uBitx distorted transmit audio
I received my ubitx in early Jan after ordering it in mid December. I have been on and off with making progress with it, but have been following this forum since then. This board has a problem and I was hoping to see someone else come up with it too. The software rev. is XX at bootup. So I have been hesitant about changing anything until I consulted with the group. It works fine on TX and RX on CW.
On SSB (l or u) the voice audio is severely distorted. When others hear me on the air or when I listen on a local receiver you can tell what the words are (callsigns) but they are distorted. I have tried several different mic elements. I have injected audio from a sig gen into the mic and also directly on the input pin. The rig will pass 300 ¨C 3K tones fine. The output of the audio/mic amp looks fine too. The tones sound good OTA.
When I look at the power out on a dummy load it drops off dramatically above 3Khz. I think the human voice has tones above 3K. The second thing seems to be that the rig is about 2Khz off frequency from the dial. When checking into a local net I¡¯m reported low as I recall.
I hope there is a simpler solution that diddling with the frequency parameters. Oh, the receiver receives fine on the designated frequency.
Any ideas?
Dan Withers ¨C WM7W
I¡¯m old too. I was hand booting a DG Nova 1200 before reset-program load was implemented. I couldn¡¯t lift a Wangco 5 mb disc drive either.
|
Mine was a Timex Sinclair with 2 k of ram. Steve kb1chu
Sent from my Verizon, Samsung Galaxy smartphone
toggle quoted message
Show quoted text
-------- Original message -------- From: "Jack Purdum via Groups.Io" <jjpurdum@...> Date: 5/9/18 11:13 AM (GMT-05:00) Subject: Re: [BITX20] Coding styles
My Sol-20 started out with a cassette interface to load the bootloader. Usually took multiple tries before it read correctly. That was replaced with a NorthStar disk drive with a single-density 90K 5 1/4" disk drive. NorthStar went to DD (180K) and I thought I had died and gone to Heaven. (I upgraded to a Horizon about then, too.) Then came a Morrow Design ThinkerToys 8" floppy drive. Finally, an IBM 5MB drive with a IBM PC where I paid $1500 for the five megabyte drive! I thought: "Man...it can't get any better than this!"
On Wednesday, May 9, 2018, 10:56:50 AM EDT, ajparent1/KB1GMX <kb1gmx@...> wrote:
My first ALTAIR as I have two was purchased December 1974 and arrived with serial number 200.
Built it and used it for two years before condemning it to storage. Replacing it with a
NorthStar Horizon.? Later a Netronics 8085 Explorer, Compupro.? Then along the way real hardware
PDP-8? and PDP-11s where more time programing than fixing was the rule.
One reason was binary blisters terrible switches.? The other was the early design (there were about
3 major versions) was such that it may take several tried to load anything from tape (audio or paper).
Mine was retired due to the green plague, the side effect of gold over copper without nickle plate
and mixing that with?solder plate edge connectors.? By time I'd gotten past that the only thing
still stock was the one-shot laden front panel (hit deposit and get two or none).? I considered that
was a sign, get something better.? I still feel the flakiness was a handicap to programming.? ?
It was a mad crazy time when 4K bytes of ram?cost about a 100$ a K.??
I keep it for the fact that it has history in my collection of systems and some fool might
offer four digits for it.? It was the first kit however some of the followers were designed
and made better.
Allison
|
Re: uBITX for sale - in Circuit Specialists Case
#ubitx
uBITX #324/1 has been SOLD.? Thanks for all the interest.
Jim, W0EB
|
I could not afford the 'Cadillac' Altair 8080.
My first computer was a used Northstar, with probably 8k.
I found out how user-antagonistic CP/M was during memory
upgrades.
Rod KM6SN
|
A $5 RasberryPi Zero might be.
The Nano, not so much.
toggle quoted message
Show quoted text
On Wed, May 9, 2018 at 08:36 am, Gerry Hull wrote:
Now the arduino?that powers our uBitX?is 10,000 times more powerful.
?
|
Almost the same experience!
My ham roommate bought an Altair 8800.? Man, those 2102 RAMs were expensive.? Also used ASR33.
Now the arduino?that powers our uBitX?is 10,000 times more powerful.
73, Gerry W1VE
On Wed, May 9, 2018 at 11:25 AM, Dennis Zabawa <kg4rul@...> wrote: My first computer was an IMSAI 8080 in kit form.? I remember toggling in the boot loader from the front panel.? Then I got the loader in ROM and thought I was living high on the hog.? My I/O device was a Teletype Model 33 with reader/punch, that I got non-working and refurbished myself.? I was able to load the OS, Basic Interpreter, Basic Code, and data using the reader and punch tape for data and programs with the punch.? Then, another advance, I got a? dual,cassette tape interface.? How far we have come!
|
My first computer was an IMSAI 8080 in kit form.? I remember toggling in the boot loader from the front panel.? Then I got the loader in ROM and thought I was living high on the hog.? My I/O device was a Teletype Model 33 with reader/punch, that I got non-working and refurbished myself.? I was able to load the OS, Basic Interpreter, Basic Code, and data using the reader and punch tape for data and programs with the punch.? Then, another advance, I got a? dual,cassette tape interface.? How far we have come!
|
The SWR/power measurements are better done off the Jackal board to keep the RF routes and the processor away from each other.? Just access to an I2C or SPI port is all that is needed. ? ? Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ ? Owner - Operator Big Signal Ranch ¨C K9ZC Staunton, Illinois ? Owner ¨C Operator Villa Grand Piton ¨C J68HZ Soufriere, St. Lucia W.I. Rent it: Like us on Facebook! ? Moderator ¨C North American QRO Group at Groups.IO. ? email:? bill@... ? ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Jack Purdum via Groups.Io
Sent: Wednesday, May 9, 2018 9:24 AM
To: [email protected]
Subject: Re: [BITX20] SWR? We were going to put an SWR meter on the JackAl board, but have backed away from it for this iteration. The main reason was because of the board size. The nano-acres it would take on the board would raise the PCB cost above the 100x100mm size to do it right for the possible power levels that might be involved. That doesn't mean you can't add one... ? The good news: Right now, there are about a dozen "empty" pins available on the board for experimenting. We are currently using less that 15% of the 1MB of flash memory and less than 10% of the 256K of SRAM, and that includes code space for some features that we've coded for (e.g., a RTC) but have not implemented yet (e.g, adding a button battery to power the Teeny's RTC in sleep mode). The Teensy is a 3.3V device, so we have an onboard regulators for 5V and 3.3V. Al and I think this will up the "fun level" for hackers considerably...at least that's our intention. ? The bad news: We got caught in some kind of Chinese holiday and other "delays" to get the PCB. Al ordered the board since he did all the work on the EE design. It seems our Beta PCB order got pushed to the back of the line. When I wrote to them and pointed out that I was disappointed in their service, especially after ordering more than 1000 boards from them last year, our order suddenly went from "In line" to "shipped" in less than 24 hours. I was a little more than PO'ed, but at least we got the board this week. We discovered several errors on the board (2 ours, 1 theirs), so the Beta board will have some "hairs" on it. Long story short, we will be demoing JackAl at FDIM, but not all of the features will be implemented. The order for the new board will be sent this week, and I think it will be done in a more "timely" fashion this time. We'll immediately send it to our Beta testers and then make it available via an announcement here. ? Al and I look forward to seeing some of you at FDIM! Jack, W8TEE On Wednesday, May 9, 2018, 9:49:27 AM EDT, Kees T <windy10605@...> wrote: I think Jack, W8TEE, is waiting for FDIM to tell us all the correct way to solve the problem.
Just found an article from January 2011 QST which uses LF398N parts and a PIC16F876A by Bill Kaune, W7IEQ.
73 Kees K5BCQ
|
My Sol-20 started out with a cassette interface to load the bootloader. Usually took multiple tries before it read correctly. That was replaced with a NorthStar disk drive with a single-density 90K 5 1/4" disk drive. NorthStar went to DD (180K) and I thought I had died and gone to Heaven. (I upgraded to a Horizon about then, too.) Then came a Morrow Design ThinkerToys 8" floppy drive. Finally, an IBM 5MB drive with a IBM PC where I paid $1500 for the five megabyte drive! I thought: "Man...it can't get any better than this!"
On Wednesday, May 9, 2018, 10:56:50 AM EDT, ajparent1/KB1GMX <kb1gmx@...> wrote:
My first ALTAIR as I have two was purchased December 1974 and arrived with serial number 200.
Built it and used it for two years before condemning it to storage. Replacing it with a
NorthStar Horizon.? Later a Netronics 8085 Explorer, Compupro.? Then along the way real hardware
PDP-8? and PDP-11s where more time programing than fixing was the rule.
One reason was binary blisters terrible switches.? The other was the early design (there were about
3 major versions) was such that it may take several tried to load anything from tape (audio or paper).
Mine was retired due to the green plague, the side effect of gold over copper without nickle plate
and mixing that with?solder plate edge connectors.? By time I'd gotten past that the only thing
still stock was the one-shot laden front panel (hit deposit and get two or none).? I considered that
was a sign, get something better.? I still feel the flakiness was a handicap to programming.? ?
It was a mad crazy time when 4K bytes of ram?cost about a 100$ a K.??
I keep it for the fact that it has history in my collection of systems and some fool might
offer four digits for it.? It was the first kit however some of the followers were designed
and made better.
Allison
|
Like the PDP8 and Imsai 8080.
Back before EEPROM or fuse link ROM or Flash.
The alternative was to build enough memory by hand to boot the machine
using an array of diodes, a diode wherever you needed a zero bit.
But a row of switches on the front panel was cheaper, and the hourly rate for
an operator to flip switches every morning was pretty cheap too.
If the operator did it enough, they could key in the boot sequence from motor memory
while thinking about some coding problem to be tackled that day.
I had a friend who spent enough time around a PDP8 and ASR-33 that he could?
look at the holes in punched paper tape and know what was being said.
Jerry
toggle quoted message
Show quoted text
On Wed, May 9, 2018 at 07:36 am, Jack Purdum wrote:
OMG! I built one of the Altair's for a friend and had to program it to test whether I had built it correctly. For those who don't know, those switches are for setting the binary bits for each byte of the program. When you had that byte set, you hit a "Deposit" switch which moved that byte into RAM. You could tell early Altair programmers by the "binary blisters" on their index finger!
Jack, W8TEE
|
My first ALTAIR as I have two was purchased December 1974 and arrived with serial number 200.
Built it and used it for two years before condemning it to storage. Replacing it with a
NorthStar Horizon.? Later a Netronics 8085 Explorer, Compupro.? Then along the way real hardware
PDP-8? and PDP-11s where more time programing than fixing was the rule.
One reason was binary blisters terrible switches.? The other was the early design (there were about
3 major versions) was such that it may take several tried to load anything from tape (audio or paper).
Mine was retired due to the green plague, the side effect of gold over copper without nickle plate
and mixing that with?solder plate edge connectors.? By time I'd gotten past that the only thing
still stock was the one-shot laden front panel (hit deposit and get two or none).? I considered that
was a sign, get something better.? I still feel the flakiness was a handicap to programming.? ?
It was a mad crazy time when 4K bytes of ram?cost about a 100$ a K.??
I keep it for the fact that it has history in my collection of systems and some fool might
offer four digits for it.? It was the first kit however some of the followers were designed
and made better.
Allison
|
Re: #ubitx replaced Arduino Nano gives a lot of I2C interference during reception
#ubitx
Very cool, sounds like you added a 4mhz crystal filter for CW in parallel with? the 12mhz crystal filter for SSB as suggested by Farhan here: ? ?? /g/BITX20/message/36041Further discussion here: ? ?? /g/BITX20/message/36549? ?? /g/BITX20/message/36947You might describe exactly what you found to work, posting code and schematics. Maybe a quick description of what crystals you used and how they were matched. Did this in any noticeable way affect performance when using the 12mhz SSB filter? There's a bunch of CW operators in the forum who might be interested in duplicating your efforts. Jerry, KE7ER? ??
On Wed, May 9, 2018 at 02:42 am, Alex - PA1FOX wrote:
Hi ubitx fans,
This is a follow up from my original message of February 22nd.?
I finally managed to find the root cause of the interference. Having replaced the Nano serveral times, the SI5351 Adafruit breakout board and ending up building half of the digital circuit next to the ubitx in breadboard style. I then eliminated all functions in the firmware of the nano one by one to figure out if I just could set the frequency on the display and the SI5351 without any interference. This worked.
The cause was a software issue. I have added a menu function to change the CW band width from wide to narrow and back. During all the test work I had not linked this software addition to the noise I got in the receiver. During the tuning process, at every frequency change, the function was called and set the correct? IF oscillator frequency to 4Mhz or 12Mhz (dependant on wide or narrow). As this frequency was set again and again with every tuning step this was causing the audio clicks. I now changed the function to check if the frequency is already correct and need not to be set again.
All clicks gone. Pfew, that one got me puzzled for quite some weeks. As often, the issue and solution are very simple as soon as you've discovered them.
I will throw in a youtube video shortly showing the effect of wide and narrow in CW with the ubitx. it's huge!
73, Alex
PA1FOX
|
Re: New RF AGC modification
#ubitx
Indeed there was. Thank You!
The new graphic sketch had the relay labels reversed from Rx to Antenna. Schematic and text were correct, as were the graphics in the RF Gain pages. I just got turned around and TOTALLY missed it. All is corrected now. I hope. Good eye!
73,
Don
|
OMG! I built one of the Altair's for a friend and had to program it to test whether I had built it correctly. For those who don't know, those switches are for setting the binary bits for each byte of the program. When you had that byte set, you hit a "Deposit" switch which moved that byte into RAM. You could tell early Altair programmers by the "binary blisters" on their index finger!
Jack, W8TEE
On Wednesday, May 9, 2018, 9:40:35 AM EDT, John P <j.m.price@...> wrote:
Here are some of the early computers I dealt with. #1 at Stevens Tech, the Univac 1105; a big room full of 12AX7s!  In my early days at AT&T, the Datapoint 2200. A discrete component version of the Intel 8008! Ours had a 9-track tape drive!  Later at AT&T, the TRS-80  And of course, I had one of these at home:  I saw somewhere online that someone has an Arduino based version of it; even looks the same! -- John - WA2FZW
|
Can't wait to see the FDIM presentation detail or other JackAl schematic/functional detail.?
BUT......I'm still focused on the minimalist approach? What can I do for the most function and least cost and minimum UBITX ripup.
73 Kees K5BCQ
|
The terminal looks like a Friden "Flex-O-Writer". ?Had one of those once a HUGE number of years ago - LOL
W0EB
------ Original Message ------
Sent: 5/9/2018 8:40:16 AM
Subject: Re: [BITX20] Coding styles
Here are some of the early computers I dealt with. #1 at Stevens Tech, the Univac 1105; a big room full of 12AX7s!
In my early days at AT&T, the Datapoint 2200. A discrete component version of the Intel 8008! Ours had a 9-track tape drive!
--
John - WA2FZW
_._,_._,_
|
We were going to put an SWR meter on the JackAl board, but have backed away from it for this iteration. The main reason was because of the board size. The nano-acres it would take on the board would raise the PCB cost above the 100x100mm size to do it right for the possible power levels that might be involved. That doesn't mean you can't add one... The good news: Right now, there are about a dozen "empty" pins available on the board for experimenting. We are currently using less that 15% of the 1MB of flash memory and less than 10% of the 256K of SRAM, and that includes code space for some features that we've coded for (e.g., a RTC) but have not implemented yet (e.g, adding a button battery to power the Teeny's RTC in sleep mode). The Teensy is a 3.3V device, so we have an onboard regulators for 5V and 3.3V. Al and I think this will up the "fun level" for hackers considerably...at least that's our intention. The bad news: We got caught in some kind of Chinese holiday and other "delays" to get the PCB. Al ordered the board since he did all the work on the EE design. It seems our Beta PCB order got pushed to the back of the line. When I wrote to them and pointed out that I was disappointed in their service, especially after ordering more than 1000 boards from them last year, our order suddenly went from "In line" to "shipped" in less than 24 hours. I was a little more than PO'ed, but at least we got the board this week. We discovered several errors on the board (2 ours, 1 theirs), so the Beta board will have some "hairs" on it. Long story short, we will be demoing JackAl at FDIM, but not all of the features will be implemented. The order for the new board will be sent this week, and I think it will be done in a more "timely" fashion this time. We'll immediately send it to our Beta testers and then make it available via an announcement here. Al and I look forward to seeing some of you at FDIM!
Jack, W8TEE
On Wednesday, May 9, 2018, 9:49:27 AM EDT, Kees T <windy10605@...> wrote:
I think Jack, W8TEE, is waiting for FDIM to tell us all the correct way to solve the problem.
Just found an article from January 2011 QST which uses LF398N parts and a PIC16F876A by Bill Kaune, W7IEQ.
73 Kees K5BCQ
|
uBITX for sale - in Circuit Specialists Case
#ubitx
|
Re: For those keeping track of uBITX shipping times
#ubitx
Ordered mine on April 20th, shipped May 2nd, arrived here in Nova
Scotia at my QTH on May 7th.
Very happy with the turn around time and the packaging. I have
not yet set it up as I am waiting for my enclosure and don't have
a lot of spare time right now.
On 2018-05-09 12:38 AM, Tom Christian
wrote:
toggle quoted message
Show quoted text
uBITx #2 arrived
today in 8 days.? Swapped it out with #1 to test and it worked
fine for a QSO on 20 meters.? Good deal.? Fast!
Tom
AB7WT
|
I actually removed the jack.. and just ugly bugged the resistor.. still no joy..?
73 de Joe KB5VJY
|
Jim Sheldon
Gerry Hull
K9HZ
Don, ND6T