High Voltage spike , disrupts Display functionality

[Rohit1203]

High voltage spike on 5V IO line disturbs 5.0 display functionality. all lines are protected using io buffer and SMBJ5.0A TVS, still, a high voltage spike enters the main 5V and 3.3V power lines. Spike creates spurious interrupts.

How I can stop passing high voltage spike to internal main Digital core section and save the Internal functionality and Display.

 

[bertus]

Hello,

Do you have a schematic to post?
Then we can have a look how to proceed.

I see that the mentioned TVS has a clamping voltage of 9.2 Volts.
https://www.littelfuse.com/products/tvs-diodes/surface-mount/smbj/smbj5_0a.aspx

You might want to read this:
https://www.digikey.com/en/articles/protecting-inputs-in-digital-electronics

Bertus

 

[WHONOES]

You could try a large ferrite bead on the cable.

 

[Rohit1203]

Thanks for the reply.
See below schematic for ESD protection and grounding connections.

As shown in the below image high voltage spike on WDATA_IN0 & IN1 line creates a spike on 5V and 3.3V Power lines and which creates issues in main processor functionality and LCD display functionality.

 

[Rohit1203]

You could try a large ferrite bead on the cable.

Do you mean ferrite core shown in the below image on the IO cables?

 

[Harald Kapp]

Do I see +774 V and -871 V on the IO lines? It will be difficult (not impossible but difficult) to decouple these spikes while having a galvanic connection.
I suggest you decouple the IO lines from the µC using photocouplers (also known as optocouplers)or equivalently any form of isolated data coupler, e.g. icouplers (which is a brand name by Analog Devices, but similar componnets are available from other manufacturers, too).

 

[WHONOES]

I suppose the first question is... where are the spikes coming from?

 

[Rohit1203]

I suppose the first question is... where are the spikes coming from?

trying to simulate the field issue in the inhouse lab using an electric spark generator.
Applying spark from electric gas lighter which generates high voltage spikes on digital io lines.

 

[Rohit1203]

Do I see +774 V and -871 V on the IO lines? It will be difficult (not impossible but difficult) to decouple these spikes while having a galvanic connection.
I suggest you decouple the IO lines from the µC using photocouplers (also known as optocouplers)or equivalently any form of isolated data coupler, e.g. icouplers (which is a brand name by Analog Devices, but similar components are available from other manufacturers, too).

Is any way to catch the spike and discharge on earthing.
Any ready-made modules are available in the market which provides Opto isolation. now not possible to modify the board and provide the opto-isolation, but if any readyly available io isolators are available then we can use between our device and other systems.

 

[Harald Kapp]

Any ready-made modules are available in the market which provides Opto isolation

Use the search engine of your choice, look for "optoisolator module".They come in 1-channel or 4-channel flavors (that's what I've seen) from different vendors.

 

[WHONOES]

trying to simulate the field issue in the inhouse lab using an electric spark generator.
Applying spark from electric gas lighter which generates high voltage spikes on digital io lines.

What is your rational for testing in that manner.

 

[Rohit1203]

What is your rational for testing in that manner.

We got a field issue that burns series resistors, ESD chip, and buffer chip.

So trying to simulate why it got damaged and how?

 

[Rohit1203]

Use the search engine of your choice, look for "optoisolator module".They come in 1-channel or 4-channel flavors (that's what I've seen) from different vendors.

Thank you will try it.

 

[WHONOES]

So, what conditions is your device subjected to in the field? Have you taken any measurements in your device whilst operating in the field?

 

[Proton]

maybe, the generation of hi voltage creates a kickback voltage after discharge? in other words; collapse of the magnetic field from the secondary cuts thru the primary winding creating a backfeed which is often very high but extremely fast. Most High energy ignition systems overload my digital voltmeters, while analog meters happily read less than 12v on the primary.
a fast diode across switching transistor, perhaps would help.
digital meters in some cases get erratic within 3-4 feet implies to me that its also a very noisy enviroment