Maker Pro
Maker Pro

Analog Isolator

H

hrh1818

Jan 1, 1970
0
Any suggestions for an analog isolator? Requirements are 0 to 5 VDC
in, 0 to 5 VDC out, 1% of full scale accuracy, monotonic output,
output ripple less 0.25% of full scale, 0 to 10 Hz frequency response,
input impedance greater than 10,000 ohms, load on output is 1 megohm,
quantity one.

Howard.
 
J

Joerg

Jan 1, 1970
0
hrh1818 said:
Any suggestions for an analog isolator? Requirements are 0 to 5 VDC
in, 0 to 5 VDC out, 1% of full scale accuracy, monotonic output,
output ripple less 0.25% of full scale, 0 to 10 Hz frequency response,
input impedance greater than 10,000 ohms, load on output is 1 megohm,
quantity one.

Check out these:
http://www.analog.com/UploadedFiles/Data_Sheets/AD215.pdf

Make sure you sit down before looking at the price. Digikey has them in
stock, singles cost over $80.
 
J

Joerg

Jan 1, 1970
0
ian said:
Not to worry though, there are several application notes floating around the
web for analogue isolators using ordinary opto-isolators.

Yes, but it depends on Howard's expertise, available time and whether
it's for production. For a one-off unit I'd just buy it.
 
S

Scott Seidman

Jan 1, 1970
0
Yes, but it depends on Howard's expertise, available time and whether
it's for production. For a one-off unit I'd just buy it.

It does have the advantage of providing isolated power.
 
J

Joerg

Jan 1, 1970
0
Scott said:
It does have the advantage of providing isolated power.

Yes, that's a neat feature. Although it's quite easy to do. Just take
the output of a logic gate or if more oomph is needed take a gate driver
chip and capacitively couple into one side of a LAN transformer, rectify
and filter on the other side. In a pinch just wind some wire on a
Fair-Rite #77 core.

I have done stuff like "borrowing" a LAN transformer out of a hub that
didn't have all ports occupied (after getting client permission...). I
bet it still has duct tape on that port :)
 
J

Joerg

Jan 1, 1970
0
Jamie said:
You know that make them ready to go
din rail mountable.
They aint cheap however.!

If you're looking for something to make then
use the Photo feed back optical couplers with
2 supplies and 2 separate op-amps. those track
as close as the tolerance of the feed back diode.


Or chop, send over a toroid and clamp to restore the DC-level on the
other side. That what I always do. No feedback :)
 
J

Jamie

Jan 1, 1970
0
hrh1818 said:
Any suggestions for an analog isolator? Requirements are 0 to 5 VDC
in, 0 to 5 VDC out, 1% of full scale accuracy, monotonic output,
output ripple less 0.25% of full scale, 0 to 10 Hz frequency response,
input impedance greater than 10,000 ohms, load on output is 1 megohm,
quantity one.

Howard.
You know that make them ready to go
din rail mountable.
They aint cheap however.!

If you're looking for something to make then
use the Photo feed back optical couplers with
2 supplies and 2 separate op-amps. those track
as close as the tolerance of the feed back diode.
 
J

Jamie

Jan 1, 1970
0
Joerg said:
Or chop, send over a toroid and clamp to restore the DC-level on the
other side. That what I always do. No feedback :)
Well yeah, a PWM scheme works. We have some ready made units from KB
that were originally designed to be used as a signal isolator to drive
their cheap DC drives which don't have isolation from the bridge and mains.
I find them all over the place in machines put there before my time
at my current place of employment.

http://webpages.charter.net/jamie_5"
 
H

hrh1818

Jan 1, 1970
0
Check out these:http://www.analog.com/UploadedFiles/Data_Sheets/AD215.pdf

Make sure you sit down before looking at the price. Digikey has them in
stock, singles cost over $80.

Thanks Joerg the AD215 looks very promising. I was looking for an
analog isolator to replace a $349 isolator sold by National
Instruments. Their isolator also required a $349 case. So $80 plus is
a bargain. However, the isolator will be used by a novice and I am
wondering if wiring to a SIP socket. adding gain resistors and a power
supply might be to challenging. I would prefer an isolator that came
in a package with screw terminals even it if costs twice what the
AD215 costs. Once again this is a one of a kind item.

Howard
 
J

Joerg

Jan 1, 1970
0
hrh1818 said:
Thanks Joerg the AD215 looks very promising. I was looking for an
analog isolator to replace a $349 isolator sold by National
Instruments. Their isolator also required a $349 case. So $80 plus is
a bargain. However, the isolator will be used by a novice and I am
wondering if wiring to a SIP socket. adding gain resistors and a power
supply might be to challenging. I would prefer an isolator that came
in a package with screw terminals even it if costs twice what the
AD215 costs. Once again this is a one of a kind item.

In that case take a look at companies like this:
http://www.measurementcomputing.com/cbicatalog/directory.asp?dept_id=251&top_id=25&dept_name=Analog

Much of this will be rack mounted though but in a heavy duty industrial
environment that might actually suit better. Makes it easier to swap if
something got hit by the forklift ;-)
 
H

hrh1818

Jan 1, 1970
0
You know that make them ready to go
din rail mountable.
They aint cheap however.!

Please name a couple of suppliers of DIN rail mountable analog
isolators.
I want to give the user a choice between a more user required wiring
type like the
AD215 and a DIN rail mounted unit.

Thanks Howard
 
J

Joerg

Jan 1, 1970
0
J

Jamie

Jan 1, 1970
0
hrh1818 said:
Please name a couple of suppliers of DIN rail mountable analog
isolators.
I want to give the user a choice between a more user required wiring
type like the
AD215 and a DIN rail mounted unit.

Thanks Howard
http://www.wici.com/



--
"I'd rather have a bottle in front of me than a frontal lobotomy"

"Daily Thought:

SOME PEOPLE ARE LIKE SLINKIES. NOT REALLY GOOD FOR ANYTHING BUT
THEY BRING A SMILE TO YOUR FACE WHEN PUSHED DOWN THE STAIRS.
http://webpages.charter.net/jamie_5"
 
Well yeah, a PWM scheme works.

PWM works, but you need a reference voltage source on the sending side
with which to compare the incoming voltage and and another voltage
reference on the receiving side to turn the mark-to-space ratio back
into an absolute voltage.

Joerg's transformer transmits an absolute voltage.

I once put together a scheme around a transformer with three windings
- you drove one winding and used local feedback to get exactly the
desired voltage on second (local) winding, and monitored the voltage
across the third (isolated) winding with a high-input impedance
follower. Since there was very little current flowing through the
second and third windings, the voltage was transferred very accurately
(and cheaply).

You did need a sample and hold on the output to hold the output
voltage while you wound down the flux through the core from time to
time, so the bandwidth wasn't great.
 
J

Joerg

Jan 1, 1970
0
PWM works, but you need a reference voltage source on the sending side
with which to compare the incoming voltage and and another voltage
reference on the receiving side to turn the mark-to-space ratio back
into an absolute voltage.

Not really. I never had voltage references on either side. Ok, except
for the path that transferred isolated power. And even then only very
occasionally.

Joerg's transformer transmits an absolute voltage.

On the other side you can clamp the DC level and, voila, there's the
signal again.

I once put together a scheme around a transformer with three windings
- you drove one winding and used local feedback to get exactly the
desired voltage on second (local) winding, and monitored the voltage
across the third (isolated) winding with a high-input impedance
follower. Since there was very little current flowing through the
second and third windings, the voltage was transferred very accurately
(and cheaply).

You did need a sample and hold on the output to hold the output
voltage while you wound down the flux through the core from time to
time, so the bandwidth wasn't great.

I've transferred signals with spectral content up to 50MHz that way.
 
H

Hammy

Jan 1, 1970
0
Check out these:
http://www.analog.com/UploadedFiles/Data_Sheets/AD215.pdf

Make sure you sit down before looking at the price. Digikey has them in
stock, singles cost over $80.

The AD202 sounds like it might work and it's $40.00 cheaper.

http://www.analog.com/UploadedFiles/Data_Sheets/AD202_204.pdf

What are they using to obtain such high common mode ranges,with so
high an input impeadance? All the difference amp's I've seen with high
common mode ranges have kohm input impeadances.
 
Top