Need a 0 to 5VDC Analog Isolation Circuit 2

[rica]

I am designing a isolation circuit for an analog signal that will range from 0 to 5VDC and is referenced to a 28VDC power return. The other side of the isolation circuit will also be 0 to 5VDC referenced to a logic return. The isolation voltage needs to be greated than 600V. The signal accuarcy requirement is 4%. Operating temp range is -40 to +85C. Board space is very limited.

I was planning to use an HP HCNR200 analog optocoupler with an op amp controlling the LED (on the 28V side) and an op amp interfaced to the phototransistor on the output (logic side). I have +5VDC, +15VDC, and -15VDC available on the logic side. 28VDC is the only voltage available on the power side.

Does anyone know of a more integrated device or approach to analog isolation?

 

[Lewish]

You ain't gonna get there that way with 4% accuracy.

Use the optocouler for isolation. You can get them all the way up to 2500volts iso quite cheaply.

Then use a V-to-F on the one LED side and an F-to-V on the other side. You can pretty readily get down to 1% accuracy.

 

[IRstuff]

Analog Devices makes a variety of isolation amplifiers:

http://www.analog.com/Analog_Root/p...6metaId%3D202%26resourceWebLawID%3D13,00.html

This one has 4% max error.

TTFN

 

[biomedicalEng]

try Analog devices AD202, it can do, what exactly you want.

but the only problem is it's cost, it's a lil bit expensive.

the another way s using linear opto couplers such as CP Clare's LIA100, they send 2 or 3 samples of their ICs if you want.

I think HP no longer produces any semiconductor including HCNR200. they sold all of their product lines to agilent technology.

If you want, I can send you some opto Isolation circuit schematics.

Kasra_Ravanbakhsh@hotmail.com

 

[pebe]

How about chopping your analogue waveform at a frequency that can handle its highest frequency component.

Then AC couple the resulting sqare wave with a capacitor to your logic circuitry and DC restore to your logic return.