Linear optocouplers and applications

[@Chris_Grossman]

Thank you for this simple analysis. I will refer to this if I need to use a linear isolator in a design.

[@sambenyaakov]

Thanks

[@freedomisfood6966]

Best electronic master in youtube

[@sambenyaakov]

🙏👍

[@ccshello1]

In late 80s and early 90s, these are popular replacement of isolation transformers and were used in VBD (voice band data) modem's DAA (direct access arrangement) function block. Typically, a dual T-R pair is used for two uni-directional analog data paths coupling. They are popular in PCMCIA modem card for its slim form factor. AT&T Microelectronics, CP Clare, Siemens, SGS made them. SGS and Siemens made the IL300 variant which later transitioned to Vishey made. Was a fun memory.

[@sambenyaakov]

Thanks for the interesting history lesson.

[@charlesdorval394]

Wow thank you! That's exactly what I needed

[@sambenyaakov]

👍

[@pieters286]

Always much appreciated!

[@sambenyaakov]

👍🙏😊

[@byronwatkins2565]

At 7:20, using a negative power supply on the photodiode's anode and the opamp on the cathode will also generate a positive gain=R2/R1.

[@sambenyaakov]

You are absolutely correct. Good thinking. But where do you get the negative voltage and why not connect the amplifier as a follower?

[@JacquesMartini]

A more simple explaination for the better option to place the isolation barrier AFTER the compensator is, that any static error or drift is compensated by the control loop, as long as the comparion between the reference and the output has a (very) low error and the loop has enought gain or a minimum integral component (PI compensator).

[@sambenyaakov]

Thanks for the input. I am not sure that this is a simpler explanation

[@sXAPOLLls]

Guys, don't use linear optocouplers (even optocouplers) for isolation of digital and analog signals BY DEFAULT. As any diode, optocouplers are very unlinear due to temperature change. That is why you don't see much solution like this on market. Industry use digital isolators based on transformers or capacitors (TI, AD). Use linear optocouplers only as big tradeof between solutions from above.

[@sambenyaakov]

Thanks for your input.

[@t1d100]

👍👍👍👍👍

[@sambenyaakov]

👍🙏

[@_Jkobson]

I will add a remark not mentioned in the video: it makes sense to use this solution with a linear optocoupler only when you need an isolated analog signal and a digital signal is unacceptable. because it's much easier to send a digital signal through the insulator. or if you still need an analog signal, then you should stick to solutions like with an optocoupler in smps feedback with tl431 on the secondary side.

[@sambenyaakov]

Thanks for your input.

[@JacquesMartini]

So funny! First explaining how a linear servo optocoupler works (academic solution) to finally explain, why it is better and cheaper to use a nonlinear, cheap optocoupler in a clever way (engineer solution, viva la TL431)! 😂

[@sambenyaakov]

👍😊🙏

[@gaozihan4]

The Linear optocoupler was very helpful to me back to college to do small stuff! But after graduating right now we are simply using VFC110 plus fiber optics to travel long distance between pretty high power converters and the controller. XD I suppose the isolated amplifier uses similar optical coupling method, or otherwise modulated through an air core transformer?

[@sambenyaakov]

Thanks for comment. Long distance is another story. V/F is a bot archaic, isn't it? I used it 50 (!) years ago to transmit information from deep ocean probes, down to 5km

[@gaozihan4]

​@@sambenyaakov Yes, stupid and simple, but not too cheap. I heard that TI even lost the design files of VFC110 through the years. They promised to continue producing because industry still uses it.

[@sambenyaakov]

@@gaozihan4 👍

[@tamaseduard5145]

👍🙏😍

[@sambenyaakov]

👍😊🙏

[@robr8554]

I'd imagine they suffer aging effects where the led intensity decreased over time. Is there a way to avoid these optocouplers and use a modulator with a transformer instead to send the dc info?

[@JacquesMartini]

If the LED degrades over time, the servo loop will compensate. Just as for any thermal drift of LED or photo diode transfer function. The "trick" is, that this drift or aging affects both reciever photo diodes equally! As long as this holds true, any error is compensated. See the thermal drift of K3! Almost zero, 0,005%/K, 0,5% for 100 K temperature variation! Not bad!

[@robr8554]

​@JacquesMartini this is for power supplies in space systems. Optos are frowned upon since they don’t last in harsh environments. Thank you for the reply

[@biswajit681]

Sir could you please make a video on servoing and margining of power supply?

[@sambenyaakov]

Sorry for the ignorance. What is "servoing and margining"

[@JacquesMartini]

Objection! The transimpedance amplifier has ALWAYS a higher bandwidth (and lower noise) than a simple resistor in photoconductive operation! The gain of the amplifer does the "trick".

[@sambenyaakov]

Go Fight with VIshay " PHOTOCONDUCTIVE ISOLATION AMPLIFIER The photoconductive isolation amplifier operates the photodiodes with a reverse bias. The operation of the input network is covered in the discussion of K3 and as such will not be repeated here. The photoconductive isolation amplifier is recommended when maximum signal bandwidth is desired. Bipolar photoconductive isolation amplifier"

[@JacquesMartini]

@@sambenyaakov I assume you refer to application note 50. There is some confusion. They are using the terms losely or even wrong. In this document as well as in the normal data sheet, photo voltaic mode is NEVER used due to its serious disadvantages! Only photoconductive mode (using a simple resistor) or an transimpedence amplifier.

[@sambenyaakov]

@@JacquesMartiniWhat do you call photovoltaic mode? when the photo diode is clamped to zero voltage?

[@JacquesMartini]

@@sambenyaakov The same as you in your video. When the voltage across the diode is directly measured and used. Clamping the voltage to zero required an transimedance amplifier.

[@Sajad-e6r]

Hi. Sir could you please make a video on SVPWM inverter and their control loop? Thank you

[@sambenyaakov]

Will try

[@Edmorbus]

Nice work Can you help me with a circuit schematics for 4x 12.6v Lithium battery to measure separate the voltage with ESP32 i have the ADS1115 I300 ic realize galvanic isolation

[@sambenyaakov]

Sorry, I am just too busy.

[@Edmorbus]

@@sambenyaakov No problem thanks

[@biswajit681]

Hi sir when are you going to release new webseries 😅

[@sambenyaakov]

What do you mean by webseries?

[@biswajit681]

@sambenyaakov next video of our power electronics webseries:)

[@sambenyaakov]

@@biswajit681 In an hour or so😊😊

[@biswajit681]

18 more to go 😅

[@sambenyaakov]

😊

[@alexchan4226]

for IQ test.

[@freedomisfood6966]

Hi master

[@sambenyaakov]

🙏😊