Help me understand the Micro Vibe

[jlo]

in the Univibe R11,R17,R23 R29 is 4K7.  Ive read that the resistors in series to the LDRs will affect the light resistance.   In the Microvibe are R10,12,14,16 (47K)  in series with the LDRs and can that value be lowered to get a lower light resistance? 

I understand that Im comparing op amps to a discrete circuit.   In the Easy Vibe it looks like 10K is the value chosen.  Whats the reasoning behind all of this?

[antonis]

Maybe just because we don't deal with identical optocouplers..??

[duck_arse]

in both the microvibe and the easyvibe, those resistors are not in series w/ the ldrs. the values are chosen as a match for the opamp feedback resistors to limit gain to unity each stage, and to work against the cap at the other input. in the univibe, the 4k7's .... dunno, balance out the ldr's and limit their swing?

[jlo]

in both the microvibe and the easyvibe, those resistors are not in series w/ the ldrs. the values are chosen as a match for the opamp feedback resistors to limit gain to unity each stage, and to work against the cap at the other input. in the univibe, the 4k7's .... dunno, balance out the ldr's and limit their swing?

Thx for the explanation.  So if I change opamps do I need to change the resistors?  The stock Microvibe has the TL062.   What if I sub in a TL072 or OPA2604 or OPA2134?

[duck_arse]

opamps is opamps. you tell them what to do with the resistors, and they do it, or go in the bin.

no, no need for opamp part number dependant resistors.

[jlo]

Yes I misunderstood.  The resistor at the output matches the feedback resistor.   And what effect would lowering or raising the value of the resistor have?  Comparing the Micro Vibe and the Easy Vibe why was 47k chosen vs 10k

[antonis]

And what effect would lowering or raising the value of the resistor have?

None, by maintaining equal resistor ratio..
Actual resistors values only counts  for noise & off-set issues..

@ Stephen: 4k7 resistors in the univibe are actually set inside NFB loop but are placed on Darligton Emitters rather than Bases for not upsetting bias bootstrapping..
(rough guess..)

[jlo]

Ok. Things are starting to make sense.  Now how about the gain staging?  According to RG Keen the preamp of the Univibe has a gain of 4.  And the phasing stages are unity? 

For the Micro Vibe what is the gain staging?   U1-1 is the input buffer?  Is the gain 2 if the input and feedback resistors are 47K? Thats for U1-1/2,U2-1/2 and U3-1 each have a gain of 2?   And then what happens with U3-2? 

[antonis]

It's a bit complicated topology..

e.g. U3-2 is wired as non-inverting configuration with its own stage gain of (1 + R19/R21) BUT actuall signal comming into pin5 is what comes out of U3-1 divided with R22/[(R20//(R18+R9+R8)] (or multiplied with the reverse fraction - to be more true-blue voltage divider formula..)

U1-2, U2-1, U2-2 & U3-1 are wired as inverting configuration with unity gain BUT their non-inverting inputs are biased via LDRs and toghether with respective capacitors (C5, C7, C8 & C9) high-pass filter configuration form all-pass filters..
At high frequencies caps are shorts so we have unity gain voltage buffers (no phase lead)..
At low frequencies caps are open circuits and we have unity gain inverting amplifiers (180^o phase lead)
At corner frequency of HPFs (0.159/R*C) we have 90^o phase lead..
(I let you guess the reason for it.. hint: at corner frequency LDRs resistance and capacitors impedance are equal..)

LDRs resistance varies with light intensity which variation is militated by lower right circuitry (LFO) resulting into LFO frequency depended "moving" phase lead..

[ElectricDruid]

The resistor at the output matches the feedback resistor.   And what effect would lowering or raising the value of the resistor have?  Comparing the Micro Vibe and the Easy Vibe why was 47k chosen vs 10k

No, the resistor at the *input* matches the feedback resistor. I know it looks like that because it's connected to the output of the previous op-amp, but you're dividing the parts up wrong. That pair of resistors are an inverting op-amp arrangement.
If you look at the equation for the gain of an inverting op-amp, it's

gain = -Rf/Rin

and since 10K/10K = 1 and 47K/47K = 1, both give unity gain and unless you only change *one* of the resistors, it doesn't much matter what the value is within reason, at least not for the gain.

[jlo]

It's a bit complicated topology..

e.g. U3-2 is wired as non-inverting configuration with its own stage gain of (1 + R19/R21) BUT actuall signal comming into pin5 is what comes out of U3-1 divided with R22/[(R20//(R18+R9+R8)] (or multiplied with the reverse fraction - to be more true-blue voltage divider formula..)

U1-2, U2-1, U2-2 & U3-1 are wired as inverting configuration with unity gain BUT their non-inverting inputs are biased via LDRs and toghether with respective capacitors (C5, C7, C8 & C9) high-pass filter configuration form all-pass filters..
At high frequencies caps are shorts so we have unity gain voltage buffers (no phase lead)..
At low frequencies caps are open circuits and we have unity gain inverting amplifiers (180^o phase lead)
At corner frequency of HPFs (0.159/R*C) we have 90^o phase lead..
(I let you guess the reason for it.. hint: at corner frequency LDRs resistance and capacitors impedance are equal..)

LDRs resistance varies with light intensity which variation is militated by lower right circuitry (LFO) resulting into LFO frequency depended "moving" phase lead..

Thank you for the detailed explanation.  Yes its complicated to someone like me that has little to no electronic knowledge. 
What about U1-1?  How does it compare in gain to the Univibe preamp?   I see that Wampler recommends changing R8 from 47k to 39K. 

[antonis]

What about U1-1?  How does it compare in gain to the Univibe preamp?   I see that Wampler recommends changing R8 from 47k to 39K.

U1-1 is wired as non-inverting configuration (that's when signal goes into + marked Input)
Gain of such a configuration is what Tom said above, without minus sign (-) plus unity.. (1+Rf/Rin with no phase reversal - although Rin should be named Rgain 'cause there isn't any input signal through it..)
So, U1-1 stage gain (ignoring R18 & R20) is 2 (1 + 47k/47k).. By making R8 39k you just raise gain 10%.. (from 2 to 2.2)

P.S.
Any "complicated" circuit consists of a number of more "plain" blocks so you firstly have to locate (isolate) them, understand their function and then cascade their individual facilities..

[jlo]

What about U1-1?  How does it compare in gain to the Univibe preamp?   I see that Wampler recommends changing R8 from 47k to 39K.

U1-1 is wired as non-inverting configuration (that's when signal goes into + marked Input)
Gain of such a configuration is what Tom said above, without minus sign (-) plus unity.. (1+Rf/Rin with no phase reversal - although Rin should be named Rgain 'cause there isn't any input signal through it..)
So, U1-1 stage gain (ignoring R18 & R20) is 2 (1 + 47k/47k).. By making R8 39k you just raise gain 10%.. (from 2 to 2.2)

P.S.
Any "complicated" circuit consists of a number of more "plain" blocks so you firstly have to locate (isolate) them, understand their function and then cascade their individual facilities..

Ok thats what I thought re U1-1.  But the original Univibe preamp has a gain of 4.   So the gain in the MV is made up in U3-2?  Which has a gain of 2.27 “ BUT actuall signal comming into pin5 is what comes out of U3-1 divided with R22/[(R20//(R18+R9+R8)] (or multiplied with the reverse fraction - to be more true-blue voltage divider formula..)”.  What does that mean? 

Also does it matter where the gain is in the circuit?  I get that the Univibe is not true bypass vs the Micro Vibe which is.

[antonis]

So the gain in the MV is made up in U3-2?  Which has a gain of 2.27 “ BUT actuall signal comming into pin5 is what comes out of U3-1 divided with R22/[(R20//(R18+R9+R8)] (or multiplied with the reverse fraction - to be more true-blue voltage divider formula..)”.  What does that mean?

Maybe I talk (write) too much resulting into listener (reader) confusing..

U3-2 STAGE (isolating it from whatever is set in front of it) gain indeed is 2.27..
For shake of simplicity, ignore any resistor on the left side of R20 upper leg..
Signal coming out of U3-1 (pin1) reaches U3-2 pin5 by the half of its previous amplitude due to voltage dividing effect..
(Vpin5 = Vpin1 X R20/R22)
Now, if you substitute everything goes to AC GND for the single R20 you have the equivalent resistane of voltage divider lower leg..

Also does it matter where the gain is in the circuit?

Hmmm.. That's a, more or less, philosophical query..
(in the mean of: "Better attenuate then amplify" or vice-versa..??)

[duck_arse]

@ Stephen: 4k7 resistors in the univibe are actually set inside NFB loop but are placed on Darligton Emitters rather than Bases for not upsetting bias bootstrapping..
(rough guess..)

are they? circuits like the magnavibe don't have the series resistor [or, usually, the darlington], I thought it was just a mixing function.

[antonis]

It depends on particular point of view..

You may also face it as a variable impedance Emitter decoupling cap for individual Darlington stage variable Gain..
(wired on Collector for more effective "grounding" - apparent LDR + 4k7 value lowered by stage gain..)

[jlo]

Ok.  So the gain of U3-2 is 2.27 but the signal coming from U3-1 has been halved.  Is this because we are combining the dry and wet signals?  So in the end the whole thing is unity?   The gain of U1-1 is 2 but we are splitting the signal into the wet and dry paths.  Phase stages are unity.  Am I on the right track?

Thx again for all the detailed explanations.  They may be over my head at first but it is helping me learn alot!

[antonis]

Ok.  So the gain of U3-2 is 2.27 but the signal coming from U3-1 has been halved.  Is this because we are combining the dry and wet signals?  So in the end the whole thing is unity?   The gain of U1-1 is 2 but we are splitting the signal into the wet and dry paths.  Phase stages are unity.  Am I on the right track?

No..!!
(with no further explanation..) 

[jlo]

 

Ok.  So the gain of U3-2 is 2.27 but the signal coming from U3-1 has been halved.  Is this because we are combining the dry and wet signals?  So in the end the whole thing is unity?   The gain of U1-1 is 2 but we are splitting the signal into the wet and dry paths.  Phase stages are unity.  Am I on the right track?

No..!!
(with no further explanation..) 

[jlo]

Oops I mean a gain of 2.27.  When combining the dry and wet signals we are back to unity