Author Topic: Schematic Compared to Physical Components on Routed Board Question  (Read 1912 times)

0 Members and 1 Guest are viewing this topic.

Offline OilsFanTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: us
I have routed a board following my schematic but noticed that as I arranged the components on my PCB that I connected them in such a way that they are 'out of order' so to speak. Look at the picture of the schematic and you'll see that coming off a pin of an opto (OC2) I have C4 to ground followed by R7 but the way I routed it (blue trace), I have the pin of OC2 physically between C4 and R7. Does it make a difference in the circuit? Will the software (DIPTRACE) stop me from routing things in a way that doesn't make sense? Having a hard time wrapping my head around that.



« Last Edit: February 20, 2016, 11:59:43 am by OilsFan »
 

Offline lem_ix

  • Regular Contributor
  • *
  • Posts: 192
  • Country: cs
Re: Schematic Compared to Physical Components on Routed Board Question
« Reply #1 on: February 20, 2016, 12:49:05 pm »
Can't really see much on your schematic but assuming the pin of the opto is connected to that node between pin 1 of C4 and pin 1 of R7 it's fine. The schematic could be redrawn in many ways but electrically stay the same like you did in the actual routing. As long as they stay connected to the nodes they're supposed to their order doesn't matter. Ofc it's a smart idea to keep components as close as possible but i doubth it matters in this case.
 

Offline OilsFanTopic starter

  • Regular Contributor
  • *
  • Posts: 52
  • Country: us
Re: Schematic Compared to Physical Components on Routed Board Question
« Reply #2 on: February 20, 2016, 07:00:40 pm »
Can't really see much on your schematic but assuming the pin of the opto is connected to that node between pin 1 of C4 and pin 1 of R7 it's fine. The schematic could be redrawn in many ways but electrically stay the same like you did in the actual routing. As long as they stay connected to the nodes they're supposed to their order doesn't matter. Ofc it's a smart idea to keep components as close as possible but i doubth it matters in this case.

Thanks for the reply. I'm thinking it doesn't matter in this case but being new to this I sometimes wonder if it's right.  :o
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3667
  • Country: us
Re: Schematic Compared to Physical Components on Routed Board Question
« Reply #3 on: February 20, 2016, 07:26:05 pm »
All those connections are part of the same net, or node (think wire or trace), and most of the time doesn't matter exactly where the connection is actually placed. One thing to keep in mind though, is that capacitors are generally intended to be placed "close" to something, such as the power connection(s) to chip. On some schematics this can be very non-obvious, as the capacitors may be all grouped together and drawn way off in some random corner of the sheet for the sake of "neatness" or something.
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 21686
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Schematic Compared to Physical Components on Routed Board Question
« Reply #4 on: February 20, 2016, 09:52:22 pm »
I think I've heard that PADS supports sub-nets, i.e., chaining parts in exact or constrained sequence.  Altium has From-Tos, though I haven't seen them used.

All programs, as far as I know, by default, treat all pins in a node as exactly equivalent.  As long as those pins are connected by some copper, it doesn't matter how, or where, or with what size.  (Okay, most/all programs support design rules for trace size, but that's probably done as a separate check, not when checking connectivity.)

As for whether it really matters -- it depends.  If you place components out of order, you can get high frequency effects, like stub transmission lines and peaks and valleys in the impedance or frequency response.  This only shows up when the frequency is very high (length of stub > wavelength/10, say), and when the impedance ratio is very different (system impedance very low or very high relative to the trace impedance: switching circuits being the most common example, as impedances of \$\sim 1 \Omega\$ are common, and impedances under \$ 20 \Omega\$ are hard to construct).

That shouldn't be an "or", actually.  More precisely, the ratio of impedance (let r = Z / Zo; the ratio of impedance is then: (r + 1/r) / 2) multiplies the critical stub length, so that, say, for an operating frequency of 1MHz and an impedance ratio of 20, you need stub length < 1m.
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf