Ground / Shield / Chassis / Return Paths

Thread Starter

lienPCB

Joined Nov 15, 2023
2
After watching some videos on grounds / return paths / shielding etc I came up with a design that I was wondering if someone could assist in assessing below simplified diagram with notes to see if it is a good grounding strategy also considering PCB EMI/EMC ?Screenshot_20231115-203257_WhatsApp.jpgSimplified diagram attached.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
17,790
The short answer is that "it depends". Really, the level of shielding and isolation depends mostly on the application and the amplitude and impedance of the signals and circuits involved. The other consideration is how much electrical noise is present in the surrounding area.
 

Thread Starter

lienPCB

Joined Nov 15, 2023
2

Ya’akov

Joined Jan 27, 2019
8,973
Welcome to AAC.

I am a bit baffled that you chose to post this question in a forum on the Internet. Are you really depending on advice from random folks to make decisions concerning a military application?

I realize free is a lot cheaper than paying a legitimate consultant—at least on the front end. The cost of doing it is wildly unknown, though, when you actually apply the advice.

No matter my level of knowledge and experience* I wouldn’t want to be responsible for advice based on this low-quality interaction concerning a critical application. It doesn’t make sense.

Can you explain why you are looking for this kind of help in this forum so it doesn’t take on the appearance of something suspicious?

*In fact, the likelihood of getting involved is inversely proportionate to knowledge and experience.
 

MisterBill2

Joined Jan 23, 2018
17,790
Welcome to AAC.

I am a bit baffled that you chose to post this question in a forum on the Internet. Are you really depending on advice from random folks to make decisions concerning a military application?

I realize free is a lot cheaper than paying a legitimate consultant—at least on the front end. The cost of doing it is wildly unknown, though, when you actually apply the advice.

No matter my level of knowledge and experience* I wouldn’t want to be responsible for advice based on this low-quality interaction concerning a critical application. It doesn’t make sense.

Can you explain why you are looking for this kind of help in this forum so it doesn’t take on the appearance of something suspicious?

*In fact, the likelihood of getting involved is inversely proportionate to knowledge and experience.
Looking at the claimed credentials it is even more a puzzle. The required "Tempest Level" will certainly make a big difference, although I am not sure if it applies to non-data systems. And it does not apply to non-US equipment, although other similar standards may apply.
 

drjohsmith

Joined Dec 13, 2021
850
Looking at the claimed credentials it is even more a puzzle. The required "Tempest Level" will certainly make a big difference, although I am not sure if it applies to non-data systems. And it does not apply to non-US equipment, although other similar standards may apply.
Fyi.
In the world of none us mil stuff I live with, Tempest is a almost universal term , like a hoover is not always made by hoover .. we adjust to the local requirements.
As for it only meaning data ,
In my experience ,
Just about anything now days is classed as having data / tempest et all.
But we await further info from the poster.
 

nsaspook

Joined Aug 27, 2009
12,778
Welcome to AAC.

I am a bit baffled that you chose to post this question in a forum on the Internet. Are you really depending on advice from random folks to make decisions concerning a military application?

I realize free is a lot cheaper than paying a legitimate consultant—at least on the front end. The cost of doing it is wildly unknown, though, when you actually apply the advice.

No matter my level of knowledge and experience* I wouldn’t want to be responsible for advice based on this low-quality interaction concerning a critical application. It doesn’t make sense.

Can you explain why you are looking for this kind of help in this forum so it doesn’t take on the appearance of something suspicious?

*In fact, the likelihood of getting involved is inversely proportionate to knowledge and experience.
Tempest, My God, don't even go there. The horrors that have been dumped on practical engineering because this witch-craft is the stuff of nightmares.
https://www.navy-radio.com/tty-tempest.htm
 

MisterBill2

Joined Jan 23, 2018
17,790
Certainly what can be verified is that FULLY TEMPEST COMPLIANT systems do not emit any RFI or EMI at all. and even more certain is that the required effort is great. (Quite an understatement, if ever there was such.)
 

MisterBill2

Joined Jan 23, 2018
17,790
Responding to post #1, still no mention of the signal types. Low level ( micro-volt amplitude) analog signals from high impedance sources may need all of those precautions will need that sort of shielding to prevent noise entrance.
 

nsaspook

Joined Aug 27, 2009
12,778
Certainly what can be verified is that FULLY TEMPEST COMPLIANT systems do not emit any RFI or EMI at all. and even more certain is that the required effort is great. (Quite an understatement, if ever there was such.)
As you read the old NSA TEMPEST info you can see that low-level keying (aka RS-232) was basically invented by NRL and NSA in the 50's as a TEMPEST countermeasure.
https://www.navy-radio.com/crypto/TEMPEST_Boak_NSA_1973.pdf

1700148996826.png
Cost was a great factor. TEMPEST Systems still emit RFI or EMI, but the intelligence in those emissions is undetectable beyond some distance.
To the OP: Assessing the grounds / return paths / shielding on your design depends on the specification of what you need to suppress. If it's just bulk RFI or EMI it's likely good, if it's detectable intelligence that needs to be suppressed, that's another story entirely.
 
Top