Kees & Team: ? I know my opinion may matter little since I am not an engineer but only a Ham for 44 years. I still have and use projects I built many decades ago. And when the sunspot cycle once again peaks, many of us who are now investing into perfecting our uBitx projects will regret if 15, 12, and 10 meters are not included in the final solution.? I want to encourage your team to think long-term and so strive for the most-excellent solution possible, that we all might be proud of our radios and enjoy them for many decades to come. ? ? David A Posthuma, WD8PUO 1 (616) 283-7703 ?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]> On Behalf Of Jerry Gaffke via Groups.Io
Sent: Saturday, September 01, 2018 12:48 PM
To: [email protected]
Subject: Re: [BITX20] uBitx Unfiltered ? Kee's
Are you talking filters at the output of the rig, for use with current v3 and v4 boards?
That's likely the best bet for getting current rigs compliant.
Me, I'm ok with dropping the high bands, maybe stick with just 80,40,20m.
Killing spurs at 15m and 12m is difficult, and perhaps not worth the trouble
at our latitudes given where we're at in the sunspot cycle.
The current four transmit LPF's are sufficient for harmonic suppression if the board routing and relay scheme is clean.
Not sure what a "basic dual relay" scheme is.
I think the relay scheme of post 58293 is sufficient.
Regarding where this rig should go on v5:
Allison has said repeatedly that we need band specific band-pass filters
where we currently have one 30mhz LPF at L1,2,3,4.
She's built lots of this stuff and I respect her opinion.
That's how pretty much all reg-compliant radios are built.
But I am still very curious if we can tame the objectionable spurs with these changes:
1)? ?Use a PX1002 86.85mhz SAW filter instead of the 45mhz crystal filter (and appropriate transistors and transformers)
2)? ?Bring the local oscillators up to the required 7dBm.going into the mixers, keep mixer input signal levels 10dB below that
3)? ?Avoid having stuff from the power amp get into the IF amps.
If that is sufficient to bring the spurs into compliance, we keep the radio almost as simple as it is now.
The primary added complexity is additional gain required in the power amp to deal with the reduced signal from the mixers.?
Somewhat related:? Clean up layout so the BFO does not jump across the 12mhz filter.
Jerry,? KE7ER
On Sat, Sep 1, 2018 at 09:20 AM, Kees T wrote: Since most keep talking about the need for BPFs for certain bands, I "guess" the 80m-10m uBITX direction should be a board with 6 filter capability. How about a?basic?dual relay?switched filter board (3-1/4" x 3-7/8") with?up to?6 pluggable filters, all?manually?selected with a 6 position rotary switch --AND-- you can add a small mux board with 2N3904 relay drivers for the filters and a CD4028B mux to go from the 3 Raduino pins to the filters. Start with the existing 111,110,100,000 and add two others when you have the code modified.
|
Kee's
Are you talking filters at the output of the rig, for use with current v3 and v4 boards?
That's likely the best bet for getting current rigs compliant.
Me, I'm ok with dropping the high bands, maybe stick with just 80,40,20m.
Killing spurs at 15m and 12m is difficult, and perhaps not worth the trouble
at our latitudes given where we're at in the sunspot cycle.
The current four transmit LPF's are sufficient for harmonic suppression if the board routing and relay scheme is clean.
Not sure what a "basic dual relay" scheme is.
I think the relay scheme of post 58293 is sufficient.
Regarding where this rig should go on v5:
Allison has said repeatedly that we need band specific band-pass filters
where we currently have one 30mhz LPF at L1,2,3,4.
She's built lots of this stuff and I respect her opinion.
That's how pretty much all reg-compliant radios are built.
But I am still very curious if we can tame the objectionable spurs with these changes:
1)? ?Use a PX1002 86.85mhz SAW filter instead of the 45mhz crystal filter (and appropriate transistors and transformers)
2)? ?Bring the local oscillators up to the required 7dBm.going into the mixers, keep mixer input signal levels 10dB below that
3)? ?Avoid having stuff from the power amp get into the IF amps.
If that is sufficient to bring the spurs into compliance, we keep the radio almost as simple as it is now.
The primary added complexity is additional gain required in the power amp to deal with the reduced signal from the mixers.?
Somewhat related:? Clean up layout so the BFO does not jump across the 12mhz filter.
Jerry,? KE7ER
toggle quoted message
Show quoted text
On Sat, Sep 1, 2018 at 09:20 AM, Kees T wrote:
Since most keep talking about the need for BPFs for certain bands, I "guess" the 80m-10m uBITX direction should be a board with 6 filter capability. How about a?basic?dual relay?switched filter board (3-1/4" x 3-7/8") with?up to?6 pluggable filters, all?manually?selected with a 6 position rotary switch --AND-- you can add a small mux board with 2N3904 relay drivers for the filters and a CD4028B mux to go from the 3 Raduino pins to the filters. Start with the existing 111,110,100,000 and add two others when you have the code modified.
|
Re: CW PTT attack time to slow
#ubitx
Op 1 sep. 2018, om 04:10 heeft W2CTX < w2ctx@...> het volgende geschreven:
On webpage is NANO software that implements interrupt driven CW generation.? Also
eliminates the voltage divider scheme.? Also eliminates the need to switch between straight key
and paddle as both are always active.
rOn
Okay! Sounds good, will have a look. I¡¯m afraid the stock firmware is unusable for serious CW transmissions.
Thanks, Joop PG4I On August 31, 2018 at 9:41 PM ajparent1/KB1GMX <kb1gmx@...> wrote:
Joop,
IF you don't mind grinding code where are many possible fixes.
I would suggest looking at V4.3 as it is a bit cleaner and more current.
The biggest issue is the software is a big loop and CW ke is checked
a few times in the loop.? Speeding up the loop is one thing.? Or making
the input interrupt driven so when there is an event it gets the processors
attention.
Allison
_
|
Re: CW PTT attack time to slow
#ubitx
Op 1 sep. 2018, om 03:41 heeft ajparent1/KB1GMX <kb1gmx@...> het volgende geschreven:
Joop,
IF you don't mind grinding code where are many possible fixes.
I would suggest looking at V4.3 as it is a bit cleaner and more current.
Thanks Allison, I have now tried both the recent V4.3 from GitHub and KD8CEC¡¯s version. Ian¡¯s version works a bit better, attack time is improved but I am still missing part of the first dit to dah. The biggest issue is the software is a big loop and CW ke is checked
a few times in the loop. Speeding up the loop is one thing. Or making
the input interrupt driven so when there is an event it gets the processors
attention.
Allison Thanks for responding Regards Joop
|
Makes more sense in this thread.............
Since most keep talking about the need for BPFs for certain bands, I "guess" the 80m-10m uBITX direction should be a board with 6 filter capability. How about a?basic?dual relay?switched filter board (3-1/4" x 3-7/8") with?up to?6 pluggable filters, all?manually?selected with a 6 position rotary switch --AND-- you can add a small mux board with 2N3904 relay drivers for the filters and a CD4028B mux to go from the 3 Raduino pins to the filters. Start with the existing 111,110,100,000 and add two others when you have the code modified.
73 Kees K5BCQ?
|
Answer. The comment means- apparently someone is successful in the realm of electronics and ham radio. Another coincidental BITX success? It is simply newsworthy and I wish him well. Just like when Ashhar Farhan was inducted into CQ Amateur Radio Hall of Fame and noted here as well. Both men deserving.
|
Re: Which xmit transistor is blown?
Sean,
Its been several days since you posted likely its eben seens and everyone
is going "I dunno".
? ?Not enough description of whats going on or not going on.
? ?Not sure of the mods done.
? ?Are you sure the power supply is working ok.
Everyone running FT8? without big fans seems to fry the amp and yet people
run FT8 and likely at max power possible.? ?While technically it ignoring spurs
and harmonics can there is insufficient cooling for sustained long transmit
sessions.? HINT: cooling required.
FYI hot is how hot?? You get a blister?? You don't what to hold your finger
there long?? You feel warmth but its not painful?
Many of the devices will heat to the very warm (about 140-150F).? That is not
deadly to the devices.? It will make things like bad soldering stand out and
maybe other marginal components fail.? Some have noted board level
issues like open vias.
Allison
?
|
Re: UBITX V3 TX GAIN VARIATIONS
Jerry KE7ER,
>>I'll be buying that amp, once Hans starts selling it..?Then cut off the back half of my uBitx with a hacksaw.
And you have half a radio....? That amp expect about 12-14dbm drive from the radio its attached to.
That means for ubitx you still need about 32-34 db of gain hopefully not 3904s from the mixer output
to the amp...? and still need a band pass filters for the bands above 17M and low pass filters,
and there are issues with ground loops in the modulator and IF area.
Its total gain is about 26-28DB.? To get from the mixer output though filters and result in 10W you
need about 60DB of gain.? ? If you want even performance 3-30mhz you need to design that
predriver section to be flat.? ? Which by the way is half the reason why the current amp
has low power at 10M.
As to the comment > Bag, cat, QSX amp.? The cats out of the bag, and its a QSX amp.
Allison
|
Dennis Yancey
???#58339??
They are marginal at best... save your Money and get a better tuner...
What is marginal at best?? Does it have a name or ?
Allison
|
Re: VK4PP uBitx LPF BANK add-on Board
Timothy,
Have you ever looked at the board???
Fixing the driver moved the problem sage back to the predriver and first amp.
Pulling bands at the time of its design was not considered as it was desired to
have a 3-30mhz radio unbounded by "bands".? ?I agree a bounded transmitter
is a must? but at that time it wasn't heard.
Allison
|
Re: VK4PP uBitx LPF BANK add-on Board
If it is done cleanly it could look quite nice. Specifically small copper lugs can be soldered on the bottoms of the existing control relay pins to attach to braided core Cu wires which are then hot melt glued to the underside of the PCB for strength and brought out to a? green plastic header male plug? that is why I suggested the click in header carrying all voltage GND and the three signals to the PCB designer.
If there is a known good solution to the existing issues that does not require a masters in EE to source and solder up then A Farhan has a prayer at moving? the remaining uBitx boards in Inventory.? Some people will chose to operate the thing unmodded and some with legal sensitivities will want a fix.
After that inventory is moved, I suspect a full rework of the PCB will be out of the Eastern Horizon. My guess. Hopefully with AGC? and so on. Maybe with BS170 x 2? in the driver stage .?
But what is set up as a standard as to what the relay control signals mean now will have to be carried through,? or all hope of? forward software compatibility will fly out the window. That is why a one of ten decoder to use those three lines makes a whole lot of sense even if it adds dollars to the price of the finished board. If you start pulling 15 and 17 metres off the table,? (combined filter) the multiband nature of the beast is compromised.? ? Timothy E. Fidler : Engineer BE Mech(1) Auckland , NDT specialist AINDT UT /RT3 , MT2?
Telephone Whangarei?? 022? 691 8405
e: Engstr@...
----- Original Message -----
To: <[email protected]>
Cc:
Sent: Sat, 1 Sep 2018 08:29:31 -0500
Subject: Re: [BITX20] VK4PP uBitx LPF BANK add-on Board
My thinking is that this whole discussion of? butchering the uBITx board and cobbling in another board that, in itself, will have some sensitivities to things like cable runs and connections etc. will scare the unseasoned builder away from the uBITx completely.?
? I used what I thought was a better solution for some of the hams near me¡ by simply jumpering/cutting out the whole LPF network¡ so there is raw straight from the PA RF at the output.? I then used one of Hans¡¯ BPF modules outside the box with a minimal BNC to pins ¨C BPF ¨C pins to BNC that is about $5 worth of parts including the filter.? Now granted you must buy all of the filters for bands you want to operate¡ but it¡¯s the same with any LPF replacement? A complete board in an outboard box is even less scary to most.? Yes the rest of us will be at the uBITx endlessly with our hack saws, hot glue, and soldering irons¡ but that will only work for us hackish engineering students.
--
¡_. _._
|
They are marginal at best... save your Money and get a better tuner...
--
72 and God bless
KD4EPG
|
Re: VK4PP uBitx LPF BANK add-on Board
Yes, it would be a lot easier to simply disable the LPF's and use external filters. As per the final sentence here from long ago, keeps things very simple:?? ? ? /g/BITX20/message/30530Given the uBitx board layout, trying to get more power on the upper bands will likely add to the spurs due to coupling from power amp to the IF stages. Board layout around the IRF510's is not that good for 30mhz, and the upper bands have more trouble with spurs.? I'd be happy with 80m,40m,20m, which the current uBitx can do a reasonable job of using external filters. But replacing the power amp would solve most of those issues.? No need for the hacksaw, just disable the uBitx power amp. Use coax for signals, I doubt the cabling will be much of an issue. Can experiment with band specific filters in that coax from exciter to power amp if necessary. Still need to address mixer levels, IF gain distribution, and a layout issue that allows the carrier to sneak around the 12mhz crystal filter. But the uBitx does give us a very simple analog SSB exciter/receiver to work with. Hacking up a uBitx to perform well with minimal cost and complexity might help give direction to future versions from hfsignals. Note that the G0UPL power amp will require nearly 20 dBm of power from the exciter.? Final mixer of the uBitx exciter should have about -3dBm max going in (10 dB below the 7dBm local oscillator), and loses 6dB.? So will want up to 30dB of gain between the uBitx exciter?and the G0UPL power amp, perhaps two MMIC's built ugly style on the back of the uBitx.? Not exactly trivial. Jerry
On Sat, Sep 1, 2018 at 06:29 AM, MadRadioModder wrote:
My thinking is that this whole discussion of? butchering the uBITx board and cobbling in another board that, in itself, will have some sensitivities to things like cable runs and connections etc. will scare the unseasoned builder away from the uBITx completely.?
?
I used what I thought was a better solution for some of the hams near me¡ by simply jumpering/cutting out the whole LPF network¡ so there is raw straight from the PA RF at the output.? I then used one of Hans¡¯ BPF modules outside the box with a minimal BNC to pins ¨C BPF ¨C pins to BNC that is about $5 worth of parts including the filter.? Now granted you must buy all of the filters for bands you want to operate¡ but it¡¯s the same with any LPF replacement.? A complete board in an outboard box is even less scary to most.? Yes the rest of us will be at the uBITx endlessly with our hack saws, hot glue, and soldering irons¡ but that will only work for us hackish engineering students.
?
|
Re: VK4PP uBitx LPF BANK add-on Board
My thinking is that this whole discussion of? butchering the uBITx board and cobbling in another board that, in itself, will have some sensitivities to things like cable runs and connections etc. will scare the unseasoned builder away from the uBITx completely.? ? I used what I thought was a better solution for some of the hams near me¡ by simply jumpering/cutting out the whole LPF network¡ so there is raw straight from the PA RF at the output.? I then used one of Hans¡¯ BPF modules outside the box with a minimal BNC to pins ¨C BPF ¨C pins to BNC that is about $5 worth of parts including the filter.? Now granted you must buy all of the filters for bands you want to operate¡ but it¡¯s the same with any LPF replacement.? A complete board in an outboard box is even less scary to most.? Yes the rest of us will be at the uBITx endlessly with our hack saws, hot glue, and soldering irons¡ but that will only work for us hackish engineering students. -- ¡_. _._
|
Re: VK4PP uBitx LPF BANK add-on Board
The flaming obvious and yet trivial? has? just occurred. How are people going to mount the PCB as shown ?? Three hole posns at a minimum will be required for say 3.4mm holes.? Edge distances have to be allowed.? As long as it can be mounted on a bit of AL with a return on one side it can be bolted anywhere. As it stands it will have to be (hot melt glued ??) to something ..a bit like skyhooks required on some engineering projects in a way.? ?Even if people have to drill out prescribed rondels for the screw down on Rx of the board to keep the price down,? they still need to be nominated.?
|
On Fri, Aug 31, 2018 at 10:41 AM, Lawrence Galea wrote:
Google?picatune ATU
I dust did. An interesting article with lots of good information. Tom, wb6b
|
He is getting a laser cut acrylic version. I offered to print a few if he provided the material to produce the object . Instead they went with laser cut acrylic.?
Ill get s copy of that object in case it comes up again and some one is ok with white or whatever color I happen to have.?
|
I am sure I missed this topic, but my keyer worked fine until I upgraded the firmware to KD8CEC's V. 1.095 for use with my 4.3-inch Nextion?display. Since the upgrade, I can only use a straight?key. What did I miss??
|
Re: CW PTT attack time to slow
#ubitx
On w0eb.com webpage is NANO software that implements interrupt driven CW generation.? Also
eliminates the voltage divider scheme.? Also eliminates the need to switch between straight key
and paddle as both are always active.
rOn
On August 31, 2018 at 9:41 PM ajparent1/KB1GMX <kb1gmx@...> wrote:
Joop,
IF you don't mind grinding code where are many possible fixes.
I would suggest looking at V4.3 as it is a bit cleaner and more current.
The biggest issue is the software is a big loop and CW ke is checked
a few times in the loop.? Speeding up the loop is one thing.? Or making
the input interrupt driven so when there is an event it gets the processors
attention.
Allison
|
Re: UBITX V3 TX GAIN VARIATIONS
I'm convinced Hans did a good job.
Especially after seeing this in his post:
? "I am very grateful to Allison KB1GMX for her advice, encouragement, support and testing during the development of this Linear PA."
I'll be buying that amp, once Hans starts selling it..
Then cut off the back half of my uBitx with a hacksaw.
I had thought of the BS170 in regard to the uBitx in months past,
but figured it might be too lightweight when used as a linear.
And didn't fancy dealing with two more bias pots.
Will be interesting to see how Hans went about it
> Bag, cat, QSX amp.
I've heard of bag phones, though not bag HF linear amps.
And radio has been explained through the metaphor of a very long cat (or lack thereof)
But still seems a bit cryptic.
Jerry, KE7ER
toggle quoted message
Show quoted text
On Fri, Aug 31, 2018 at 06:20 PM, ajparent1/KB1GMX wrote:
Jerry,
If you have been reading I've mentioned I have several amps that actually do work...
Bag, cat, QSX amp.? I've been beating the crap out of one and its by far one of
the most solid designs to date.? Simply put, its that good.? For those that say?
it can't be done the answer is its done at last count everyone of the ones built
in SA YOTA 2018? its not a fluke.
Hans did some fine work.
Allison
|
David Posthuma
MadRadioModder