# Collector feedback bias...

Discussion in 'The Projects Forum' started by rougie, Sep 2, 2012.

1. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
hello,

wow...wow...wow

I did my first collector feedback bias circuit and I was impressed as to how close the measured and calculated values were. See attached file.

BUT THEN.... I swapped the transistor with another one with the same part # and wooops!

The Ie was now 1.36ma and Vc was 3.74!!!

I can't believe how volatile these things really are!!!!

But I guess this circuit is alot more stable that the simple base bias one!

Okay... I will now learn the other 2 types!!!!

thanks guys!
r

File size:
208.1 KB
Views:
69
2. ### #12 AAC Fanatic!

Nov 30, 2010
14,988
5,572
I think a 2 resistor base bias is more stable, as in, not depending on the gain of the transistor to decide the operating point. Eagerly awaiting you next post when you tell us what you have learned.

3. ### Sensacell Active Member

Jun 19, 2012
874
191
The hfe or beta of transistors varies widely, in general, if your design requires a specific value of hfe to work correctly, it's not a very good design.

It's often kind of horrifying to see the data sheet spread of hfe and realize what this means for your design- for it to work correctly with such a huge spread of gain. For example, the 2N4123 data sheet says it goes from 50 to 200!

4. ### debjit625 Well-Known Member

Apr 17, 2010
790
186
Except for Voltage Divider Biasing all other biasing with negative feedback circuit for a BJT in common emitter configuration will be $\beta$ dependent.

Note in Voltage Divider Biasing, the effect of $\beta$ is less but its not obsolete

I will prefer you to use some simulator software to perform these experiments,before doing them in real.

LTSpice will be good.
http://www.linear.com/designtools/software

Good Luck

5. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
</
I think a 2 resistor base bias is more stable, as in, not depending on the gain of the transistor to decide the operating point. Eagerly awaiting you next post when you tell us what you have learned.
/>

One thing though, asides from showing better stability than
a feedbackless base bias common emitter circuit, there isn't
much use for this circuit .... Right?

As you said, I can't even control it !!!

</
The hfe or beta of transistors varies widely, in general, if your design requires a specific value of hfe to work correctly, it's not a very good design.
\>
Yes Indeed!

Thanks

6. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
I will eventually look into LTSpice...

7. ### MrChips Moderator

Oct 2, 2009
11,970
3,184
If you wish to insert other people's comments into your post, use the QUOTE tag.
Instead of the </ and /> that you used, end the quote with [/quote].
Begin the quote with

8. ### DickCappels Moderator

Aug 21, 2008
2,234
512
If you reduce the value of the (470k) feedback resistor to something like 47k, the collector voltage and current will be much less dependent on the current gain of the transistor.

If taken to the limit -that is shorting the base to the collector, it will be even more stable, but less useful

#12 likes this.
9. ### ramancini8 Member

Jul 18, 2012
414
97
The calculated emitter current for Hfe=200 is 1.35mA, and for Hfe=50 is .671mA. The Hfe varies by a factor of 4 while the emitter current only varies by a factor of 2. Drop Rb to 47K and the numbers are Hfe=200, Ie=1.97mA and Hfe=50, Ie=1.69. Hfe varies by a factor of 4 while emitter current varies by a factor of 1.16.

This shows the dramatic effect that feedback can have on DC stability, but the circuit gain is reduced proportionately. If the gain required is AC it can be restored by bypassing Rb with a capacitor. A popular configuration splits Rb with a bypass capacitor to control DC and AC gain.

The two resistor base bias is useless unless there is an emitter resistor involved. Take the Thevinin looking from the base to the junction of the two base resistors and you get one resistor in series with a source; no feedback.

10. ### ramancini8 Member

Jul 18, 2012
414
97
Not true unless you are using emitter degeneration (an emitter resistor). What don't I see????

11. ### #12 AAC Fanatic!

Nov 30, 2010
14,988
5,572
Well, of course you have to use an emitter resistor. I think you're not seeing that most of us know you will need an emitter resistor.

12. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
thank you all ... your help is very generous...

r

13. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
okay,

I have been looking at this collector feedback biased circuit. But just wondering based on this quote:

"If the emitter current were to increase, the voltage drop across RC increases, decreasing VC, decreasing IB fed back to the base. This, in turn, decreases the emitter current, correcting the original increase."

1) there is no way for us to control the gate current as it is established via rb which is connected to rc!

2) Therefore, where would I use such a circuit?

I have no issues with the math shown here:

but what I don't understand is what would be the use for a circuit like this where it's gate cannot be controlled?

PS. I hate to bring this post back up.... but I would like for someone to bring up some usable practical reasons besides the fact that it is more stable than a simple common emitter base bias circuit.

r

Last edited: Sep 4, 2012
14. ### #12 AAC Fanatic!

Nov 30, 2010
14,988
5,572
Please stop using the word, "gate" when you mean, "base".
There is a way to control base current. You control the base current when you design the circuit. Also, you can use this configuration as an amplifier. That means you can control the base current by adding a signal which needs to be amplified.

Why use this configuration? It uses one less resistor than a 2 resistor base bias circuit.

15. ### ramancini8 Member

Jul 18, 2012
414
97
We correct newbies for not giving enough information, so we should be true to ourselves.

If you use an emitter resistor the stage gain for a large Hfe transistor is Rc/Re, and that is a lot less than is usually derived from a collector-base feedback resistor. There are many ways to skin a cat, and all must be tried before we can say which way is best. Of course you can choose to bypass the emitter resistor with a cap to improve AC gain.

16. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
Hello,

Mr #12, you have replied:

And I will, here it is (read till the end):

Now, one thing though, before I start, I would like to make something clear.... The way I see it according to what I have read here:

is that there are 3 main ways to bias a transistor. The first way is the BASE BIAS method. I have evaluated it's math, I have calculated it, I have built it, I have measured it and I have compared real readings to the calculated ones.

Having said this, here is what I have learned about the bias base method.

BIAS BASE METHOD:

1) The HFE varies from transistor to transistor however the manufacturer gives us a min/max HFE range variation for the part.
2) This method is mostly appropriate to use as an on/off switch.... (saturation / cut off mode)
3) This method can also be used in active region mode, however, the Vc and Ie readings will vary from one transistor to another transistor since there is no feedback resistor to stabalize the HFE.
4) In saturation mode, we calculate Ib so that it is 1/10th of Ie. Hence: Ib=(Vcc/Rc)/beta.
5) In active region mode, we calculate ic by:
ic = beta(ib) where beta needs to be measured for that particular transistor. Also we calculate Vce by doing Vce = Vcc - Vrc.

The second way to bias a transistor, is by the collector bias feedback method. I have evaluated it's math, I have calculated it, I have built it, I have measured it and I have compared real readings to the calculated ones.

Having said this, here is what I have learned about the collector bias method:

COLLECTOR BIAS METHOD:

1) Although the HFE varies from transistor to transistor at manufacturing, we use a feedback resistor connected from Rc to the base of the transistor allowing us to better control the beta of the transistor based on the following fact described at the link above which happens at the speed of light as:

2) Because of this collector feedback configuration, we have to take Rb into consideration when calculating Ie... and mathematically we have:
Ie= Vc-Vbe/(Rb/Beta)+Rc

as opposed to the simple emitter bias configuration which was:
Ie = Vce-Vbe/(Rb/beta)

rending this configuration much more stable than the previous one.

Now, Mr #12, when you say:

In all due respect, and I am not saying you are wrong, I am simply confused by the statement here..... How can you state that a 2 resistor base bias configuration is more stable than the feedback collector configuration which offers the convenience of controlling the beta of a transistor. I have been exposed to transistors for approximately one week, and all I read in all my posts was that a feedback resistor *MUST* to be added in to achieve better stability..... Honestly.... confused!

EMITTER BIAS FEEDBACK

Getting back to what I was saying, there is also a 3rd way of biasing a transistor which is the emitter bias feedback resistor method which is the next one I did... by the way I will be posting a question on that one since something isn't right with the REE!!!

I have learned that in the emitter bias feedback resistor method, there is more of the same however, the KVL loop is concentrated on the base to emitter part of the circuit... whereby I don't understand as to why the KVL loop doesn't involve the collector branch too... I will have to re-read the document to better understand this.

VOLTAGE DIVIDER BIAS

A fourth way to bias a transistor is an off-spring of the emitter bias feedback method which replaces the VBB by a voltage divider. Thevenin theorem is used to calculate the voltage divider resistors.

Understood!

I did not recalculate this so to try it on my circuits... this is only because, I have about 20 sheets with circuits on them as I was doing tests...I have sheets and examples all over the place here ... LOL but nonetheless, for me, this is an interesting observation! Thanks!

Ah, I think this makes sense to me now... LOL

I have a question about Emitter bias feedback which I will post in another thread.... in the mean time thanks all for your help!
r

Last edited: Sep 5, 2012
17. ### #12 AAC Fanatic!

Nov 30, 2010
14,988
5,572
Consider that I have a 9 volt battery running an amplifier with about 1 milliamp of collector current. I want to get as much output voltage as I can. My transistor might have a gain of 50 and it might have a gain of 200. What size restance will I use for the bias resistor and how much will the DC collector voltage be when the gain is 50, when the gain is 200?

If I used 2 resistors to set the base bias and an emitter resistor to set the idle current, how much would the DC voltage be on the collector if the gain of the transistor is 50, and when the gain is 200?

File size:
4.2 KB
Views:
23
18. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
r

Last edited: Sep 7, 2012
#12 likes this.
19. ### #12 AAC Fanatic!

Nov 30, 2010
14,988
5,572
I expressed a personal preference and you spent a week on it?
Amazing!

20. ### rougie Thread Starter Active Member

Dec 11, 2006
410
2
"Personal preference"? LOL!

Beginners don't need personal preferences... they need sincere help!

Personal preferences are hard to catch on to when you learning this stuff you know!