开云体育

--- In BITX20@..., "eternalesquire" <eternalsquire@c...>
wrote:

Allison,

No, I intend this to be used as 20M. I just use a colorburst crystal
oscillator for signal tracing, and I used a scope to compare the
incoming and outgoing signals to measure gain.

Ok, just curious.

Either those are very poor quality versions of MPS2222 (defective
lot?) or something is very wrong. Could it be measurement error?

I wired up a second stage and checked it against the schematic as I
was doing it, every trace is accounted for.

I had used ground-plane component support construction. Could that
technique reduce the gain?

That works well you unless you use really long leads.
I do groundplane AKA "dead bug", Ugly or Manhattan style mostly
as I can fabricate faster than creating a PCB first. Also ground
loops are unheard of with deadbug. I've used this method to
frequencies up to 1296mhz!

Allison
Kb1GMX

--- In BITX20@..., "ajparent1" <kb1gmx@a...> wrote:

Inline and edited.

--- In BITX20@..., "eternalesquire" <eternalsquire@c...>
wrote:

All,
Next question: I have now finished building the first bidirectional
amplifier after the preselector. I injected a 3 mV signal at
3579 KhZ in each direction, and got out a 20mV phase-inverted
copy of the same signal when power in the correct direction was
applied. The fidelity is good, but the gain stages are not
even a factor of 10. Is this the expected performance?

This is very low gain. The expected should be far higher in the
order of of around 16DB. I would expect to see around 100mv
for 3mv in.

I'm curious why 3.579nmhz? IS thise going to be 80M version?
If so I'd suggest using T50-2(red) ferrite rings for the bandpass
filter or the required turns will be very high.

I am using all parts as specified, for transistors I am using
MPS2222 and for switching diodes I am using 1N4148.

I know these are probably not the best of transistors, but I
want to attempt to duplicate the results per the web article.

Those should be more than adaquate at 4mhz. However, I'd suggest
rechecking your wiring. Make sure the capacitors are in the correct
places.



Allison
KB1GMX

All,

Here is the new archive for BITX20 Engineering diagrams.

I have reposted a new archive, EGR&#92;BITX_REV_C, containing all the
corrections for Revision B diagrams, plus GIF translations done by Rod.

The Eternal Squire

Hello,

This email message is a notification to let you know that
a file has been uploaded to the Files area of the BITX20
group.

File : /EGR/BITX_rev_C.zip
Uploaded by : eternalesquire <eternalsquire@...>
Description : BITX20 Engineering Diagrams, Rev C, 21 August 2005. Rev B diagrams + GIFS

You can access this file at the URL:


To learn more about file sharing for your group, please visit:


Regards,

eternalesquire <eternalsquire@...>

Allison,

No, I intend this to be used as 20M. I just use a colorburst crystal
oscillator for signal tracing, and I used a scope to compare the
incoming and outgoing signals to measure gain.

I wired up a second stage and checked it against the schematic as I
was doing it, every trace is accounted for.

I had used ground-plane component support construction. Could that
technique reduce the gain?

Hi,

Orange, pink and purple/violet are common to 10.7 IF coils.

Allison
KB1GMX

For calculate resonant frecuency in used TOKO or FI cans can be used a
practical old know formulae:

F(MHz)= sqrt(25330/(L(uH) * C(pF))) or L(uH)=25330/(F*F(MHz) * C(pF))
or C(pF)=25330/(F*F(MHz) * L(uH)).

73s
Manuel
EA7ARX

--- In BITX20@..., "bambazonke01" <jimandruth@s...>
wrote:

Does anyone know the colour code for TOKO 10.7 Mhz printed citcuit
transformers?
Which are the IF , oscillator and dicriminator colours?
Jim VE3DDY

They are in 10,7MHz:
orange 1? IF 10,7 (This is the can to be used in BITX)
(100pF capacitor in parallel)
green 2? IF 10,7(uncommun) (100pF capacitor in parallel)
pink (discriminator) (without capacitor as usuall)
blue (disciminator used in old radios)
(without capacitor as usuall)

In 455KHz:
yellow 1? IF 455KHz (1nF capacitor in parallel)
white 2? IF 455KHz (1nF capacitor in parallel)
black AM detector (with secondarie) (1nF capacitor in parallel)
black FM narrowband discriminator (without secondarie)
(1nF capacitor in parallel)

Some others Blue used in 27-34MHz (30pF capacitor in paralell)
All other used TOKO Cans with o without paralell capacitor.

Good Luck
Manuel
EA7ARX

Does anyone know the colour code for TOKO 10.7 Mhz printed citcuit
transformers?
Which are the IF , oscillator and dicriminator colours?
Jim VE3DDY

Inline and edited.

--- In BITX20@..., "eternalesquire" <eternalsquire@c...>
wrote:

All,
Next question: I have now finished building the first bidirectional
amplifier after the preselector. I injected a 3 mV signal at
3579 KhZ in each direction, and got out a 20mV phase-inverted
copy of the same signal when power in the correct direction was
applied. The fidelity is good, but the gain stages are not
even a factor of 10. Is this the expected performance?

This is very low gain. The expected should be far higher in the
order of of around 16DB. I would expect to see around 100mv
for 3mv in.

I'm curious why 3.579nmhz? IS thise going to be 80M version?
If so I'd suggest using T50-2(red) ferrite rings for the bandpass
filter or the required turns will be very high.

I am using all parts as specified, for transistors I am using
MPS2222 and for switching diodes I am using 1N4148.

I know these are probably not the best of transistors, but I
want to attempt to duplicate the results per the web article.

Those should be more than adaquate at 4mhz. However, I'd suggest
rechecking your wiring. Make sure the capacitors are in the correct
places.



Allison
KB1GMX

Hello,

Two weeks I finished my experiment with the 9MHz crystal filter design
and reported the results of it in several files in the File-box Ladder
Crystal Filter Design. I dropped a message in the messages box about
it.

Maybe you can start selecting the crystals the way several amateurs
suggested to you. Look at the Excel file to see how I did this for the
9 MHz filter. Selecting the filter is only the first step and doesn't
result in a good filter automaticaly. Select good C's and take care of
matching the filter to the amplifier stages is also very important.

Ashhar Farhan discribed the filter design in "A simple SSB trenceiver",
this article is in the File box. An other (but more difficult) way is
discribed by G3JIR.

What you could do is finish the BITX20 with the calculated C's. After
that you can see the shape by putting a noise source at the antenna and
send the audio to the computer. You can use SpectrumAnalyzer or an
other program to draw the shape of the filter and do some "fine
tuning" with the C's on the filter. A lot of information about this can
you find in earlier messages of this group.

Please show us the result of the filter shape after you finished the
job. Not many amateurs have done so far. Succes.

73

Ruud
PE2BS

--- In BITX20@..., "eternalesquire" <eternalsquire@c...>
wrote:

Hi,

I never saw anything about the original article about this. I've used
the BFO and a frequency counter to measure all my crystals. How
closely do they need to be matched for SSB reception? for CW

reception?

Thanks,

The Eternal Squire

P.S. I am using 10.7 Mhz as the IF

All,

Thanks for the info re crystals, it will come in handy when I am ready.

Next question: I have now finished building the first bidirectional
amplifier after the preselector. I injected a 3 mV signal at 3579 khZ
in each direction, and got out a 20mV phase-inverted copy of the same
signal when power in the correct direction was applied.

The fidelity is good, but the gain stages are not even a factor of 10.
Is this the expected performance? I am using all parts as specified,
for transistors I am using MPS2222 and for switching diodes I am using
1N4148.

I know these are probably not the best of transistors, but I want to
attempt to duplicate the results per the web article.

Thanks,

The Eternal Squire

All,

Thanks for the info re crystals, it will come in handy when I am ready.

Next question: I have now finished building the first bidirectional
amplifier after the preselector. I injected a 3 mV signal at 3579 khZ
in each direction, and got out a 12mV phase-inverted copy of the same
signal when power in the correct direction was applied.

The fidelity is good, but the gain is only a factor of 4. Is this the
expected performance? I am using all parts as specified, for
transistor I am using PN2222.

Thanks,

The Eternal Squire

My comments at bottom.

--- In BITX20@..., Rahul Srivastava <vu3wjm@y...> wrote:

Hi!

Some one once mentioned , thumbrule for xtal matching was within 10%

of the required BW. However I try to keep it within 100hz for SSB filters.

Till date all appear to have worked fine.

73

Rahul VU3WJM

PS: I make the BFO first and with 33pf in series with xtal measure

all available with me. Select ones within 100hz of each other . Take
the lowest one for carrier osc for BITX , takes care of xtal pulling.

This is a good thing to do. Having done lower sideband by picking
a high crystal makes pulling it high easier.

One other trick. Using the BFO to pick the crystals is a good first
step. Once the filter is fabricated you can use the BFO to sweep the
filter to assess the filter shape and bandwidth. Usually that
requires tacking on a coil and larger tuning capacitor temporarily.


Allison
KB1GMX

Hi,

I never saw anything about the original article about this.? I've used
the BFO and a frequency counter to measure all my crystals.? How
closely do they need to be matched for SSB reception?? for CW reception?

Thanks,

The Eternal Squire

P.S.? I am using 10.7 Mhz as the IF

---

NEW - crystal clear PC to PC

Hello,

Experience and also literature suggests less than 100hz for SSB
bandwidth (2.2-2.8khz) and less than 1/10th the bandwidth
or >30hz for CW filters (200hz to 800hz bandwidth).

Practical filters also depend on crystal Q and series resistance
among other factors for sucess.

My experience with many published recipie design filters is if they
are just built with any crystal they will likely work but bandwidth
and passband shape will be unpredictable. Usually the passband shape
is degraded the worst. So simply matching for frequency is only one step.

Allison
KB1GMX
--- In BITX20@..., "eternalesquire" <eternalsquire@c...>
wrote:

Hi,

I never saw anything about the original article about this. I've used
the BFO and a frequency counter to measure all my crystals. How
closely do they need to be matched for SSB reception? for CW reception?

Thanks,

The Eternal Squire

P.S. I am using 10.7 Mhz as the IF

Hi,

Have a look at messages 124 and 131.

I found a set of 4 crystals within 30 Hz and they are doing fine for
SSB. I have posted a plot of the resulting filter passband in the
files section. Look for BITX20_10MHz_crystalfiler_2.sxc.pdf

73 Heinz, OE5EEP

--- In BITX20@..., "eternalesquire" <eternalsquire@c...>
wrote:

Hi,

I never saw anything about the original article about this. I've used
the BFO and a frequency counter to measure all my crystals. How
closely do they need to be matched for SSB reception? for CW reception?
P.S. I am using 10.7 Mhz as the IF

Hi,

I never saw anything about the original article about this. I've used
the BFO and a frequency counter to measure all my crystals. How
closely do they need to be matched for SSB reception? for CW reception?

Thanks,

The Eternal Squire

P.S. I am using 10.7 Mhz as the IF

Hello,

This email message is a notification to let you know that
a file has been uploaded to the Files area of the BITX20
group.

File : /BITX 15 + PLL/MULTIBAND_BITX_B1.PDF
Uploaded by : eb7hpm <msraya@...>
Description : Multiband Design for BITX FI TX/RX

You can access this file at the URL:


To learn more about file sharing for your group, please visit:


Regards,

eb7hpm <msraya@...>

Hello,
I just dropped the schema how I measured the curves of the ladderfilter
in the Files box, folder: "Ladder Crystal Filter Design". One is in
GIF the other one is in SPL. The last one can be viewed and printed
with: SPLANViewer50.exe, also in the Files box, folder "PE2BS".

73
Ruud
PE2BS

Hi Chris,

Thanks for your comment. When the filter is part of the new BITX80/20
I will have to match the filter in the trcvr again. With you I wonder
how it will be, special because the filter is used in two directions.
I will drop the pictures by that time in the group.

73

Ruud.
PE2BS

--- In BITX20@..., "vdberghak" <vdberghak@z...> wrote:

Very nice shape, nice experiment too.
Like to mentioned, matching is important. I wonder how the shape

looks

when it is part of the BITX but I am sure you will add it.
Chris.

--- In BITX20@..., "Ruud Jongeling" <pe2bs@t...> wrote:

Hi fellow BITX-ers,

About one or two months ago I promised to report of my ladder

filter

experiment on 9MHz. I just dropped the files with shape-pictures

in

the

Files box. See the folder "Ladder Crystal Filter Design".
There is also an Excel file that can be used for your own

experiments.

To give you a short summary: it can be done but without measuring

the

shape it is a bit of lotery, like Max said in earlier messages.

73

Ruud
PE2BS