|
New to Ham Radio?
My Profile
Community
Articles
Forums
News
Reviews
Friends Remembered
Speak Out
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
|
|
1-10 of 15 messages
|
Page 1 of 2
Next
|
|
frequency push and pull with VCTCXO
|
Reply
|
|
by KD4YKW on September 19, 2009
|
Mail this to a friend!
|
Hello,
i am using an AIO single chip solution for FSK+PLL+VCO.
i would like to do crystal push pulling to implement the AFSK modulation (when selected).
however i do want the frequency stability and have been looking at VCTCXO's...
i see i can put in a sine wave to the control pin of the VCTCXO and vary the VCTCXO output by approx 10 PPM.
i know 10 PPM isnt much....
the other idea i had was to use the crystal oscillator which had the greater push pull range of ppm and then have a separate TCXO then i could manually compare the outputs of both and adjust for the crystal osc in the FSK AIO chip which has a crystal adjust register (for manual compensation).
any thoughts??
the 10 PPM adjustment of the VCTCXO isnt much..... but it will get a varying freq in there.... will it work with 10 ppm sin variance?
thanks for your input.
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by KD4YKW on September 19, 2009
|
Mail this to a friend!
|
ok i answered my own question :)
i will use the regular crystal implementation and modulate it with a varactor cap. with a sine wave to pull it up and down. that will get me my audio tone.
then i will divide that frequency to a lower one and measure it with input compare and an elapsed timer source.
i will use a TCXO on the processor, so i will be able to do the math once a minute to update the crystal adjust register on the FSK / PLL / VCO chip.
a good solution!!!!
thanks :)
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by WB6BYU on September 19, 2009
|
Mail this to a friend!
|
No, I think you are going about it all wrong.
First, AFSK is Audio Frequency Shift Keying. That means
it is generated by feeding two AUDIO tones into a SSB
transmitter.
If you are trying to shift the frequency of the oscillator,
that is regular FSK, not AFSK.
If you feed a sine wave into your oscillator the output
will be frequency modulated by the sine wave signal.
That's not what you want: you want the frequency to
shift back and forth between two values. What you want
is to provide a change in DC level on the control line
so that the oscillator frequency shifts by the desired
amount when the voltage changes, but it is constant and
unmodulated when the DC level is steady at either value.
10PPM means you can shift the frequency 10 Hz per MHz.
With 170Hz shift you can operate as low as 17 MHz. For
850 Hz shift the oscillator would have to be operating
over 85 MHz.
But an easier method should be to modulate the VCO in
your PLL. That is where modulation is generally
applied in such a circuit.
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by KD4YKW on September 19, 2009
|
Mail this to a friend!
|
thanks for your input,
its not wrong, its correct...
modulate the crystal feeding the PLL with the AUDIO freq's based on the digital FSK bit....
got it all worked out already just didnt go into detail.
there will be a DAC feeding the "network" and varactor which will modulate the crystal freq going into the PLL. it will work. i will keep the FSK chip data line low while i do that (so there is no FSK modulation).
now when i want to switch to FSK modulation, then i output a middle value steady value from the DAC to stop modulating the crystal, then i just feed my FSK bits into the FSK bit pin and the FSK chip will switch between the 2 freqs for regular FSK modulation (200 hz shift.)
itlll work :)
now my crystal temp compensation method will work too... since i have to use one in another place in the design, i will use input compare on the MCU which also runs off the TCXO, so ill get a time rate off of that (its fast mcu) and make adjustments as necessary in the FSK chip crystal adjust.
itll work :)
thanks again.
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by KD4YKW on September 19, 2009
|
Mail this to a friend!
|
yah your thinking of a VCXO im not using a VCXO.....
i am using a regular crystal and an external varactor to push and pull the crystal.
if i find a nice VCXO may use that instead..... but the ones ive seen only do 100 ppm.
with push and pulling a crystal you can go like 10% :) in some cases. so much better to do it this way. change the load capacitance of the normal crystal and the output freq changes.
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by KD4YKW on September 19, 2009
|
Mail this to a friend!
|
yah the VCTCXO just wont be enough push and pull... i need 1 Khz change approx... in the audio tone. and that can be adjustable in software too.
so yah ill go with the normal crystal at 10 Mhz.
Thanks for your help!
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by G3RZP on September 20, 2009
|
Mail this to a friend!
|
|
You need to watch the PLL loop bandwidth. Also bear in mind that unless you DC couple, then you will get a bias appearing with certain character strings. The PLL loop bandwidth will affect the '1' to '0' (and vice versa) transition times - ideally, the loop bandwidth should be about ten times the modulation rate. You can then filter the transition of the signal modulating the oscillator to get reasonable edges without spreading too far.
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by WB6BYU on September 20, 2009
|
Mail this to a friend!
|
Are you using AFSK on FM?
On VHF it isn't uncommon to apply AFSK to a standard FM
rig, which would give the affect that you are designing.
(Frequency modulation using the AFSK tones.)
Most amateur HF AFSK operation feeds the audio tones into
a SSB transmitter, in which case the output is equivalent
to regular FSK with two distinct frequencies (except for
spurious outputs like the suppressed carrier, etc.) To
get that you would have to apply switched DC levels to
the varactor instead of the audio signals.
Are you using the crystal as the reference for your PLL
with some sort of divider applied to the VCO frequency?
If so, the required shift in the crystal will depend on
the divider ratios.
The "conventional wisdom" is that a crystal can be
"rubbered" (shifted) about 1% on the fundamental.
(Crystals oscillating in the overtone mode don't pull
well at all.) There are some newer circuits that will
do better than this, especially by using two identical
crystals in parallel, but 10% is very difficult to
achieve with good stability. That shouldn't be a
problem, however, if you only need 200 Hz shift.
(And the required shift is reduced by the dividers in
the PLL. If you are operating at 200 MHz using a 10 MHz
reference crystal you only need 10Hz of shift for FSK,
but 150Hz of modulation capability for AFSK - FM.)
If you can explain more about the desired modulation
you are trying to produce and the application it will
be easier for us to give you good advice.
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by KD4YKW on September 20, 2009
|
Mail this to a friend!
|
tnx G3,
yes you are entirely correct!
nope this is not on a rig...
this is custom hardware.
i have used analog's simulation software to design everything (and simulate the loop bw etc) and it even shows component values and what the output will look like on a spectrum analyzer...
very good simulation software, so that is all done.
actually this APRS transmitter will be available for sale after some testing in a month or soo.
it will do AFSK and FSK..
has tons of features including:
-40 MIPS MCU 128K flash / 16K mem with DMA
-1 Mbit eeprom for holding data points, data values (alternative to SD card... should suffice for 1 minute logging intervals)
-switching power supply 3.3 and 6 v rails upto 1.5 amps each
-voltage input approx 7-20 Vdc
-I2C port
-SPI port
-UART port
-Analog inputs (3.3v) and TTL IO at 3.3 V
-Battery voltage inputs upto ~33 volts DC to log / xmit the values
-External change notification / interrupt pins
-USB 2.0 port
-GPS input on DB9 connector (CMOS level) and on pins
-ability to power 6 V servos
-4 output compare pins for PWM control (for servos)
Onboard sensors:
-pressure sensor 10 Pa accuracy from sea level to over 100K feet
-temp sensor
RF section:
-PLL'd VCO tunable from 130 Mhz to 180 Mhz
-RF amp power level settable in software from 0-8 Watts (approx)
-can turn off RF amp in software
-FSK transmitter
-SMA connector
I am able to control the RF output power level through the use of a DAC... yah 8 watts is overkill but if you dont have any digipeaters anywhere near you it may be necessary, however it is not intended for someone to drown out other stations.. the transmitter will be well filtered.
it is primarily intended for balloon sats :)
i am pretty excited about the project..
it will be proven with a balloon launch hopefully past 90K feet if we are lucky :)
will post here if anyone is interested in purchasing...
thanks for all the great suggestions!!
i am new to RF but i am learning tons!!!
|
| |
|
RE: frequency push and pull with VCTCXO
|
Reply
|
|
by WD4HXG on September 25, 2009
|
Mail this to a friend!
|
"with push and pulling a crystal you can go like 10% :) in some cases."
If you are pulling a crystal around by 10% then you are not gaining the benefits of the crystal any longer.
You would be better off to AM the carrier then try to jerk the signal around for AFSK. If you persist the advantages of using a crystal will be lost with the loss of much lower phase noise, frequency stability etc.
|
| |
|
Email Subscription
You are not subscribed to this topic.
Subscribe!
My Subscriptions
Subscriptions Help
Check our help page for help using
Forum, or send questions, comments, or suggestions to the
Forum Manager.
|
|
|
