Secret Radio from World War II

The maker of this movie, Richard, PE0RIG, found this tin canned radio on a fleemarket in Eelde.The previous owner informed that he had cleaned the attic and the tin can was used by his father but he did not know for what purpose. It appeared to be a nice designed secret radio (radio receivers were not allowed in Holland during World War II). Apart from the Selenium rectifier which has been replaced by Richard, the radio was in good shape. It operates with a ECH21 tube (heptode and triode in one tube). A small transformer takes care of the 6.3 Vac filament voltage and provides 75Vac which is used (after being rectified) for the hi anode voltage. The carefully designed spiderweb coils control the amount of feedback to operate the reflex circuit. The audio level is loud enough to fill a small room.There is a multiposition switch that selects the fixed frequency broadcast stations. It appears that more of this type of radios were produced in the South of The Netherlands in 1944 and 1945 by Philips employees. . .

Some sporadic E on 50 MHz

Today I ran the WSPR software during the afternoon. This time I choose the 6 m band
to do the propagation analysis. 
Adjusted the FT450 to 50.293.. MHz and set the TX interval to 20%. In the end 
only 2 Italian stations were found in the database; IW1PAK and IK1WVQ. Their 
QTHs are close to each other in the north of Italy, near the border with France (Monaco), 
QRB appr. 930 km. Although about 20 stations were active on 6m WSPR 
during the afternoon those stations were the only ones I could copy.

                                              Propagation path on 50.2 MHz this afternoon

X1M QRP Transceiver kit

On various weblogs we can read that the QRP transceiver X1M, a 5-bander, has been introduced. Recently the rig came available in the Wimo webshop. De transceiver is delivered as a DIY kit.

The kit must be assembled. All components have already been solderded onto the PCBs. It is just a matter of sliding the PCBs into the housing and attach some connectors.

The transceiver is capable of working on 80, 40, 20, 15 and 10 meter bands. As well as in CW as in SSB. Max output RF power is 5 Watts (QRP limit). As an option one can add a CAT-interface for 18 euroos in order to control the rig from your PC. Dimensions are 97 x 40 x 155 mm. Total weight is 650 grams. Price is just below 300 euroos.

http://wimo.de/cgi-bin/verteiler.pl?url=x1m_qrp_transceiver_e.html

http://www.rys.nl/nl/producten/16-amateurzendontvanger/36-kortegolf/38-mobiel/1200-x1mtransceiver-kit-qrp-5-band.html


.

2N3904

Until some years ago for regular general circuits I used European NPN transistors like BC547, BC548 and BC549. I found that in circuits from the US or UK often the 2N3904 was used for general low-power purposes. Since I started to use the 2N3904 I have good experience with this transistor. The transition frequency is a surprising 300 MHz which makes this component suitable for RF experiments. Since I found out the transistor is low priced, RF capable and reliable I keep a little stock of those generalists. I ran out of stock lately and ordered qty 100 2N3904 for 1.80 euro ! (including free shipping) on eBay.

Qty 100   2N3904's arrived this week from Hong Kong. Price 1.8 eurocent each transistor

The 2N3904 is a common NPN bipolar junction transistor used for general purpose low-power amplifying or switching applications. The type was registered by Motorola Semiconductor in the mid-sixties, together with the complementary PNP type 2N3906, and represented a significant performance/cost improvement, with the plastic TO-92 case replacing metal cans. It is designed for low current and power, medium voltage, and can operate at moderately high speeds. This transistor is low cost, widely available and sufficiently robust to be of use by experimenters. When looking at the flat side with the base pointed downward, the three wires emerging from the base are, left to right, the emitter, base and collector leads.
It is a 200 mA, 40 volt, 625 milliwatt transistor with a transition frequency of 300 MHz, with a minimum beta or current gain of 100 at a collector current of 10 mA. It is used in a variety of analog amplification and switching applications. The 2N3904 is optimized for currents of around 10 mA thus minimizing thermal heating.

                                                                       A 2N3904

Electrically similar devices are available in a variety of small through-hole and surface mount packages including TO-92, SOT-23, and SOT-223, with package-dependent thermal ratings from 625 milliwatts to 1 Watt.

A 2N3906 is a complementary (PNP) transistor for the 2N3904. The 2N2222 is an NPN transistor that can safely switch three times as much current as the 2N3904 but has otherwise similar characteristics. 
Nevertheless, in many applications such as variable frequency oscillators where lower currents are used to minimize thermal heating and consequent thermal drift of the fundamental frequency, the greater current capacity of the 2N2222 gives it no advantage. Whereas the 2N2222 is optimized to reach its highest gain at currents of around 150 mA, the 2N3904 is optimized for currents of around 10 mA.

The 2N3904 is used very frequently in hobby electronics circuits including home-made ham radios, code practice oscillators and as an interfacing device for micro-controllers.

Mosfet headphone amplifier

IRF Mosfets are known for its use in HF amplifiers up to about 20 MHz. These Mosfets can also be used for LF applications. Greg Szekeres has designed a straight forward headphone amplifier that operates in class-A mode. Therefore, good audio quality is garanteed.

Using only 12 components you can build your own 1-channel headphone amp. You can use it to connect  it to the audio output of your CD- or MP3- player. Also it can be used as a final stage in your receiver designs.
An N-channel Mosfet IRF513 is used. Similar types as IRF510 or IRF520 can be used instead. Two diodes are present as means for protection. In principle this circuit has been designed to be used with 30 ohms headphones. Greg informs that the quality of C2 is important for the overall quality of this amplifier. Since this amplifier operates in class-A about 2.5 Watts electrical power is dissipated even when no audio is presented to the input of this amplifier.

In case the audio output appears to be too low, a 1-transistor input pre-amp stage should
be added.


http://headwize.com/?page_id=31

More DIY headphone amplifier designs:

http://www.headwize.com/

                          My version of the 1-channel headphone amplifier on VERO board

1000 pageviews

Yesterday, I saw the blog counter indicated 1000 pageviews since the the launch of this blog July 25, 2013.

Thanks visitors for giving me a good headstart. Your comments are welcome.

73 Ron

 

.. --

70 MHz Whispering

Yesterday evening it was too hot (25 degrees C at half past 9) to stay inside or even worse, sit behind the rig in the attic. A perfect moment for unmanned radio beacon operation while sitting in the garden. The homemade 70 MHz transverter and FT450 rig were switched on and the WSPR program was enabled on the PC. Only two Dutch 4m stations were active on 70 MHz yesterday evening (PA0TBR at 61 km and NL8992 at 138 km).

 

In spite of this low amount of WSPR participants on 4m it was good to see that both stations picked up my 10 Watt WSPR signal at 70.0925... MHz transmitted via an indoor inverted-V antenna in the attic. Later in the evening I found a nice eQSL card from Anthonie NL8992 from Ommen (Drente) - showing his farm in the snow - in my eQSL inbox.

 

.

70 MHz WSPR activity Aug.1/2013 in the evening

 

Confirmation reception MW broadcast station VAHON Radio

Triggered by an article in the latest Electron, I tuned in to VAHON AM Radio last Thursday. I used my 30 years old Kenwood tuner/receiver on the base floor to listen to this local medium wave AM broadcast station which broadcasts in Hindustan language on 1557 kHz.
In the morning the reception at my QTH was better than in the evening. VAHON transmits with a 1 kW AM transmitter. The antenna is a T-shaped type at appr. 27 meters height about 15 km’s from my house near the village of Stompwijk. Because of the small distance to the transmitter it must have been the ground wave I received.
Just for fun I sent in a SWL report to the QSL manager of VAHON-AM and was surprised to found a reply with an electronic QSL in return:

http://www.vahonfm.nl/site/index.php

                                              Electronic QSL from VAHON

80 mtr DC receiver

About 12 years ago I built a nice DC receiver of which I found the description a couple of weeks ago. Good enough for a blog entry and... the receiver still works fine.

The DC receiver has the advantage that, with an absolute minimum of components, one
can build a reveiver with a fine performance. In this case, the abbreviation DC does not mean " direct current" but Direct Conversion, refering to the principle this machine is based upon. In fact, this type of receiver is nothing else but a " super" having an Intermediate Frequency (IF) in the audible part of the frequency spectrum.

In general one can say: F_MF = F_ANT - F_OSC

For example: 2 KHz = 3580 KHz - 3578 KHz.

So, the difference between the antenna signal frequency and the oscillator signal frequency (both signals offered to a mixer) is a signal which can directly (without any additional detection circuitry) be heard. Because a numerous amount of components can be ignored (no dectection circuitry required), this type of receiver invites for homebrewing.

A DC receiver can be recognized by its characteristic sound, being produced when the oscillator is set 2-3 kHz above the missing carrier frequency of an SSB signal. Reception of AM signals is possible. However, a microscopic mistuning (from the carrier frequency of the received signal) immediately results in a loud tone !

Another advantage of the DC receiver is the fact that, besides receiving CW (Continuous Wave; morse - keyed carrier wave) and AM (Amplitude Modulated) signals also SSB (Single Side Band) signals becomes possible. Receiving SSB signals using the conventional method with a superheterodyne receiver (' super' ) is possible; however additional circuitry is required: a BFO (Beat Frequency Oscillator) and a product detector. DC receivers do not need those add-ons to convert SSB signals into normal speech. A sensitivity of approximately 0.3 m V (without any additional high frequency pre-amplification) with commonly used regular mixer devices is feasible.

With every advantage there comes a disadvantage. This type of receiver can be blamed of the fact that it has a limited dynamic range. HF signals with more than average amplitudes can easily generate unwanted mixing products (as produced by the mixer itself). Processing of the high fequency signal, just after the antenne stage, is a requirement. A signal divider (potentiometer) with a bandpass filter can do miracles. When a filter is added, the intruding strong broadcast signals ( ' short and medium wave') can be taken away. Now nothing can stop a fine reception of the proper shortwave signals.

Also, the sensitivity to pick up 50 and 100 Hz rumble is greater than compared to that of the ' super'. When carefully asembled (sufficient spacing between power supply transformer and receiver and proper shielding) this disadvantage can be neglected. In one case I could not cope with the 50/100 Hz rumble and noise. I have replaced the bridge rectifier by 4 seperate diodes; over each diode I soldered a 10 nF capacitor. The result was asthonishing. To obtain the best result one have to do some experiments now and then.

The design as shown by photo 1 is suitable of receiving CW-, AM- and SSB signals with frequencies within the 80 meter HAM radio band. Frequency range of this receiver exceeds 1300 KHz (3680-3810 KHz).

If you like to know more about DC receivers and design philosophies I can recoomend the book of Joseph J. Carr, ' The secrets of RF-technology. Beautiful book with very useful hints and tips.

Description and details

The receiver is designed around an NE602. This ic contains amongst others a double balanced mixer, an oscillator and a voltage regulator. The mixer is designed to handle signals up to a frequency of approximately 500 MHz (!) and the oscillator is capable of generating signals with frequencies till about 200 MHz. The low frequencies (appr. 3500 KHz) used here are no big deal for the NE602 whatsoever. The dynamic range of the NE602 is something which can be improved. A later version of this ic, the NE602AN, has improved dynamic range characteristics. Also available is the NE612. This ic, being pin-compatible with the NE602, has, like the NE602AN (which is difficult to obtain), an enlarged dynamic range.

 

Circuit 1 80 m DC receiver

The suppresion of unwanted harmonics by the mixer device inside the NE602 (Gilbert cell) takes care of the presence of the sum- and difference frequencies of the antenna and oscillator signals only on the balanced output of the ic (pin 4 and 5 ). In this case, only the difference signal is a useable audio signal and will be used for further LF amplification.

The input filter stage resonates at a frequency (exactly adjustable by the ferrite cores of the TOKO coils) of 3.7 MHz. The more the frequency of the antenna signal diverts from the frequency on which the filter resonates, the more the input signal is suppressed. In this way strong broadcast signals will sufficiently be blocked. The potentiometer in the input circuitry brings the amplitude of too large input signals to proper level, making sure the mixer device inside the NE602 is able to handle them. To enable the reception of weak stations, a one stage (selectable) RF amplifier forms part of the design. Thus an additional + 6 dB gain is obtained. The manufacturer of the NE602 (Signetics) advises to, if the power supply voltage is +9V, use a 1000 ohm resistor in the power supply feed line. As was done in this design.

Both varicaps BA125 enable the internal VFO to exactly be tuned by an adjustable voltage. For this design it was choosen to obtain this control voltage by a multiple turn potentiometer (Bourns). Precise tuning, an absolute must to listen to SSB transmissions, has been realized in this way. For an acceptable ' low-budget' solution, one might use two (regular) 1 turn potentiometers (e.g. 10 k + 470 ohm in series). Tuning becomes less comfortable, yet remains possible. A small trimmer capacitor is used to make the coarse frequency adjustment. Fine tuning is done with the multiple turn potentiometer. For LF amplification the LM386 has been selected. Depending on the type, the LM386 can deliver an LF output power of 250…750 mW. (The LM386N-1 makes about 325 mW where the LM386-4 achieves a huge 750 mW).

Following add-on circuitries have been mounted into the DC receiver housing:

Power supply unit, signal strength meter, frequency indicator

Circuit 2 shows the S meter suitable for looking at the relative signal strength changes. The output of this circuit can be connected directly behind the 100 m F electrolytic capacitor ( LF output). The circuit was built up on a seperate VERO-board.
-The complete receiver has been built onto a VERO-board. The VFO frequency determining coil uses 30 turns (diameter of 0.35 mm copper wire) on an Amidon ring core T50-2. Wen using a different core, the number of turns should be determined by trial-and-error (frequency counter recommended!).
-TOKO KANK3333R coils are used in the input circuitry. This HF transformer is designed for the 3-4 MHz frequency range. Lots of alternatives can be used (e.g. KANK3334).
-A disadvantage of the circuit here described is the small amount of LF output power when receiving weak signals. I did some experiments using two LM386's in series to increase the output level. The results were very dissappointing. Because of the very high amplification factor a low-frequency oscillation can not be avoided. The amplification of the LM386 is adjustable; if a value of 10m F for the elco between contacts 1 and 8 is used, the amplification is 46 dB (200 x). When the elco is deleted the amplification is only 25 dB (about 20 x).

 

Standing Wave Ratio

Recently heard on the 2 meter band: "My Standing Wave Ratio currently shows a 1 to 1.5 . "Now, an antennetuner is really required to improve my antenna network in order to obtain a better value!". In case the OM realizes his idea, his station will produce a weaker instead of stronger RF signal. In fact, the loss at s=1.5 is 0.18 dB. With the incorporation of an antennetuner or other antenna adaptation circuit an additional loss of 1 dB will be added. At least remarkable !

When explaining the theory related to the "Standing Wave Ratio", authors often use nice graphs e.g. showing the radiated and reflected power versus the SWR. Of course those relations do make sense, however the question "What does this mean for my transmitted signal at the receiving station?" is neglected in most cases. The graph shown in this article shows the relation between losses (indicated as a in dB) and the SWR (indicated as s). By the way, the graph is a presentation of the formula:   a = -10.log (4s/(s+1)2)

Now take a look at the graph: A power loss of 1 dB appears at a SWR of s=2,7 (!) A 1 dB difference is hardly recognizable in the HAM radio bands during regular radio traffic as we use it. The difference between signals at a perfect SWR (s=1) and a value between s=1 and s=2 cannot be recognized. Also between s=2 and s=3 the antenna system performs fine. A higher degree of awareness might be necessary when the SWR exceeds s=3. Now the amount of reflected power becomes serious. We do not want to operate a chronic overloaded RF Power Amplifier. Another observation: in order to lower the S-meter deflection at the receiving station by 1 S-point (which is equivalent to a power loss of 6 dB) we really have to force things into the wrong direction: Starting at s=14 (!!) we can see the S-meter at the receiving station moving to a 1-point-lower S-value. Very important when signals just above the noise level are observed of course, but when taking regular transmissions into consideration it hardly does make sense. In fact the accuracy offered by most SWR meters is far too good. A device with e.g. 3 LEDs should do fine; s smaller than 2 (green), smaller than 3 (yellow ) and greater than 3 (red) provides sufficient information. However, e.g. antenna builders do use the accuracy of one number after the dot.

 

 

Example of a simple, 1 LED, SWR indicator (according to the balanced-bridge principle)

More important than getting a better SWR, is improving the attenuation
of the antenna (coaxial) cables. In the case your SWR measuring device
shows a value of 1 to 2 (or better) and you still want to improve your
station’s performance, you better spend your money and time on improving
your antenna cabling and stay away from your aerial !

Source: (1) FUNK amateur November 2002, seite 1155 (2) Rothammels Antennenbuch, A. Krichke