|
Hi Everyone, I was just wondering if I could use a Code Oscillator such as this:? by taking the audio out and feeding it into the mic port? Would that work?
Thanks for your help Ben KD2NOU? --
|
If the tone is a clean sine wave and not over driven the CW will be
perfect!
Raj
At 14-11-2017, you wrote:
toggle quoted message
Show quoted text
Hi Everyone,
I was just wondering if I could use a Code Oscillator such as this:?
by taking the audio out and
feeding it into the mic port? Would that work?
Thanks for your help
Ben
KD2NOU?
|
It can work, it's been done. ?
The archives of this forum show that most attempts at CW on a Bitx have done something similar.
But Allard's CW mod is very simple and will give a cleaner CW signal.
I have my doubts than any ebay code practice oscillator gives a clean sine wave,
especially one that consists of nothing more than an 8 pin microcontroller.
Their "clean audio tone" is almost certainly a square wave, which is not suitable here.
Assuming the oscillator gives a sine wave, you still have distortion in the mike amp and modulator,
plus you are depending on carrier and opposite sideband suppression in the crystal filter.
Allard's method of sending CW avoids all of these potential sources of trouble.
At 5w it's not so much an issue, but keep in mind that abrupt transitions in the keying envelope
result in key clicks at adjacent frequencies. ?
The cap at C107 should somewhat suppress this for Allard's method.
I'm not sure exactly what the result would be of giving hard-keyed audio to that mike amp,?
would be interesting to measure someday.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Tue, Nov 14, 2017 at 12:52 am, Raj vu2zap wrote:
If the tone is a clean sine wave and not over driven the CW will be perfect!
Raj
?
At 14-11-2017, Ben wrote:
Hi Everyone,
I was just wondering if I could use a Code Oscillator such as this:? by taking the audio out and feeding it into the mic port? Would that work?
|
On closer inspection they have an LM555 timer IC, not a uC.
But output is still a square wave, not suitable.
toggle quoted message
Show quoted text
On Tue, Nov 14, 2017 at 09:49 am, Jerry Gaffke wrote:
I have my doubts than any ebay code practice oscillator gives a clean sine wave,
especially one that consists of nothing more than an 8 pin microcontroller.
Their "clean audio tone" is almost certainly a square wave, which is not suitable here.
|
Well, I went ahead and ordered one anyway. I guess I¡¯ll give it a try and see what happens.
73 Ben KD2NOU
On Tue, Nov 14, 2017 at 1:00 PM Jerry Gaffke via Groups.Io <jgaffke= [email protected]> wrote: On closer inspection they have an LM555 timer IC, not a uC.
But output is still a square wave, not suitable.
On Tue, Nov 14, 2017 at 09:49 am, Jerry Gaffke wrote:
I have my doubts than any ebay code practice oscillator gives a clean sine wave,
especially one that consists of nothing more than an 8 pin microcontroller.
Their "clean audio tone" is almost certainly a square wave, which is not suitable here.
|
I thought CW was turning the TX on for dits and dahs and off otherwise, basically an unmodulated carrier at a given frequency.
Is it common to use a modulated carrier, where the audio of code is sent instead?
Or did I get the whole concept wrong?
|
?Unbalancing the mixer so that it actually produces an OUTPUT signal which actually is the carrier, seems pretty much the idea of CW to me.? ??
However, I did some research recently to figure out if the FCC definition of "CW" included what FLDIGI does to create CW (send an audio tone into a SSB balanced mixer to create single frequency signal) -- and concluded that it was completely included and
that even novices / technicians could use it.? ?
There is more than one way to turn a carrier on and off, would be my thinking.
Gordon
toggle quoted message
Show quoted text
I thought CW was turning the TX on for dits and dahs and off otherwise, basically an unmodulated carrier at a given frequency.
Is it common to use a modulated carrier, where the audio of code is sent instead?
Or did I get the whole concept wrong?
|
Search "audio sine oscillator". There's a 2-transistor (one is an output buffer) Wien Bridge circuit there. I'd recommend keying that buffer's power (or the whole thing) with a resistor to reservoir cap for 'build-up' and let its consumption 'decay' it for shaping rather than sudden on/off. The values given are for 777Hz. Too much reliance on ICs - and Kits! In the word(s) of Sir Terry Pratchett's Pictsie character Bob Anybody: "Auchtahelweit!" Rummage the junk box ... and be real Hams, able to use what's to hand.
|
Before you try it on the air, add R904,R905,R906 and C904,C905,C906 from Farhan's minima:
? ??
That will filter the square wave, turn it into a mostly sine wave.
The minima has been superseded by the uBitx, there were a number of issues with the minima design.
Here's the post that presented the schematic above: ??
Sending a 1khz sine wave into the mike of an LSB transmitter tuned for a suppressed carrier at 7.2mhz
will create a CW carrier at 7.199mhz. ?Indistinguishable from any other CW transmission.
In a SSB transmitter, the carrier and the opposite sideband are suppressed, leaving only that single note to be transmitted.
However, Allard's method is much easier and likely cleaner.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Tue, Nov 14, 2017 at 10:26 am, Ben Piecora wrote:
Well, I went ahead and ordered one anyway. I guess I¡¯ll give it a try and see what happens.
?
|
?... Allard's method is much easier and likely cleaner.
Jerry, KE7ER
Absolutely, Jerry. Only one thought: Key-clicks. It's being driven from D6, digital on-off. Is that 10k and C107 enough R-C and reservoir to shape the RF? No 'scope here to test ... anyone? Or another R-C before that ? or (extreme) PWM D6! :)
73 de ZL2DEX
|
We started out with a 47k resistor for the CW mod. ?That worked on some rigs but for others the BFO was too far down on the crystal filter skirt. ?So we reduced the resistor to 10k. ?This drives the final very hard on rigs that had been ok at 47k, but then that's ok cuz the IRF510 need not be operating linearly for CW transmissions. With a 47k resistor into the 0.1uF cap at C107, the RC time constant is 4.7ms which is just about ideal. With 10k into 0.1uF, ?the RC time constant is 1ms, which means we are keying kind of hard but not bad. Most of the other simple methods folks have come up with in this forum switch it instantaneously which makes key clicks considerably worse, but I'm not aware of any complaints about key clicks even on those implementations. ?Perhaps because these are all running at QRP levels. When Allard moves to v2 firmware with the BFO driven from the si5351, he can move the BFO to the crystal filter center on CW transmit, and we can go back to the 47k resistor. Here's some ancient history: /g/BITX20/message/26902/g/BITX20/message/22841And some prehistoric: If building from scratch and not just trying to make an easy hack for the existing Bitx40v3,? I'd use something like?the integrator in fig 6.114-E on p663 of EMRFD. We had a long discussion in the yahoo AT_Sprint group that resulted in a very similar circuit, keying envelope was a very nice trapezoid: ?? ?Jerry, KE7ER
On Tue, Nov 14, 2017 at 01:34 pm, Dexter N Muir wrote:
Absolutely, Jerry. Only one thought: Key-clicks. It's being driven from D6, digital on-off. Is that 10k and C107 enough R-C and reservoir to shape the RF? No 'scope here to test ... anyone? Or another R-C before that ? or (extreme) PWM D6! :)
|
On Tue, Nov 14, 2017 at 02:53 pm, Jerry Gaffke wrote:
When Allard moves to v2 firmware with the BFO driven from the si5351, he can move the BFO to the crystal filter center on CW transmit, and we can go back to the 47k resistor.
Good news, Jerry. Those using the Si5351 for BFO anyway can do that too.
I (and I suspect a goodly few others here) do NOT Yahoo, so the Files there are not available. Is there a (section of) circuit diagram for that? (just to complete the picture :) )
73 de ZL2DEX
|
My best guess is you want to see the integrator used on the KD1JV rigs for CW envelope shaping.
Though just a guess.
On page 11 of this doc, lower right, cap C60 causes Q3 to integrate on rising and falling edges, creating a nice trapezoid.
The zener and R24,R35 are rigged such that the rising edge of the trapezoid is somewhat symmetrical to the falling edge.
? ??
The 3 BS170's are the final amp, like our IRF510. ?Q3 is supplying DC to that final, but ramps it up and down
to give a nice shape to the CW keying envelope.
I believe the discussion on the Yahoo AT_Sprint group pointed to previously is publicly available.
My first exposure to this trick, subsequently spotted it in EMRFD.
Easy enough to become a member of AT_Sprint, about as exclusive a club as Bitx20.
Both let me in.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Tue, Nov 14, 2017 at 03:35 pm, Dexter N Muir wrote:
I (and I suspect a goodly few others here) do NOT Yahoo, so the Files there are not available. Is there a (section of) circuit diagram for that? (just to complete the picture :) )
?
|
That item is out of stock now but when I checked it had a LM386 amp
and
a single transistor oscillator. I presumed it would be a phase shift
osc.
Raj
At 14-11-2017, you wrote:
toggle quoted message
Show quoted text
On closer inspection they have
an LM555 timer IC, not a uC.
But output is still a square wave, not suitable.
On Tue, Nov 14, 2017 at 09:49 am, Jerry Gaffke wrote:
- I have my doubts than any ebay code practice oscillator gives a clean
sine wave,
- especially one that consists of nothing more than an 8 pin
microcontroller.
- Their "clean audio tone" is almost certainly a square wave,
which is not suitable here.
|
We must be looking at different items. Not that it matters much, but here's the link from his post: > ? ?I was just wondering if I could use a Code Oscillator such as this:???
When I follow that, they have three images of the product.
The last image is a schematic, showing 2 R's and a C and an 8 pin DIP. feeding the speaker.
If I zoom it up and stare at it long enough, I can just barely read "LM555" ?in the bill of materials.
Weird that the speaker in that photo apparently has no speaker cone.
Jerry
Regardless, he could probably make it work well enough using the minima's low pass filter of R's and C's.
On Tue, Nov 14, 2017 at 07:13 pm, Raj vu2zap wrote:
That item is out of stock now but when I checked it had a LM386 amp and
a single transistor oscillator. I presumed it would be a phase shift osc.
Raj
. . .
?
At 14-11-2017, you wrote:
On closer inspection they have an LM555 timer IC, not a uC.
But output is still a square wave, not suitable.
|
On Tue, Nov 14, 2017 at 10:19 pm, Jerry Gaffke wrote:
Weird that the speaker in that photo apparently has no speaker cone.
Jerry
If there's any surface capable of moving air it will work. Take a look at a high-power sub-woofer; it will likely be 90% coil-dome, those 'bigjobs' have the large voice-coils so the excursions help dissipate the heat - and to use the massive magnet of course.
73 de ZL2DEX
|
Very odd Jerry,
That item says out of stock and nothing displays. When I checked it
originally it had a 386 and a transistor and many
passive components.
I guess someone here has bought it and can tell us more.
Raj
At 15-11-2017, you wrote:
toggle quoted message
Show quoted text
We must be looking at different
items.
Not that it matters much, but here's the link from his post:
>??? I was just wondering if I could use a Code
Oscillator such as this:
??
When I follow that, they have three images of the product.
The last image is a schematic, showing 2 R's and a C and an 8 pin DIP.
feeding the speaker.
If I zoom it up and stare at it long enough, I can just barely read
"LM555"? in the bill of materials.
Weird that the speaker in that photo apparently has no speaker
cone.
Jerry
Regardless, he could probably make it work well enough using the minima's
low pass filter of R's and C's.
On Tue, Nov 14, 2017 at 07:13 pm, Raj vu2zap wrote:
- That item is out of stock now but when I checked it had a LM386 amp
and
- a single transistor oscillator. I presumed it would be a phase shift
osc.
- Raj
- . . .
- ?
- At 14-11-2017, you wrote:
- On closer inspection they have an LM555 timer IC, not a uC.
- But output is still a square wave, not suitable.
|
Very very odd indeed.
I just cleared my browser cache, can still follow that link and get an image of the $14.95 LM555 oscillator.
Perhaps that ebay store is playing games with our geographical locations, and we see whatever they happen
to have stocked in the warehouse that's cheapest to ship from?
From this it seems we should not be trusting ebay links.
When recommending some ebay item that worked for us, might be best to include a brief description.
And perhaps a photo.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Wed, Nov 15, 2017 at 03:12 am, Raj vu2zap wrote:
Very odd Jerry,
That item says out of stock and nothing displays. When I checked it originally it had a 386 and a transistor and many
passive components.
I guess someone here has bought it and can tell us more.
|
I don¡¯t know what happened to the link. Guess I bought the last one. Here was a screenshot 73 Ben KD2NOU?
On Wed, Nov 15, 2017 at 10:10 AM Jerry Gaffke via Groups.Io <jgaffke= [email protected]> wrote: Very very odd indeed.
I just cleared my browser cache, can still follow that link and get an image of the $14.95 LM555 oscillator.
Perhaps that ebay store is playing games with our geographical locations, and we see whatever they happen
to have stocked in the warehouse that's cheapest to ship from?
From this it seems we should not be trusting ebay links.
When recommending some ebay item that worked for us, might be best to include a brief description.
And perhaps a photo.
Jerry, KE7ER
On Wed, Nov 15, 2017 at 03:12 am, Raj vu2zap wrote:
Very odd Jerry,
That item says out of stock and nothing displays. When I checked it originally it had a 386 and a transistor and many
passive components.
I guess someone here has bought it and can tell us more.
|
I did a search for the description you just gave, arrived at this link which has your board and 0 in stock:
? ??
That actually looks like a good deal.
The description given is "HAM MORSE CW CODE/TELEGRAPH PRACTICE OSCILLATOR KIT BUILD - TWIN T Oscillator"
Vendor is ?
But your original link still gives me the crummy LM555 oscillator, at $14.95
? ? >??? I was just wondering if I could use a Code Oscillator such as this:????
The description I get when I follow that link is "RAINBOW KITS OSC-3 TELEGRAPHY HAM MORSE CODE CW KEY PRACTICE OSCILLATOR"
And I get the photo of the simple LM555 board posted previously.
Vendor is ?cqradiosupply
Note that the 13 digit item number at the end of the two links above is identical.
Can anybody figure out exactly what's going on here? ??
We often recommend ebay items in this forum that work for us by just giving a link,
could be somebody has figured out how to cheat on popular kits by throwing up an inferior product.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Wed, Nov 15, 2017 at 07:30 am, Ben Piecora wrote:
I don¡¯t know what happened to the link. Guess I bought the last one. Here was a screenshot
?
|
Raj vu2zap
Arvo W0VRA
