Take the 2-minute tour ×
Electrical Engineering Stack Exchange is a question and answer site for electronics and electrical engineering professionals, students, and enthusiasts. It's 100% free, no registration required.

If I have a small signal, does routing it through vias to different layer introduce any significant noise on the signal compared to routing it within same layer without any vias? Considering all other parameters like track length, width constant.

share|improve this question
    
What web sources have you found? There are several sites which provide routing and interference information. What sort of frequency is the signal? Clearly a via makes a track longer, often 1.6mm each time, which over a couple of via's can add significant length. –  gbulmer Aug 22 at 10:22

2 Answers 2

A via doesn't add noise, per se, but it does represent an opportunity for noise to be coupled to your signal that is different from the opportunity along the traces.

Noise couples to your signal through capacitive coupling and inductive coupling. Both will happen along a trace and along a via of any length. How much depends on the aggressor signal, geometries and materials.

NB: As to the other answer indicating that vias create impedance change for high speed digital circuits: This is not really the problem. If you have different trace impedance on different layers, then obviously you will have a reflection from switching layers. Vias can be used in high speed digital circuits to above 25 GBps as described in my blog here.

share|improve this answer
1  
How is reflection and impedance tamed in modern real-life products, like PCB lines (short distance) and USB and HDMI (long). The University teaching approach is have an driver with near zero output imp, put a 150 ohms resistor in series then drive down a strip line (typically imp 150 ohms, or in that range) over ground plane. Another 150 ohms terminator at far end and pick up the signal. Why not seen this in real nowadays products. Are they all build inside the chip? Is it true that reflection is no practical issue when cable length is shorter than certain percentage of signal wavelength? –  EE developer Aug 22 at 12:39
    
Not knowing much about hi speed signal integrity, I note that many modern long distance (a few meters instead of a few inches on PCB) signal cable use differential signalling. Does it help to tam the reflection issue as single ended driver over ground plane as above. May be diff means same reflection at both line and 'cancel' out the bad effect? –  EE developer Aug 22 at 12:43
1  
@EEdeveloper, That would be a very good question for the site. In fact, I would have thought it had been asked before, but I wasn't able to find it with searching. If you post it as a new question, you should get some good answers. Or it might end up closed as a duplicate, but at least you will get your answer. –  The Photon Aug 22 at 15:51

A via doesn't add noise, per se, but it does represent an impedance change that can create extra reflections of high-speed signal edges. This is why they're avoided in the traces used for things like DDR SDRAM, high-speed USB and Ethernet connections.

share|improve this answer
    
I agree with the first 7 words. Not the rest - see my answer. –  Rolf Ostergaard Aug 22 at 10:55
    
The "A via doesn't add noise, per se, but it does represent an" exact duplicate intro wasn't pure coincidence, was it?! :o –  Rev1.0 Aug 22 at 12:59
    
Dave was first - I am the "copycat" as I think the words are spot on :-) –  Rolf Ostergaard Aug 22 at 16:10

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.