Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 3 Next   Go Down
  Print  
Author Topic: Colpitts oscillator.  (Read 7931 times)
DXTUNER
Member

Posts: 72




Ignore
« on: May 31, 2013, 09:31:21 AM »

I've come across a few templates for a Colpitts oscillator that I think a simpleton such as myself can make. But I need some advice please.

1. What (roughly) should the C1:C2, i.e. top:bottom, ratio be, for VHF? Can they be equivalent?
2. For the whole tank circuit, should the L:C ratio be rather small, since its a parallel circuit?

Thank you for answering.
Logged
KB1GMX
Member

Posts: 782




Ignore
« Reply #1 on: May 31, 2013, 09:52:20 AM »

1. What (roughly) should the C1:C2, i.e. top:bottom, ratio be, for VHF? Can they be equivalent?

Varies with active device used,  the tuned circuit, and layout (important at VHF).

2. For the whole tank circuit, should the L:C ratio be rather small, since its a parallel circuit?

Varies with overall scheme of the above.  For VHF the stability of the osc is likely not going to
be enough for most uses other than a raw signal source.  There are enough design variables
that not knowing a lot of exacts (transistor, frequency and all) that any answer is a guess.

Allison

Logged
DXTUNER
Member

Posts: 72




Ignore
« Reply #2 on: May 31, 2013, 11:42:32 AM »

Ok, I've drawn it - in case you want to look at it & then answer. Its just a real basic Colpitts oscillator meant to be an l.o. in a VHF converter I'm building. Please note: I just want to make a simple thing, for fun. I'm not trying to compete with Palomar or Palstar here.

Please note 2: I'm not asking for biasing advice or anything like that. I can already bias it. I'm just wondering generally what the C1:C2 ratio should be, at VHF. This is my first Colpitts circuit attempt.

Thank you much.

https://www.circuitlab.com/circuit/9jd4zd/proposed-colpitts-oscillator/
Logged
WB6BYU
Member

Posts: 13339




Ignore
« Reply #3 on: May 31, 2013, 12:46:10 PM »

When in doubt, make both capacitors equal for a starting point.  Then vary
the ratio once it is working to see if you need less feedback.

Values?  I'd probably start with reactances in the 100 - 500 ohm range,
depending what I have on hand.


Comments on the initial circuit:

The tank coil will conduct DC from the collector to base, throwing off your bias.

The center tap from the capacitors should go directly to the emitter rather than
to ground (though that will work if the emitter resistor is bypassed.)

A popular way to reconfigure the oscillator is to:
1) remove the RF choke in the collector lead.  Connect the collector directly to
the +12V supply with a bypass capacitor to ground

2) disconnect the tank coil from the collector and tie it to ground instead.  This
is still effectively connected to the collector because both are at RF ground.

3) use a larger emitter resistor (1K or so).  Tie the capacitor center tap directly to
the emitter instead of ground.  Take the output from the emitter through a small
coupling capacitor.

4) put a series capacitor between the base bias resistors and the tuned circuit
so the coil doesn't provide a DC path that messes up the bias.


Also note that, while the MPS-H10 is rated for 25V, some of the other VHF/UHF
transistors have lower voltage ratings, and may be best run at 6V or 9V rather
than 12V.
Logged
G3RZP
Member

Posts: 4721




Ignore
« Reply #4 on: May 31, 2013, 02:50:05 PM »

I usually look at the emitter to ground capacitor being twice the value of the base to emitter cap. But a lot depends on the transistor gain. If you are worried at all about phase noise, an un-bypassed 27 ohm emitter resistor helps.

The old W3JHR 'synthetic rock' is a useful variation, too.
Logged
KB1GMX
Member

Posts: 782




Ignore
« Reply #5 on: May 31, 2013, 08:08:53 PM »

Interesting..

two errors.  One its a Pierce variant of colpitts.  The other is the coil
shorts the bias from collector to base.

The coil should have a cap in series to block DC.  You can redraw the circuit
as a Pi network connecting the collector to the base (through a DC blocking cap)
and the transistor acting like a amplifier.

Now the collector side cap is likely to be the smaller value and the base side
the larger.  There is a lot of latitude in values and what hits frequency is
more important, the right values will tend to be more stable in frequency as well.

W6byu gives some good starting values.

Note when you say VHF that is any frequency between 30 and 300mhz, so a
more specific range is nice to know.

Allison
Logged
DXTUNER
Member

Posts: 72




Ignore
« Reply #6 on: June 01, 2013, 09:57:51 AM »

Thank you all very much. Well, maybe it was dumb luck, but I put together a simpler Colpitts-like circuit which, to my surprise, works rather good. It isn't perfect but the output signal is strong, from start up. There are a couple of drawbacks that I'd like fix, if possible.

1. There's a drift of about 1 MHz on start up, i.e., the signal will travel back and forth between 121.500 MHz and 122.500 MHz. It settles down a bit after 15 minute or so.
2. Connecting a simple J310 buffer circuit decreases the output signal, to a level that I'm not comfortable with.

https://www.circuitlab.com/circuit/q566f9/works-good/



« Last Edit: June 01, 2013, 10:04:38 AM by DXTUNER » Logged
G3RZP
Member

Posts: 4721




Ignore
« Reply #7 on: June 01, 2013, 10:04:54 AM »

What circuit for the buffer? At that frequency, you probably need a source follower - anything with a grounded source and drain load will have problems caused by Miller effect, which will decrease input impedance.
Logged
KB1GMX
Member

Posts: 782




Ignore
« Reply #8 on: June 01, 2013, 10:28:40 AM »

Again you have the coil connecting the collector to the base The only thing
limiting the current is that 1K resistor.  The fact that it oscillates is not that surprizing
but that it has no stability is not.

You need a cap between the coil and the base of the transistor.  Any value like .001uf
will help.

Use 9V and regulate it.  Any voltage change will result in frequency change.

The more output level you try for the more heating of the transistor and the more drift.

If you can get the drift down to around 50khz your doing ok.  10Khz is exceptional.
It will never be as stable as a crystal at 121mhz.

FYI: that is the aircraft band (108-135mhz) and radiating a signal there can be serious. 
An oscillator there should be well shielded.

Using a fet as a buffer is problematic due to gain and power limiting.
 

Allison
Logged
DXTUNER
Member

Posts: 72




Ignore
« Reply #9 on: June 01, 2013, 01:12:32 PM »

Again you have the coil connecting the collector to the base The only thing limiting the current is that 1K resistor.  The fact that it oscillates is not that surprizing but that it has no stability is not.
Allison

I want to be sure I follow you. Are you saying the L and C should be flipped? So that the 2 capacitors are closer to the B-C bridge? Or are you suggesting that the tank circuit should be put somewhere else entirely? Yes, sorry I wasn't clear before, but I want it to oscillate in the 115-125 MHz range. Its the only way I can see how to make a VHF converter for the air band (to SW)

My designs look weird, I'm sure, but I'll explain the method. I want to get an oscillation first, then build off of that. Initially I was constructing published circuits having 12-20 components. Such as, but not limited to:
http://www.qrp.pops.net/butler.asp
And this one, ("Starts first time, every time." See figure captioned "local oscillator".):
http://homepage.eircom.net/~ei9gq/vhf_conv.html

Well, neither of them worked for me at all. I spent considerable time tweaking the values afterward. Still, nothing. And I'm looking at those circuits, seeing parts that don't make any sense *to me*. So, I take small sections of certain circuits which make sense to me, leaving behind remainders I don't understand, and blend them into a simple hybrid that will at least oscillate. I mean, what good is an oscillator that doesn't oscillate? My goal is to use a simple circuit as a starting point. But I want to least get an oscillation, first. Which I'm getting. I hope somebody understands my reasoning, if only one person.

What circuit for the buffer? At that frequency, you probably need a source follower - anything with a grounded source and drain load will have problems caused by Miller effect, which will decrease input impedance.

It was a source follower, using a J310. From my results I'd have to agree with KB1GMX that JFET's are gain killers. I want my l.o. to have as strong a signal as possible, because I don't want any HF bleed through. On my earliest attempts making a VHF converter I got too much from the RF input, and not enough from the local oscillator.

Thank you for continuing to throw me advice and tips. Try to ignore that I appear stubborn.






« Last Edit: June 01, 2013, 01:30:37 PM by DXTUNER » Logged
DXTUNER
Member

Posts: 72




Ignore
« Reply #10 on: June 01, 2013, 02:47:08 PM »

I said too much in reply before fully absorbing all of your posts. Ok, I'm going to try this:

https://www.circuitlab.com/circuit/j4r25a/try-this/


« Last Edit: June 01, 2013, 03:26:46 PM by DXTUNER » Logged
WB6BYU
Member

Posts: 13339




Ignore
« Reply #11 on: June 01, 2013, 07:04:23 PM »

Making progress, but there are still some critical components missing.

1) The coil shunts the base bias to DC ground.  You can run an FET with zero volts
on the gate, but a bipolar isn't happy in that condition.

Fix:  add two resistors:  one from +6V to base, one from base to ground.  If you
use 33K from 6V to base and 10K from base to ground you should get about
1mA collector current in the transistor.  But you also need a SERIES capacitor
between the coil and the junction of the two resistors with the base to prevent
the coil from shorting out the bias voltage.  (There are other configurations you
can try, but that is the simplest.)

2) The emitter needs a DC connection to ground.  A 1K resistor will suffice.
In fact, take the one out of the collector circuit and use it.  Connect the
collector directly to +6V AND MAKE SURE YOU HAVE AN RF BYPASS CAPACITOR
CONNECTED DIRECTLY FROM COLLECTOR TO GROUND.  Something like 470pf or
1000pf.

3)  You still need a series output capacitor between the emitter and the
buffer stage, otherwise the DC levels will be thrown off.

4)  I'd probably choose something smaller than 22pf for the tuned circuit,
perhaps 10 to 15pf, and then roughly 1x to 3x that value in place of the 82pf.
You'd have to use a larger coil, of course, and if that is all you have then
you can live with the existing values.


For each circuit you have to make sure that both the DC and RF circuits are
set up properly, and that they don't interact.  Neither will work properly without
the other.  The DC circuit requires appropriate voltages on the collector (6v),
base (roughly 1.5V) and emitter (about 1V based on the recommended values.)
Note that a coil looks like a zero ohm resistor at DC, while a capacitor is an open
circuit.  If connecting the next stage messes up the bias then that needs to be fixed.

For RF you need to make sure that every terminal is properly and intentionally
connected as needed to make a complete circuit.  If the battery or power supply
is not properly bypassed, the connecting leads may become part of the tuned
circuit, and the frequency will vary when they are flexed.  If bypass capacitors
are too large they will look like inductors (due to the lead length, which must be
kept short).  That's why we're recommending 1000pf rather than 10nf bypass
capacitors at this frequency.
Logged
DXTUNER
Member

Posts: 72




Ignore
« Reply #12 on: June 01, 2013, 11:41:10 PM »

Like this?

https://www.circuitlab.com/circuit/xtp8m5/new-circuit-to-try/
Logged
WD4HXG
Member

Posts: 186




Ignore
« Reply #13 on: June 02, 2013, 06:59:01 AM »

Here is what I typically use. The bias values for the
resistors on Q1 may be off and have to be adjusted.
Working from a 62 year old memory here.

Q2 is a buffer (emitter follower) which minimizes the
connections downstream reflecting their load back
to the oscillator and causing frequency pulling.

I usually do not use a resistor in the collector of
the buffer or oscillator. The Colpitts as shown is
an emitter follower which provides POWER gain.

For a LC tank I usually set the tap at 50%. Thus
if you have 100 pF as the resonating cap then use
200 pF for each. Similarly if your resonating cap
is 50 pF then use 100 pF for each.

For crystals I start out with a 1:10 ratio. The
objective is to use as small amount of feedback
as possible to get the circuit started and sustain
operation over the temperature and operating
voltage range. The less feedback through the
crystal the less heating and less drift. (Yes even
the small microwatt feedback levels on quartz
create enough heating to shift the quartz
frequency of resonance).

As for getting better than the 1 MHz you are
currently experiencing over 15 minutes you are
doing pretty darn well already. LC tanks can be
fairly stable over temp with tolerable drift up to
5 to 10 MHz (my personal experience). At 100
MHz you really need either a quartz resonator
or a PLL system otherwise the drift is likely to
be way more than you want to tolerate. Radios
from the 50's through the 70's used phase locking
techniques to keep the receiver LO locked to the
incoming station the listener tuned to.

If you use a crystal you really need an overtone
rock. 5th or 7th overtone will get you between
the 100 and 200 MHz range. Don't try to use a
fundamental crystal in an overtone application.
The harmonic resonances in fundamental crystals
are much lower amplitude than the fundamental.
You will find yourself chasing your tail trying
to add selectivity in the oscillator to narrow
the amplification bandwidth so the other resonances
do not pull you off frequency. Also the harmonic
resonances are often not exact multiples of the
fundamental. They will be close but off just
enough to be maddening.

Hope this helps.

73

Chuck  WD4HXG

https://www.circuitlab.com/circuit/5rtwm4/charlie-colpitts/#menu_file_link_and_share
Logged
WB6BYU
Member

Posts: 13339




Ignore
« Reply #14 on: June 02, 2013, 08:14:27 AM »



Not quite:  the base bias resistors can't deliver any DC bias to the base because
the 47nf capacitor blocks it, and the coil still shunts the base to ground for DC.

Connect the bias resistors directly to the base lead, then put the series capacitor
(47nf is way too large:  47pf would be better) between the bias connection
and the tuned circuit to isolate the coil from DC.

Then to make it an overtone crystal oscillator, adjust the LC circuit so it resonates
close to your desired frequency and add the crystal in series in the lead between
the emitter and the two capacitors in the tuned circuit.
Logged
Pages: [1] 2 3 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!