Go Down

Topic: Different results, using same values (resistors) from beadboard to pcb? (Read 820 times) previous topic - next topic

xl97

Hey gang-

Im trying to finalize a few last tings on my custom board here.. to tweak the audio (volume and filter/range) a bit better..

I got some time to dial in the pots for both the volume and the 'filter'…. and got some odd results...

summary: (fine tuning resistors for volume level (R5/R6) and audio filter/range (R8) for final pcb placement/production)

schematic (for those that need it)


quick image to show how I wired it up... (similar for R8 too..but only used one side of pot of course)




I soldered wires from the 3 resistors   (2 wires from filter resistor (R8) pads) & (3 wires from resistors (R5 & R6) pads)

I put a 1K pot on 8R (filter) & a 100K pot on the R5/R6 (volume).. adjusted both until it 'felt' good to the ears in both good volume, but not too much as to distort,..and a good range of lows & highs, but not where it cracked or hissed..

I metered the pots..(set @ 200k o meter)

R8 (filter) resistor: was about .98 (980 ohms) is how I read it

R5/R6 (volume) resistor was:
R5: 91.0 (91k ohm)
R6: 4.9 (4.9k ohm)


Dug through my boxes.. and (still using the breadboard) I replaced the pot with:
R8 = 1k ohm   
R5 = 100k ohm
R6 = 4.7k ohm

And it still sounded good/great…

So I took my a finished board.. and used the SAME values.. but in SMD footprint for the board.. and it sounds DIFFERENT..

Volume seems to be 'ok' still.. but the 'filter/range' seems to be a bit muffled? A clear difference than on the breadboard?

I have the two board still set-up, one is my 'dev/practice' board to do changes to.. and is connected to breadboard as well for prototyping with this board…
The other is standalone like normal.. but I replaced those resistors with values above..and as per my results from the dev board/breadboard tests..

What am I missing? Doing wrong here?  How can they sound different, yet use the same values?


Thanks

WillR

Did you allow for component tolerance in your calculations?

Alternatively, measure every component in the working circuitry and SELECT components to match what works...

That's why EE is a four year program -- they eventually get to error analysis and how to deal with components that vary...  ]:D

Just another Hacker

xl97

nope.. I guess I didnt..

(must have something to do with not having that 4 year EE degree?)  :)

Although.. I guess what component tolerance could it be?

they are two, exactly, identical boards/pcbs.. (assembled, re-flowed the exact same..same parts..etc)

board A: has those 3 resistors left off.. and wires soldered in place going to breadboard/pots..etc (and went through the process above to get those static values from the metered pots)

board B: has those 3 resistors in place (after I had metered the pots.. I added those resistors to board B to test the final outcome)


thanks....

WillR

Yep... the fours years is needed... lol

OK -- every component has a tolerance. If you get a 1K resistor from your supply was it 5%, 2%, 1% or maybe shudder -- 10% or 20%?
http://en.wikipedia.org/wiki/Electronic_color_code


I may have been making a joke -- but it wasn't a joke. ...not really.
http://en.wikipedia.org/wiki/Engineering_tolerance

and here:
http://web.awrcorp.com/content/Downloads/MWEE-November-2005.pdf

cheers
Just another Hacker

James C4S


they are two, exactly, identical boards/pcbs.. (assembled, re-flowed the exact same..same parts..etc)

This might be a generalization you should remove when troubleshooting a problem.  While the boards were intended to be exactly the same, tolerances in manufacturing will cause them to be slightly different as well.  Trace widths, copper thickness, etc will all be a little bit different.

Sometimes this isn't an issue, but when dealing with filters changes in parasitics can cause noticeable changes in behavior.
Capacitor Expert By Day, Enginerd by night.  ||  Personal Blog: www.baldengineer.com  || Electronics Tutorials for Beginners:  www.addohms.com

WillR

Read what James said very carefully....

Additionally I just came back here to add a point to my post -- but James used the words parasitic.... so I'll tag on...

A wire is a resistor and a capacitor and an inductor... Really! It can sometimes add enough of one of those elements so as to completely change the circuit behavior. It becomes a parasitic capacitor or inductor or some combination. Kinky traces and wires are the worst. You can add an inductor to a PC board by making a trace a 90 degree corner...

The PDF I linked says that -- in engineerspeak -- which can only be decoded in the light of the full winter moon (at midnight) with the use of a secret decoder ring... In other words it should be completely clear to you.

And now that I made the comment about right angles (as opposed to nice smooth curved traces) -- have another look at your board. It shouldn't matter that much at audio frequencies -- but then I will bet you did not run a SPICE simulation either did you?

http://en.wikipedia.org/wiki/SPICE

One of my old profs wrote at least part of the transient analysis section -- he was weird man!
Just another Hacker

winner10920

what does the right angles have to do with anything other than looks?

AWOL

Quote
what does the right angles have to do with anything other than looks?

Quote
It shouldn't matter that much at audio frequencies

Inductance
"Pete, it's a fool looks for logic in the chambers of the human heart." Ulysses Everett McGill.
Do not send technical questions via personal messaging - they will be ignored.

WillR


what does the right angles have to do with anything other than looks?


It was right in my post...

You can add an inductor to a PC board by making a trace a 90 degree corner..

Clear?
Just another Hacker

CrossRoads

@xl97,
I would think that R5/R6 are not so critical as that is the volume.
R8, C18, variations affect the low pass filter frequency, variation there is probably what you hear.
Might have to select R8 to allow a little higher frequency so parts variation in the 'wrong' direction do not impact the sound too much.
Designing & building electrical circuits for over 25 years. Check out the ATMega1284P based Bobuino and other '328P & '1284P creations & offerings at  www.crossroadsfencing.com/BobuinoRev17.
Arduino for Teens available at Amazon.com.

z2012

Every wire or lead you put in ur bread board forms a capacitor. Try making a collpits oscillator on a bread board, u will see it operate no where near the calculations as you increase the frequency. So your bread board is your problem. 

winner10920

Other than a right angle being longer in most situations than two 45s, how does it act as an inductor? I imagine unless it very high frequeny there shouldn't be reflections and even at higher currents being at 90 degrees the fiels shouldnt affect anything i think

winner10920

And at wht frequency doea that right angle start to affect things?

WillR


And at wht frequency doea that right angle start to affect things?


That depends....

This is a well understood phenomena.
http://ocw.bib.upct.es/pluginfile.php/8158/mod_resource/content/1/Best_practice_in_circuit_board_design.pdf

What the effects might be are harder to say as the designer of a board does not have control of the operating conditions. The boards could well be placed in a strong RF field -- i.e. near a transmitter. The connections may well "inject" an RF signal from a clock signal.

Do some searching on your own.

Here is one possible search.
http://www.google.ca/#hl=en&gs_nf=1&pq=pcb%20design%20smooth%20traces&cp=34&gs_id=1nz&xhr=t&q=PCB%20design%20curved%20traces%20vs%20angled&pf=p&sclient=psy-ab&oq=PCB+design+curved+traces+vs+angled&gs_l=&pbx=1&bav=on.2,or.r_gc.r_pw.r_qf.,cf.osb&fp=e3ea13f2fea9acba&biw=1461&bih=790

Much more discussion on this point would be hijacking the thread.
Just another Hacker

winner10920

Sorry about that, cant help when I see something interesting

Go Up