¿ªÔÆÌåÓý

  1. BITX20
  2. Messages
Expanded Topics Messages Polls
Date
Re: Current Firmware

 

¿ªÔÆÌåÓý

Fix one (the first one) and re-compile. ?Frequently one error will produce multiple errors so it¡¯s likely not as bad as it seems.

Clark Martin
KK6ISP

On Jun 5, 2018, at 6:49 PM, Dennis <dennis@...> wrote:

Hi,?
So what do I do about the nearly 50 error messages? I am just learning how to program in Arduino C and this is way above my skill level. (I'm only 67 pages into Jack's excellent book).
Re: ND6T AGC implementation for uBIT-X

 

This has been an "interesting" activity so far.

As of today, there have been 403 paid orders for AGC and/or Click Mini-Kits and 399 have been shipped. I have additional boards and parts and will continue to supply these Mini-Kits as long as there is demand. Details are in the "Files" section under my call? K5BCQ

I have been somewhat surprised to be able to process such a qty over a short period of time. That indicates there are opportunities for others to join in with their ideas. The uBITX is a great vehicle to roll up your sleeves and modify/improve.

73 Kees K5BCQ
Re: Current Firmware

 

Hi,?
So what do I do about the nearly 50 error messages? I am just learning how to program in Arduino C and this is way above my skill level. (I'm only 67 pages into Jack's excellent book).

73, Dennis
W7DRW
Re: RF power chain mods and improvements..

 

I found some msa-0386 mmic's in the junk box. It's been a long time
since I've used any of these. They are much like a MAR-3.

With a 12v power supply I'm sure you want to use a 200ohm resistor to
the output pin. That will give you 5v at the mmic with a 35ma current.
But I don't remember what to use for the blocking capacitors and the
power choke. I'm guessing around .0047mf and around 33uh for the choke.

This would be quick and easy to build up if someone could confirm these
values.

I like the idea of a daughterboard underneath. I'm pretty sure the
choke could be wound on an ft37-43 or ft37-61 and still fit on the
bottom of the board.

Actually I'm not sure the choke is even required at the frequencies
used in the ubitx.

Comments? Suggestions? Critiques?

tim ab0wr

On Tue, 05 Jun 2018 13:15:01 -0700
"Jerry Gaffke via Groups.Io" <jgaffke@...> wrote:

Very cool, that's all making sense.

So the 12mhz and 45mhz IF amps on transmit should both have around
12dB of gain. Could replace Q20 with an MPSH10.
Or could reduce gain a 45mhz to 8dB, increase it at 12mhz to 16dB,
and continue with all 2n3904's??

Additional gain in the transmit drivers could be had by removing C82,
then adding a daughterboard glued to the back with connections to T8
pin 5 and the top of RV1.?

This daughterboard could allow us to reduce gain in the existing 3
stages of 2n3904 transmit drivers. I think many with working rigs
would be hesitant to replace all the 2n3904's from Q90 on. Nice if
any mods kept to a minimum and easy to reverse, the daughterboard
helps in this.?

Jerry
?
Re: RF power chain mods and improvements..

 

I was assuming most of the trouble was at the clk2 mixer, as gain doesn't really get
out of hand till you have both TIA's in series.
Rather surprising that an extra 4dB gain at 12mhz into the clk1 mixer would be such a bother.

On the other hand, we're pushing it big time with only the 30mhz LPF before going into the power amp.
As you have said, pretty much every other multiband rig has multiple bandpass filters there.
Getting by without those means being extra nice to the mixers.

If the worst of the required hack is that we need to add 17dB to the TX drivers, that's not sounding bad at all
Remove C82, add an MMIC plus two caps and a resistor to the back of the board.
Could be done in a variety of ways without a custom PC board,?
I've etched lines using a dremel cutoff wheel into scraps of pc board much like ND6T favors,
that's good for SM parts because it gives good mechanical support.

An old Swan 100MX I bought new in the early 1980's is the only commercial HF ham gear I have.
Not unlike the uBitx in many ways with simple controls, mostly discrete parts, easily understood.
Though the bandpass filters plus all the local oscillators and their filters double the complexity of the rig.?

Jerry, KE7ER


On Tue, Jun 5, 2018 at 03:57 pm, ajparent1/KB1GMX wrote:
>>Or could reduce gain a 45mhz to 8dB, increase it at 12mhz to 16dB, and continue with all 2n3904's? <<

Can't do that as the 12 to 45mhz mixer is still being overloaded as well as the 45mhz to signal frequency mixer.
Sure the 45mhz filter takes out some of that but not a lot.? ?As was the gain of the 12mhz IF was greater than
17db, 6-7 db too much.? Its the sequential overload that is bad.? Two overdriven mixers are not a good thing.
The other negative thing is more IF gain than needed puts the carrier to modulated signal lower as you have
to turn down the audio drive to get fewer spurs.
Re: Bitx 40 source file

 

I disagree with the finding that it behaves the same on the Mac.
At least on Ubuntu it just worked as advertised no renaming or removing of files requuired.

On the Mac it creates a duplicate file with again duplicate content.

For me the manual process by as shown above was what worked.
Re: Bitx 40 source file

 

¿ªÔÆÌåÓý

Misspell in the link, good ol' typos!

Get us all.



On 6/5/2018 2:35 PM, kj6etl wrote:
Thanks Karl !!!!

Finally I understud about the duplicate file. I was still focussed ont he duplicate directory that is automatically created when you select "OK" when the Arduino app to create the new directory etc etc etc Whle all I had to do was remove the redundant .ino file and only leave the "raduino_v1227.7.ino" file in place.

I have included a screenshot of my file structure containing only the data that is needed so it's clear for other Mac users what it all should look like.
The only other thing you have to do is to install the



Link to a full size screenshot:

-- 
Mike Hagen, WA6ISP
10917 Bryant Street
Yucaipa, Ca. 92399
(909) 918-0058
PayPal ID  "MotDog@..."
Mike@...
Re: RF power chain mods and improvements..

 

Been hearding electrons for over 45 years.? It always makes sense even if its not obvious on first glance.

>>Or could reduce gain a 45mhz to 8dB, increase it at 12mhz to 16dB, and continue with all 2n3904's? <<

Can't do that as the 12 to 45mhz mixer is still being overloaded as well as the 45mhz to signal frequency mixer.
Sure the 45mhz filter takes out some of that but not a lot.? ?As was the gain of the 12mhz IF was greater than
17db, 6-7 db too much.? Its the sequential overload that is bad.? Two overdriven mixers are not a good thing.
The other negative thing is more IF gain than needed puts the carrier to modulated signal lower as you have
to turn down the audio drive to get fewer spurs.

Since Farhan used TIA in a bidirectional arrangement we can tailor the gain for one direction without
bothering the other.? This way we don't break the receiver.

You understand the path to flat power gain.? Rather than forcing the transistors to
deliver nonexistent gain we limit it to what they can deliver at all frequencies and
supplement the missing gain.? But that is nearly 17db.

Allison
Re: Bitx 40 source file

 

Yes I know the feeling, hi
But sometimes fool-proof automation can have its advantages.
Anyway thanks for confirming. It means that basically the installation instructions are identical for Windows and Mac (and perhaps Linux too).

73 Allard PE1NWL
Re: Bitx 40 source file

 

The Mac version behaves the same way as what you are describing. My process is however slightly different: I clone?the project and then rename?the project folder - and then delete/rename/move anything that should not be there. Whenever the Arduino IDE asks me if I want to create?the corresponding folder, I always feel like I made a mistake :)?
--
Karl Heinz - K5KHK
Re: Github's future?

 

This is true Arv.

Terry - KB8AMZ
?
Re: Github's future?

 

This is not a good thing for many reasons.

Terr - KB8AMZ
Re: Bitx 40 source file

 

Thanks Karl Heinz
As you pointed out, there should be only one .ino file in the project folder.
The github repository incudes two ino files, but the project folder should only have one ino file in it.
Normally the IDE will take care of this automatically. It shouldn't be neccessary to delete or rename files manually.

But perhaps the IDE behaves slightly differently on a Mac than under Windows or Linux?

This is what I normally do under Windows:
1. on the github page, press the green button "clone or download"
2. Open the downloaded file "bitX40-master.zip" and extract all files to a location of your choice
3. Go to the extracted folder "bitX40-master" and open it
4. Inside this folder there are several files. Find the file named "raduino_v1.27.7.ino" and double click it
5. The Arduino IDE will start
6. A message box will pop-up: The file "raduino_v1.27.7.ino" needs to be in a sketch folder named "raduino_v1.27.7". Create this folder, move the file, and continue?. Click YES.
7. This will create the sketch folder with ONLY one ino file in it.
8. When you go to Sketch => Verify/Compile (or press ctrl-R), then the sketch should be compiled without errors.

When you follow exactly these steps, does this not work on a Mac?

73 Allard PE1NWL
Re: Bitx 40 source file

 

Thanks Karl !!!!

Finally I understud about the duplicate file. I was still focussed ont he duplicate directory that is automatically created when you select "OK" when the Arduino app to create the new directory etc etc etc Whle all I had to do was remove the redundant .ino file and only leave the "raduino_v1227.7.ino" file in place.

I have included a screenshot of my file structure containing only the data that is needed so it's clear for other Mac users what it all should look like.
The only other thing you have to do is to install the



Link to a full size screenshot:
Re: Current Firmware

Jack Purdum
 

I have cleaned those up, and changed the *.ino files to *.cpp files, except for the primary sketch file (i.e., the one with setup() and loop() in it) and created an appropriate header file for type checking on function calls. The two advanatages of doing this is:

1) type checking of parameter and return types on function calls
2) incremental compiles
3) it's easy to see what the sketch directory should be named

While I have cleaned up most of the warnings/errors, I have not touched the warnings coming from the library files, and there are a lot of those. To my knowledge, none are fatal. However, it do not want to go messing around with library files. Those should be fixed by the authors of those files, not me.

Jack, W8TEE

On Tuesday, June 5, 2018, 5:14:13 PM EDT, Gary Anderson <gary.ag5tx@...> wrote:


Green Button 'Clone or download' then download the zip file.
You put the folder containing multiple files into your Arduino "sketch" directory/folder
.. on a pc it can be found under the 'file -> preferences' tab.? I bet something similar on a mac.

Yes it may be a little confusing that there are multiple .ino files in one 'sketch'.
... and I'm pretty sure that the primary .ino file needs to have the same name as the folder in Arduino land.
Anyways, I put in a folder called "ubitx_v43" and renamed "ubitx_v4.3_code.ino" to "ubitx_v43.ino"

It compiled and linked, albeit quite a few warnings for someone so driven to clean-up :)

Regards,
Gary

Re: Current Firmware

 

Green Button 'Clone or download' then download the zip file.
You put the folder containing multiple files into your Arduino "sketch" directory/folder
.. on a pc it can be found under the 'file -> preferences' tab.? I bet something similar on a mac.

Yes it may be a little confusing that there are multiple .ino files in one 'sketch'.
... and I'm pretty sure that the primary .ino file needs to have the same name as the folder in Arduino land.
Anyways, I put in a folder called "ubitx_v43" and renamed "ubitx_v4.3_code.ino" to "ubitx_v43.ino"

It compiled and linked, albeit quite a few warnings for someone so driven to clean-up :)

Regards,
Gary
Re: Github's future?

 

On Mon, Jun 4, 2018 at 04:44 pm, <groupio@...> wrote:
The mass migration away from Sourceforge a while back should serve as a cautionary tale to them.
Microsoft has never encountered a 'cautionary tale' that it couldn't ignore.? (remember what Gates said 25 years or so ago, about the Internet being a passing fad)

The ancient recipe of money, ambition, and hubris never goes out of fashion.? Only the names are changed to protect the guilty.? Micro/Google/Book/tube is nothing more than the currently-visible cohort of its adherents in the corporate realm...? we simply need to be wise enough to see the situation for what it is - a cautionary tale that -we- need to heed.
Re: Current Firmware

 

Actually, all of them are "the file":?You need all?the .ino files in one folder, which needs to be named after the base name of the main file. In this case, when you go to?, you will see that the main file is ubitx_v4.3_code.ino, which means that all these files need to be in a folder named ubitx_v4.3_code?

How do I know that this is the main project file??Because it is the only one that contains the setup() and loop() functions. They are an integral part of an Arduino "sketch".?
--
Karl Heinz - K5KHK
Re: Bitx 40 source file

 

I went back to your original error messages, and it looks like you have _TWO_ .ino files in your raduino_v1.27 folder - the raduino_v1.27.ino file and the?raduino_diag_v3.ino file - both of them are in the GitHub repository for Allard's code, but only one of them can be included in your project folder. You have two ways to accomplish this: Remove it (and potentially other files) from?the project folder (as suggested by Jack: "1) All of the program files need to be in the same directory. Rel 1.27.7? only has one file, the ino file named?Raduino_v1.27.7.ino"), or you can rename that file as done in my project folder in?the following screenshot:



The way the Arduino IDE works is that it will add _ALL_ .ino files in your project folder to your project, and will therefore compile all of them. This is why you ended up with the errors about things that are getting redefined.

The Arduino IDE is pretty straight forward to work with, but you have to follow a few rules.?
--
Karl Heinz - K5KHK
Re: RF power chain mods and improvements..

 

Very cool, that's all making sense.

So the 12mhz and 45mhz IF amps on transmit should both have around 12dB of gain.
Could replace Q20 with an MPSH10.
Or could reduce gain a 45mhz to 8dB, increase it at 12mhz to 16dB, and continue with all 2n3904's??

Additional gain in the transmit drivers could be had by removing C82, then adding a daughterboard
glued to the back with connections to T8 pin 5 and the top of RV1.?

This daughterboard could allow us to reduce gain in the existing 3 stages of 2n3904 transmit drivers.?
I think many with working rigs would be hesitant to replace all the 2n3904's from Q90 on.
Nice if any mods kept to a minimum and easy to reverse, the daughterboard helps in this.?

Jerry
?


On Tue, Jun 5, 2018 at 12:07 pm, ajparent1/KB1GMX wrote:
Jerry is right.
?
However on design review the reason for that is interesting.
?
Simple answer too much gain in the low level parts of the radio for the TX path.
This is not a requirement of the architecture of the transceiver.? ?Turns out that
loss total is about 18db (two mixers at about 7db each and the filters at about
4db each) so a total of 18+7=25db is exactly what we need. so each stage
at about 12.5db is enough, we have easily 35db or more.

This is the core of the issue, carrier leakage plus signal at 45mhz?
makes the 45mhz mixer very rich in spurs. unless we keep the levels down.
?
The transmitter has much less need for gain to get clean signals.
If we have too much we end up with the first mixer the modulator leakage
as carrier.? we want as much or that mixer as we can get or about -17dbm?
as it works out.
?
The total loss (mixers, filters) is about 18db, theres about 34db of gain!
To both mixers are being pushed hard with even 50mv(70mvPP) of audio
at the mixer not the mic input.? The balanced mod should be able to take?
200mv PP (140mV RMS) of audio and if run at that level its output is clean?
with headroom but the 12 and 45 mhz mixers are being run due to the existing?
gain at very high levels hence the spurs.? And we needed to be close that?
limit for full power resulting in high order spurs that seemed random and?
well as low order spurs.? I had to really lower the audio drive (well?
below 50mV)to drop the spurs and it resulted in carrier showing even with?
a reasonable BFO offset.?
?
That resulted in an experiment. With TIA style amps the gain can be easily?
dialed down with feedback.? So with a few measurements in hand I dial the?
first amp (12mhz) down to? insure that and full modulator input the input
to the 12mhz to 45mhz? mixer is not more than -10DBM.? The 12mhz crystal?
filter has a loss of 4db so I need 12db of gain to hit -10dbm at the mixer.? ?
?
Then I repeat that and its about 4db for the 45mhz filter with matching?
network so we need another 11-12 Db of gain to keep the 45mhz mixer?
happy at -10dbm in.? The result is -17dbm output and the spurs are?
not less than 53Db down and most are more like 60-70, much cleaner.
Also the carrier to signal ratio is improved by about 10db, also better.
Yes we also need more audio drive but only little and its easy to do and?
we get a bit more headroom this way.
?
Here the annoying part, to get full power at 80m we are about ok.? At 10M
we are about 11db short of gain.? Maybe a tad more as the 30mhz filter?
eats a db or so.? So fixing the spurs by improving the gain distribution helps
clean things up but power is still wanting.
?
We go from about -18- to -19dbm at the mixer output to +40dbm(10W)
We must have 58 to 59db of gain and a stock power amp chain does?
that only on 80m with a little margin about 3db.? As we go up the amp?
as it exists has less gain.
?
So now we have killed the spurs mostly and improved the carrier leakage.
But the amps is back in line as "the project".
?
First step is to flatten the gain to less than 3dB slope and then add gain
to get the power up.? Adding gain will be some form of added stage.
at the RV1 point.
?
That is the long story of why we have spurs.? We just put too much?
into the mixers.??
?
Allison
65901 - 65920 of 117000