Misbiased Opamp to 1 meg to Vref. [period]

[petemoore]

  http://aronnelson.com/gallery/melanhead/HeadCase
 
  It's probably just needs debugged some more but, Has this project been verified?
  It all looks 'right' on the schematic, thanks Melanhead for your layouts contributions...

[Melanhead]

Yup ... I built 3 of 'em with this layout and a friend of mine also built one .... I haven't had any issues building it other than forgetting to cut the traces under the op-amp ... It's a simple build so it's probably something silly ... I'd love to hear your comments on it once you get her going. Both friends that have it love it!

[petemoore]

   now both opamp + inputs, the ones the signal comes into, through a DC blocking cap only and a 1meg resistance to Vbias...
  Vbias reads 1/2v, but going through the 1meg [measured to the +input [verified as isolated bias-wise], [both OA's], I'm getting readings of 2.42Vdc exactly on them...I can't figure out what is pulling them toward ground, the 1m's measure ~1m, the Vbias coming in at a nice ~1/2v, nothing else in the picture except the DC Blocking cap, which of course won't influence DC Bias. 
  The rest is also quite simple and biases nicely, all the neg inputs and outputs are ~1/2v. It sure looks ok...
  The only part I couldn't really 'find, looking at various mammalian pest schematics, is the tie from the neg input through an R/C [ITCase a 47ohm to .68uf to Gnd.], well there's a split R/C on the Rat...topic for other interesting {OT ITCase] discussion...why a split R/C there....perhaps it has to do with current /vs. frequency, one 'side' appears 'easier' when current rises or something.
  @@ Rate I spent more time for verifications, [the 'ol continuity, resistances, voltages, non-continuity tests] concentrating on the +input, having a hard time fathoming because it looks, measures and seems so simply right, just two connections, one DC Blocked, the other to 1/2v...tried a buncha OA's, nothing seems to want to bring up the voltage on the +input...of course I was at a loss as for things to actually try on it.

[petemoore]

  Just wierd.
  No reason why, silly really for me to type about verified...as shown, the problems I'm having can't/shouldn't exist...the -inputs there, both go to ... nothing but 1/2v via a 1 meg, [+ the 1k on the input OA] all measured, continuity checked, non-continuity checked to all other points, oh...the DC blocking caps, but they don't mess with DC Bias then.
  The soundclip was cool, but the thing that should not be turned very wierd, V+ Started measureing some fluxuating voltages around 2, and at the same time continuity between V- and V+...which of course is impossible, so I doubled the checks there just to see it a few more times, and checked the battery voltage and continuity, measuring 9.68v battery and continuity between V- and V+ on the battery...I need some kinda break...

[Melanhead]

THat is odd! ....

Hmmm, so you're using a 1M res at each OA + input ? ... The schematic calls for 1 at the first stage, I used at 510K but a 1M shouldn't be an issue. The second stage doesn't need a bias resistor to V ref as there is no blocking cap between the stages and VRef should travel through ( just like Jack O's Son Of A Screamer ... )  ... Not sure if this is causing any issues.... I'm getting more confused when you speak of the R/C network of 47 ohm and .068uf, 'cause this is from another project, "The Headcase" ... which also has two 1M bias resistors.
I've only bilt one using a vero layout I made from the schematic so it should work. are you sure it's the MOD that you've built ?

tungngruv built both the Mod and Headcase in one enclosure and I dont' think he had any issues ...

This is one of those things that's gonna drive you crazy, just like my Phase45 I can't seem to get to work 

[tungngruv]

Yep, I built both and they fired right up. I even had to "fake" stripboard by just using ends of components to create the strips on perfboard. The layouts for the M.O.D., Distortion and Headcase were all first time starters.

[petemoore]

  I built the Headcase..yupp yupp yuuppp
  they looked so similar I didn't notice..
  Yes that one has the DC cap and 1 meg to vref thing going on, and that's the trouble, I cant figure what's pulling the +input of the second...and first OA's down to around 2 volts...I went through, around, over, and checked...
  that 1 meg is 1 meg
  that it connects only to Vref on one end and the + input on the other, also the cap
  That it connects to nowhere else but the +input...both +inputs are too near ground.

[Melanhead]

 

Ahhh! ... 

What about the blocking caps ? Maybe one's funky ! ?

[petemoore]

  I thought of that, and the couplers are hard to get to...but not the 'one leg, perhaps I pull that leg and see if it doens't let the +input 'float up' to a nicer bias point...I looked from all angles...can't be...but the thing that should not be...is!!
  LoL...I'll check that out...

[Melanhead]

Don't drive yourself crazy but it's definitely worth building! ... It's now my Fav pedal

[petemoore]

yeap,,,clips are cool, schematic looks good...
  still...it's:
  "JUst WIerd"...I go check that cap soon...thing is it's both +inputs...so...a real 'go figure' item...

[petemoore]

  Battery alone 8.07 [all readings..are VDC] and connected to cct. 7.84
  Opamp pins
  1{3.96
  2{3.95
*3{1.96
  4{0.0
  5{7.89
  6{3.94
  7 {3.93
*8 {1.96
  The only thing connected to pin 5 [2nd OA's +input] is a measured 1meg, to Vbias, which is real near 1/2v.
  Pin 3 [Neg input of OA1] is still connected through the gain loop to pin 1 [output...which is biased ok...and @ 18k -118k >I used an 18k stop resistor series with 100k pot],  should bring bias closer if anything]...pin 3 also has a measured 1k to 1meg to Vbias].
  Thisis super staranged out.
  I clipped to pin 5's 1meg connection [voltmeter set to beep mode, black lead clipped to pin five and verified to beep to pin 5...and only to pin 5] I went and touched every other conductive surface near/around/far/on, and offboard points to verify that Pin 5, after disconnecting the capacitor connection shown on the schematic, Pin 5 goes ONLY to a measured 1meg, which connects ONLY to Vbias, which is very close to measuring 1/2V.
  So...hate to be the party killer, but I'm at a loss, me the 'expert debugger', cause I make mistakes...I'm at a total loss as to how to proceed...other than
redouble-Blingbling my supertripled-doublecheck efforts....
  That or start over...but...I kind of got a nasty aftertaste, and cannot isolate what ingredient is illiciting it, so it's unlikely...unless I can somehow isolate the problem and find the total nonsense...that I'll be repeating the recipe...looked real nice and the soundtaste had me goin'...Unless I find an alternative debuggin' technique/solution, and I can't imagine how there is one...audio probing isn't necessary 'cause I've found the problem...it's gonna have to go down as a 'failed'...one must count ones 'winnings' at some point in the game, I must concede: it's got me beat.

[MartyMart]

Pete, looks very "wierd" to me .... obvious I know, but have you swopped out the
op-amp at all ??  or checked for bent pin ....
Is pin 8 really not at battery voltage !!

MM.

[petemoore]

  Yupp, tried all the unusual as well as usual debugstuff like OA swaps...all OA's exhibiting closeley the same results...voltage and otherwise...Bad OA...doubtful.
  The two battery voltages shown :
  8.07VDC...with battery connected to DMM set to 20v range
  7.84...with battery connected to circuit And DMM from V+/V-...pin 4 and pin 8.
  This...I don't quite recall how to interpret...seems like borderline normal voltage drop [maybe a little much?] that the circuit would drop voltage by .23v ... from 8.07 to 7.84...
  I kind of disregard that for the time being...but perhaps I should try that angle of approach if there is one...ascertaining why the battery drops voltage when connected to the circuit.
  I disregard a power supply cap problem, being as the 1/2v is real near 1/2v.
  Not knowing what could happen inside the opamp...I figure there's a DC blocking cap or something separating the + from the - input, also -input would be isoalted from the output...I mean...I dunno...there's not much there except that '1/2v'd 1meg'' ...just a resistor goin' to 1/2v...that IS it....
  both +inputs misbehaving, final point for ponderances...ring any bells?
 

[Johan]

addmitedly, i didnt read through all replies, so it might allready been covered..but people tend to forget that bias is not only a voltage, but a current too. at 9volts, it might be hard to get any bipolar opamp to bias properly through a 1M resistor. it simply cant pull enough current through it...reduce it to 100k for 4558 or even 10k for 5532...or replace the opamp with a J-FET type like TL-072, they need MUCH less current to bias..

johan

[MartyMart]

addmitedly, i didnt read through all replies, so it might allready been covered..but people tend to forget that bias is not only a voltage, but a current too. at 9volts, it might be hard to get any bipolar opamp to bias properly through a 1M resistor. it simply cant pull enough current through it...reduce it to 100k for 4558 or even 10k for 5532...or replace the opamp with a J-FET type like TL-072, they need MUCH less current to bias..

johan

That's right and Pete has less than 8v to start with !
Had something similar with a self designed 2x dual opamp circuit using 4559's
used 100k's as stated to get enough current for circuit to "work" / bias.

MM.

[petemoore]

  Just so happens I just did just that.
  I started with a 470k across the 1meg to Vref, that brought the second oa's +input to 2.87v, Then with a 100k, I got 4.61v, from a battery...what was it...9.20V [I've been charging my 9v 'brick' batteries.
  Reducing the resistance between the OA pins and Vref pulled the bias [up] closer to 1/2v. Just a resistor, connected to Vref, with the other end floating...a 1meg to Vref, other end 'open' measures 2.2~V from Ground, or a 470k from Vref with other end floating measures 2.87v, the 100k...well I tried that with the opamp out, and with the opamp in...and it *Was riding at 4.61v, or very reasonable bias.
  And the circuit seemed to be working...for a time
  Now I'm getting .8v V+ when connected to a battery that measures 9.+v when alone..the circuit is pulling the voltage wayy down...so...whatever...back to point '-B'...or a few below the 'starting point '.