|
And first of all, this uBITX works perfectly. It sounds great. Its very sensitive and has good tone quality. Literature says this :?"To invert the sideband between USB and LSB, the second oscillator is switched between 33 MHz and 57 MHz. " . If usb matches with 33 and lsb matches with 57 which it does at the site here : (clearly on the diagram where it clearly says 33 usb, 57 lsb, then its time to do the math ? So I checked very clearly the radio in the next video. Block diagram is here. Ashhar says we should do the math. Now, if I can figure out how to do that math, its another video lol.
|
I hate sitting through videos.?
The math involved is addition and subtraction,? no need to solve any differential equations.
The 12mhz crystal filter has a 3dB passband of around 11.996 to 11.998 mhz, something like that.
The BFO on clk0 should be roughly 500hz away from the passband, so either 11.9955 or 11.9985 mhz.
Will vary from rig to rig, crystals get sorted by frequency so each rig has a matched set of crystals for the filter.
Original plan had been to switch the BFO between high and low side when selecting USB vs LSB.
But the 11.9985 mhz BFO had harmonics that beat with the 16mhz harmonics from the Nano processor,
so Farhan elected to freeze the BFO at 11.9955 mhz.? Instead moves the second oscillator
between 33 and 57 mhz when selecting between USB and LSB.
The VFO is always 45mhz higher than the operating frequency.
Here's an old discussion of how it worked on the Bitx40, using very round numbers to make it easy to follow:
? ??/g/BITX20/message/24724
Jerry, KE7ER
?
toggle quoted message
Show quoted text
On Wed, Mar 7, 2018 at 12:24 pm, Michael Shreeve wrote:
And first of all, this uBITX works perfectly. It sounds great. Its very sensitive and has good tone quality. Literature says this :?"To invert the sideband between USB and LSB, the second oscillator is switched between 33 MHz and 57 MHz. " . If usb matches with 33 and lsb matches with 57 which it does at the site here : (clearly on the diagram where it clearly says 33 usb, 57 lsb, then its time to do the math ? So I checked very clearly the radio in the next video. Block diagram is here. Ashhar says we should do the math. Now, if I can figure out how to do that math, its another video lol.
|
Ok, checked out the BITX40 explanation, and clearly the uBITX is a completely different giraffe and requires different math. If someone even wanted to try to do that I'd be grateful. I guess I might be able to wrap my head around it.
|
OK. Let's see if I can make it easy.
Lets say that we have a 12 MHz carrier and we want to convert it to transmit to the 28 MHz band. We could either mix the 12 MHz carrier with a 16 MHz signal (12 MHz + 16 MHz = 28 MHz) or a 40 MHz (40 MHz - 12 MHz = 28 MHz).
Let's say that now we have a 12 MHz filter which passes the lower side band and want to mix the filter output signal to transmit on the 28 MHz band. Let's say that the carrier frequency is exactly 12 MHz and we modulate the carrier with a 1 kHz audio tone.
In a normal AM transmitter, we end up with the carrier and two side bands, one which is 12 MHz + 1 kHz = 12,001,000 MHz which is the Upper Side Band (USB) and the other 12 Mhz - 1 kHz = 11.999,000 MHz which is the lower side band
Now let's take the same example as above but the signals therough a filter which passes the lower sideband (LSB) (11,999,000 MHz in the example) and filters out the Upper Sideband (USB) (12,001,000 MHz) in this example.
The carrier is also attenuated somewhat by the filter as it is placed down the slope of the filter curve to cut the lower voice frequencies which are not re and further heavily attenuated in the balanced modulator. Theoretically, if the balanced modulator was perfectly balanced and everything screened, the carrier attenuation will be infinite, but practically it is attenuated to a very very low level as to be considered as being suppressed. Now let's say that the carrier was not suppressed but the lower sideband was still passed by the filter and the upper sideband was rejected we mix them with a 40 MHx oscillator.? The output of the mixer will contain the 40 MHz oscillator, the 12 MHz carrier and the 11.999.000 MHz lower sideband including other mixer products which for the present purpose we shall ignore. However, by using a balanced mixer as used in the 12 MHz sideband generator the 40MHz signal can be cancelled (attenuated to a very low level) and we will have an output signal from the mixer of 40 MHz + 12 MHz = 52 MHz and 40 MHz - 12 MHz = 28.000,000 MHz and also 40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz. However, since we used a balanced modulator at 12 MHz, the 12 MHz carrier was suppressed and only the lower side band was allowed to pass through, which simplifies the matter as when we mix the lower side band from the filter with the 40 MHz oscillator will will only get?
40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz.
At the output of the mixer we put a filter to select our wanted 28 MHz signal and strip the unwanted 52 MHz signal. This clearly shows how the lower sideband from the filter has been turned into an upper sideband by the mixer when we used an oscillator frequency for mixing higher then the output frequency required.. Now if we take the lower side band which is 11.999.000 Mhz and mix it with a 40 MHz oscillator, we can get?
On Wed, Mar 7, 2018 at 10:13 PM, Michael Shreeve <shreevester@...> wrote: Ok, checked out the BITX40 explanation, and clearly the uBITX is a completely different giraffe and requires different math. If someone even wanted to try to do that I'd be grateful. I guess I might be able to wrap my head around it.
|
Sorry for the incomplete email.? Pushed the wrong button on the keyboard while still compiling this draft before any corrections..
OK.Let's see if I can make it easy.
Lets say that we have a 12 MHz carrier and we want to convert it to transmit to the 28 MHz band. We could either mix the 12 MHz carrier with a 16 MHz signal (12 MHz + 16 MHz = 28 MHz) or a 40 MHz (40 MHz - 12 MHz = 28 MHz). In the case of mixing the 12 MHz carrier with the 40 MHz signal we will also get 52 MHz which we do not require and which we will filter out.
Let's say that now we have a 12 MHz filter which passes the lower side-band and we want to mix the filter output signal to transmit on the 28 MHz band. Let's say that the carrier frequency is exactly 12 MHz and we modulate the carrier with a 1 kHz audio tone. In a normal AM transmitter, we end up with the carrier and two side bands, one which is 12 MHz + 1 kHz = 12,001,000 MHz which is the Upper Side Band (USB) and the other 12 Mhz - 1 kHz = 11.999,000 MHz which is the lower side band?
Now let's take the same example as above but the signals are passed through a filter which passes the lower side-band (LSB) (11,999,000 MHz in the example) and filters out the Upper Side-band (USB) (12,001,000 MHz).
The carrier is also attenuated somewhat by the filter as it is placed down the slope of the filter curve to cut the lower voice frequencies as we are not after generating a hi-fi signal and are not really required for intelligibility apart from other considerations while the carrier is further heavily attenuated in the balanced modulator. Theoretically, if the balanced modulator was perfectly balanced and everything screened, the carrier attenuation will be infinite, but practically it is attenuated to a very very low level as to be considered as being suppressed.
Now let's say that the carrier was not suppressed but the lower side-band was passed by the filter and we mix them with a 40 MHx oscillator. (Remember that the upper side-band was suppressed by the filter and theoretically is not present, leaving us with the 12 mHz carrier and the 11.999,000 MHz signals)
The output of the mixer will contain the 40 MHz oscillator, the 12 MHz carrier and the 11.999.000 MHz lower side-band including other mixer products which for the present purpose we shall ignore.
By using a balanced mixer as used in the 12 MHz side-band generator the 40 MHz signal can be cancelled (attenuated to a very low level) and we will have an output signal from the mixer of 40 MHz + 12 MHz = 52 MHz and 40 MHz - 12 MHz = 28.000,000 MHz and also 40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz.
However, since we used a balanced modulator at 12 MHz, the 12 MHz carrier was suppressed and only the lower side band was allowed to pass through, which simplifies the matter as when we mix the lower side band from the filter with the 40 MHz oscillator will will only get??40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz.
At the output of the mixer we put a filter to select our wanted 28 MHz signal and strip the unwanted 52 MHz signal.
This clearly shows how the lower side-band from the filter (11.999,000 MHz) in relation to the original carrier frequency of 12 MHz has been turned into an upper side-band by the mixer when we used an oscillator frequency for mixing higher then the output frequency required.
Now if we take the lower side band which is 11.999.000 Mhz and mix it with a 40 MHz oscillator, we can also get on the 50 MHz band, but in this case it is easy to see that we will still end up with a side-band transmission if you work out the simple mathematics.
Now let's say that instead of mixing our crystal filter output with 40 MHz, we mix it with 16 MHz to get on the 28 MHz band (16 MHz + 12 MHz = 28 MHz). But since we have suppressed the carrier and we only have an 11.999,000 MHz signal, 16 MHz + 11,999,000 = 27.999,000 signal which in this case will be outside the 28 MHz band but serves to illustrate our purpose. 16 MHz - 11.999,000 Mhz will also give us 4,001.000 MHz which we can easily filter out.
If our side-band filter passed the upper side-band, the opposite will be true
As you can see it is simple mathematics but may be difficult to understand without some numbers.
Hope this explained the difficulty.
Regards Lawrece
On Fri, Mar 9, 2018 at 12:39 PM, Lawrence Galea <9h1avlaw@...> wrote: OK. Let's see if I can make it easy.
Lets say that we have a 12 MHz carrier and we want to convert it to transmit to the 28 MHz band. We could either mix the 12 MHz carrier with a 16 MHz signal (12 MHz + 16 MHz = 28 MHz) or a 40 MHz (40 MHz - 12 MHz = 28 MHz).
Let's say that now we have a 12 MHz filter which passes the lower side band and want to mix the filter output signal to transmit on the 28 MHz band. Let's say that the carrier frequency is exactly 12 MHz and we modulate the carrier with a 1 kHz audio tone.
In a normal AM transmitter, we end up with the carrier and two side bands, one which is 12 MHz + 1 kHz = 12,001,000 MHz which is the Upper Side Band (USB) and the other 12 Mhz - 1 kHz = 11.999,000 MHz which is the lower side band
Now let's take the same example as above but the signals therough a filter which passes the lower sideband (LSB) (11,999,000 MHz in the example) and filters out the Upper Sideband (USB) (12,001,000 MHz) in this example.
The carrier is also attenuated somewhat by the filter as it is placed down the slope of the filter curve to cut the lower voice frequencies which are not re and further heavily attenuated in the balanced modulator. Theoretically, if the balanced modulator was perfectly balanced and everything screened, the carrier attenuation will be infinite, but practically it is attenuated to a very very low level as to be considered as being suppressed. Now let's say that the carrier was not suppressed but the lower sideband was still passed by the filter and the upper sideband was rejected we mix them with a 40 MHx oscillator.? The output of the mixer will contain the 40 MHz oscillator, the 12 MHz carrier and the 11.999.000 MHz lower sideband including other mixer products which for the present purpose we shall ignore. However, by using a balanced mixer as used in the 12 MHz sideband generator the 40MHz signal can be cancelled (attenuated to a very low level) and we will have an output signal from the mixer of 40 MHz + 12 MHz = 52 MHz and 40 MHz - 12 MHz = 28.000,000 MHz and also 40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz. However, since we used a balanced modulator at 12 MHz, the 12 MHz carrier was suppressed and only the lower side band was allowed to pass through, which simplifies the matter as when we mix the lower side band from the filter with the 40 MHz oscillator will will only get?
40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz.
At the output of the mixer we put a filter to select our wanted 28 MHz signal and strip the unwanted 52 MHz signal. This clearly shows how the lower sideband from the filter has been turned into an upper sideband by the mixer when we used an oscillator frequency for mixing higher then the output frequency required.. Now if we take the lower side band which is 11.999.000 Mhz and mix it with a 40 MHz oscillator, we can get?
|
Nicely done...
Jack, W8TEE
From: Lawrence Galea <9h1avlaw@...>
To: [email protected]
Sent: Friday, March 9, 2018 7:03 AM
Subject: Re: [BITX20] Raduino oscilators. 33mhz and 57mhz. Documentation says one thinng, but this is what I measured.
Sorry for the incomplete email.? Pushed the wrong button on the keyboard while still compiling this draft before any corrections..
OK.Let's see if I can make it easy.
Lets say that we have a 12 MHz carrier and we want to convert it to transmit to the 28 MHz band. We could either mix the 12 MHz carrier with a 16 MHz signal (12 MHz + 16 MHz = 28 MHz) or a 40 MHz (40 MHz - 12 MHz = 28 MHz). In the case of mixing the 12 MHz carrier with the 40 MHz signal we will also get 52 MHz which we do not require and which we will filter out.
Let's say that now we have a 12 MHz filter which passes the lower side-band and we want to mix the filter output signal to transmit on the 28 MHz band. Let's say that the carrier frequency is exactly 12 MHz and we modulate the carrier with a 1 kHz audio tone. In a normal AM transmitter, we end up with the carrier and two side bands, one which is 12 MHz + 1 kHz = 12,001,000 MHz which is the Upper Side Band (USB) and the other 12 Mhz - 1 kHz = 11.999,000 MHz which is the lower side band?
Now let's take the same example as above but the signals are passed through a filter which passes the lower side-band (LSB) (11,999,000 MHz in the example) and filters out the Upper Side-band (USB) (12,001,000 MHz).
The carrier is also attenuated somewhat by the filter as it is placed down the slope of the filter curve to cut the lower voice frequencies as we are not after generating a hi-fi signal and are not really required for intelligibility apart from other considerations while the carrier is further heavily attenuated in the balanced modulator. Theoretically, if the balanced modulator was perfectly balanced and everything screened, the carrier attenuation will be infinite, but practically it is attenuated to a very very low level as to be considered as being suppressed.
Now let's say that the carrier was not suppressed but the lower side-band was passed by the filter and we mix them with a 40 MHx oscillator. (Remember that the upper side-band was suppressed by the filter and theoretically is not present, leaving us with the 12 mHz carrier and the 11.999,000 MHz signals)
The output of the mixer will contain the 40 MHz oscillator, the 12 MHz carrier and the 11.999.000 MHz lower side-band including other mixer products which for the present purpose we shall ignore.
By using a balanced mixer as used in the 12 MHz side-band generator the 40 MHz signal can be cancelled (attenuated to a very low level) and we will have an output signal from the mixer of 40 MHz + 12 MHz = 52 MHz and 40 MHz - 12 MHz = 28.000,000 MHz and also 40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz.
However, since we used a balanced modulator at 12 MHz, the 12 MHz carrier was suppressed and only the lower side band was allowed to pass through, which simplifies the matter as when we mix the lower side band from the filter with the 40 MHz oscillator will will only get??40 MHz + 11.999,000 MHz = 51.999,000 MHz and 40 MHz - 11.999,000 = 28.001,000 MHz.
At the output of the mixer we put a filter to select our wanted 28 MHz signal and strip the unwanted 52 MHz signal.
This clearly shows how the lower side-band from the filter (11.999,000 MHz) in relation to the original carrier frequency of 12 MHz has been turned into an upper side-band by the mixer when we used an oscillator frequency for mixing higher then the output frequency required.
Now if we take the lower side band which is 11.999.000 Mhz and mix it with a 40 MHz oscillator, we can also get on the 50 MHz band, but in this case it is easy to see that we will still end up with a side-band transmission if you work out the simple mathematics.
Now let's say that instead of mixing our crystal filter output with 40 MHz, we mix it with 16 MHz to get on the 28 MHz band (16 MHz + 12 MHz = 28 MHz). But since we have suppressed the carrier and we only have an 11.999,000 MHz signal, 16 MHz + 11,999,000 = 27.999,000 signal which in this case will be outside the 28 MHz band but serves to illustrate our purpose. 16 MHz - 11.999,000 Mhz will also give us 4,001.000 MHz which we can easily filter out.
If our side-band filter passed the upper side-band, the opposite will be true
As you can see it is simple mathematics but may be difficult to understand without some numbers.
Hope this explained the difficulty.
Regards Lawrece
|
Thankyou Jerry for trying and fairly successfully explaining what Ashhar is doing by using the ACTUAL frequencies used in the uBITX. This is VERY important to this discussion. Any other explanation using completely different frequencies is re-inventing a completely different wheel in my opinion, very confusing. Hopefully discussions about this will continue to use Ashhars frequency plan, thus eliminating much of the confusion.? Thanks again for not being confusing.
On Wed, Mar 7, 2018 at 12:57 PM, Jerry Gaffke via Groups.Io <jgaffke@...> wrote: I hate sitting through videos.?
The math involved is addition and subtraction,? no need to solve any differential equations.
The 12mhz crystal filter has a 3dB passband of around 11.996 to 11.998 mhz, something like that.
The BFO on clk0 should be roughly 500hz away from the passband, so either 11.9955 or 11.9985 mhz.
Will vary from rig to rig, crystals get sorted by frequency so each rig has a matched set of crystals for the filter.
Original plan had been to switch the BFO between high and low side when selecting USB vs LSB.
But the 11.9985 mhz BFO had harmonics that beat with the 16mhz harmonics from the Nano processor,
so Farhan elected to freeze the BFO at 11.9955 mhz.? Instead moves the second oscillator
between 33 and 57 mhz when selecting between USB and LSB.
The VFO is always 45mhz higher than the operating frequency.
Here's an old discussion of how it worked on the Bitx40, using very round numbers to make it easy to follow:
? ??/g/BITX20/message/24724
Jerry, KE7ER
?
On Wed, Mar 7, 2018 at 12:24 pm, Michael Shreeve wrote:
And first of all, this uBITX works perfectly. It sounds great. Its very sensitive and has good tone quality. Literature says this :?"To invert the sideband between USB and LSB, the second oscillator is switched between 33 MHz and 57 MHz. " . If usb matches with 33 and lsb matches with 57 which it does at the site here : (clearly on the diagram where it clearly says 33 usb, 57 lsb, then its time to do the math ? So I checked very clearly the radio in the next video. Block diagram is here. Ashhar says we should do the math. Now, if I can figure out how to do that math, its another video lol.
|
Here's a current and more complete summary of what's going on with the uBitx.
For the few who really really want to know.
Not many, judging from the response to post 44278.
Actual frequencies used in the original uBitx code are to have clk0 (bfo) fixed at 11996500 hz,
maybe 500hz below the 12mhz filter's 2000 hz wide 3dB passband.
Oscillator clk1 (second local oscillator) is fixed at 56995000 hz for USB and 32995000 hz for LSB.
You can find those three numbers in file ubitx_20.ino at lines 166, 163, 164 respectively.
I'm looking at the version dated Dec 6, 2017:??? ??
The vfo is used to select the operating frequency Fop according to these formulas.
For USB:? ? Fop? = vfo - (clk1-bfo)? ? ? ?so? ? vfo = Fop + (clk1-bfo)? ? ? ?where clk1 is around 45mhz+12mhz
For LSB:? ? Fop? = vfo - (clk1+bfo)? ? ? ?so? ? vfo = Fop + (clk1+bfo)? ? ? where clk1 is around 45mhz-12mhz
To receive a 7.2mhz LSB signal (where 7.2mhz is the frequency of the suppressed carrier),
the VFO gets set to? ?7200000+(32995000+11996500) = 52191500 hz.
That formula gives an exact result, not an approximation.
Regarding USB vs LSB:
The BFO corresponds to the carrier frequency of the station being received or transmitted.
The 12mhz filter is always above the BFO, so within the 12mhz IF it allows through only the upper sideband.
The VFO is always above the 45mhz first intermediate frequency, and so always inverts the sidebands:
? ? A carrier at 7200000 would get translated to vfo-Fop = 52191500 - 7200000 = 44991500 hz
? ? A lower sideband at 7198500 would get translated to 52191500 - 7198500 = 44993000 hz
In this example we assume the lower sideband is generated from a single audio tone into the mike of 1500 hz.
I have chosen 1500 hz because it will land in the middle of the 12mhz filter's passband,?assuming the filter
has a 3 dB passband that's 2000 hz wide and the BFO is 500 hz below that passband.
The actual range of frequencies passed will be 500 to 2500 hz.
Those assumptions of 2000 hz and 500 hz might be off by a couple hundred hz.
Likewise, a high side clk1 of 56995000 hz? for USB always flips the sidebands when translating to 12mhz,
however the low side clk1 of 32995000 hz we use to receive the 7.2mhz LSB signal does not:
? ? Our 7.2mhz carrier:? ? 44991500 - 32995000 =? 11996500 hz? ? (exactly equal to our BFO frequency)
? ? Our 7.2mhz lower sideband:? 44993000 - 32995000 = 11998000 hz? (in the middle of the crystal filter passband)
That's how the original uBitx code works.
I believe there is a problem because where an LSB signal hits the 45mhz filter will be 4khz removed from
where a USB signal hits it, resulting in a different audio quality between the two.
A solution to this is found in post 44278, review the previous posts in that thread to see why.
Even with the fix, there will be differences in the audio unless the 45mhz filter response is flat
across the 2000 hz at the center.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Wed, Mar 14, 2018 at 09:07 am, Michael Shreeve wrote:
Thankyou Jerry for trying and fairly successfully explaining what Ashhar is doing by using the ACTUAL frequencies used in the uBITX.
|
Jerry, have you actually measured the frequencies of ck 2, ck 1, and ck 0 at specific frequencies, say 7.2 lsb if you would like. So far no one has reported their results.?
|
Nope.
But that's what the code says it does.
I'll be very surprised if it's wrong.
toggle quoted message
Show quoted text
On Wed, Mar 14, 2018 at 04:21 pm, Michael Shreeve wrote:
Jerry, have you actually measured the frequencies of ck 2, ck 1, and ck 0 at specific frequencies, say 7.2 lsb if you would like. So far no one has reported their results.?
|
Well, wrong or right isn't what I'm looking for, but the basic detail I was interested in here:?
Likewise, a high side clk1 of 56995000 hz? for USB always flips the sidebands when translating to 12mhz,
however the low side clk1 of 32995000 hz we use to receive the 7.2mhz LSB signal does not:
corresponds to what I measured basically. The high 57 is used for usb, and the low 33 is used for lsb.
And of course, that doesn't match (at least last I looked) what Ashhar says. The description text is a little vague, but the block diagram specifically says 33 for usb, 57 for lsb. Maybe just a typo, but not sure.?
If my counter is on freq, which I still haven't determined, the frequencies I'm measuring seem to indicate that possibly each raduino is matched to the radio, and the master clock actual frequency wouldn't be so important if some of the calibration routines are used. The radio would still be fairly accurate. I should know soon. thanks again.
A part of Ashhars writeup that I did not notice before is down below,?
"?VHF/UHF coverage?With the 45 MHz IF, it is trivial to build band-pass filters with microstriplines for 144 MHz, 220 MHz and 432 Mhz frequencies. The Si5351’s clock may not high enough for the first conversion directly at 432 Mhz but a sub-harmonic mixer that works with only half the local oscillator frequency can easily scale this rig for VHF/UHF work. MMICs like the MAR6 series and power modules from Mitsubishi can easily scale this radio to reasonable performance level for weak signal and satellite work."?
WOW. Has this always been there ??
|
Plenty of errors to be found in comments and documentation, including mine.
Better off obsessing with how it works than what the documentation says.
> WOW. Has this always been there ??
Been there for awhile.
I doubt it's "trivial".
And for UHF, I'd use an Si5338 or Si5341 instead of messing with doubling the Si5351,
But yes, seems reasonable to do.
You can get years of tinkering out of that $109 investment,
if that's what you're looking for.
toggle quoted message
Show quoted text
On Wed, Mar 14, 2018 at 05:54 pm, Michael Shreeve wrote:
Well, wrong or right isn't what I'm looking for, but the basic detail I was interested in here:?
Likewise, a high side clk1 of 56995000 hz? for USB always flips the sidebands when translating to 12mhz,
however the low side clk1 of 32995000 hz we use to receive the 7.2mhz LSB signal does not:
corresponds to what I measured basically. The high 57 is used for usb, and the low 33 is used for lsb.
And of course, that doesn't match (at least last I looked) what Ashhar says. The description text is a little vague, but the block diagram specifically says 33 for usb, 57 for lsb. Maybe just a typo, but not sure.?
If my counter is on freq, which I still haven't determined, the frequencies I'm measuring seem to indicate that possibly each raduino is matched to the radio, and the master clock actual frequency wouldn't be so important if some of the calibration routines are used. The radio would still be fairly accurate. I should know soon. thanks again.
A part of Ashhars writeup that I did not notice before is down below,?
"?VHF/UHF coverage?With the 45 MHz IF, it is trivial to build band-pass filters with microstriplines for 144 MHz, 220 MHz and 432 Mhz frequencies. The Si5351’s clock may not high enough for the first conversion directly at 432 Mhz but a sub-harmonic mixer that works with only half the local oscillator frequency can easily scale this rig for VHF/UHF work. MMICs like the MAR6 series and power modules from Mitsubishi can easily scale this radio to reasonable performance level for weak signal and satellite work."?
WOW. Has this always been there ??
?
?
|
From what I read its not a bitx20 or 40, so the math shown appleis to a single conversioon receiver.
The microbitx ubitx is very different as its DUAL CONVERSION. So any math shown is just plain
wrong to a degree.? ?Yes you can flip the 3rd oscillator for sideband change or the second of for
that fact is the oscillator can do? it the first!
First all signals are converted to the 45mhz first IF, makes no different if its 1.8mhz or 144mhz save for the?
1st LO is going to be above the IF or below it.? The second mixer and associated 2nd LO are being switched from?
45MH-12mhz=~33 or 45+12mhz=~57mhz and can be used to switch what sideband is in use just as easily as
flipping the 11.9978(give or take) and 12.500(also give or take).? The trick is if you take 45-33 your get -12?
(the minus means inversion in frequency in this case) or its 45-57 or 12 note the sign is positive or no inversion.
Also the fact that the first IF filter is 45mhz means images are not an issue all the way up to 432/450mhz or higher
as the image frequency is 90mhz away and easily filtered.? So with a decent RF (LNA) and appropriate filters?
VHF and UHF are possible with the same basic design though the Local osc the 1st may requires a multiplier
stage to reach UHF.? ? Its an positive artifact of using a high IF so the radio can tune the whole HF range without
switching filters other than the required low pass output filters.? Its a simplification of what most commercial
(YaIcKen) HF radios do.
Allison/kb1gmx
YaIcKen is Yaesu/Icom/Kenwood.
|
On Wed, Mar 14, 2018 at 06:20 pm, ajparent1 wrote:
>>>The trick is if you take 45-33 your get -12?
(the minus means inversion in frequency in this case) or its 45-57 or 12 note the sign is positive or no inversion.
Got it backward and could not go back and edit.? should be... ?The trick is if you take 45-33 your get 12?(the minus means inversion in frequency in this case) or its ?45-57 or -12 note the sign is positive or no inversion. For a multi conversion radio keep track of the sign to know what sideband is active. Allison
|
So, exactly what math is this that was shown? Perhaps post 44560? Michael is only describing the conversion from 1'st IF to 2'ed IF, he's quite aware the uBitx is double conversion. Take a look at the math in this post of our current thread: ? ? ?? /g/BITX20/message/44515See anything wrong there? I really am looking for somebody to vet it. Especially the improved code in post 44278 Jerry, KE7ER
On Wed, Mar 14, 2018 at 06:20 pm, ajparent1 wrote:
From what I read its not a bitx20 or 40, so the math shown appleis to a single conversioon receiver.
|
Jerry, you are correct as long as you don't alter your BFO frequency. Many of us do, in order to maximize our audio performance.
In my sketch, I therefore set my lower sideband value as my BFO and my upper sideband value as 2*BFO. I then use this snippet:
void program(){
?if(L2==upper) L1 = FQ+L2-BFO;//Calculate 1st LO frequency(USB)
?else(L1=FQ+L2+BFO);?????????????? //Or for LSB
?si5351.set_freq(L1 * 100, SI5351_CLK2);//Program it to CLK2
?post=millis();?????????????????????????????? //Return and display?????
}
where L2 is the second LO, L1 is the first LO, and FQ is my operating frequency.
It places the signals spot-on, allowing AM reception on both sidebands without adjustment. The ultimate test. This algorithm has been in use for a couple of months now. 73, Don
|
I would have to think he is using the incorrect word. By "trivial" perhaps he means "easy" ? Trivial , to us, would mean unimportant. Not sure how easy it would be to do this, but I'm thinking this is what he meant.
|
This BFO calculate code may be all that is needed but do isn't it possible that Ashhar also may use a master clock calibration routine when he aligns uBITX raduinos in the factory ??
|
My Raduino reference was not when it arrived, and my BFO certainly was not (for my ears, at least). It worked OK but I enjoy being able to tune in any of the WWV stations at their correct frequency and be able to listen on either sideband without adjustment. It also reinforced my understanding of how it works. As I have said before, works for me. Your experience may differ. Looking forward to our next uBITX to uBITX QSO, by the way! -Don
|
I understood "trivial" to mean "easy" in that context without a second thought.
Not so sure what "us" means though.
toggle quoted message
Show quoted text
On Thu, Mar 15, 2018 at 09:02 am, Michael Shreeve wrote:
I would have to think he is using the incorrect word. By "trivial" perhaps he means "easy" ? Trivial , to us, would mean unimportant. Not sure how easy it would be to do this, but I'm thinking this is what he meant.
|
Michael Shreeve
Don, ND6T