Thursday, September 30, 2010

Discriminator out of MC3361


Now that the VHF FM receiver part is almost done, nothing like giving it a little "professional" touch...said that, on to make a "packet" dedicated output, also know as discriminator out...

Basically this output is a low pass filter from the recovered audio of the MC3361. The same signal goes to the speaker amplifier, so it's just a matter of filtering and preserving the signal in it's more pure form.
I add an amplifier because the sound card wasn't liking the very low output of the MC3361. The "filter" part is just the 5k6 resistor and the 4.7nF capacitor.


Before making the amplifier I tested for some time with the my PIC aprs decoder but not even one packet was decoded (it's working/well connected since I get the boot message and the blinking led...)

Then I proceeded to "multimon" software, now with the amplifier working, had limited success, about 20% of all packets were decoded (remember, I am using a peace of wire as antenna inside the shack). By this time the TM-D700 was decoding around 95% (I was beaten by a kitchen electronics manufacture).

Bellow, "multimom" decoding:


Also tested with "soundmodem" software but no success. I didn't proceeded with further testing since I installed and tested "multimon" and "soundmodem" today for the first time. Could be my mistake in the configurations.
Hope to get better results in the future with the outside antenna.
I also tweaked the IF frequency a little bit and the quad coil to try improving reception but not much success.

Incidentally I found a PIC aprs decoder project with an LCD that migh work a litle bit better since is based on the MX614 chip, unfortunately is very hard to find the MX614, so for now I will put on standby this project. Here's one of the links:

http://www.ke4nyv.com/peekpack.htm


Further reading:

* http://www.xastir.org/wiki/HowTo:SoundModem

not in documentation but port on "Xastir" should be configured at 4800 and probably you need to "chmod" a+rw /dev/pts/"whateveryours"

* http://www.baycom.org/~tom/ham/linux/multimon.html

Download link for "multimon" is at the bottom!

* http://www.discriminator.nl/

Sunday, September 26, 2010

Akihabara

Normally this blog is just about building or experimenting but I will open an exception that got me depressed.... a lot depressed...you know, when you want someting and realise that it will take for ever to get...or never!

"Akihabara" is from what I know an area in Tokyo devoted to the electronics market.
I've seen, in the past, some comments and some photos about it but never had seen a more inside look.

Please visit http://www.tokyohackerspace.org/akihabara/

and browse from the shops on the right of the video. If that doesn't get you depressed then you must live there. This is something like Friedrichshafen times 1000 and on steroids.

Isn't the world unfair?.... sometimes even a 1N5711 diode is difficult to find locally and this guys have all the components in a metro station!

I am still depressed..... cheapest fligh to Tokyo is allmost 3000 Eur :)

Friday, September 24, 2010

From junk to a power supply

I've been given an old VCR....I hope it was not working since it was immediately opened for parts.
One of the nice things is; this one had a linear power supply with a real transformer and not that interference causing switching thing's :)
Said that, time to remove the transformer. I didn't care for reverse engineering the power supply diagram, just cut'd the transformer away and the rectifier bridge dissipator wich had also a 7812. I didn't reused the filtering capacitors since I had a new one to put giving a little more confidance.
And here's my new bench power supply, half done. There are still holes for other voltage outputs. For now only 15 on the left side and 12V on the right.

You can see in the background the VHF RX/TX being built and far away the RF power meter

The internal mess:

Need to put a fuse outside, for now it is internal on top of the transformer (also reused).
If you need the schematic....go search the net :) There are a billion power supply schematic's around...

And here's the trusty old bench power supply!


Is this junk with style or what? :)

Wednesday, September 22, 2010

Georg Simon Ohm law

Now, after all this years, this is the first time I know Ohm's first name (never met him dough), some other authors/researchers/inventors I learned the full name at school but I don't recall any teacher referring to Georg Ohm... we are always learning.

Why a topic on Ohm's law, well, just because I needed some relax from more complicated maters (for instance an inverter gate oscillator that is refusing to start with a recycled 4.5Mhz ceramic filter. It starts with a transistor oscillator configuration!).

Now let's do a simple demonstration about Ohm's law:

Place a 9 Volt power supply feeding a 2.2K Ohm resistor in series with a red led (the led is just to brighten things a little bit)

Now measure the voltage across the resistor. In my case I measured the total voltage applyed

then removed the led voltage.


To get: 8.82 - 1.90 = 6.92 Volt across the resistor, measured directly should give the same result.

So, let's calculate the current passing trough the resistor (and the led because it is in series) by Ohm's law:

I = U / R

That's: current is equal to the voltage divided by the resistance.

Since the resistor is a 2200 Ohm one...

I = 6.92 / 2200
I = 0.00314 Ampere
I = 3.14 mili Ampere

let's then measure the current to confirm our calculation:


Now that's a 60 uA (micro Ampere) difference... maybe Mr. Ohm was wrong although 60 uA in 3140 uA is a very, very, very small difference.

But what would happen if my resistor was a 2150 Ohm, as I measured.

Then the current would be 6.92/2150 = 3.21mA that's rounded to 3.2mA and yes...we can confirm that Mr. Georg Simon Ohm was correct!

The very small difference is due to the precision on the multimeter. That's another subject just by it self...

Thank you Mr. Georg for all the fun!

Incidentally I came to the conclusion that I don't know any LED reference value....normally one just ask a 5mm red led...or green. Who cares!?

Tuesday, September 21, 2010

S-meter for the MC3361

Sometimes when you want something...it's better to do it yourself!

I searched and tested some s-meter circuits (in fact it were 2) for the MC3361 who doesn't has RSSI output but in one circuit I didn't had the transistor, so placed an equivalent without success, the other it didn't worked... so decided to design my own "signal meter" to connect the MC3361.
After some testing and rough calculations here's the double 455Khz amplifier and rectifier.


It is nor perfect or optimized and if you take the transistor parameter calculation, you should see the bias resistor are a big on the high side, but never mind, it works.
Other thing, I think the meter I used is a 250uA (it isn't marked) and I didn't care to verify but if you have to much displacement on yours , just place a serial resistor.

Keep the wire connecting to pin 5 as short as possible or you might start receiving some broadcast stations or hum.

The germanium diodes weren't marked and I forgot the reference (I have them for some time), maybe some OA90 or something 119 (can't recall if it isn't a 1n34). Just ask a glass diode one with a green stripe :). Any crystal radio detector diode should work and even 1n4148 might do the job, just have fun and test!

The circuit was made using a technic similar to island boards used extensively by PY2OHH, you can see some examples here: http://py2ohh.w2c.com.br/trx/bitx/bitx.htm


....ah...this is just an indicator s-meter, any similarity with a real calibrated one is purely fictional...

Sunday, September 19, 2010

Car radio PTO

Funny thing, this past week I read at some blog (can't recall which one) something about an old car radio tuner using an unusual PTO mechanism, yesterday I went to a friend's house and he wanted to get rid of some electronics junk (i.e: ham gold), end up with this car radio:

Can you see the PTO? Well, as soon as I have some time will let you know what it can be done, maybe an 50Mhz receiver or another 144...who knows.

From what I could see the schematic is similar to some old designs using extensively IF cans, there's a nice tutorial in the following link about those designs: http://www.vintage-radio.com/repair-restore-information/transistor_vhf-sets.html

I was given another radio but more modern design, anyhow one of the 10.7 Mhz crystal filter that came in it is already at the VHF receiver replacing the 10.7 Mhz simple crystal.

I am still having some trouble in the receiver, pressing the board and there's some interference and loss of signal, probably some solder joint that I didn't found yet. I am also in the process of adding an s-meter to the MC3361. Will post design as soon as it works....

Today while I was doing some tests, and with a piece of wire (70cm) inside the shack, I got some, of what looks like, meteo sat transmission! Incredible!
Here's the short video of NOAA 18 at 137.100 Mhz today:


Friday, September 17, 2010

Batery protection - An implemented idea


Already, somewhere in this blog, I mentioned my solar panel and the homebrew battery charger/regulator. This setup gives power to the 12v equipment in the shack while I don't finish a 12V power supply project I have, then it will be used for backup.
For now it has been running un-protected, that means I have a direct cable from the battery to the equipment without any interrupt point, that is not safe and I know it.
There are many ways of protecting a battery and the most used one is a fuse, why? Because is fast acting. There are also switch's breakers specially made for DC systems, those are expensive and the fuse option means you have to have some spare ones.... but... there's always a but, the AC protection switch breakers work the same way a DC one do, in fact I think the only difference is the isolation characteristics, that's my theory which was confirmed from a talk with a former school teacher. He didn't looked at me with surprise when I told him my idea... he just mentioned to get one with a fast acting curve ("shot" time versus circuit current).

So, on to buy a 230/25A protection switch (cost 3.45 Eur), which must be a fast action one but mine...is....I really don't know (didn't care to confirm), but I will let you know if I have any future problem. Anyhow for now this is just to have a break point.

The other terminal of the battery will probably get a car fuse if I can get in the scrapyard some car fuse holder (car fuses are relative inexpensive). The best of both worlds in each line.

I you look at the photo, the color code in the cable doesn't match the standard 12V system (Red and Black) but this cable was free (scrap from a datacenter I worked on) and really, you should know this, electrons are color blind!

Thursday, September 16, 2010

Reference voltage


Not enough digits to show 4.096!

What's that? You say!

It's a reference voltage IC working! Courtesy, in the interest of science, of Texas Instruments, that was kind enough to spare me some dollars and sent me for free one of "REF2940" IC. They have several versions with different output voltages but I chose this one (4.096V output) because it's a round number in binary that could be handy for calibrating or making comparations using an ADC (analog to digital converter) chip.
By having this reference voltage source I can now verify the acuracy, in voltage terms, of my equipments. With a precision and calibrated resistence I can then calibrate current meter equipment (it can be done using only a voltage reference),

By having calibrated instruments one can have more confidence in the results measured and that's specially true when most RF measuring equipment is expensive as hell and you have to build your own.
This is in no way an replacement for a serious calibration situation just an amateur solution....remember, professionals built the Titanic!

The assembly is easy, except that the IC itself it's the size of an smd resistor, nowadays every litle thing comes in smd... the IC pinout is as follows:

If your are accessing this page in say 1280*1024 resolution (in a 19" monitor) the IC is the size of the word "RE" in the above picture...whell... maybe a litle smaller :)

Here's the table of dropout voltage vs load current which is nice to print and put outside the equipment box (when it's ready, after this prototype stage).


The simple schematic follows... +5 V are provided by a little 7805 not in schema.

And the prototype (can you spot the IC? :)



It's in the intersection of a strait line from the end of the crocodile claw and an horizontal line from the body of the 7805 (the big black dot).

Have fun and "calibrate" your equipment.

Wednesday, September 15, 2010

Made a QSO

I was making (again) some changing's in the operating position at the shack. The "radio" table has two sides and a center place, the left side is for the not so used equipment, the center for the computer and the right side for the operational equipment. Now decided to change the layout; the left for the commercial equipment in the center stays the computer and on the right side will be all the homebrew stuff....being say that, I removed the FT-102 from the right side to the left side to get some space for the "Speaky" when it's finished, the FT-102 occupies now the space left by a Swan 270 (we have almost the same age :).
Anyhow, decided power on the Swan just to see if it was still working (never had made any qso since bought).
After some tuning I was listening to "9A500AA" station (500 Croatian islands) a special call of 9A2AA, somewhere around 14.210Mhz (I calibrated the dial via the built marker generator). The tube output tuning was only made in reception to a 17m band dipole). I know this works and it's safe for short qso's.


Above the temporary operation position...

Made one call... "9A500AA" answered another station call, made a second call and guess what... he returned my call!!! I just exchanged signal report and wished him good DX. It looked like my first qso since my last one according to the log was 29/6/2008 with "OE2008S", that makes 2 years and 3 months without hitting the PTT!
I was hopping making "this" qso with a homebrew radio.... maybe next one!

Tuesday, September 14, 2010

VHF FM transmiter... PLL LCD callsign

I make no absolute question in having my callsign displayed in the LCD of the PLL, first because it wasn't me who had the hard work making the software and credit must be given to the one (PY2EAJ) who did it and second just because all I want is it to work!
Anyhow Euclides (PY2EAJ) was kind enough to provide us (me and CT5JZX) the software sources, so, I ,had some time and exercised a little of my knowledge in assembly (I only programed so far the x86 family, and a long time ago). Since assembly languages are all similar, on to the business of replacing Euclides callsign for mine.
Initialy I translate Euclides callsign to Hex and then searched for the occurrence of it, which gave me the following code:


;; 197 : hd44780="P" hd44780="U" hd44780="2" hd44780="X"
movlw H'50'
call _11498__vector
movlw H'55'
call _11498__vector
movlw H'32'
call _11498__vector
movlw H'58'
call _11498__vector

;; 198 : hd44780="J" hd44780="E" hd44780=" " hd44780="R" hd44780="X"
movlw H'4A'
call _11498__vector
movlw H'45'

Euclides hex callsign is: 50 55 32 58 4A 45
Them I replaced for my own call in Hex (see the movlw instruction):

;; 197 : hd44780="C" hd44780="T" hd44780="2" hd44780="G"
movlw H'43'
call _11498__vector
movlw H'54'
call _11498__vector
movlw H'32'
call _11498__vector
movlw H'47'
call _11498__vector

;; 198 : hd44780="Q" hd44780="V" hd44780=" " hd44780="R" hd44780="X"
movlw H'51'
call _11498__vector
movlw H'56'
call _11498__vector

Well, that is it....for now... but we still need to generate new "hex" for the pic. That's easy in Linux just issue the following command:

gpasm lc72131-ct2gqv.asm

Where lc72131-ct2gqv.asm is the changed source file

Then you get for free the following files :)

lc72131-ct2gqv.cod
lc72131-ct2gqv.hex
lc72131-ct2gqv.lst

We only need the "hex" one. Now let's clean the PIC "internals":

picprog --pic /dev/ttyS1 --erase --burn

And burn the new code in it:

picprog --input-hexfile lc72131-ct2gqv.hex --pic /dev/ttyS1 --burn

No errors? Nice! I ended up with this on the display:




....We asked the wrong code to Euclides, should have been the transmit one and not the rx with 10.7 offset.....

p.s. Euclides was kind enough to send the tx code and also his new transceiver design based on the MC3362 and the LC72131 PLL. I've got mine almost finish but will send it to Pedro (CT5JZX) maybe he will test, he has a MC3362 for testing.


Friday, September 10, 2010

VHF FM transmiter... PLL PCB tests


After almost completing the PLL board assembly it was time for some testing.
I still don't have the push buttons connected and there's still missing the 4.5Mhz crystall for the LC72131 IC but anyhow decided to go with an 4Mhz just for seeing some progress. It's mandatory to use the 4.5Mhz (or 7.2) one as per LC72131 datasheet but for the PIC with exception for timers it's just a clock and a 4Mhz will do.



There were some connections missing int the PCB board due to abbreviation in the original artwork (power supply capacitors and LCD control pins), but no problem, components were connected in the cooper side.



It was also needed to put the LC72131 in the cooper side due the fact that this IC doesn't came in standard DIP22 but DIP22S, I guess the "S" is for short....



The display reads lock but no VFO connected so this is probably due the fact of the 4Mhz crystal is connected to the LC72131 instead of a 4.5Mhz one, or wrong connections between the IC and the PIC. Latter on when the VFO is connected I will investigate further. Without the PIC code it's impossible to know at the moment.

Original schematic artwork and software at: http://rf.w2c.com.br/PllLC72131.aspx
Very detailed buildup at: http://py2ohh.w2c.com.br/trx/2mpll/2mpll.htm
PCB design by: CT5JZX
The assembly mess by: CT2GQV (that's me)

p.s. maybe you prefer your own call sign in the LCD... then send an email to PU2XJE maybe he can arrange that for you.... or search for:

movlw 0x50
movlw 0x55
movlw 0x32
movlw 0x58
movlw 0x4a
movlw 0x45

....I just recalled, that last time I wrote similar instructions was in 1995 :)

Tuesday, September 07, 2010

FV-107 fixed crystal frequency

...Today it rained!

You may ask what that has to do with amateur radio...well first I don't recall last time I saw rain, probably in July, second it means the temperature will drop a little bit and it will be possible for me to do some soldering in the shack! That's the importance of rain...and of course I can stop watering my coriander plantation, the rain will do that for me :)



Continuing, as you probably know I have an FV-107, it's an external VFO for the FT-107 Yaesu transceiver but it's being used (connected) in my FT-102 with an adaptation circuit I made, the schematic and article available somewhere in this blog: (http://speakyssb.blogspot.com/2009/09/external-vfo-controller-for-ft-102.html)

The FV-107 VFO has provision for 6 fixed crystal operation. For crystal value selection you need to know your operation frequency, get a table provided by Yaesu (taking in account IF frequency scheme) make some calc's and then you have the VFO or crystal frequency, alternately you can put the radio in the frequency you want and measure the VFO output frequency, not very elegant solution....

Since I want to calc several frequencies and it's a repetitive boring task I made a small PHP online program that gives me instantly the needed crystal frequency and made it available for everyone to use, here's an usage example:
http://www.osprompt.com/ct2gqv/apps/fv-107.php?f=14200&m=usb
In the above example the frequency (f=) is 14200Khz and the mode (m=) is USB. The result VFO crystal will be 5248.5Khz. It can calculate for LSB/USB,AM and CW modes (use lsb our usb or amcw in the parameters), AM and CW have the same carrier frequency. 10m band isn't inserted in the programed frequency table just because it wasn't provided in the manual and I didn't use it. 17m band wasn't also in the manual but I calculated the IF values for the band.

Unfortunately it's difficult to find crystals in the range 5-5.5Mhz so for now my VFO will stay working as a true VFO :)

Saturday, September 04, 2010

Light dimmer

Now for something completely unusual in this blog, I present the _incandescent_ light bulb dimmer (just in case you try to use this on a fluorescent bulb...).

How it all started: My parents bough a table lamp made in in the other side of the globe. Off course many products made there, maybe because of sea transport, have many faults on arrival :P. Well in fact I think it only worked for 1 minute before exploding in smoke, so on to me for repair, typical! The initial circuit was of so low quality and lousy components that I decided to rebuild the bloody thing, including the power cable.
Some time later here's the finished product now much more dependable and expecting a long life...

and the schematic:



The tricac is a BT138 that I had since 1988 (probably last time I build something like this), the diac is a BR100 and the potentiometer is 100k in value. Observe capacitor voltage ratings if your going to duplicate this... at least 400v just in case...

Should work for controlling the soldering iron temperature but my iron problem is not excess temperature.. :) so will not build one for me. By the way, bough another 250g solder roll, I think it's the second one this year...

Friday, September 03, 2010

Updates on the VHF transceiver

Some progress in the VHF FM transceiver....well...it's more like a receiver and transmitter in the same box.

Only commented the transmitter part, the receiver I already did that in some older post.
Just after the following photo I remembered that I still needed the frequency control...

..already done 3 holes between tone selector and mic gain...

Another photo:

Thursday, September 02, 2010

Manhattan-style pad tool

After some unsuccessful tests trying to build one tool to make Manhattan-style pads I decided to buy one that I found at the hardware shop. Normally my pads are build square shape with the hack saw, not elegant and for sure not easy to make and also labor intensive, so, this is something welcomed to the tool box.
here are some preliminary results:



Although the toll is 10mm diameter the effective cooper pad (the one starting to drill on the cooper side) is 7.5mm in diameter. The tool is made for cutting holes in tiles but for sure fiber glass and thin cooper are no match for the "diamond" bits in the tool. Let's hope it last long since it's a bit on the expensive side, around 15 Eur.
I think I will exchange this one for the 8mm drill bit, for making even smaller pads.

Manufacteur is "Wolfcraft" from Germany (were else?!) and you can see specs here, in the manufacteur web site: http://www.wolfcraft.de/jcatalog_generated/en/products/product_groups/17618_product.html

Now here some progress on the VHF FM transmiter front panel:


Still need the holes for the TX led, MIC gain, Mic input and tone selector. Only the LCD and LCD contrast adjust drilled so far.