# Homework Help: What is the output of this UJT relaxation oscillator circuit?

1. Jun 30, 2012

### Femme_physics

1. The problem statement, all variables and given/known data

This is a relxation oscillator that's realized by UJT. The graph is included.

What is the output of the circuit?

http://img821.imageshack.us/img821/2274/ujtok.jpg [Broken]

2. Relevant equations

3. The attempt at a solution

Well, to find the output I need VEB or VBB which are not given to me...therefor unless I make the assumption that VEB = 0.7 volts this cannot be solved. And I can just say that

Vout = VR2

Last edited by a moderator: May 6, 2017
2. Jun 30, 2012

### I like Serena

Hey Fp!

Seems to me that without the specification of the UJT you are indeed limited in what you can do.

What you can do, is give the shape of the output signal.
You can also give the frequency of the output signal.
And you can calculate the value of the capacitor.

3. Jun 30, 2012

### Femme_physics

Yes I know the value for the capacitor :) But thanks for confirming it's a dead end-- I thought as much!

4. Jul 1, 2012

### Staff: Mentor

Everything good again with your teacher, FP? Now that the dust has settled.
When designing with active devices, you'll never be "given" precise specifications, for the simple reason that the specs vary from one device to the next, from one production batch to the next, and from manufacturer to manufacturer. Generally, the best you can hope for is a range, or, next best, a typical value. With this, the designer does the best she can. As for VBB, you are as good as given it; if you know it to an accuracy of 10% then consider it is known.
You feel that's an unreasonable assumption to make? :uhh:

It's not clear whether the question sheet referred to the allaboutcircuits site, and its figures for the 2N2647, or whether you added that on your own initiative, but I expect those figures should be fine for the calculations here. Note that the amplitude and width of pulses for triggering may not be of major importance, as the signal is either certain to be more than adequate for the need, or else it will be fed to a buffer or attenuator, so your not being able to specify an output level to an accuracy of 3 decimal places is not the calamity you may think.
10/10 :rofl: :rofl:

5. Jul 2, 2012

### Femme_physics

Thanks Nascent, I can write an explanation about the assumption I made at worst case scenario.

Pretty much, I just think he needs more experience teaching. He's still a young guy.

While we're on the issue of UJT and seeing it's such a short thread, I thought to throw in another question...

http://img403.imageshack.us/img403/5214/question99.jpg [Broken]

Given:
Ic = Ie
Vp = η x VBB + 0.63
Vv = 1.2 volts
VEB = 0.65
η = 0.4

Calculate the frequency of the circuit pulses. Neglect the discharge time of the capacitor.

I wasn't sure how to answer it, so I picked at a notebook I acquired of someone who already took my electronics course and this is what he wrote:

Where is the formula for t0 and t1 taken from? And what does it mean?

Last edited by a moderator: May 6, 2017
6. Jul 2, 2012

### Staff: Mentor

The formulae will fall into place once you have figured out how it works.

First, you'd better identify the <oh, no, not this again> EMITTER of that BJT. :tongue:

Next question to ponder, what is that ZENER doing?

You'll soon fix that; you're aging him fast.

Last edited: Jul 2, 2012
7. Jul 2, 2012

### Femme_physics

*sets hair on fire and jumps from Everest!*

Well... The zener defines the voltage at the cross-section above the zener to the ground. *cleans snow from hair*... The emitters are the arrows. There is one emitter for the NPN transistor and one emitter for the UJT

8. Jul 2, 2012

### Staff: Mentor

You have a good understanding of the UJT oscillator, so move your focus to the BJT in this modified oscillator.

Providing the zener is adequately fed, it will maintain the base of the NPN at a constant voltage. What would follow on, as a consequence of this, to that NPN's voltages and currents?

9. Jul 2, 2012

### Femme_physics

Before I answer this, let's clear the details....

This is not an NPN, this is a PNP!

And this is not a BJT, it's a UJT!

Or...am I blind?

10. Jul 2, 2012

### Staff: Mentor

You're right.

PNP and NPN transistors are examples of Bipolar Junction Transistors.
This circuit is a UJT relaxation oscillator, but it is augmented by a PNP BJT.

11. Jul 2, 2012

### Femme_physics

I'm not sure what the last part means...if there's no BJT defined in the problem, then there is no BJT. How can this logic be flawed?

Well, it will affect the amount of current passing through the circuit but this has to be calculated to figure out

12. Jul 2, 2012

### Staff: Mentor

There is sufficient information on the schematic to enable you to calculate the PNP's emitter current and voltage.

You can see that a resistor of the original UJT oscillator (that you examined in an earlier thread), has been replaced by a PNP transistor in the circuit you are examining here.

A PNP transistor is also known as a BJT. (The same goes for an NPN, it also is a BJT.)

13. Jul 2, 2012

### Femme_physics

Can't they just call things one name? :( Fine.

I did find the current IE all by myself :)

http://img600.imageshack.us/img600/4098/ieic.jpg [Broken]

And Ie = Ic according to the problem definition

The problem is that they're asking for the " frequency of the circuit pulses"... I'm unsure how to relate it all, and that new formula is really confusing

Last edited by a moderator: May 6, 2017
14. Jul 2, 2012

### Staff: Mentor

I'll accept your word that IE is 3.2mA. Your approach looks right.

As you know, IE is approx equal to IC.

So, where does the path for IC lead to, after it emerges from the PNP transistor?

15. Jul 2, 2012

### Staff: Mentor

P.S. I shall have to depart shortly (in 10 mins). Back in 9 or 10 hrs.

16. Jul 2, 2012

### Femme_physics

well, to the UJT or the Capacitor, depends on the time.

ahh...drat ;) Well, we'll rehash this tomorrow then if you will...thanks!

17. Jul 2, 2012

### Staff: Mentor

Just as with the basic UJT oscillator, for most of the time that source of current is doing nothing but charging up the capacitor. So, the only change we have essentially made here is to replace the resistor in the basic RC timing network with a constant current source. How is that going to affect the shape of the waveform at the EMITTER of the UJT?

The importance of encountering alternative technical terms was brought home to me as a student. We had an exam, and I was pleased with it as we had thoroughly covered the work just a few days before. But some of the other students were not happy with it. One of the questions asked us to "explain the need to bias a Bipolar Junction Transistor", and my classmates claimed we hadn't been taught that topic. I realized they were right, we had thoroughly explored biasing NPN transistors, but the lecturer may not have even once referred to them as BJTs. Though of course our textbooks did.

18. Jul 3, 2012

### I like Serena

Hi Fp.

Do you have formulas for a capacitor being charged?

You should have something like:
Q = C V
where Q is the charge on the capacitor, C is the capacity, and V is the voltage across the capacitor.

And you should also have something like:
ΔQ = I Δt
where ΔQ is the change of charge on the capacitor, I is the constant current, and Δt is the time interval that the capacitor is charged.

In your case the capacitor is charged from the Vv voltage to the Vp voltage.

19. Jul 3, 2012

### I like Serena

Btw, I liked how you found Ie. Good!

20. Jul 3, 2012

### Femme_physics

Last edited by a moderator: May 6, 2017
21. Jul 3, 2012

### I like Serena

Yep. That's it! :)

(Assuming your V is actually ΔV for the difference between Vp and Vv.)

Btw, the formula ΔQ = I Δt only works if the current I is constant.

Did you have those formulas? Or other formulas like it?

22. Jul 3, 2012

### Femme_physics

I had the Q = VC

I did not have the Q = IT anywhere in my notebook. Problem is we're way behind in catching up with all the material in industrial electronics before the test (poor planning of our courses), so I'm going over UJT's, PUT, LPF's and relays. Though we covered the stuff in electricity and digital electronics well. Final external electronics test is at 18th this month.

23. Jul 3, 2012

### I like Serena

Perhaps you have something like this?
$$I = {dQ \over dt}$$
This is actually the definition of current.
(Or from an SI point of view, the definition of charge.)

24. Jul 3, 2012

### Staff: Mentor

Hence, you find frequency of oscillation using t1.

You quoted also a formula for t0. Was t0 explained, either here or in the earlier circuit? It has a close resemblance to t1 ....

25. Jul 3, 2012

### Femme_physics

Yes, except you don't minus Vv in t0..since you start from zero!

BTW,

I assume as in before Vout = Vp + Veb ?