Topics

Please explain to me how works the VFO's


Gerard
 

hello Evan,
Thanks for your advice.
I had some idea that the code was good, otherwise no µBitx could work.

To night the spread seemed better (?). I caught Russians, Italians, etc
The sound was much louder, but still a bit distorted. You must play on the cursor of the pictogram representing the curve of the filter to improve.
I had recalibrated with a generator, but it makes a pure carrier and must distort the result. i will ask a French ham, if there is a reference transmitter.
In this case, i will test  this:
https://petetest2017.files.wordpress.com/2019/01/micro-bitx-calibration-instructions.pdf
I’ll also redo a good antenna. That’s 80% of the reception.

cdt


Gerard
 

On Tue, Sep 15, 2020 at 08:26 PM, Evan Hand wrote:
The code should work with the SI5351 module and the 25MHz crystal setting.

Try adjusting the BFO to the 11996795 value and see if that helps.  Also, I would remove the AGC circuit from the rig at this time so that it is not limiting the audio to the amp.

Just suggestions.
73
Evan
AC9TU

 

 


From: Gerard <kabupos@...>

 
 
 
Reply to Group Discard Private


Evan Hand
 

On Tue, Sep 15, 2020 at 10:09 AM, Gerard wrote:
When i calibrate the BFO, is good the low beat is exactly at 11998370  >> i thing it's good.
This may be where you are off.  What you calculated was the center frequency of the SSB filter if it was actually 12MHz.  The capacitors in the filter shape and lower the filter frequency.  Since those can vary in tolerances, the actual amount of shift varies from board to board.  Most would be between 1700 and 1800 lower.  You would need to run a scan on the filter to find the actual value (a NanoVNA or Antuino are good tools for that). 

The audio signal is 3kHz wide, so you need to subtract half of that or 1500 so that the BFO would be set at 12000000-1750-1500=11996750  That is very close to the BFO setting that I have for one of my v4 uBITX with a 12MHz SSB filter (11.996795).

The code should work with the SI5351 module and the 25MHz crystal setting.

Try adjusting the BFO to the 11996795 value and see if that helps.  Also, I would remove the AGC circuit from the rig at this time so that it is not limiting the audio to the amp.

Just suggestions.
73
Evan
AC9TU


Gerard
 

Hello,

i have made all the tests possibles on the hardware
1) Tested with frequency generator the first stage of amplificator (3 transistors) >>Ok
2) tested the second stage of amplificator (also 3 transistors)  >>ok
3) Tested the 45MHZ filter, the best signal is at 45007680 Htz >>> Delta of 7680Htz so tolerance of 0.0017%  >> Good for me
4) Tested the squale filter 12MHZ (I have 12MHZ crystal)  >>Best signal at 11.988370Htz Delta 11630htz (Quality of quartz)  >> good for me
5) tested the amplifier  >> ok

so some calculation:
The best LSB frequency will be of 56996050Htz  (56996050 - 11988370 =45007680
The best USB  frequency will be of 33019310Htz  (33019310 + 11998370 =45007680
When i calibrate the BFO, is good the low beat is exactly at 11998370  >> i thing it's good.
But when you made calibration, the zero beat is done too fast, just a few rotation movements. the sound also seems slightly more strident.
Frequency can by ajust at 33019310, I CAN't GO more  HIGHER in frequency. (EXmax 35 mhz)
I don't know how is made this calibration (On 33 or 56mhz)
THE PB in the recption:
When i put a counter, to be LSB or USB, frequency seem be the same.(Clock1)
The frequency of clock 0 seem also less....

All this to ask you if what is below is normal or not.
look on at the beginning of SI5351 code: (It's the same on V3, V6 for example)
 
#define BB0(x) ((uint8_t)x)             // Bust int32 into Bytes
#define BB1(x) ((uint8_t)(x>>8))
#define BB2(x) ((uint8_t)(x>>16))
 
#define SI5351BX_ADDR 0x60              // I2C address of Si5351   (typical)
#define SI5351BX_XTALPF 2               // 1:6pf  2:8pf  3:10pf
 
// If using 27mhz crystal, set XTAL=27000000, MSA=33.  Then vco=891mhz
#define SI5351BX_XTAL 25000000          // Crystal freq in Hz
#define SI5351BX_MSA  35                // VCOA is at 25mhz*35 = 875mhz

Should lines not be unlocked? like the definition of the crystal?
I have a SI5351 module (Cristal 25mhz) plugged into my main PCB. Does the app match?
I think the reception pb is there, following misalignment.

When do you think of all that?

cdt


Gerard
 

hello,

I re-checked the pcb. The cabling is correct.
For me, there is only one solution: Replace the 2n3904. they may be of poor quality. I don’t have a bfr at my fingertips, but I’m going to do a test with 2n2222.

cdt


Gerard
 

hello,
I did some tests again today. The change on the inductances coming out of the 45 mhz filter did nothing more.
I think it’s at its limits and it lacks sensitivity
I compared the shematic (see pdf)  and that of a V3 --->
https://www.hfsignals.com/wp-content/uploads/2017/12/ubitxv3.pdf             ---->(With also 2N3904).
I see only a pb on schematic on the 1st transistor and the resistance of 220ohm.

The collector is directly connected on the power RX, on µbitx V3, it's pass trought this resistance.
Schematic error and good on the PCB. I will check tomorrow. 
Or it doesn’t matter how it works. A transistor specialist will tell us that. In any case, as already said, we will make the modification if necessary.
There is also a small difference in the 45mhz filter and impedances. (see schématic)
If I test with RFsim 99 the filter made of 2 inductances of 0.8µH and capa of 6.8pf, it indicates a loss of 10db at 45mhz. This is not negligible.
So there’s still a test to be done.
Otherwise, on the 7MHZ, I have a lot of conversations,English, Italian, Deuch, but still weak.You need to stick your ear to the speaker.
Amplificator works fine, I tested it with another source
cdt


Gary Anderson
 

Hi Gerard,
You don't want the inductors in parallel, especially at close distances.
If trying the axials, still don't mount them in parallel.

With your layout, use the Z-Axis to be orthogonal.

| -----|
L     |
L     |
L     |
 |     |----
 | PCB  |

--LLLL--
| PCB  |

Beyond my expertise, but I would venture that you would be better off with rewinding one toroid  with longer leads and mounting it 90 degrees off in either the Y or Z plane.
You can experiment, and others with more experience can give advice.  Electrically, no reason why both are on the same side of the PCB.  (e.g. 1 toroid flat on the bottom of the PCB)

Regards,
Gary


Gerard
 

Evan, jack
At first I improved my provisional antenna, it is very slightly better.
I think I’m on to something, and that would be more in the design.
The 45MHZ filter is a little different on the M0thy model. However, for the 2 selfs, it uses axial inductances. Me, I wound these 2 inductances on tores. So the tores are glued together next to each other. There must be an induction effect.
Look at the pictures, it’s very airy on the µBitx and much less on mine.
This may be the solution!!
I’m going to desold and replace the tores. It doesn’t cost anything to do it.
cdt


Evan Hand
 

Gerard,

I would try using the IF Shift function in the KD8CEC software to see if that helps.  There could be an alignment issue between the 45 MHz roofing filter and the second IF vfo.  The IF Shift changes the second IF vfo frequency a small amount in either direction.

The other thing that I would try is to remove the AGC function with Q1 by removing the gate signal.  Q1 could be drawing down the input to the audio amp too much.  I do not see a way to adjust that level.

Just suggestions for you to try.
73
Evan
AC9TU


Gerard
 

Hello,

It's version V4 + nextion
All is ok for the soft i think. i can go in all the options of the menu + calibration + BFO
The Nextion work 
The main menu works with the frequency display, etc
I hear conversations, but the sound is weak.
It falls after the 12Mhz filter.
I need to recontrol this part.
cdt


Evan Hand
 

The other question is which display are you using?

As far as I can tell from the schematic that is referenced in the link you provided, it is equal to a v3 or v4 board.  You should use one of the schematics from HFSignals for one of those boards for reference.

The Ashhar Farhan stock software from either the v4 or v4 should work. The alternative is the KD8CEC software.   As I had stated before, I would go with the KD8CEC software and the Nextion display.  You could use the 1602 parallel input display as well.  Both can work with that software.  You will need to select the version for the display that you choose. 

The v6 software will not work without a lot of modifications.  I do not know of a version that allows adjustment for different (earlier) boards.  This includes the version from Reed N.

73
Evan
AC9TU


Jack, W8TEE
 

So, are you using the software distributed by HF Signals or some 3rd party?

Jack, W8TEE

On Friday, September 11, 2020, 3:13:58 PM EDT, Gerard <kabupos@...> wrote:


Jack, 
Remenber. It's an home made. Pcb, tricoils, the case  and others. All is made with my hands!!!
It's a big project, but beautifull to do.
I bought components with Amazon in china, great britain with Ebay. Also in France.
You’re going to say it’s expensive, but it’s a choice
i think we are getting close to the goal.we need to understand how things work and right settings. We make progress day by day.

To see how to do the home made:  https://ubitx.net/2018/07/02/homebrew-ubitx-with-through-hole-components/

cdt

--
Jack, W8TEE


Gerard
 

Jack, 
Remenber. It's an home made. Pcb, tricoils, the case  and others. All is made with my hands!!!
It's a big project, but beautifull to do.
I bought components with Amazon in china, great britain with Ebay. Also in France.
You’re going to say it’s expensive, but it’s a choice
i think we are getting close to the goal.we need to understand how things work and right settings. We make progress day by day.

To see how to do the home made:  https://ubitx.net/2018/07/02/homebrew-ubitx-with-through-hole-components/

cdt


Jack, W8TEE
 

Are you using the µBITX or the JackAl?

Jack, W8TEE

On Friday, September 11, 2020, 2:09:01 PM EDT, Gerard <kabupos@...> wrote:


Evan, Jack
I took over the calibration and adjustment of the BFO with a generator and an oscilloscope.
I checked the entire amplification chain and did 0 beats more to the oscilloscope than to the ear.
I think he’s all lined up now.
I compared the M0thy schema and the Ubitx v6 schema.
I have for the moment a 50 ohm resistance which was directly adjacent to the antenna amplifier input winding. This resistance does not exist on the V6. I’ve undone it for now
I did a test, but still, the sound is too low (even the background noise). It is especially after the 12mHZ filter
I did a test with a temporary antenna that’s more of a long wire.
I have conversations (A little deformed) and morse on 3 and 5 mhz, but as said before the sound is weak.
First I have to make a real antenna.
I think a square loop antenna will be easier to do. But first, I have to find out why the signal is too low when the 12MHZ filter is released. I think that even without an antenna, I must have a "loud" background sound.
By the way, smeter’s still on nine, maybe it’s a lead

cdt



--
Jack, W8TEE


Gerard
 

Evan, Jack
I took over the calibration and adjustment of the BFO with a generator and an oscilloscope.
I checked the entire amplification chain and did 0 beats more to the oscilloscope than to the ear.
I think he’s all lined up now.
I compared the M0thy schema and the Ubitx v6 schema.
I have for the moment a 50 ohm resistance which was directly adjacent to the antenna amplifier input winding. This resistance does not exist on the V6. I’ve undone it for now
I did a test, but still, the sound is too low (even the background noise). It is especially after the 12mHZ filter
I did a test with a temporary antenna that’s more of a long wire.
I have conversations (A little deformed) and morse on 3 and 5 mhz, but as said before the sound is weak.
First I have to make a real antenna.
I think a square loop antenna will be easier to do. But first, I have to find out why the signal is too low when the 12MHZ filter is released. I think that even without an antenna, I must have a "loud" background sound.
By the way, smeter’s still on nine, maybe it’s a lead

cdt



Evan Hand
 

Gerard,
The vfoA and vfoB are two places where you can set the frequency of operation.  They can be two different values to allow to switch between them, or they can be set to receive on vfoA and transmit on vfoB (for RIT operation).  They will change if you adjust the frequency.  

I am not sure what you are trying to do.  It seems that it is working correctly, so what is the problem that you are trying to solve?

Here is the operation manual for the v6 software:
https://www.hfsignals.com/index.php/ubitx-v6-operations/

Operations manual for the KD8CEC software:
https://ubitx.net/wp-content/uploads/2018/04/ug1072_087.pdf

Original v2 to v5 one of the operation manual:
https://ubitx.net/download/ubitx-v4-mini-manual/

There are many more that you can find if you search on the internet.

Hope the above helps.

73
Evan
AC9TU


Gerard
 

Evan, Jack,

Evan, jack,
May you confirm one thing.
Are these two values fixed and never changed??

I don't understood:  i fixed these values with the manager, after close normally the connexion.
i reboot the µbitx. VFOa and b values coorect.
I tried some bands, modify the frequency with the encoder.
If i try the screen touch VFOA/VFOB , the values changed and i confirm that with the manager. (Values changed!!)
Is an error of manipulation? or other? 
if you should never touch the VFO button, then why display it?
If I have to, I’ll test another version to see the behavior.
Thank's

If you should never touch the VFO button, then why display it.


version is 4.3


Gerard
 

hello,
Thank's for the answers
The view plays tricks on me, I saw a zero more. thanks for the explanations. I’ll do tests this afternoon
cd


Evan Hand
 

Gerard,
I believe that there is a typo.  Go back and count the zeros in the second capture you provided.

To answer your question, vfoA and vfoB are memory locations where the program is storing the values that the operator has selected.  Depending on which one is active, that value is used to adjust the internal oscillator (Clock2 for main frequency) which is APPROXIMATELY 45MHz + the vfoA or vfoB depending on which is selected.

So if the vfoA = 7.100200 MHz (7100.200kHz) 
Then the Clock2 VFO of the Si5351a will be set to approximately 7.1002 + 45 MHz = 52.1002 MHz

If the vfoB is 14.19735 MHz (14197.35 kHz) and selected then Clock2 of the Si5351a is approximately 59.19735 MHz.

Hope this helps.
73
Evan
AC9TU


Jack, W8TEE
 

Everything in the tabs is compiled and uploaded as part of the program. The compiler generates the code and the linker stitches everything together correctly. All of this is hidden behind the IDE.

Jack, W8TEE

On Wednesday, September 9, 2020, 4:24:32 PM EDT, Gerard <kabupos@...> wrote:


By the way, silly question about the Arduino software.
When uploading, does it only take into account the selected tab or does it do it for everything?
If it only takes the selected tab, then I will miss a piece of program!! LOL
My faith, interested to know anyway.
I don’t know?

--
Jack, W8TEE