开云体育

I have a R-PI3+ built up as laptop and running Raspbian on a 64GB uSD.
I call it my steampunk craptop (laptop based on Pi and battery powered).
Its been fairly easy to take most anything that runs under Debian linux
(mint, ubuntu and others) and compile it for the Pi. Helps that many
already have done that for me.

I plan to get a PI-4 as an upgrade as the pi-3 has only 1gb ram
and it tends to slow it down for big applications. With 4GB the
Pi-4 solves that.

For smaller stuff I have a PI-2 and Pi-zero for that but I find I
can get STM32F4 even cheaper for dedicated applications.

As a long time linux user to me the Pi is a cheaper smaller platform
that does it well and stays familiar

Allison
--
Please reply on the group, no private emails we as a group get to share info that way.

Thanks Vojtech for pointing out KD1JV design but the idea I have in mind is to recover parts or modules of old cellphones to be reporposed for ham cw encode and decode.
Today I dont know enough of cellphone block diagrams if its simply disconnecting the 900mhz transmitter block and adding a simple vfo plus transmitter chain.
The logical step of course is to make research on cellphones diagrams.
Cheers,Emi ,DW3 CBO
On Wed, Sep 18, 2019 at 3:46 PM, Vojtech Bubnik<bubnikv@...> wrote: In winter 2015 I bent my old code from 2010 to a "value line" MSP430 processor that comes with the MSP430 launch pad kit.





The original design is here.


I never published the update seen on the videos, but I can if there is an interest. I will be happy if anybody is willing to kit it.

73, Vojtech OK1IAK

I handle a lot of metal over the years and a few tricks.

Improvised brake, two pieces of steel angle iron and the material to be cut
between them and score along the edge with a really sharp sheetrock knife.
the metal will bend along the score and snap off (works well for brass and
aluminum). With a block of wood it can do good bends with practice.

While I'e used a table saw for thicker aluminum the blade is a plywood blade with
candle wax or alumacut applied to help keep it cool. Works well for .090 and
.100 materials.

However for most cuts a band saw with a suitable and quality blade
and coolant is the favorite.

Allison
--
Please reply on the group, no private emails, we as a group get to share info that way.

Hi guys, New to the group. You can use a carbide tipped Formica scoring tool you will never wear it out, just use a straight edge and pull it to you (clamp straight edge to part and table) 73

In winter 2015 I bent my old code from 2010 to a "value line" MSP430 processor that comes with the MSP430 launch pad kit.





The original design is here.


I never published the update seen on the videos, but I can if there is an interest. I will be happy if anybody is willing to kit it.

73, Vojtech OK1IAK

Hi Juanjo,

Thanks for your interest.

The project will be published completely on github. The firmware is just a working sketch as of now, and there is no printed board. I am in the process of studying details the ATtiny13a datasheet to fully make use of the power saving and A/D noise reduction features, while Dan OK1HRA (who runs ) is working on the PCB.

The circuit is so simple that you can basically read it out from the photos, hi.

Vojtech OK1IAK

No route to host

73! Mark KA6WKE

Website:

The really expensive machine!!


Monty Python's Meaning of Life reference.....

ping

A lot of resources are wasted on tons of out of fashion? but still good cell phones!!?

Why not rescue the text mode together with the key board and the lcd display for use as CW keyer and?decoder?
Old nokia phones lcd displays are already in use in dds vfos.
Has any one started this project?Will be very glad to hear someone have made a success out of this idea!!
73Emi,? ?DW3 CBO

Hi Vojtek,

Would love to see the full document write up for the project. My guess is
you are using a resistive absortion bridge for direct/reverse wave sampling
and then a small microcontroller but I would love to see more info. TIA!

73, 72 de Juanjo, EC5ACA. EA-QRP #104, G-QRP #9742, QRP-L #1662,
FP #899.

Juanjo Pastor
C/San Roque, 4-1?
46460 Silla
SPAIN

e-mail: ec5aca@...
web:
web del club:
Tel.: +034 96 120 17 67
Móvil: 651 35 35 11


-----Mensaje original-----
De: [email protected] [mailto:[email protected]] En nombre de Vojtech
Bubnik
Enviado el: lunes, 16 de septiembre de 2019 9:26
Para: [email protected]
Asunto: Re: [qrp-tech] powering an SWR meter from transceiver output

I have a prototype working, see the attached photos showing the infinite SWR
with open antenna terminals by lighting up the last LED, and showing a SWR
roughly 1:1.1 with a 50 Ohm load. The bridge is powered with around 1.5W
input power at 40m.

The six orange LEDs are interpolated by PWM, showing smooth transitions
between levels. The SWR scale is following:
1st LED - 1:1
2nd LED - 1:1.5
3rd LED - 1:2
4th LED - 1:3
5th LED - 1:5
6th LED - 1:8 and higher

therefore the 1st LED being illuminated significantly more than the 2nd LED
indicates SWR somwhere between 1:1 and 1:125.

On key up, the SWR meter is powered from a large 1000uF capacitor for around
1 second to blink the power detected on the 6 LED scale. I think that is all
the functionality I can manage to squeeze into the 1kB of FLASH of this tiny
$1 micro.

The device works somehow, but it is far from optimal. The micro is powered
by a full wave rectifier with an artificial ground in the middle of the
resistive bridge. As long as the current drawn is low, such arrangement
should not cause significant read out errors or harmonics generated. I plan
to do simulations of harmonic content in ltspice. The SWR bridge electronics
is grounded to the virtual ground with a low impedance wire, therefore it
shall be physically as tiny as possible to produce as little as possible
capacitance against the other poles of the bridge, and this capacitance
shall be balanced. The other three connections from the electronics to the
SWR bridge are high impedance (1kOhm for power, 39kOhm for the forward and
reflected power read outs).
I do not consider myself an expert in RF electronics, though I have a
feeling that such a setup will be less than optimal above 20m due to the low
impedance path between the electronics ground and the virtual ground at the
bridge, but with today's tiny SMD circuitry the capacity may be lower than
expected.

There is a lot of work to be done on fine tuning the firmware and possibly
the circuit. The firmware should utilize power saving features of the micro
to a full extent, it should run the A/D conversion with all the processing
disabled to lower the noise, it should use averaging to improve read out
SNR, the diode detector read out shall be linearised. Currently the
electronics draws around 2.5mA from the transceiver with 1mA flowing through
the LEDs. To lower the micro current below the planned 1.5mA, as low as
possible Vcc shall be used, likely around 2.2 to 2.5V. The micro would work
from 1.8V, but a little bit higher voltag is required to have enough of
current regulation on the resistors powering the low voltage red LEDs. The
CPU shall be operated at as low as possible CPU frequency.

It is an interesting engineering optimization task, it is a lot of fun, and
it is manageable due to the simplicity of the circuit and the ATtiny13A
micro. As always, I will be thankful for constructive criticism of my
endeavors.

Vojtech OK1IAK

I have a prototype working, see the attached photos showing the infinite SWR with open antenna terminals by lighting up the last LED, and showing a SWR roughly 1:1.1 with a 50 Ohm load. The bridge is powered with around 1.5W input power at 40m.

The six orange LEDs are interpolated by PWM, showing smooth transitions between levels. The SWR scale is following:
1st LED - 1:1
2nd LED - 1:1.5
3rd LED - 1:2
4th LED - 1:3
5th LED - 1:5
6th LED - 1:8 and higher

therefore the 1st LED being illuminated significantly more than the 2nd LED indicates SWR somwhere between 1:1 and 1:125.

On key up, the SWR meter is powered from a large 1000uF capacitor for around 1 second to blink the power detected on the 6 LED scale. I think that is all the functionality I can manage to squeeze into the 1kB of FLASH of this tiny $1 micro.

The device works somehow, but it is far from optimal. The micro is powered by a full wave rectifier with an artificial ground in the middle of the resistive bridge. As long as the current drawn is low, such arrangement should not cause significant read out errors or harmonics generated. I plan to do simulations of harmonic content in ltspice. The SWR bridge electronics is grounded to the virtual ground with a low impedance wire, therefore it shall be physically as tiny as possible to produce as little as possible capacitance against the other poles of the bridge, and this capacitance shall be balanced. The other three connections from the electronics to the SWR bridge are high impedance (1kOhm for power, 39kOhm for the forward and reflected power read outs).
I do not consider myself an expert in RF electronics, though I have a feeling that such a setup will be less than optimal above 20m due to the low impedance path between the electronics ground and the virtual ground at the bridge, but with today's tiny SMD circuitry the capacity may be lower than expected.

There is a lot of work to be done on fine tuning the firmware and possibly the circuit. The firmware should utilize power saving features of the micro to a full extent, it should run the A/D conversion with all the processing disabled to lower the noise, it should use averaging to improve read out SNR, the diode detector read out shall be linearised. Currently the electronics draws around 2.5mA from the transceiver with 1mA flowing through the LEDs. To lower the micro current below the planned 1.5mA, as low as possible Vcc shall be used, likely around 2.2 to 2.5V. The micro would work from 1.8V, but a little bit higher voltag is required to have enough of current regulation on the resistors powering the low voltage red LEDs. The CPU shall be operated at as low as possible CPU frequency.

It is an interesting engineering optimization task, it is a lot of fun, and it is manageable due to the simplicity of the circuit and the ATtiny13A micro. As always, I will be thankful for constructive criticism of my endeavors.

Vojtech OK1IAK

Unless you expect your antenna impedance to change radically while you are transmitting, there is no need to continue monitoring your SWR once you have properly adjusted your antenna tuner.? There is a circuit now referred to as a Tayloe SWR bridge (Google it) which uses a resistive bridge during measurement of the SWR. While measurement is in progress it also supplies a protective resistance in series with the transmitter output, so the transmitter cannot see a dangerously mismatched impedance before adjustments are completed.
Once the SWR is measured you switch the bridge out of the circuit entirely, so every milliwatt of output goes to your tuner and antenna.
I? will modestly point out that I described this circuit in an April 1977? QST Hints and Kinks article. But I'm sure I'm not the first person to have thought of it.?

Al Woodhull?? N1AW ex-W1GSJ
Leyden, MA

Here is another version of an AD9833 signal generator. There are several other variations availablr.


DuWayne KV4QB

There are two reasons to try it: First, it is challenge therefore fun. Second, it shall be an external box and it is great to not to worry about flat batteries.

1 mA flowing through a clear high intensity red LED makes my eyes hurt indoors.

Since the diodes are switches they generate

harmonic waveforms that will be contributed to the output of the
transmitter.

I will draw the current before the resistive bridge, therefore the harmonics will be attenuated. And the harmonics will be generated during the tune procedure only. I suppose that at 1.5% power consumed by the rectifier these harmonics will be quite weak.

I think it is time to test it. If I fail, at least I will learn something. Going shopping for the linear regulators.

Vojtech

I get emails from instructables.com for all kinds of projects.
Some of them are electronics and today I got one on building
a signal generator.



I leave it for someone in the group to build and review this
for the start of the building season this fall and winter.

FYI

chuck, k7qo, he who stares at 01001000 combos all day :0)

PS here is a sample

00000000: 7f45 4c46 0201 0100 0000 0000 0000 0000 .ELF............
00000010: 0200 3e00 0100 0000 6007 4000 0000 0000 ..>.....`.@.....
00000020: 4000 0000 0000 0000 a041 0000 0000 0000 @........A......
00000030: 0000 0000 4000 3800 0900 4000 1e00 1b00 [email protected]...@.....
00000040: 0600 0000 0500 0000 4000 0000 0000 0000 ........@.......
00000050: 4000 4000 0000 0000 4000 4000 0000 0000 @.@.....@.@.....
00000060: f801 0000 0000 0000 f801 0000 0000 0000 ................
00000070: 0800 0000 0000 0000 0300 0000 0400 0000 ................
00000080: 3802 0000 0000 0000 3802 4000 0000 0000 8.......8.@.....
00000090: 3802 4000 0000 0000 1c00 0000 0000 0000 8.@.............
000000a0: 1c00 0000 0000 0000 0100 0000 0000 0000 ................
000000b0: 0100 0000 0500 0000 0000 0000 0000 0000 ................
000000c0: 0000 4000 0000 0000 0000 4000 0000 0000 ..@.......@.....
000000d0: 0c37 0000 0000 0000 0c37 0000 0000 0000 .7.......7......
000000e0: 0000 2000 0000 0000 0100 0000 0600 0000 .. .............
000000f0: 183e 0000 0000 0000 183e 6000 0000 0000 .>.......>`.....
00000100: 183e 6000 0000 0000 5c02 0000 0000 0000 .>`.....\.......
00000110: 68fe 4f0f 0000 0000 0000 2000 0000 0000 h.O....... .....
00000120: 0200 0000 0600 0000 403e 0000 0000 0000 ........@>......
00000130: 403e 6000 0000 0000 403e 6000 0000 0000 @>`.....@>`.....

PSS It's really not that bad. Just a joke for those who may have
to some day recover long lost valuable data and programs.

If your running a micro 1.5ma seems very low and LEDs you can see in
daylight want a few ma to themselves.

Also don't forget the transformer and rectifier is a load ont he transmitter and
that is in parallel with the antenna or dummy load which contributes to
SWR . That whole DC recovery has an impedance and what is that?
How will the transmitter behave with that addded (loss in power
and stability with a mismatched load).

There is no something for nothing in RF if your using power from the
transmitter you also contributing load and possible reactance to the
transmitters total load. Also the transformer, rectifier and power
converter do not deliver efficiency as components have loss.

That and the spec for the MPU used may hit low power but
soon as the IO and A/D does something that goes up.

If the power needed is so low a few coin cells would power
it for a very long time. A CR3020 cell is about 225mah.
That would save messing with RF and dealing with
inefficiencies of power conversion. You already have a
battery to run the transmitter why suffer the cost in
efficiency and losses (transfomer loss, rectifier loss,
regulator loss) to convert DC to RF and then back to
DC. Keep in mind that a CW rig running class E might
be 90% efficient but if it s class C its only 70% and
it can easily be lower. If SSB those linear amps are maybe
55% at full power and at less than half power 25% would
be good. So converting precious battery to RF then back
to DC is at best inefficient.

You are correct save for your also incomplete in your loss projections.
Transformers are not 100% efficient, same for diodes (.2V at 1.5ma is
0.3 mw of loss per diode). Since the diodes are switches they generate
harmonic waveforms that will be contributed to the output of the
transmitter. That's harder to predict as it varies with power in and
load power needs. In any case it needs to sty below the requirent
but the telecommunications rehulatory (FCC, PTT, other)
agencies as well as good engineering practice.

In the end you get pecked to death a little at a time by ducks.

I do hardware, software, as well as RF for the last 50 years.

Allison
--
Please reply on the group, no private emails we as a group get to share info that way.

Even if the suggested electronics uses a watt on a 5W radio that's a lot of power not going to the antenna.

The suggested SWR meter shall draw roughly 1.5mA using a 1:2 or 1:3 transformer from the transceiver. Using a 1:2 transformer, with the assumption that only the positive polarity peak will be rectified, to source a 1.5mA voltage regulator, the circuit will draw 3mA from the source at the positive voltage peaks. Transformed by the 1:2 transformer, the current requirement will be halved, that is 1.5mA.

At 5W the maximum peak voltage at the transmitter terminals when loaded with an open resistive bridge will be 29.81V. That corresponds to a 45mW power loss mostly in the linear regulator, that is roughly 1% of the input power.
At 10W it will be 63mW.
At 1W the power drawn to source the SWR meter electronics will be 10V * 1.5mA = 15mW, that is still 1.5% of the input power, that corresponds to -18dB power lost.

The SWR bridge current requirements may likely be even a bit lower, the 1.5mA is quite a conservative estimate using lower CPU speed and highest efficiency LED diodes.

I would be thankful if you correct my thoughs if I am wrong. I am more a software engineer than an electronics engineer.

Vojtech



Vojtech

I pondered the idea an using a QRP radio or even a 20W radio where
some larger chunk of power out goes to SWR/power meter tp power it
seems wrong. Even if the suggested electronics uses a watt on a 5W
radio that's a lot of power not going to the antenna.

Also Tayloe bridge SWR meters are a source of some 6-10db of
insertion loss. This is good while tuning the antenna as it
protects the transmitter from a bad swr but unacceptable
for inline use. So the SWR detector (directional coupler) for inline
use should be of the Tandem match type or similar and even
that sampler eats a little transmitter power. The only point
for it is now reflected power so a simple display works fine.
For power a meter will be far higher resolution.

Why not just use power from the radios 9-15V battery or an internal one?
Saves a lot of extra electronics and doesn't hurt the power out of the TX.
For intermittent use the battery would be small.

As to the EFHW antenna matching unit make it outboard in case
you use a different antenna. For 40-202m a T50-2 or T68-2 should
do the job there are lots of schemes on the net for doing this.

Allison
--
Please reply on the group, no private emails, we a group get to share info that way.