Multilayer boards are for hiding your work, not simplifying fanout or clean power planes

Well, no, I don't actually believe this. But it makes about as much sense as the "assemblers are for hiding your work" thread currently raging.

"werty" is clearly a few medications short of a pharmacy.

Reply to
larwe
Loading thread data ...

I have been involved with boards that had ground planes on both top and bottom layers, all the signals inside. Done for a combination of noise reduction and IP security on a classified project.

I'm strongly considering adding filtering for that other thing...

Scott

Reply to
Not Really Me

In normal stripline circuits with ground planes on the top and bottom layer and the actual transmission lines, resonators and impedance matching in the middle layer, you can not figure out the connection by just putting a strong light source behind the board, but you would have to use X-ray etc. to figure out the actual connections and functions :-)

Paul

Reply to
Paul Keinanen

This is routine in reverse-engineering. In fact, there's software you can acquire that will take a set of X-ray layer photographs and create a netlist.

Reply to
larwe

really? any links? how does the software differentential between different traces in different layers, the xray just gives you all layers at once, I would think

Reply to
steve

Well, I used the word "acquire" advisedly, it's not exactly an off-the-shelf product. I have seen it in use but I don't know what it's called. The time-consuming part is telling the software where the packages are in the board geography.

For a 2-layer board you can simply use scans of the layers. For multilayer you need to take delta pictures - remove one layer at a time and X-ray what's left. Then you align the layers onscreen and the rest is automatic.

Reply to
larwe

I thought most engineers understood that filters for noise are only a bandaid. Most noise is better dealt with at the source... if you get my meaning ;^]

Reply to
rickman

I once told a senior marketing VP that I while I had read his drivel, I hadn't comprehended a single word of it because it was all out-of-band noise. He didn't take it well. Luckily, I was indispensable.

Reply to
larwe

Im EE , we dont do it for noise , we do it for characteristic Z . .....Transmission line has 2 conductors , one can be a gnd as in unbalanced . Coax cable is unbalanced .

Reply to
werty

ha ha ha Reverse engineering !!!

Reverse and find its a reverse ,, of a reverse ,, and its worse than you could have done by simply ENGINEERing right , by yourself ! -------------------------- Its like trying to learn software by disassembling the bloat trash , today .

They did such a bad job , you learn nothing from disassembling / tracing others code ,

so i just create my own code ...

Reply to
werty

That's the trick, isn't it ? (:

Reply to
Jim Stewart

Hilarious.

I've done boards with 28 layers on admittedly quite thick boards, although 16 layers is the max I have managed for a 2.4mm board that had lots of impedance control requirements ;)

As to dealing with noise - we can control it, not eliminate it, but I agree it's best dealt with at the source (and at the design stage ;)

Cheers

PeteS

Reply to
PeteS

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.