Date   
Re: The SAGA is COMPLETE!

Jack, W8TEE
 

Reed:

There's two kind of documentation, and I think you're confusing the two.

1. There is the documentation that goes along in the source code. That includes not only comments, but the style used. You can write an if-else with those two keywords or you can write it using a ternary operator. To me, the if-else better documents what you're doing simply because its easier to read. If you look at the assembly code for the two types of expressions using an optimizing compiler, the generated code is the same, so there's no real cost to using the more-easily read form. Given that about 80% of the development time is spent testing/debugging, anything you can do to make the code easier to read and understand is going to pay off.

I also try to use a consistent style, like function header:

/*****
   Purpose: This function is used to.....

   Argument list:
        char *msg        the message to be sent
        int count           the length of the message

    Return value:
        int                    the final message length

    CAUTION:    none
*****/
int MyFunction(char *msg, int count)
{
   // function body
}

Using a fixed style like this, you can write a filter looking for " /***** " and copy everything up through the function signature to a file, print it out, and you have an up-to-date DOC file of all the functions you wrote for the app. You can use the __FILE__ and __LINE__ macros to even pinpoint where the function resides. In this case, the function header style "documents" the code at the function level. If you're writing Open Source code, you need to do this kind of documentation. If not, then you document so you (or your team) can read it.

2. The second type of documentation is the user's manual. Most programmers are not real good at this, probably for the reasons you mentioned. Large software houses rarely use programmers to write user documentation. There are exceptions (e.g., Hans), but they are rare.

Finally, time is less important to me now since I'm not writing commercial software, but programmers fighting the market face a different set of parameters.

Jack, W8TEE

On Saturday, February 15, 2020, 1:18:23 PM EST, Reed N <greenkid336600+groupsio@...> wrote:


I'm not sure I agree with this. I don't mind documenting, and I don't think many developers mind it either. However, we also have finite time, and given the choice between documenting existing code and fixing or writing new code, the latter wins. It's a lot of work to document a changing code base, so development speed is drastically reduced if the programmer has to formally document every change as it happens. Doubly so if they discover an issue with their change and have to undo or re-change it.


Reed

--
Jack, W8TEE

Re: Reed's latest uBITX software on CW

Reed N
 

Hi Jim,

Thanks for the confirmation! I've PM'd you the details of the icon to make it easy to "steal" :)


Reed

Re: The SAGA is COMPLETE!

Reed N
 

I'm not sure I agree with this. I don't mind documenting, and I don't think many developers mind it either. However, we also have finite time, and given the choice between documenting existing code and fixing or writing new code, the latter wins. It's a lot of work to document a changing code base, so development speed is drastically reduced if the programmer has to formally document every change as it happens. Doubly so if they discover an issue with their change and have to undo or re-change it.


Reed

Re: uBITX Ver. 5 Completion Report

Phil Karn
 

On 2/15/20 08:55, Ora Smith wrote:

Calibration: The calibration instructions mystified me, so I ended up
tuning the carrier frequency to 10 MHz on my Flex, which is pretty
accurate.  BFO adjustment was far off initially. I found that it also
affected carrier suppression. Tuning the BFO by the "sounds good"
technique resulting initially in only 25 db of suppression. QST
reported 49 db suppression, so I suppose I could tweak some more, but
at 35 db it's OK, about what I see with my S-line. Opposite sideband
suppression looks very good on the Flex scope - I can't see anything. 
Spectral purity (carrier spectrum) is not as good as expensive
commercial rigs, but I have had no complaints on the air.
In my case it would help if I had an *exact* description of how the
various generated clocks change with control adjustments (USB/LSB/CW,
tuning knob, calibration, BFO, etc). I know this depends on the firmware
running on the Arduino.

I don't see any carrier suppression adjustments on my v6, just a
balanced modulator that in theory (but not practice) should remove it
entirely. So I suspect the actual carrier suppression depends on where
you park the carrier relative to the crystal filter passband. That's why
it changes with the BFO frequency setting. Better suppression involves
moving the carrier farther away from the crystal passband, which means
cutting off the low frequency response.

This is where I'm really spoiled by SDRs and their razor-sharp filters.
A 'next generation bitx' really ought to go digital here. The crystal
filter, BFO and balanced modulator/product detector would move into
software.

(Has anybody noticed just how *good* SSB can sound when generated and
received by an SDR with 100 Hz cutoffs instead of 300 Hz, and with good
frequency references?)

The lack of AGC on the receiver is bothersome. I have to ride the
volume control constantly when there is QSB or a roundtable with
greatly varying signal strengths.
Yeah, but I'm actually surprised at how usable it is without one.

I do get the impression that there's at least one opinion per ham on how
an AGC *should* work, so by omitting an AGC altogether Ashhar cleverly
avoided that argument. :-)

73, Phil

Re: The SAGA is COMPLETE!

Jack, W8TEE
 

I'm kinda different because I really do enjoy writing--even documentation--very much.

Jack, W8TEE

On Saturday, February 15, 2020, 12:40:44 PM EST, MadRadioModder <madradiomodder@...> wrote:


Documentation never seemed to be a problem for you…

 

 

From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of Jack, W8TEE via Groups.Io
Sent: Saturday, February 15, 2020 8:54 AM
To: BITX20@groups.io
Subject: Re: [BITX20] The SAGA is COMPLETE!

 

As a rule, most programmers don't like to write documentation.

 

Jack, W8TEE

 

On Saturday, February 15, 2020, 9:48:49 AM EST, Dean Souleles <dsouleles@...> wrote:

 

 

Hi Dan - 

To my knowledge Dr. Lee never wrote a manual for the firmware.  I believe the document you have is the most complete document that there is.  The rest of the info I have found in various blog posts on Dr. Lee's KD8CEC blog - http://www.hamskey.com/ and right here on the forum.   He hasn't posted in a while but in the past he has been responded to questions.   I learned about the CW offset from the posts on here and then on his blog.  I've also poked around in the firmware source a bit.

73,
Dean
KK4DAS


--
Jack, W8TEE


--

…_. _._

--
Jack, W8TEE

Re: The SAGA is COMPLETE!

Jack, W8TEE
 

Yeah, I've run into a few of those, too. Still, I've never seen a new user of a piece of software load up its source code, read it, and then start using the app.

Jack, W8TEE

On Saturday, February 15, 2020, 12:58:05 PM EST, Phil Karn <karn@...> wrote:


On 2/15/20 06:53, Jack, W8TEE via Groups.Io wrote:
> As a rule, most programmers don't like to write documentation.
>
>
Actually, most programmers will say that their code IS the documentation.

(I've been guilty of this myself.)

Phil







--
Jack, W8TEE

Re: The SAGA is COMPLETE!

Phil Karn
 

On 2/15/20 06:53, Jack, W8TEE via Groups.Io wrote:
As a rule, most programmers don't like to write documentation.

Actually, most programmers will say that their code IS the documentation.

(I've been guilty of this myself.)

Phil

Re: The SAGA is COMPLETE!

MadRadioModder
 

Documentation never seemed to be a problem for you…

 

 

From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of Jack, W8TEE via Groups.Io
Sent: Saturday, February 15, 2020 8:54 AM
To: BITX20@groups.io
Subject: Re: [BITX20] The SAGA is COMPLETE!

 

As a rule, most programmers don't like to write documentation.

 

Jack, W8TEE

 

On Saturday, February 15, 2020, 9:48:49 AM EST, Dean Souleles <dsouleles@...> wrote:

 

 

Hi Dan - 

To my knowledge Dr. Lee never wrote a manual for the firmware.  I believe the document you have is the most complete document that there is.  The rest of the info I have found in various blog posts on Dr. Lee's KD8CEC blog - http://www.hamskey.com/ and right here on the forum.   He hasn't posted in a while but in the past he has been responded to questions.   I learned about the CW offset from the posts on here and then on his blog.  I've also poked around in the firmware source a bit.

73,
Dean
KK4DAS


--
Jack, W8TEE


--

…_. _._

Re: Made Mods to V3 uBitx - no longer transmitting #ubitx-help #v3

Steve - KE0VCD
 

20mV to antenna jack when keyed-up

Re: Made Mods to V3 uBitx - no longer transmitting #ubitx-help #v3

Steve - KE0VCD
 

K3 pins 8&9 0v and 20mV

Reed's latest uBITX software on CW

Jim Sheldon
 

Nice Job Reed,
I finally got a good build of your last. Got it calibrated exactly using my 10 MHz GPS disciplined oscillator and a 111 dB attenuator. Just got done working an old friend, Art, KA9ZAP on 40 meters using my uBITX test set. He was exactly on 7.022.000 both on my Elecraft K3S and the uBITX V6 and when I had him tuned in, matching the sidetone to his incoming tone, I watched the CW TX frequency on my K3S/P3 (panadapter) station rig. The CW TX carrier was dead superimposed on his, both with USB and LSB selected (yes they do switch, I can hear a change in background noise to confirm that as well.

Again, finally someone other that our TSW group has gotten the CW TX frequency right as it should work and does on all normal transceivers (QRP OR QRO/Commercially built).

Got to admit I like the "Gear" indicator for the setup menu. Might have to have our programmer "steal" that one - LOL.

Jim Sheldon,
W0EB

uBITX Ver. 5 Completion Report

Ora Smith
 

I recently finished (? - is it ever done?) a uBITX version 5.1 kit with the nice blue case that I received as a gift.  I must admit that when I opened the box and realized it was mostly a bag of parts with no instructions I nearly quit before I started - really glad I kept going. Though occasionally frustrating (it's not a Heathkit!), I now have a very capable 80-10 meter SSB/CW transceiver. It brings back the magic of old time radio.  I find myself using it to the exclusion of a bunch of other fancy radios. I discovered a few things along the way I thought I would share. Perhaps none of this is new, but here goes anyway.

Calibration: The calibration instructions mystified me, so I ended up tuning the carrier frequency to 10 MHz on my Flex, which is pretty accurate.  BFO adjustment was far off initially. I found that it also affected carrier suppression. Tuning the BFO by the "sounds good" technique resulting initially in only 25 db of suppression. QST reported 49 db suppression, so I suppose I could tweak some more, but at 35 db it's OK, about what I see with my S-line. Opposite sideband suppression looks very good on the Flex scope - I can't see anything.  Spectral purity (carrier spectrum) is not as good as expensive commercial rigs, but I have had no complaints on the air. 

I have made a number of barefoot contacts (first contact was with Curacao on 20 meters) but the rig also will drive my Ten Tec Titan (pair of 3CX-800's) to 1000 watts on 80, about 500 on 40 and 20, and 300 on 17 meters. Have not tried other bands.  The trick is how to key the amp. I installed a keying relay inside the case - $1 for a relay, 6 cents for diodes, and a couple of bucks for a piece of solderable breadboard.  However, the keying relay I installed is activated by PTT instead of by the radio T/R switch.  In order to key the amp on CW, I Rube Goldberged a hand PTT switch in parallel with the amp keying line. That way I can key the amp in CW mode to tune it. Operating CW this way would be cumbersome, but mostly I am on SSB. I do see a noticeable power output spike on first key down, but have not investigated, and have not looked at the keying characteristics.

I get great audio reports using the included mic element.  It doesn't seem to have polarity marked, so if it doesn't work, reverse the wires. That happened to me. I put the little element in a D-104 mic head with a grip-to-talk stand. Now the mic is almost as big as the radio.

The radio came with a few defective parts that gave me some grief: the shaft encoder was bad, but I found an exact replacement at Digikey for about $1.60. Also, the fuse holder was defective (may have been my fault), and the fuse that came with the radio was high resistance and intermittent. I ended up bypassing the case's fuse holder and putting an automotive type in-line fuse cartridge in the power lead. The on/off switch that came with the case was defective, but I ended using the switch in one of the included volume controls.

I found the internal speaker to be tinny sounding, so I just plugged an amplified speaker I already had into the ear phone jack.

The lack of AGC on the receiver is bothersome. I have to ride the volume control constantly when there is QSB or a roundtable with greatly varying signal strengths.

By the time I got to installing the T/R LED it was too much trouble to pull the main board out of the case, so I just skipped it.  

Overall it's an amazingly good radio. Wish I had this in 1961 instead of an S-38 and DX-35.  The radio is a great QSO conversation topic.  I've only been harassed by one group (75 meters, of course), HI!

73,

Ora
W9DJ


Re: Made Mods to V3 uBitx - no longer transmitting #ubitx-help #v3

Steve - KE0VCD
 

The last 2 lines of my above comment were the "Corrected" info.

Re: Made Mods to V3 uBitx - no longer transmitting #ubitx-help #v3

Steve - KE0VCD
 

CORRECTION:
Original comment: I tested voltage out of TXA TXB and TXC (when keyed-up and talking) at R151-3.  Nada

R151-3 voltage unkeyed was 6mV, keyed 22mV  (to be clear I mean milli-volts, not exactly sure mV is correct)

I'm just "stumbling around in the dark, trying to find a flashlight"  :)  Matches?  Glow stick?  

Re: Made Mods to V3 uBitx - no longer transmitting #ubitx-help #v3

Jack, W8TEE
 

This one can handle up to 30W and has a small wattmeter:

Inline image

I probably costs around $15 to build, depending on your junk box.

Jack, W8TEE


On Saturday, February 15, 2020, 10:22:00 AM EST, Steve - KE0VCD <sksshel@...> wrote:


I am not sure what you mean by "go low".

How can I build a "Small Dummy Load"?  I assume I would attach it at the antenna jack, correct?

--
Jack, W8TEE

Re: Made Mods to V3 uBitx - no longer transmitting #ubitx-help #v3

Steve - KE0VCD
 

I am not sure what you mean by "go low".

How can I build a "Small Dummy Load"?  I assume I would attach it at the antenna jack, correct?

Re: The SAGA is COMPLETE!

Jack, W8TEE
 

As a rule, most programmers don't like to write documentation.

Jack, W8TEE

On Saturday, February 15, 2020, 9:48:49 AM EST, Dean Souleles <dsouleles@...> wrote:


Hi Dan - 

To my knowledge Dr. Lee never wrote a manual for the firmware.  I believe the document you have is the most complete document that there is.  The rest of the info I have found in various blog posts on Dr. Lee's KD8CEC blog - http://www.hamskey.com/ and right here on the forum.   He hasn't posted in a while but in the past he has been responded to questions.   I learned about the CW offset from the posts on here and then on his blog.  I've also poked around in the firmware source a bit.

73,
Dean
KK4DAS



--
Jack, W8TEE

Re: The SAGA is COMPLETE!

Dean Souleles
 

Hi Dan - 

To my knowledge Dr. Lee never wrote a manual for the firmware.  I believe the document you have is the most complete document that there is.  The rest of the info I have found in various blog posts on Dr. Lee's KD8CEC blog - http://www.hamskey.com/ and right here on the forum.   He hasn't posted in a while but in the past he has been responded to questions.   I learned about the CW offset from the posts on here and then on his blog.  I've also poked around in the firmware source a bit.

73,
Dean
KK4DAS


Re: The SAGA is COMPLETE!

Daniel Flanagan
 

Hi Dean,

Yes, I bought the Amateurradiokits case for my uBITX.

I did a simple test with the uBITX V5 stock firmware on CW a while
back. I assumed that when I changed the CW tone that the frequency
would shift by the amount that I changed it. I like the tone around
500 Hz, but I did not detect any change in the transmit frequency
after I made the change. Now that I think about that.... once while
listening to a LSB SSB signal I changed to the upper sideband to see
how good the rejection was but I still heard a normal LSB SSB signal.
I then discovered that I had to change the vfo frequency (just one
tick) before the change took affect. This is probably what was
happening on CW also.... this was with the stock V5 firmware. I'll
have to check the V1.2 firmware.

I found a manual online for KD8CEC's V1.072 firmware, hoping that I
would learn about the firmware's features in detail, but it was not
very detailed and sections of the manual were incomplete. I am hoping
to find a manual for KD8CEC V1.2 firmware.

73, Dan (W3DF)

On 2/15/20, Dean Souleles <dsouleles@...> wrote:
Dan -

Fantastic! Glad you are on the air. I'm with you - I love being able to
experiment - but primarily want to use the rig for QRP and it has been fun!
I assume by the mention of the front panel PCB you have the
Amateurradiokits.in case.  That is the one I have and I had a similar
problem with the PCB- but the short was in the mic in input. I think there
may have been a problem with a run of those PCBs.  Sunil was good enough to
send me a replacement PCB - but for the speaker it should be fairly
straightforward to wire around it.  I have an AGC mod sitting on the bench
but haven't gotten around to installing it.  That's probably the only
additional thing I will do for now.

For CW you should be aware that by default the CW transmit frequency is
shifted from the displayed frequency by the frequency of the side tone,  so
you may be transmitting 700-800Hz off the frequency displayed on user
interface.  You can change that behaviour using CEC's UbitX Manager
software.  The details are
here: https://ubitx.net/2018/07/02/kd8cec-firmware-hint-cw-frequency-display/
-   here is the relevant part:

"Update: Dr. Lee addressed this issue in the latest version of the software
and Memory manager! If you select the boxes “Enabled Adjust CW Frequency”
and “Shift Display Frequency on CWL, CWU Mode” and leave the “CW Freq Adjust
Value” at zero (this adds to the CW offset or CW tone selected elsewhere),
then the receiver will automatically shift up or down by the amount of the
selected CW tone and appropriately for CWL and CWU such that the transmitter
will be transmitting on the displayed frequency. " (from Vic WA4THR.)

I really enjoy getting on the air with the uBitX - I'm in Northern Virginia,
near Washington DC.  Last week had good SSB phone QSO with the El Salvador
Dxpedition, Portugal, Nicaragua, Canada and Italy.   Using WSPR and heard
and been heard in Antarctica, South Africa and New Zealand. It was my first
kit of any kind - and it got me over the hump of building things.  I'm
working on a completely homebrew 40 meter transceiver now.

73,
Dean - KK4DAS



Re: The SAGA is COMPLETE!

Dean Souleles
 

Dan -

Fantastic! Glad you are on the air. I'm with you - I love being able to experiment - but primarily want to use the rig for QRP and it has been fun!  I assume by the mention of the front panel PCB you have the Amateurradiokits.in case.  That is the one I have and I had a similar problem with the PCB- but the short was in the mic in input. I think there may have been a problem with a run of those PCBs.  Sunil was good enough to send me a replacement PCB - but for the speaker it should be fairly straightforward to wire around it.  I have an AGC mod sitting on the bench but haven't gotten around to installing it.  That's probably the only additional thing I will do for now.

For CW you should be aware that by default the CW transmit frequency is shifted from the displayed frequency by the frequency of the side tone,  so you may be transmitting 700-800Hz off the frequency displayed on user interface.  You can change that behaviour using CEC's UbitX Manager software.  The details are here: https://ubitx.net/2018/07/02/kd8cec-firmware-hint-cw-frequency-display/ -   here is the relevant part:

"Update: Dr. Lee addressed this issue in the latest version of the software and Memory manager! If you select the boxes “Enabled Adjust CW Frequency” and “Shift Display Frequency on CWL, CWU Mode” and leave the “CW Freq Adjust Value” at zero (this adds to the CW offset or CW tone selected elsewhere), then the receiver will automatically shift up or down by the amount of the selected CW tone and appropriately for CWL and CWU such that the transmitter will be transmitting on the displayed frequency. " (from Vic WA4THR.)

I really enjoy getting on the air with the uBitX - I'm in Northern Virginia, near Washington DC.  Last week had good SSB phone QSO with the El Salvador Dxpedition, Portugal, Nicaragua, Canada and Italy.   Using WSPR and heard and been heard in Antarctica, South Africa and New Zealand. It was my first kit of any kind - and it got me over the hump of building things.  I'm working on a completely homebrew 40 meter transceiver now.

73,
Dean - KK4DAS