Neovibe mis-bias??

[Arno van der Heijden]

bump 

(I don't have a clue what to look for....)

[R.G.]

Ok, let's see if we can round up some of these issues.

Bipolar transistors **normally** have a base-emitter voltage of close to 0.6V. It's only "close to" because the current that flows in the base-emitter junction is an exponential function of the junction voltage. That's a fancy way of saying that as you increase from 0V, almost nothing happens until you hit a voltage where the current starts to take off and then the current increases, well, exponentially. It's where we get the term.

However, there is some current that flows well under the nominal turn on voltage. With a high gain transistor, the transistor -may- be in conduction because the base is allowing nano-amperes through. That can happen down at 0.3V and up. It is common to see high gain low signal transistors that only have 0.5V on the base-emitter as a normal condition in operation. The first transistor in the Univibe has only a very low current in its collector, so it -may- have a very low base current and -may- have a base emitter voltage down in the confusingly-low region and be working fine.

It is quite difficult to measure low voltages accurately at high impedance nodes. The base-emitter of a bipolar transistor can be such a beast.

With that as background, I don't consider the B-E of Q1 being measured at 0.3-0.5V to be necessarily abnormal; what would matter is the collector and base voltages in that event, and whether the unit sounded good or not. If the unit works OK otherwise, and sounds OK, chances are it is OK and the measurement is at fault, not the unit.

Then there's that clipping thing. There are always pickups which can make anything clip. Is it abnormal? Who knows? Depends on the voltages and the pickups.

So:
All of you - measure and post all of your voltages on C, B, and E for Q1, 2, 3, 4, 5, 6, 7, 8, 9 and 10. The other voltages will tell us whether the hard to measure ones are at fault.

[Arno van der Heijden]

Hi R.G.,

Thanks for your help. Here are my measured voltages, as well as a pic of my build:
(without the phase caps)

Q1
C = 2.09
B = 0.85
E = 0.96

Q2
C = 4.16
B = 2.09
E = 1.50

Q3
C = 10.95
B = 4.17
E = 3.16

Q4
C = 14.74
B = 4.16
E = 4.20

Q5
C = 11.09
B = 4.20
E = 3.61

Q6
C = 14.74
B = 4.17
E = 4.21

Q7
C = 11.08
B = 4.22
E = 3.63

Q8
C = 14.74
B = 4.17
E = 4.22

Q9
C = 11.07
B = 4.22
E = 3.63

Q10
C = 14.74
B = 5.30
E = 5.37

[R.G.]

Just out of curiousity, what brand and model of measuring instrument did you use to measure the voltages?

[Arno van der Heijden]

Just out of curiousity, what brand and model of measuring instrument did you use to measure the voltages?

I use a Topcraft / Toledo TMMH-930, which are similar, rebranded units. They are super cheap <10 euro, but have never let me down before.


As far as the pickups, they're just plain Gibson/Epiphone alnico humbuckers in a ES-335 model. I wouldn't expect them to be particulary high output.

BTW, I made a mistake, Q3 should read:

Q3
C = 10.95
B = 4.17
E = 3.62

[R.G.]

OK. Can you go back to the board and measure the Vbe on Q1, 2, 3, 4, 6, 8, and 10 by putting the black lead on the emitter and the red lead on the base?

[Arno van der Heijden]

OK. Can you go back to the board and measure the Vbe on Q1, 2, 3, 4, 6, 8, and 10 by putting the black lead on the emitter and the red lead on the base?

Ok here we go:

Q1  = 0.511
Q2  = 0.591
Q3  = 0.621
Q4  = 0.461
Q6  = 0.461
Q8  = 0.461
Q10 = 0.591

Surprising? 

[R.G.]

That's kind of what I thought we'd find. The "misbias" was all an artifact of the method of measurement. Those measurements all show the base and emitter are well within the OK range of Vbe.

When making measurements, if you measure two things that are nearly equal and then subtract to get the difference, the difference contains all of the errors of both measurements it was taken from. You can get big errors this way. By measuring the Vbe directly, we avoided the measurement error.

Why don't I tell people to do it that way all the time in my "What to do when it doesn't work"? Because you often want base and emitter voltages separately for other things, and it's unusual for this issue to come up. Yes, three (?) people have had it recently. Maybe I should tell people to measure ALL of Vbase, Vemitter and Vbe directly, explaining this same thing to them from the other direction.

In any case, your neovibe is biased OK, and if it sounds good, it's working as designed. The distortion on humbuckers is a separate issue.

[Arno van der Heijden]

In any case, your neovibe is biased OK, and if it sounds good, it's working as designed. The distortion on humbuckers is a separate issue.

That's what I expected... kind of....

Anyway, thanks for your help!

BTW, maybe it's time for me to buy a better quality DMM.

[Solidhex]

Yo

  Wow! I'm using a Craftsman DMM ... http://www.sears.com/shc/s/p_10153_12605_03482139000P?keyword=multimeter. Testing the voltage between emitter and base I get:
Q1 .56
Q2 .57
Q3 .6
Q4 .45
Q5 .61
Q6 .45
Q7 .61
Q8 .44
Q9 .61
Q10 .57
Q11 .45
Q12 .62

--Brad

[R.G.]

You're good. You wound up in the same measurement pit.

The voltage on the bases of some of those transistors can be pulled down even by high-ish impedance meters sometimes. The Vbe voltage is much lower impedance, though, being buffered by the gain of the transistor itself. This reduces measurement error by both the method (nor measuring and then subtracting) and by lower sensitivity.

[yeeshkul]

OK. Can you go back to the board and measure the Vbe on Q1, 2, 3, 4, 6, 8, and 10 by putting the black lead on the emitter and the red lead on the base?

Ok here we go:

Q1  = 0.511
Q2  = 0.591
Q3  = 0.621
Q4  = 0.461
Q6  = 0.461
Q8  = 0.461
Q10 = 0.591

Surprising? 

I had the same problem and my voltages were the same as yours +-0.05V (i mean the same trannies misbiased the same way). I measured Vbe voltages and again got the same results like you.
BRILLIANT, everything must be ok then . What a helpful thread!