Input impedance of a transistor at class C amplifier

[goldsmith]

Dear LvW
Thank you very much for your reply .
It's efficiency is good . and it's operation point is ok ( the base has a little negative bias , that provided by input capacitor and negative half cycle and the inductor ). ( as i know .)
So what appear that circuit in your idea ? i guess that you have doubt about it's operation ? isn't it ?
Respectfully
Goldsmith

 

[LvW]

Hi Goldsmith,

If you are really satisfied ("efficiency is good") and if the output signal meets your requirements (which are unknown to me) - what is the problem?
Perhaps my simulation (showing some parasitic oscillations) was not correct.

 

[goldsmith]

Dear LvW
Hi
Thank you very much for reply .
Can you show me the result of your simulation , please?
I know that your simulator is LTSpice , and i'm not professional at LTSpice , and my simulator is PSPICSE , and i know that LT spice is more better to show realities .
With Appreciate and best Regards
Goldsmith

---------- Post added at 13:58 ---------- Previous post was at 13:54 ----------

By the way ; my oscilloscope is just 20MHZ , and i couldn't see it's waveform at real . i just hear my voice from FM radio what transmitted with this circuit .

 

[FvM]

I doubt that LTSpice is anything better than PSpice in analysis of RF circuits. You should however consider, that BD135 isn't designed for RF applications and that even it can be used at 100 MHz with some restriction, the supplied transistor model surely doesn't exactly represent all parameters of interest. Package parasitics aren't included in standard models, but the play an important role in this circuit. In a short, you can't expect the simulation circuit to be close to reality.

Regarding simulation results, you didn't show e.g. Vbe or Ic waveforms. Looking at them immediately clarifies that you can't calculate impedances based on linear circuit analysis methods.

 

[LvW]

Hi Goldsmith,

the above comments from FvM confirm my doubts as mentioned in my former postings (in particular #6 and #13).
As I am using also PSpice (rather than LTspice) I am sure you also will observe some kind of self-oscillations (Vin=0) for the circuit as shown at the beginning of this thread.

 

[FvM]

It seems like the circuit doesn't show self sustained oscillations, I guess due to the low ft of BD135 which doesn't provide sufficient gain for an oscillator. But the point is different to the determine due to the missing DC bias. It's e.g. unlikely to get oscillations in the #12 circuit.

But as said, Vbe and Ic have large harmonic components.

 

[LvW]

Of course, in reality the circuit #12 will never oscillate due to real L and C components (losses).
However, circuit #12 contains ideal elements - and, thus, self-oscillations will disturb normal amplifier operation.
The attached diagram is the result of a tran simulation (circuit #12): Base voltage in green and collector voltage in red.

 

[goldsmith]

Hi Dear LvW and Dear FvM
Thank you for your replies , and attentions.
Here are my waveforms :


And why i can't see those self oscillations ? is my edit simulation wrong ? if yes , how can i select it to see those oscillations ?
So , what transistor i can use for this frequency ? is 2n2219 good for this aim ? what about it's power ? Or are BF series ok ?
So how can i improve it ?
And how can i calculate the complex impedance of transistor ?
Is it possible that you guide me , please?
Best Regards and Best Wishes for Dear LvW and Dear FvM .
Goldsmith

 

[goldsmith]

Dear LvW and FvM
Again Hi
I think those bad hackers deleted some of your posts here ! and i'm angry !
Anyway : What's your prediction about the input impedance of that circuit ? consider , please that you want predict it , what will you do ? I'll do that too.
And What should i do , to see those oscillations ? ( self oscillations without input signal )

Thanks in Advance and best regards
Goldsmith

 

[FvM]

There were some considerations about impedance. LvW mentioned, that impedance as v/i ratio is defined for linear circuits only.

I said that you can calculate the ratio of individual harmonics voltage and current for a specific operation point to get a large signal impedance equivalent. The value for the fundamental can be used as a starting point to dimension the matching network. But by changing the matching, the operation point and thus the impedance will change as well.

Adjusting the matching network empirically for maximum output is probably the better way. In your simulation circuit, the input voltage source with zero inner impedance corresponds to infinite power capability, surely not a realistic model for a real impedance matching problem.

 

[goldsmith]

Dear FvM
Again Hi
Thank you for your attention and reply and help to me.
As LvW mentioned at past , this circuit without , input , should has , self oscillations . but i couldn't see them at my simulator . How can i see them ? what should i do ?
Is it possible that you guide me please?
Thanks in advance
Goldsmith

 

[FvM]

There are no oscillations in the present simulation. I also mentioned, that they are unlikely with a low fT transistor like BD135 (~250 MHz). To determine the stability of amplifiers in a simulation, you need to refer to models with a realistic representation of package parasitics, particularly bond wire and terminal inductance.

 

[LvW]

There are no oscillations in the present simulation.

Hi Goldsmith - it may be important for you or not, however, I have simulated the circuit as contained in your posting#12 and I got the results as shown in my posting#27.
Because there is no input signal the output signal of this (idealized) circuitry must be considered as the result of self-oscillations.

 

[goldsmith]

Dear LvW
Hi
Thank you for your attention and reply.
I believe your simulation . and i want simulate it again . i understood , that my simulation isn't correct , when you said , it should has self oscillations . but i don't know , that where is the problem in my simulations ? i want learn proper simulation of that .
Is it possible that you guide me to achieve the simulation with oscillation , please? ( i think the problem could be the wrong edit simulations )

Appreciate
Goldsmith

 

[FvM]

I understand that the waveform in post #27 is achieved with input voltage applied. I can't explicitely detect self oscillations.

 

[goldsmith]

Dear FvM
Thank you very very much for your help .
But As LvW mentioned , that wave form is for the circuit without input signal . do you think isn't it ?
Appreciate
Goldsmith

 

[LvW]

I understand that the waveform in post #27 is achieved with input voltage applied. I can't explicitely detect self oscillations.

That's surprising. I have checked again all parts values of the circuit as given by Goldsmith in post#12 - but the result is as before: Self-sustained oscillations as shown in my former post#27.
My simulation profile: Tran analysis for 5usec with max. time step of 0.1nsec (initial bias point calculation skipped).
Another confirmnation: With an additional damping R=1k in parallel to the 45pF at the left side of the circuit oscillations die out after 2.5...3 usec.

Goldsmith.
"Is it possible that you guide me to achieve the simulation with oscillation , please? ( i think the problem could be the wrong edit simulations )"

What really is your problem? When a circuit oscillates it produces an oiutput signal without any input. That means, you only have to check if a TRAN simulation reveals such a signal or not. In order to savely start oscillations (if any) it is recommended to either
* Switch-on of a power supply at t=0 , or
* Inject a short voltage puls (at t=0) into a node that is already grounded, or
* to activate the mode "skip initial bias point calculation".

 

[FvM]

Sorry for not asking clearly. I was referring to the waveform in post #27. What's your criterion for detecting self-sustained oscillations in this waveform?

Looking on the post #27 waveform again, I'm also confused about the signal frequency. The original circuit has a 100 MHz input source, but I see about 1.5 MHz in your plot.

Very clearly, the class C amplifier can't oscillate without some kind of input signal or an additional DC bias. The input matching network has also a high Q, resulting in a slow attack. So it isn't easy to distinguish between regular amplifier operation and self oscillations.

Low frequency (a few MHz) self-oscillations are more easily to achieve with a fT=250 MHz transistor.

 

[LvW]

Sorry for not asking clearly. I was referring to the waveform in post #27. What's your criterion for detecting self-sustained oscillations in this waveform?
Looking on the post #27 waveform again, I'm also confused about the signal frequency. The original circuit has a 100 MHz input source, but I see about 1.5 MHz in your plot.
.

Hello FvM, I am afraid, you have overlooked that the circuit in post#12 has no input signal at all. Thus, the "criterion" for self-sustained oscillations is rather simple for the present case:
Periodic output without any input signal.
I only can repeat that - as mentioned earlier - the circuit from Goldsmith with ideal reactice components produces a periodic output signal if the input signal is replaced by a short.

Very clearly, the class C amplifier can't oscillate without some kind of input signal or an additional DC bias. The input matching network has also a high Q, resulting in a slow attack. So it isn't easy to distinguish between regular amplifier operation and self oscillations.

Yes, exactly this was the reason I have asked Goldsmith in one of my earlier postings to check and verify if his circuit guarantees clean class-C operation (because I still have some doubts).

LvW

 

[FvM]

I didn't overlook it, but I didn't believe that you made the simulation with a circuit, that has no transistor bias at all. I see now, that it oscillates when skipping the initial transient solution.

Although the oscillations can be only sustained in large signal operation, I agree that the simulation indicates serious problems for stable circuit operation. They are mainly brought up by the inappropriate high Q of the input network and the additional low frequency resonance with the 30 uH inductor. Thus I think, they are no typical for a class C amplifier in general.