Capacitor dyslexia

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Capacitor dyslexia

Frank_NH
So last night, I thought I'd do a quick build of a circuit I've been meaning to try out for a while, the Jack Deville Deuce Coupe:



I just got some charge pump chips last week, and this was a simple project.  Everything went together very smoothly.  So I fired it up and...no overdrive!  Just a unity gain signal.  

I then spent the next two hours probing, testing, measuring voltages, changing chips...nothing.  WTF???

I was sitting down later looking over the schematic when it suddenly dawned on me.  That "1u" capacitor on the left??  Yep - I had used a "1 nF" capacitor instead!!  Oooops....

It so happens that this capacitor is in the ground leg of the op amp feedback loop.  The op amp gain for the non-inverting configuration in the Deuce Coupe is:  Gain = 1+Zf/Zg, where Zf is the impedance of the feedback loop and Zg is the impedance of the ground leg attached to the inverting (-) input.  Using a 1 nF cap made Zg increase by several orders of magnitude (over the normal effect frequency range of 50 Hz - 10000 Hz)!    Hence Zf/Zg ~ 0 --> Gain = 1 --> no amplification.  Duhhh.  

So if you ever have a problem build and can't figure out what's going on, keep an eye out for capacitor dyslexia...  

(There is also a related malady known as "resistor dyslexia"  where 470K ohm resistors are used instead of 47K, 4.7K, or 470 ohm resistors...)
Reply | Threaded
Open this post in threaded view
|

Re: Capacitor dyslexia

fenderguy79
Hahaha
I've definite been there a bunch of times.  
Last one I gave up pretty quick and went to bed.  The next morning I found I put a 220pf in place of 220nf because the 220pf cap picture on the board looked like a pf cap.  
I've also done the resistor lysdexia thing
Reply | Threaded
Open this post in threaded view
|

Re: Capacitor dyslexia

cylens
with every build I invent new ways to fuck up: my personal favorite is just to omit one resistor or one cap on a really crowded vero...

the last one, though, was quite funny: on a Maestro Brassmaster (layout by sabro), I mentally numbered a 3PDT switch "horizontally"...

it was a PITA to debug, since there's almost nothing at the output of the transformer and I was persuaded that it was the culprit...
Reply | Threaded
Open this post in threaded view
|

Re: Capacitor dyslexia

motterpaul
How many times have you charted out your cuts to start  project, and turn the VERO over and suddenly see you were holding it 90-degrees off (so the lines flow up & down, not across)?  I haven't done it a lot but I did it twice in a row one day.
Reply | Threaded
Open this post in threaded view
|

Re: Capacitor dyslexia

Frank_NH
Paul, I've gotten done prepping a vero and noticed I had one hole way off!  Grrrr.   "Snap" goes the vero - into the trash.  Fortunately, veroboard is cheap...
Reply | Threaded
Open this post in threaded view
|

Re: Capacitor dyslexia

Travis
Administrator
In reply to this post by Frank_NH
God only knows how many otherwise healthy boards have hit the bin due to this epidemic..
Reply | Threaded
Open this post in threaded view
|

Re: Capacitor dyslexia

Frank_NH
BTW - the Deuce Coupe sounds great.  The 18V (actually -9V to +9V) rails give it a much cleaner, less compressed overdrive sound than similar 9V op amp medium gainers.  Could use a touch more treble, but I won't mess with it.  I actually built the non-boost version, so it'll be easy to box in a 1590B.