Author Topic: Analog circuit analysis discussion – Colpitts oscillator  (Read 1759 times)

0 Members and 1 Guest are viewing this topic.

Offline AnneRanchTopic starter

  • Contributor
  • Posts: 26
  • Country: us
Analog circuit analysis discussion – Colpitts oscillator
« on: September 17, 2019, 02:48:45 pm »

I am looking to discuss circuit details of Colpitts oscillator.
I have found several circuit versions and having a difficult time to model them using Ltspice .
My objective is to fully understand individual components functions.

I am not looking for troubleshooting , design help , tutorials, u-tube or other references  etc.

I would like to engage  in a meaningful discussion with  someone who “been there, done that”  on component level.
For starters
“the circuit configuration should have minimal gain , just to self-start  the oscillation”
If I select 2N2222 as an active device – how to I make sure it has such “minimal gain “ ?

I would appreciate forum members to carefully read my post and reply only when they  have a meaningful contribution to post.
Cheers
 

Offline ledtester

  • Super Contributor
  • ***
  • Posts: 3279
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #1 on: September 17, 2019, 03:08:33 pm »
Since you mention a specific active device, it would greatly aid the discussion if you posted the particular Colpitts implementaton you are talking about.
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #2 on: September 17, 2019, 03:09:29 pm »
I am looking to discuss circuit details of Colpitts oscillator.
I have found several circuit versions and having a difficult time to model them using Ltspice .

Try setting the minimum maximum time step for the transient analysis in LTSpice to something on par with the period of oscillation if your oscillator isn't oscillating.

I would appreciate forum members to carefully read my post and reply only when they  have a meaningful contribution to post.
Cheers

That's a bit spicy dontchathink?



EDIT - oops, mistakenly wrote "minimum time step" instead of maximum.
« Last Edit: September 17, 2019, 05:05:02 pm by MagicSmoker »
 
The following users thanked this post: daqq

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20357
  • Country: gb
  • 0999
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #3 on: September 17, 2019, 03:58:48 pm »
I am looking to discuss circuit details of Colpitts oscillator.
I have found several circuit versions and having a difficult time to model them using Ltspice .

Try setting the minimum time step for the transient analysis in LTSpice to something on par with the period of oscillation if your oscillator isn't oscillating.
Skip initial operating point solution or start DC supply voltages at 0V can also help.
Quote
I would appreciate forum members to carefully read my post and reply only when they  have a meaningful contribution to post.
Cheers

That's a bit spicy dontchathink?
They've probably been trolled in the past. Just ignore posts which don't meaningfully contribute to the discussion or report them to the moderator, if they're that bad.
 

Offline AnneRanchTopic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #4 on: September 17, 2019, 04:13:21 pm »
Whops
I did reply then it asked me to log in AFTERWARS.
Not sure why.
So hopefully here is the current LTspice file.
The model is NOT tuned properly.
 

Offline AnneRanchTopic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #5 on: September 17, 2019, 04:14:39 pm »
Sorry, posted WRONG file.
 

Offline ledtester

  • Super Contributor
  • ***
  • Posts: 3279
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #6 on: September 17, 2019, 04:35:31 pm »
Skip initial operating point solution or start DC supply voltages at 0V can also help.

Yes, introducing a small perturbation somewhere is usually enough to get things going. For instance, adding the initial condition:

Code: [Select]
.ic I(L1) = 1m
 

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #7 on: September 17, 2019, 05:04:28 pm »
Circuit oscillates at about 14.7MHz for me just by setting the maximum timestep to 10ns. /thread

 

Offline ledtester

  • Super Contributor
  • ***
  • Posts: 3279
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #8 on: September 17, 2019, 05:30:37 pm »
Circuit oscillates at about 14.7MHz for me just by setting the maximum timestep to 10ns. /thread

Indeed, the circuit oscillates for me with just the original .trans .5m directive. It does take about 300ns to get going, though, which suggests that there is just enough perturbation caused by round-off error to start it up.

838191-0

With an initial condition like .ic I(L1) = 1m oscillation starts immediately.

FWIW, I'm using LTSpice XVII(x64), the September 11 2019 14:02:52 build.

« Last Edit: September 17, 2019, 05:38:59 pm by ledtester »
 
The following users thanked this post: MagicSmoker

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #9 on: September 17, 2019, 05:40:49 pm »
Indeed, the circuit oscillates for me with just the original .trans .5m directive. It does take about 300ns to get going, though, which suggests that there is just enough perturbation caused by round-off error to start it up.
...
With an initial condition like .ic I(L1) = 1m oscillation starts immediately.

I didn't even try it as-is; I just took the OP's claim it wouldn't simulate at face value. The .ic directive is a good one, too.

 

Offline AnneRanchTopic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #10 on: September 17, 2019, 05:54:03 pm »
OK, gentlemen
apparently  my "spicy" initial comment did not take.

Please read my posts.
I have no issues with the circuit oscillating.

However I cannot keep up with "adjust this  and that" posts here, do my stuff and "honey-dooes".

I have modified , annotated the TI circuit to match the book and changed some values.
Attached.

Perhaps I need to post SPECIFIC questions to get the discussion going in direction I was initially after.

From the TI book -
they determine the ratio of C2 / C3 experimentally as .47
from that I get the real combined value of C2/C3  - standard Colpitts calculation for oscillator frequency


Then there is a lengthy calculation involving C1
They use the initial C2C3 /C2+C3 AKA standard frequency value....

My question is - where did the C1 calculation come from and WHAT is the function of C1?

Cheers












 

Offline ledtester

  • Super Contributor
  • ***
  • Posts: 3279
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #11 on: September 17, 2019, 10:08:44 pm »
Please read my posts.
We are reading your posts.

Quote
Perhaps I need to post SPECIFIC questions to get the discussion going in direction I was initially after.

That's always a good thing to do on an internet forum.

Quote
From the TI book -

What TI book???

Quote
My question is - where did the C1 calculation come from and WHAT is the function of C1?

I don't see a C1 designation on anything you've posted in this thread.
 
The following users thanked this post: Someone, nugglix

Offline AnneRanchTopic starter

  • Contributor
  • Posts: 26
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #12 on: September 18, 2019, 12:32:31 am »
My question is - where did the C1 calculation come from and WHAT is the function of C1?

Per TI
C1 is AC bypass of R2  C1 = 300 pf  - value chosen to avoid self resonant frequency near 90 MHz 
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #13 on: September 18, 2019, 03:53:03 am »
Curiouser & curiouser...  I'm also having trouble with the reference designators, the Spice model & your questions.  Is the schematic posted mayhaps not what is desired?

Also, the schematic looks oddly like a regenerative receiver. ie., antenna on the left, headphone on the right and a control that will vary emitter current.

« Last Edit: September 18, 2019, 03:54:58 am by duak »
 
The following users thanked this post: nugglix

Offline MagicSmoker

  • Super Contributor
  • ***
  • Posts: 1408
  • Country: us
Re: Analog circuit analysis discussion – Colpitts oscillator
« Reply #14 on: September 18, 2019, 11:23:44 am »
The OP doesn't deserve help, I've concluded. Not only because the questions asked are nonsense (re: the non-existent C1), but mainly because of this gem of a thread: https://www.eevblog.com/forum/repair/464-delayed-sweep-switch-worn-out/

Keep demanding we help you for free and see where that gets you.

 
The following users thanked this post: nugglix, archnemesis


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf