DF0WD/DL4YHF's Longwave Station

last updated: December 10th, 2000

The main equipment for 136kHz is a linear(!) transverter with a maximum output of 100 watts (to compensate the loss in the matching system .... ..). The transverter can be seen on the photo below in front of the antenna coil. It converts the 136kHz-signal to 10.136MHz and vice versa. The small Icom IC706 (old model) on the left side is almost deaf on 136kHz, but it works fine on 10.136 as RX and (QRP-) TX. It does not even require a good IF filter because of the transverter's preselector (we have an extremely strong commercial signal on 138.8 kHz in Germany, which is S9+60dB on the longwire. That signal would totally block the Mixer (and the IC706) without its "sharp" LF-preselector.

The 136kHz signal is filtered by a very narrow 2-stage FET preselector using "Q-multiplication" (feedback to increase the Q factor). Its bandwidth is less than 200Hz. The preselector-frequency has to be adjusted by a 2-chamber variable capactor which can be found in old AM radios. This makes tuning a bit "tricky" because you have to rotate the preselector knob together with the VFO of the SW transceiver to hear anything at all. ( The schematics of this important part are also available as EAGLE-file now).

The transverter uses a 10.000Mhz-LO with a cheap crystal, a Schottky ring mixer, a driver, and a double push-pull final amplifier with four matched (!) IRF540 MOSFETs operated at 13 Volts DC . The final stage has an SWR-protection using an SWR-bridge which is also used to display the SWR (Fwd+Rev) on the front panel. This makes the PA virtually "undestructable" by load mismatch. The output power drops below 10 watts if the load is not purely resistive 50 Ohms. Because this is a linear transverter, the output power can be controlled by changing the output power of the HF-transceiver.
A critical part is the final's step-up transformer which must have an extremely low stray-inductance. The ferrite pot-core can bee seen in the center of the photo below. The small white toroid on the right is just a primary transformer that divides 1 watt from the driver stage for the four MOSFETs.

A PI-network is at the lower side of the final board. A part of the large Amidon T200-2 powdered-iron toroid can be seen. 50 turns of enameled wire give about 30 uH. Two Wima "FKP" capacitors (47nF each) together with the 30uH-Inductor are used in a PI-filter which also transforms 13.5 Ohms from the large pot-core-transformer into 50 Ohms for the output. The efficiency of the final stage is only about 70 percent as long as it remains a real "linear", but that doesn't hurt at the transverter's maximum output.

If there will be more time, I will convert some of the circuit diagrams from my ugly hand-drawn sketches into EAGLE drawing which you may find here in future.

Back to the Main Index

More LF information