ISO7741-Q1: Galvanic isolation

Hi Stefan,

Thanks for reaching out.

Apologies but your inputs were not very clear with respect to what is the issue. Can you please illustrate your ask using a simple diagram, showing connections and scenarios for us to review and comment.

I do have one input on your take of galvanic isolation

GHM Messtechnik said:

But galvanic isolation also means that input and output may be directly connected.

This is not correct understanding. Galvanic isolation actually means opposite.

You can get more details on this tutorial video: What is Galvanic Isolation? | Video | TI.com

Regards
Varun

Hi Varun,

excuse for the late answer but I was absent the last days.

So to my problem:

On the input an ourput side of ISO7741 an USB-Master can be connected for maintance.

When they are connectet at once to the same PC or notebook you have a simple DC-bridge between input and output sides because both USB-Ports of the PC are on the same Ground potential.

When you have an isolation by optocouplers or transformers this would be no problem at all.

And my thought is to repcace this with the ISO-ICs with capacitive coupling because this way is cheaper an takes less place.

Between input and output of ISO7741  I tried the lowest resistence may be 100 Ohm to keep the transmition functioning.

The higest capacity may be 10nF.

But this doesn't help me because I can not place an 100R resistance in the USB-Lines. Then the USB-Connection don't work.

Regards

Stefan

Hi Stefan,

Apologies but your application requirement is still not clear to me and how ISO7741 connections look like in your application.

Please provide a connection diagram with each pin of ISO7741 connected to your signals for us to review and comments. Without that, we'll not be able to comment on this.

Since you mentioned USB, we also have a family of USB isolators which might be helpful for your cause - ISOUSB111 and ISOUSB211. Do check them out on ti.com

Regards
Varun

Hello Varun,

please also understand my side of the situation:

We have no approval to publish any schematics.

So I try to make a description:

Our device should be used for measuring environmental data in an analog format.

The ISO7741 is meant to isolate the main controller from the analog section.

For that, in the analog section the measuring data is digitalized to a SPI interface.

The ISO7741 should isolate the SPI lines MISO and MOSI and two control signals with a speed of 1MBaud

The supplies of ISO7741  (3.3V on both sides) come from 2 isolated switching regulators

The power pins of the ISO are buffered with 2x 100nF ceramic capacitors.

On the supply pins you see very low interferences

For maintenance and (and only for this case) both sides can be connected via USB to a PC

But in this case the communication over ISO7741 must be still possible.

And this is not possible because both GND lines are bridget , when both usb interfaces are connected to the same PC

Isolating the USB-Lines is much to expensive. You know the price for your suggested chips.

So I can make an optical isolation i.e with 4x TLP2361  which cost 1,5x the ISO7741 and the problems are away.

(I made this several times before)

But the occupied space on the pcb is very much higher.

By the way: Why do the optocouplers always have such big cases, when they should only isolate peaks of 1,5kV.

Perhaps I have an idea for the reason of this problem:

When a loaded capacitor is switched to GND potential on the input side and input and output GNDs are bridged (via USB)

the data line on the output side becomes a negative potential.

Perhaps the ISO7741 is protected against such situations.

Of course this can only happen when the reference potential of the isolating capacitors is GND and not VCC/2

But this will also explain the rising current consumption  when I simply connect Pins 2 and 15 for testing.

Thanks before

Stefan

Hi Varun,

on our location it's the start of the working day.

As you said, the problem can be testet from anybody.who as a working hardware

Simply connect Pin 2 and Pin 15 of an ISO77.. IC.

When the communication over the ISO77 is still ongoing in this case there must be something wrong with my schematic.

The bridge simulates the condition in maintance case. The normal case (with no bridge) is working well.

What I said at the end of my last reply is only an assumption, because I don't know the detailed working principle of this device

Thanks

Stefan