Re: SSB filter - adjust width to 23kHz

#48998

Yep, I had some wide digi modes to try on 10m, so it appears I'd have
to use SDR for those.
Anyway, thanks for reply.

Show quoted text

On Sun, May 13, 2018 at 9:42 PM, ajparent1/KB1GMX <kb1gmx@...> wrote:
No.  And did you really mean 23KHZ?  That the width used for the old wide FM
15khz deviation.
Generally ladder filters for even 6-8khz are difficult at 12mhz, between
ripple and notches at
the edges its hard to get a decent filter.  For 23Khz I'd go to the drawer
and get a 10.7mhz
or 21.4mhz filter as they can be had with 25khz bandwidth.  However its not
going to fit well.
The small 4pole filter for those frequencies you would need two cascaded
with noticeable loss.

You would also have to get a wider filter for the 45mhz spot too.

Allison

Variable IF

#48997

Thanks Allison.  I forgot about the K2 and will check the schematic when I return home.  One nice thing about the old MXM circuit was that it was continuously variable rather than the step system used in the K2.  Both work beautifully.  I was thinking that the continuous shift might work, until I went back to read your comment about adjusting the BFO as well.  I should have realized that if there was a simple solution, someone would have done it already.  Life is never quite as simple as I would like it to be and my self-taught electronics instructor is nearly always behind the curve.

Gary, KJ5VW

Re: SSB filter - adjust width to 23kHz

#48996

No.? And did you really mean 23KHZ?? That the width used for the old wide FM 15khz deviation.
Generally ladder filters for even 6-8khz are difficult at 12mhz, between ripple and notches at
the edges its hard to get a decent filter.? For 23Khz I'd go to the drawer and get a 10.7mhz?
or 21.4mhz filter as they can be had with 25khz bandwidth.? However its not going to fit well.
The small 4pole filter for those frequencies you would need two cascaded with noticeable loss.

You would also have to get a wider filter for the 45mhz spot too.

Allison

Re: Oops - variable bandwidth IF

#48995

Once the proper capacitance has been determined for 2800 Hz and 500 Hz

bandwidth, it should be possible to use fixed capacitors with diode, transistor,

MOSFET, or relay switching to select bandwidth.

Arv? K7HKL
_._

Show quoted text

On Sun, May 13, 2018 at 1:34 PM, ajparent1/KB1GMX <kb1gmx@...> wrote:

The hyperabrupt diodes can easily go from a 400pf down to 50 and there are smaller value diodes
for the 40- 250pf range.? They are not all that cheap.

The alternate is switching between two filters each tuned as desired.? Diodes or relays can do
the switching.? The advantage of doing that is the end transformers can be optimized for the filter
in question as the terminating impedance changes with bandwidth.

Allison

Re: Oops - variable bandwidth IF

#48994

The hyperabrupt diodes can easily go from a 400pf down to 50 and there are smaller value diodes
for the 40- 250pf range.? They are not all that cheap.

The alternate is switching between two filters each tuned as desired.? Diodes or relays can do
the switching.? The advantage of doing that is the end transformers can be optimized for the filter
in question as the terminating impedance changes with bandwidth.

Allison

Re: Variable uBitx IF bandwidth question

#48993

Gary,? if you need large type let us know.

Using Varicap diodes is do able as the Elecraft K2 does exactly that.

Allison

Oops - variable bandwidth IF

#48992

Managed to get my thinking on the caps.backward...the current filter is already designed for SSB, so we would have to lower the cap value for CW filter, not increase.  Reminds me I shouldn’t do anything from memory.  Basic issue remains, finding a varicap with sufficient shift in capacitance.

Gary, KJ5VW

Variable uBitx IF bandwidth question

#48991

开云体育

I apologize if this topic was covered in an early message string, but I have only been actively reading the list for a couple of weeks and noticed a couple of posts about it.

In the mid-1990s, I built the MXM transceiver kit and it used a 100 ufd or 150 ufd panel-mounted air variable capacitor to vary the IF band width of a four crystal Cohn filter where all the caps were the same value. If I remember correctly (at my age that isn’t always a safe assumption), the IF band width would vary from about 100 hertz to 500 hertz, maybe as much as 1 kHz.? Sorry, but I am out of town and can’t check the schematic.

Based on some of my early QRP kit building experience, the change in capacitance you could get from varying the voltage was fairly limited.? Wasn’t ?the shift around 25-50ufd??Also, the change wasn’t linear across the entire range, though you could restrict the voltage and keep in a linear range.

My question is this:? Are there varactor diodes today with enough capacitance range to shift from a 2 khz bandwidth SSB signal to a 500 Hz CW signal?

I have not done the math to calculate how much capacitance shift would be required and in what range (...and not sure I even remember where to start - - so little time, so many dead brain cells). ?I assume it would need to go from the current 100ufd caps upward by at least another 200ufd ?I think my White Mountain 20 meter SSB transceiver from the 1990’s used around 330ufd in the 8 MHz SSB filter.

Also wonder if we could use a polyvaricap instead of diodes?

I agree with Bill that this would be a pretty easy solution if the right varactor diodes could be found.

Gary, KJ5VW

Re: can I bypass the CP2102 chip on the NANO with an FTDI based serial adapter? #arduino #cp2102 #ftdi #nano

#48990

the FTDI interface board has selectable 5 or 3.3 volt TTL logic levels, along with lead for the reset pin. it's intended for Arduino boards that do not have the Serial interface chip.?

-Justin N2TOH

Wiring schematic uBITX

#48989

Has anybody drawn out a schematic for the USB & D9 connectors supplied with Inkits metal chassis kit? I know it has been suggested to them, but I don't think they've posted anything yet. Thanks.
?????????????????????????????????????? Bill A.

Re: BITX QSO Afternoon/Evening, Sunday, May 13, 3PM & 7PM Local Time, 7277 kHz in North America, 7177 kHz elsewhere.

John P

#48988

On Sun, May 13, 2018 at 11:23 am, kc1at wrote:

I don't hear a single station on 40M. CW or SSB

Same here! Only 2 stations on FT-8, and my 100W FT-8 signal is only getting out about 700 miles.
?
--
John - WA2FZW

Re: SWR

#48987

Hello

Discussions about reading FWD and REV power and SWR bring to mind a

possibility that I have not yet seen in BITX software.? Would it be advantageous

to periodically check RF measurements and compare with some pre-set

threshold to drive an alarm function?? This way you would know if something

changed while you are using the rig.

Arv? K7HKL
_._

Show quoted text

On Fri, May 11, 2018 at 4:10 PM, Jerry Gaffke via Groups.Io <jgaffke@...> wrote:

Problem with the two Nano readings goes away if you tune up the antenna in CW mode.
Or whistle into the mike, assuming you can whistle a pure single frequency note.

It might sort of work well enough on regular SSB phone transmisssions with large enough caps on the detector outputs.
I'd have my doubts about the SWR figure in that case, but the forward and reverse power readings might be useful.

Here's a couple old posts:
? ??/g/BITX20/message/48408
? ??/g/BITX20/message/48422

So the bias offset could be read once immediately before we start transmitting?
at the outputs of the two diode detectors, the same points we read during transmit.
Alternately, we could read the offset once during calibration and store it to eeprom.
Or read it once during development and hard code that offset into the firmware.
I think that last one would work well enough, better than most such detectors which have no offset adjustment.

Then whenever we read the diode detectors for power and swr, we subtract that offset.

I'm shooting for a max reading of 25 Watts into 50 ohms.?
So best case resolution of the detectors using the 10 bit ADC on the Nano is 25000/1024 or about 25mW
Could have a low range by going to the 1.1v reference voltage on the Nano ATMega38P's ADC, bringing that to around 5mW.
? ??

So when measuring 100mW, the 5mW resolution means we might see 95mW, 100mW, or 105mW.
If everything works perfectly and the ADC's are really 10 bit, this could be useful down to around 100mW.

Jerry, KE7ER

On Fri, May 11, 2018 at 02:36 pm, Kees T wrote:

OK, sounds good to me and eliminates the external ADC and S/H circuitry. I thought there was a problem syncing up the two Nano readings here also ? I really like those HSMS-2815 matched diodes. Where are you going to read the bias offset to subtract from the ADC reading ? ....may change over time.

Re: PA breaking into oscillation? (uBitx)

#48986

Many things....? Layout being the most important.

Copying the board layout with leaded parts on copper (dead bug) is not the same
and leads will get you.? Also in that format you build power stages in a straight line
with no folding to make the input and output close.

The other can be that your balanced modulator is not!? You will leak carrier as a result.
Again lead length will be a problem, short is a must.

Any band switching such as the output filters need to be far from the input end of the amp.
also the leads to the relays short an may even require shielded wire COAX.

How about a picture?

Allison

SSB filter - adjust width to 23kHz

#48985

Hi,
can you extend input SSB filtering to from 2.5kHz to 23kHz?

Thanks,
J.

Re: Broadband HF Folded Dipole Antennas

#48984

Gordon,

The trick with non resonant antennas is to insure the lengths are not resonant.??
I never said it was easy hence Cebik? going with two one 44ft and 88Ft
for 3-30 coverage.

Another trick is a resistor say 200 ohms at the series 1 wave high current point
(2 for a doublet) as then the non resistor part is not terminated.? Since part of the antenna is
not terminated the efficiency holds for higher frequency as the resistor is high enough
and the pattern is less broken.? At lower frequencies the resistor can be situated at
a higher voltage point to invoke low loss.? Inductive chokes can be used to do that too.
However you modeling software has to be able to run many cases (optimizer) to
find the right points.? This is a case wher going for maxgain is not the case but for
even feed point impedance or maybe less lobes in the pattern.

Like said....? Class three antenna it works, not optimum, but meets other criteria.
That is pretty much true for most antennas.

Allison

Re: BITX QSO Afternoon/Evening, Sunday, May 13, 3PM & 7PM Local Time, 7277 kHz in North America, 7177 kHz elsewhere.

#48983

I don't hear a single station on 40M. CW or SSB...some digital around 7104 but NOTHING else....Is it me ??? Maybe all out celebrating Mothers Day.????
We had heavy thunderstorms roll thru here this morning around 3AM. Antennas were disconnected..

Don

Re: PA breaking into oscillation? (uBitx)

#48982

I hope you are testing into a dummy load. The circuit is quite stable over many reported builds. I also hope that everything is laid out in a line.

- f

Show quoted text

On Sun, May 13, 2018 at 10:03 AM, Mike <msmith@...> wrote:

I could have a balance?issue with the mixer diodes, I suppose. I'm using discrete BAT85 . . . not necessarily matched but from the same lot (Digi-Key).
The BAT85 is electrically identical to the BAT54.

Re: boosting the power on 28 MHz #ubitx

#48981

开云体育

Allison,

concerning the bifilar balun choke to supply the DC voltage to the drains of the IRF510 is one big step forward in equalizing the gain!

I got a large improvement on the famous 20W G6ALU amp for my PIC A STAR after inserting this choke.?

I also agree that the ft of the 2N3904 for a high gain stage is not sufficient-- my LTSPICE simulations already show that the W7ZOI bidi amp in the receive chain drops by about 3 dB @45 MHz i.e. the first If.? So the MPSH10 or its SMD equivalent may be the better solution.

I just have ordered this transistor from a chinese supplier.

Henning Weddig

DK5LV

Am 13.05.2018 um 19:36 schrieb ajparent1/KB1GMX:

Show quoted text

Seems like the variation between radios is significant.

For certain I nee to look at the power chain.? My breadboard is better behaved but there are a handful of significant
differences.? The first of which is that based on other radios I've built the 2n3904 is not adequate above 20mhz.? I've
used 2n2369 for the first two stages and 2n2222A for the driver.? I think the MPSH10 in the driver might help further
but I'm waiting for some..? Significantly different transformers for t10 and t11 and used? a bifilar balun at l8/l9.
That amp is standalone and for a given output power (enough for 10W at 80M) that same input power at 10M
gets me 6w, thats ok but I know it needs more work.? That was the point it was at about 2 years ago. I got side
tracked by other projects so now it the time to revisit.

Generally the firmware makes no difference unless you have moved the various oscillator frequencies out of range.

Allison

Re: boosting the power on 28 MHz #ubitx

#48980

Seems like the variation between radios is significant.

For certain I nee to look at the power chain.? My breadboard is better behaved but there are a handful of significant
differences.? The first of which is that based on other radios I've built the 2n3904 is not adequate above 20mhz.? I've
used 2n2369 for the first two stages and 2n2222A for the driver.? I think the MPSH10 in the driver might help further
but I'm waiting for some..? Significantly different transformers for t10 and t11 and used? a bifilar balun at l8/l9.
That amp is standalone and for a given output power (enough for 10W at 80M) that same input power at 10M
gets me 6w, thats ok but I know it needs more work.? That was the point it was at about 2 years ago. I got side
tracked by other projects so now it the time to revisit.

Generally the firmware makes no difference unless you have moved the various oscillator frequencies out of range.

Allison

Re: Broadband HF Folded Dipole Antennas

#48979

For what its worth
Ulf Sm0cgl
http://www.hard-core-dx.com/nordicdx/antenna/wire/t2fd.html