and D/A chips...  but I haven't
> >found any that use the same data format over the serial link...
> 
Another way to do it would be to send a pulse width modulated signal through
an opto-isolator to a low pass filter on the other end.  The LPF will convert
the PWM to an eqivalent analog signal.  You will most likely have to linearize
this circuit, but that can be easily done in software.  I don't know what kind
of accuracy you need, but 0.5% or so should be attainable.

rick

-- 
Structured Electronic Systems, Inc.	| ...uunet!mailrus!sharkey!cfctech
650 Red Oak Lane                        | !rphroy!trilby!sesi!rick
Rochester, MI 48063			|
(313) 652 4809				|