DC to 25Hz Isolated amplifier - 10KV Dielectric strenght

Hey Bloggs, Why is it you must attack the person asking for help rather than critic the proposed circuit and possibly be even suggest what you think is a better solution? You know, be helpful. Mikek

I think he maybe meant to use a discrete LED and a discrete photodiode, with a big air gap between. Or a light pipe. Or maybe plastic fiber optics.

PWM can be extremely accurate. People used to do process control math, like analog multiplication and square roots and stuff, with PWM circuits.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Sure - up to a point. Sigma-delta A/D converters run on PWM.

But what you are transferring is a precise ratio, not a voltage, so the output is only as accurate as the two voltage references on the input and output sides.

Transformer-based isolators don't need the voltage references.

--
Bill Sloman, Sydney

No opto device on the market could fulfill that requirement. An LED shooting across one inch of free space to a phototransistor may work dandy. Or, since you didn't specify size, how about an LED, a mile of fiber optic cable, and a phototransistor?

PWM-ing a transformer or LED/phototransistor pair should work, then.

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

I have much problem with opto device in the past,they derate with temperature, did not maintain initial spec with ageing ...etc, etc, the reason why there is a LED in retro-action in the good IL300 you mentioned.

PWM the transformer should work certainly, please take a lokk at mine if you have time, i believe it's more simple than a PWM based solution.

Habib.

Looks good as drawn, but not useful in practical situation. That RC from output low side tied to driver AKA input ground kills any possible HV use.

Making it run at ~2% for zero and ~98% for full scale will avoid nonlinearity at the edges due to limited slew rate--the slew effect is the same for all duty cycles, so it just turns into an offset.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

This R//C is a mandatory requirement to improve EMC ruggedness in all industry class equipements (in that case it will be installed in a power plant control system). It is a self healing capacitor (Class Y2). If you considerer the intrinsic inter-winding capacitor, then this R//C helps the high energy surge to flow to ground from the source. Believe me, the IEC61000-4.5 and IEC 61000-4-5 are the mothers of all EMC tests !

Anyway thanks for the topic, Habib.

Lemme see... Want isolated supply like that as shown,and wish to use it (say) +10KV. That poor resistor would have to be rated at 12KV breakdown minimum, and the power (mmm...apple whatsie or eye-eye are or ee/r

144*10^6/10^6=144 smoking whats) And that capysater poofie inless 15KV rating and you can afford it.

Sho-nuff "improve EMC ruggedness", boss.

Robert,

In that sort of thing as well as many other subjects, best to think before speaking. Please don't be offended Robert.

What makes you think the 1Meg resistor has to be a 1KV rated ? I guess you never designed i.e. Flyback DC/DC for the industry. Please re-read my previous post.

Habib.

r u e .

Habib is a typical delusional and arrogant know-nothing idiot. His own LTSp ice sim shows the BAT54s avalanching, he's such a friggin idiot he has no c lue about transformer core discharge. And I don't know what kind of clueles s idiot chops a unipolar voltage into magnetics, and that input voltage buf fer with its "idiotpathic" pseudo compensation is almost as laughable as th e pseudo-transformer feedback. Every bit of his "design" approach and circu it is completely pointless.

Then read what i said (NOT 1KV). IF one needs the drive power at (say) +10KV, then that resistor gots tew dizzypate lotza power - and for no reason. The 10KV supply in and of itself would most likely be references to ground in a more satisfactory manner.