PCB preference of putting 2 components to a trace

[epilot]

Hi guys,

I have got an important question when I was trying to implement the PCB of a pre-amplifier based up on 3 stages of TL074 op-amp chip:

Is there any rule saying the preference of 2 components connected to one trace? I do not know if I am clear or not, but I'll try to explain my mean by an example. for instance in an inverting op-amp one pin of the Rf is connected to one pin of Ri, so do I have a freedom to connect either the Ri's pin first to the "-" input of the op-amp or the Rf's pin first when designing the PCB? what about the other lead of Rf and the load's pin?
Is my mean Makes sense?

Thanks a lot

[epilot]

No response please?

[Mooly]

Always best to draw what you mean but [think] I understand...

In your example it makes no difference. The theoretical aspect is that the print resistance will add in such a way as to modify the gain... but is rf/ri+ say 0.001ohms going to make a difference. Of course not. And depending on whether its the rf or ri end of the PCB you connect to means the gain could go either way depending... and its a non problem. Think of each conductive trace on a PCB as a resistance and you won't go far wrong. This is where many designs fall down with interaction/hum issues, and its so easy to get right. You just have to think, and its good that you are

[marce]

Further to Mooly's comments, placement is critical in PCB design, generaly the small amount of resistance added by a trace is negligable...but they also add a small amount of inductance, this may be problematic with RF interferance. This can be a problem where you have high impedance inputs such as the feedback pin of a lot of voltage controllers.
The calculator from this site is a PCB designers best friend:
PCB Via Current | PCB Trace Width | Differential Pair Calculator | PCB Impedance

[epilot]

Originally Posted by Mooly

Always best to draw what you mean but [think] I understand...

In your example it makes no difference. The theoretical aspect is that the print resistance will add in such a way as to modify the gain... but is rf/ri+ say 0.001ohms going to make a difference. Of course not. And depending on whether its the rf or ri end of the PCB you connect to means the gain could go either way depending... and its a non problem. Think of each conductive trace on a PCB as a resistance and you won't go far wrong. This is where many designs fall down with interaction/hum issues, and its so easy to get right. You just have to think, and its good that you are

Hi Mooly,

thanks a bunch for your reply,

Actually It is not so simple for me to show the PCB with the components so that could explain waht I mean, So I use a simple schematic for a Smoothing cap and the load so that I show you what I meant.
I do not know why I think that in the B circuit there is a problem and the load would not be smoothed Just right because the upper side load is much near tan the bridged diodes instead of the smoothing cap, I myself think that the upper side of the load would not receive the correct DC voltage which is generated or smoothed by the cap because the unsmoothed current from the bridged diodes FIRSTLY goes to the load instead of going to the cap to be smoothed and then go to the load SO it does not get the proper time to be smoothed PERFECTLY (maybe I am wrong and I think that I am but thats what I see..). Is it makes sense to you?

[epilot]

Originally Posted by marce

Further to Mooly's comments, placement is critical in PCB design, generaly the small amount of resistance added by a trace is negligable...but they also add a small amount of inductance, this may be problematic with RF interferance. This can be a problem where you have high impedance inputs such as the feedback pin of a lot of voltage controllers.
The calculator from this site is a PCB designers best friend:
PCB Via Current | PCB Trace Width | Differential Pair Calculator | PCB Impedance

Good software marce, Thanks for sharing it, do you know another one more too?

Actually I would like to know the formulas that explain the relationship of the dimensions of a trance with the inductance/capacitance plz. Can you help me out in this regard plz?

Thanks

[Lazybutt]

Hi!

I think there is no universal answer to your initial question. It depends on the application / component.

In the particular case of the power supply you are showing, what you want is (indeed) to have the lowest resistance between the cap and the load, so you'd rather have the cap as close as possible to the load.
besides, the trace between the rectifier and the cap will act as a RC filter, so will provide better smoothing.

Note that we're talking mOhms or less here, so in the case of a power supply, it will not make a big difference in real life...

If you're designing for RF, then yes there is some optimisation to make.

In many cases, to get a "feel" of what is better, just imagine the traces are resistors or inductances (try to think extreme cases where traces have a farily high resistance) Then it will make it very obvious what is best on the PCB.

[katiyar]

The earth loop must be avoided for stability.

[Lazybutt]

I don't see a ground loop in this particular example ...?????

[marce]

Epilot, in your case B is not good practice, and where PSU smoothinbg caps are concerned we work to the rule that the PSU output goes through the smoothing capacitor first (A) always without exception, for both linear and Switched mode PSU's. This is done mainly for EMC and noise purposes, you are minimising the loop for the ripple by doing this (and also the switching loop for SMPS,s), amonst other benefits. I would make it one of your design rules and never break it. On SMPS we use the minimum copper we can get away with through that node, ie the route is not much bigger than the pad. This smoothing capacitor must be near the power supply, not near the load, the other way (ie near the load is WRONG).

Impedance calculator basic, I have a Signel Integrity and High speed router add on for my PCB package wich calculates all these figure for me, but use this one when I need to work out the basic topography at the begining of a design.
Microstrip Impedance | Electrical Engineering Tools | EEWeb

Oh and heres a creepage and clearance calculator as your playing with mains.
http://www.creepage.com/