Hello Vic, 
Thank you for your comments
Yes, I will improve all that.
I will change the potentiometer cables and put shielded. Then I will test the Smeter with the manager Ubitx.
On the diagram, as M0THY says, we can improve the sound by modifying the cabling of the potentiometer.
See what it says in the pdf. I put you the plan of its construction which I inspired.
This should make the AGC work a little better.

I also made a modification on the mosfet for the famous "pop"
Add diode+ resistor and capacitor
It is true that in this amplification story, we could use a pc amplifier  directly as we had on the sound card.
See you soon for news
cdt

Nice that it is working, Gerard, although I couldn't see the Nextion screen well enough to see the S-meter in your video. Here are the settings I used on my V4 uBitX when I was using the same amplifier to drive the S-meter. This agreed well with a commercial rig and with the logic of what the S-meter readings are supposed to represent (6dB per unit):

I think you are modeled on a V4, right? The audio levels is pretty poor on that rig, but all the adjustments to improve are typically after the high side of the volume pot so they won't affect the readings. I added a AGC-controlled audio amp between the high-side of the pot and the wiper and it has fixed all the audio levels nicely.

=Vic=

Hello Evan,
I tested the amplifier. It’s working fine. I need to put shielded cables on the potentiometer because we’re picking up the 50Htz and it’s picking up a little bit. For Smeter, it works, but doesn’t move much.
2 cases, see settings with the manager, either connect behind the 1st amp to recover more signal.
cdt

Gerard,
That looks good to me.

73
Evan
AC9TU

Hello Evan, hello Mick and all others

Evan,
It’s crazy, I ordered the amplifier on Monday night and I’ve already received it. Quite a logistics at Jeff!!!
I’m thinking of installing it directly on the circuit board, there’s just the space.
See the picture. near alim and sound stage.
Mick, Evan, For the Smètre, I was given the solution.
you must install an lm358  and connect to the A7 of the nano
See the post about it. "What a solution for the Smeter"
Tomorrow, I’m going to mount all this
I’m gonna be able to put the amp on the top left of the PCb, and to smeter it there’s only a 358 and a few components. I’ll also be able to install it on my PCB, bottom right.
This is the advantage of having made the circuit myself, I can drill where I want.
As you can see, I don’t have everything for the power emision yet
Thanks for your help
cdt

hello Mick,
Thanks for the information. Yes, today I worked on the ubitx manager for the Smeter and saw that there was an option to save these settings IFS
My calibration and BFO setting must not be too bad, because I am in the center of the IF shift.
I still have to adjust the pb of the Smeter on the Nextion screen, because it is always full.
The story of the wire on the signal A7 of the arduino is a bad track. to connect it to what? No, I think it’s soft and an option to activate correctly the smeter  as Evan said.
Moreover I look forward to an amplifier because, even with an output on HP of 0.5W, it is too weak to understand.... except for a Ham ... Victoria Bravo ...who passed very hard over everyone. Besides, he is in Mexico, so for me it is a satisfaction to have heard  it from France with a tempory installation.
There is still work to do
I use the ubitx in reception mode. Power stage is not cabled

cdt

Evan, Gerard,
If you use Ian’s KD8CEC software you can save the ATT and IF shift values and up load them to memory. I’ve found if you get the BFO and frequency correct you don’t need to use IF shift.
The best process for fine tuning the BFO (after following Ashhar’s directions) is to use Roberts N8RT tone generator suggestion. See below

 

“A way to Check the audio SSB frequency response of your microphone. 

There is a an audio ONLINE Tone Generator that is free and tunes covers 1Hz to 20,000 Hz .

The Generator is available here:

 https://www.szynalski.com/tone-generator/

Works on ANDROID and PC’s maybe others.

Connect your transmitter into a DUMMY LOAD, Connect a Wattmeter appropriate to your transmitter power. 

Key the transmitter in CW Mode to make sure your wattmeter is showing at least ½ scale on the watt meter, this is not super important and is just a reference.

Now switch back to SSB and place the device you have downloaded the tone generator speaker next to your microphone, key your transmitter and on the Tone, Generator slide the button left and right and you can now see the PEAK frequency response of your SSB audio. ( The distance from the device speaker and the mic will only increase or decrease the power reading) 

I wanted to check my UBITX Ver 6 , SSB Rf output power, frequency response of my  Modified Baofeng mic. (posted here) https://groups.io/g/BITX20/message/81778

 This frequency  measurement is very dependent on your UBITX  BFO adjustment, make sure that you have calibrated that before you do this test. My UBITX SSB power,   mic frequency peaked at about 900 hz , and fell off on each side of the 900hz frequency. 

Rob N8RT“

 

--
 

73
Mick VA3EPM 

Evan,
I think I found a solution to a pb
I said that sometimes I had to reinstaller ubitx manager to communicate.
In fact, I think the ubitx should not be on, but simply make the connection with the USB socket. in this case the arduino is self-powered and it works.
So for me it’s settled.
For the Smeter, reading in manager gives low 252 and hight 250!!!!
But here too, I think have found
We have to make a connexion in the Ubitx 
Thank you for confirming that you need to go through this step
cdt
on hamskey:
When uBITX is shipped and assembled according to manual, one purple line is left. (A7 Pin)



The purple line in the picture below. Input the signal for S.Meter to this line.
There are various ways to get S.Meter signals from uBITX.
What should be noted is that you must have enough current to be able to measure in the Arduino.

After connecting S.Meter signal line to A7, it must be set in uBITX Manager. This way we can apply most S.Meter modifications completely to uBITX.
cdt

Gerard,
I have one of these installed in one of my v4 uBITX.  It works very well.  As others have said, you will need to be careful with earphones, as it can generate a very high sound level.

I have not used the ATT and IF Shift much, but do believe that the settings are not saved after shutting down the uBITX.  It would be nice, however, that would require a lot of reads and writes to the EEPROM that could cause an early failure.  The buttons are on the main screen.  There are sliders for each of them, however they are small.  Maybe just modifying the Nextion screen to enlarge them would work for you.

Most hams use a stylus to operate the small touch screen.  Plastic with a rounded and smooth tip so that you do not scratch the screen. 

73
Evan
AC9TU

I was wondering, if it would be simpler to keep these 2 buttons on the main screen of the nextion, and by clicking on it make appear a bargraph all screen for setting. But this requires some modifications in the soft. 
Cdt

I installed the TDA 2030 audio amp for my speaker connection. It works very well. I use headphones normally and I don’t need the extra volume for them. The TDA has a volume pot on it so you can adjust so that switching between headphones and speaker is “balanced”.
--
 

73
Mick VA3EPM 

Gerard,
Sorry, I forgot that the audio amp sections were different.  

I think the pop problem would not be solved with the C38 change.  The way it was fixed was between the v3 and v4, where Q74 was added with the gate being driven by the T/R signal from the Nano.


What is the audio level if you remove Q1 on your schematic?  If it does not impact the audio level, then the AGC is working.  If it increases significantly, then the AGC adjustment is off.

To solve the pop problem I would suggest adding the diode, resistor, and connecting to pin 25 of the Nano (drives K1 through a 4.7k resistor).

As to changing the windings on t16, that would upset the impedances of the mixer and may cause bleed-through of the BFO into the later transmit stages.  I doubt that will increase the signal level.  The BFO is low enough in frequency that the mixer should be OK.  The one mixer that is most in question in my mind is the first, RM1.  That has the highest LO requirement (75MHz at the high end) and would push the Si5351a for drive level.  Again, I still think that you have an audio issue, not an RF one.

I hope this helps
73
Evan
AC9TU

Evan,
I did note the smeter with the Ubitx manager. this afternoon, I modified the input of the signal by adding the filter that exists on the ubitx v6. he L31, C205, L32 together form a single low pass filter that attenuates the 2nd harmonic of 45 MHz from getting into the diode mixer (during the transmit state). 
I had already modified the filter before and after the 45mhz quartz.
For the pop noise, I modified a capacitor, it reduce but does not make it disappear.
See this point on this file. There may still be things to do for that.
see file file, no10

It is true that we can add a lot of modules such as mixer, amplifiers, etc. in this case, we would have to invent a Ubitx with a motherboard with central bus and modules that we could evolve. it’s a different story.
I don’t change too much, otherwise it’ll look like a spagetti dish

cdt

Gerard

That mixer circuit is intended to work with very large signals. It requires more LO power to work well.

Most of us don't have huge antennas. The normal ubitx mixers are adequate. But yet I did not try it in a major contest.  one would mostly use an exotic circuit for first mixer, then use a good agc amplifier.

73 curt

I should have started with the fact that the audio in the v4 (which is what your board is modeled on) was weak.  This was one of the mods in the v5 and v6 HFSignals boards.  There the discrete transistors were replaced with an LM386 amp.  I would recommend changing that out first or go to an external audio amplifier.  Look at the audio section of the current v6 boards as a reference. 

73
Evan
AC9TU

Gerard,

Interesting matching networks.  Different from what I have seen on other sites.

On Sun, Oct 11, 2020 at 05:11 AM, Gerard wrote:

The µbitx is made of 2 rings demodulator. I read on a French forum, that we can increase performance by perfectly balancing the demodulator with a capa (Fixed or variable) for now I do not know where it should be installed. If anyone knows about this...

Not sure that I understand what a "capa" is.  Most of the balancing that I have seen use a variable resistor in place of the 50.1 resistors between the two diode bridges.  As far as I know, this is to reduce unwanted signals in the output, not improve the sensitivity or reduce the loss.


https://www.qsl.net/va3iul/RF%20Mixers/RF_Mixers.pdf


In your case, I would not start there.  Rather look to the Si5351a output.  If the trace from the Si5351a is too long, or not impedance matched (width and spacing), then the LO signal to the mixer will not be a square wave with enough amplitude to overcome the diode bias until later in the cycle.  I would see what the signal looks like on TP12, 15, and 18.  The more square the better.

You could also change to the BAT54SL or equal part for the diodes  There the pair is matched.  Or you could just go with a mixer part like the this on eBay (NOTE: I have not used this, just did a search to find one that might work):
https://www.ebay.com/i/153650913655?chn=ps&norover=1&mkevt=1&mkrid=711-117182-37290-0&mkcid=2&itemid=153650913655&targetid=934793860936&device=c&mktype=&googleloc=9021457&poi=&campaignid=10877432047&mkgroupid=112821775411&rlsatarget=pla-934793860936&abcId=9300402&merchantid=113735811&gclid=Cj0KCQjw2or8BRCNARIsAC_ppyamoeZI3jaNkY35f3Bvst67WJphBvLJUTuA-PDPUSPtbqV8HWetIu4aApTiEALw_wcB

The last option may be to replace the entire mixer with a module like this (NOTE: I have not used this, just did a search to find one that might work):
https://www.ebay.com/itm/SBL-3-RF-FREQUENCY-MIXERS-Double-Balanced-7-dBm-0-025-200-MHz-1PCS/132595966932?_trkparms=aid%3D1110006%26algo%3DHOMESPLICE.SIM%26ao%3D1%26asc%3D225086%26meid%3Db52513b5ceec484eb2d85ec8dbe7a16b%26pid%3D100005%26rk%3D6%26rkt%3D12%26mehot%3Dpf%26sd%3D352803666830%26itm%3D132595966932%26pmt%3D1%26noa%3D0%26pg%3D2047675%26algv%3DSimplAMLv5PairwiseWebWithDarwoV3BBEV2b%26brand%3DMini-Circuits&_trksid=p2047675.c100005.m1851

The above are just suggestions.  Please do your research and verify before spending money or modifying your board.  Others may have alternate and or better suggestions.
73
Evan
AC9TU