Basic Oscillator Question

[at2marty]

I have built up the attached circuit on my breadboard, but I don't see the results that I expect.  It should be a 100kHz oscillator.  I am looking at the output at JP2 with my o-scope with the ground connected to JP1.

With the values shown, it doesn't seem to oscillate at all.  If I replace C1 with a .01uF (ceramic) I start getting some oscillation, but it is only  25.5kHz and only oscillates at around 1.6V P-P at the top of the source voltage (between 5V and 3.4V).

Is this what I should be seeing?  How would I go about modifying this to achieve a 100kHz square wave signal at the output?

[rqsall]

Not sure if it's your problem, but the schematic shows NPN transistor symbold, whereas the 2N3906 is a PNP. If you're actually using a 2N3906, my gut says it won't work.

[at2marty]

My mistake, they are actually 2N3904's.  I also tried PN2222's with pretty much the same results.

[Zero999]

I admit, I've never seen this circuit before so I did some research. It appears to be an emitter coupled astable.
http://www.daenotes.com/electronics/digital-electronics/astable-multivibrators-working-construction-types

How did you build it? Was it on a breadboard?

[at2marty]

I admit, I've never seen this circuit before so I did some research. It appears to be an emitter coupled astable.
http://www.daenotes.com/electronics/digital-electronics/astable-multivibrators-working-construction-types

How did you build it? Was it on a breadboard?

Yes it is on a breadboard.  After looking at the link that you provided I may be able to get a better understanding of how it works.  Thank you.

[KJDS]

If you start with a 100n or bigger ceramic cap then you'll have an oscillator at a frequency where most of the strays are negligible. Reducing that cap by small increments will let you determine how much of an effect the strays are having.

The one thing that circuit is missing is supply decoupling.

[at2marty]

O.K.  I've corrected my schematic.  I am actually looking at the output off of the collector Q2.  I am in fact using 2N3904 NPN transistors.

This circuit comes from the ESR meter design by J_Diddy_B.

If I use much less than a .01uF capacitor, I don't seem to see any oscillation at all.

Here is a photo of what I see on my o-scope if I use .01uF.

[eejake52]

I put your circuit into LT Spice (a simulator) and it oscillates nicely at 100 kHz. So, perhaps your results are due to stray capacitance (as already hinted by previous responders), or perhaps you didn't build it exactly per the schematic. Perhaps you could post a picture of the breadboard. Meanwhile, here are some things to check:
- are the 2n3904 transistors connected correctly; i.e. Collector, Base, and Emitter in the right place?
- is the supply 5 Vdc?
- all the grounds connected as shown?
- is the base of Q1 at a DC level of 2.5V (ignore the small transient for now)?

Jake

[dannyf]

Tie the left transistor's base to ground with a capacitor.

It is too complicated. Take a look at the differential oscillator I did - it requires two transistors, one resistor, one capacitor and one inductor / transformer.

[at2marty]

I put your circuit into LT Spice (a simulator) and it oscillates nicely at 100 kHz. So, perhaps your results are due to stray capacitance (as already hinted by previous responders), or perhaps you didn't build it exactly per the schematic. Perhaps you could post a picture of the breadboard. Meanwhile, here are some things to check:
- are the 2n3904 transistors connected correctly; i.e. Collector, Base, and Emitter in the right place?
- is the supply 5 Vdc?
- all the grounds connected as shown?
- is the base of Q1 at a DC level of 2.5V (ignore the small transient for now)?

Jake

I kind of suspected that the breadboard could be part of the problem.  I double-checked everything and it all seems right.  I am not real proficient with LtSpice, but I might try to put this in there and see what I should be looking at.  After all, this whole exercise is about me learning.   

- I double-checked all of my connections, and verified the supply voltage.
- I double-checked all grounds, and they are tied together correctly.
-  The base of Q1 is right at 2.5V.

Tie the left transistor's base to ground with a capacitor.

And BAMN!  I placed a random capacitor between Q1 base and ground.  The signal looked so much better.  I then tried a smaller value capacitor (.001uF) and have some oscillation at around 160kHz or so.  I have been messing around with resistor values, so I need to go back and make sure that I am using the correct values.

Thank you both for your help and suggestions.  I'll see if I can work this out.

[Zero999]

And BAMN!  I placed a random capacitor between Q1 base and ground.  The signal looked so much better.  I then tried a smaller value capacitor (.001uF) and have some oscillation at around 160kHz or so.  I have been messing around with resistor values, so I need to go back and make sure that I am using the correct values.

Thank you both for your help and suggestions.  I'll see if I can work this out.

It's important to know why that worked.

Q1 is configured as a common base amplifier. The capacitor connected Q1's base to 0V at high frequencies, bypassing the bias resistors and therefore increasing the gain.

[at2marty]

And BAMN!  I placed a random capacitor between Q1 base and ground.  The signal looked so much better.  I then tried a smaller value capacitor (.001uF) and have some oscillation at around 160kHz or so.  I have been messing around with resistor values, so I need to go back and make sure that I am using the correct values.

Thank you both for your help and suggestions.  I'll see if I can work this out.

It's important to know why that worked.

Q1 is configured as a common base amplifier. The capacitor connected Q1's base to 0V at high frequencies, bypassing the bias resistors and therefore increasing the gain.

Interesting...  Is there an equation related to this or something more that I can read up on?  I don't want to just "make it work", I really want to understand "how and why" it works.

My next step is to breadboard the driver part of the ESR tester circuit, and explore what it does.

Please keep in mind, I have some electronics background, but I'm pretty much just an "old dog learning new tricks".  I have quite an affection with analog design and would like to learn and understand what makes certain chips work (ie. a 555 timer chip or an op-amp).  My goal is to achieve the same functionality using basic analog parts.  That is what got my attention when it came to this particular ESR Tester design.

I really do appreciate all of the help and pointers from all of the forum members.

[Zero999]

BJTs amplify the current flowing through/voltage across the base-emitter junction and convert it to a current.

There are three basic configurations for a BJT amplifier:

Common emitter
The emitter is connected to a constant voltage, the load is taken from the collector and the input is injected into the base.

This configuration is inverting, has a high voltage gain but low bandwidth, due to the Miller effect.

Common base
The base is held at a constant voltage, the load is taken from the collector and the input is injected into the emitter.

This configuration is non-inverting, has a high voltage gain, a wide bandwidth but no current gain - the input current is slightly higher than the output current.

Common collector
The collector voltage is fixed, the load is coupled to the emitter and the input is injected into the base.

This configuration has a voltage gain of just under 1 but a wide bandwidth, high input impedance and low output impedance.

See links below:
https://coefs.uncc.edu/dlsharer/files/2012/04/C5.pdf
http://www.seas.upenn.edu/~ese206/BiasingLab.pdf
http://www.seas.upenn.edu/~ese319/Lecture_Notes/Lec_9_CCandCBDesigns_08.pdf

In your circuit two of the above configurations are used. Common base is one of them, can you work out what the other one is?

[at2marty]

First of all Hero999, thank you very much for the links and the explanation.

From your description, I believe that the following configurations are being used.

Q1 is a common base since the base is being held at a constant voltage (2.5V) and the output is being taken from the collector.  My question regarding the input.  I assume the input is from the way that R4/C1 is connected... correct?

Q2 is a common emitter configuration.  This one was a bit tougher to decipher (guess), but here is my reasoning.  The output is being taken from the collector, and the input is being injected into the base.  I believe that the emitter is connected to a constant voltage from the voltage drop between R5/Q2 and R6.

I am anxious to get home and experiment with this and see if I can work it out.  I also want to take the time to go through the links that you provided and really try to understand what is going on.  Please correct me if I am wrong, and again thanks for your help.

[Zero999]

It was kind of a trick question.

The output is taken from Q2's collector but the feedback path is via the emitter, via the capacitor.

[dannyf]

the following configurations are being used.

It is one of those "feedforward" designs.

You will see them in some linear applications as a way to increase input impedance. The risk, as your experimenter shows, is oscillation (because of the positive feedback).

[dannyf]

BTW, the same technics can be used on opamps to push their input resistance to over Gohm range.

[at2marty]

BTW, the same technics can be used on opamps to push their input resistance to over Gohm range.

Dannyf, I really like reading your posts because of your knowledge.  Thank you very much for that.

Can you expand or point me to somewhere that would educate me?