by  IK0MOZ

Starting date  Nov. 04 2001
Updated on : Nov 20 2001 : Added photos and drawings
Updated on Dec. 5 2001 : Added drawings and photos
Added second page on Feb.25 2002

Warning :  Due to large quantity of photos and drawings the speed of the page is quite low.
I beg your pardon, but to keep an acceptable images quality no other possibility.

What is the  PARASET ?

The PARASET is a small British  "Spy set" transceiver  supplied during the WW2 to the Resistence forces mainly in France, Belgium and Netherland.
It uses 3 metal tubes  :
1) 6SK7 -  Detector
2) 6SK7 - AF Amplifier
3) 6V6 - Cristall Oscillator/P.Amplifier
The range covered from the receiver is  3.2 - 8 MHz on AM and CW modes.
The transmitter is cristall controlled and works on the same receiver range.
The power output is around 5-7 W on CW mode only.
The set were supplied by a 6 Vdc vibrator type PSU.

For Its semplicity the set can be easily " replicated" and many replica are already in use.

Usually this kind of work starts using the ON5LJ series of drawings, in progress some differences are due to different availability of the necessary parts.

We will do the same, and will follow step by step the construction of 3 sets. In these steps will be possible find some suggestion like to solve possible problems.

I hope that this idea will be of interest and if you decide to start with this adventure let us know, your experiences will be added to these pages.

We are thinking also to open a PARASET FAQ section
and mainly to activate a frequency entirely devoted to PARASET and Paraset replica tests.

To end  we have to thank  Jo Scholtes - ON9CFJ that supplied to us a lot of documentations,  photos and details of his original Paraset.

Jo - ON9CFJ in full action


June  2001

The story starts during the ADMM when in a QSO  Jo (ON9CFJ) asked to me if for the competition were valid also a "Replica" set assembled with surplus parts.
After my negative answer being the argument exciting I asked for details because I were personally interested.
Jo clarified to me that the replica  were a Paraset and by return he will send to me some drawings showing all details for the set construction.

The ON9CFJ  complete original Paraset, on the left the Vibrator PSU

After some days arrives by mail an ON9CFJ envelop containing
1) Series of Paraset drawings (originator ON5LJ) :

                                        1a)  Panel drilling layout
                                        1b)  Panel Silkscreen
                                        1c) Case drawing
                                         1e) TX section layout
                                         1f) RX section layout
                                         1g) TX/RX common parts layout
2) Detailed drawings for key, coils, impedances. (Originator ON5LJ)

3) The same Paraset from component side (Originator ON9CFJ)


4)  F5XM Paraset Replica  (Originator F5XM)


With the above mentioned material the construction is possible, the missing schematic we found on
 Dutch  SRS   "Surplus Radio Bulletin"  n°11/98.

In our saturday radio net I explain the idea and 3 OM (I0BR, IK5FUZ, I7KVG) are interested to the project and I send immediately to them all the drawings.

After some days I7KVG asks to me some more details about the case, I ask by e-mail to ON9CFJ and after only a couple of  days Jo send to me a very detailed drawings taken directly from his original set showing :

7)  Botton case (Originator ON9CFJ)
 8)  Top case (Originator ON9CFJ)

JULY 2001

Around the end of July I made a check about the project status  :

IK0MOZ : Panel made in 0.7 mm zinc plated iron sheet. The holes on this panel are more  then I see on the drawing. The 0.7 thikness of the panel is too low it is better choose 1 mm. I start to collect the parts, this is the situation :  Variable capacitors, crystal holder, antenna/ground connector, RX coil and knobs recovery from a BC610 Tuning Unit. Terminals board  and lamp holders from RS catalog (433-755) and (    ). Problems  RF and AF Impedance, problem also on tuning indicator and slow motion system.

I0BR : Panel made in 1.5 mm aluminium. Variable capacitors, crystal and antenna connectors from the BC610 TU.  Knobs from BC312. Key completed (very fine). 36 H impedance done using a AF output transformer adding  0.1 mm wire to maximum possible. HF choke he will use 2 Geloso parts. Components  : many of surplus origin and other using modern components but put into surplus case. Tuning disk original, slow motion done using a potentiometer spindle.

br componenti pronti

Panel and collected parts from I0BR

IK5FUZ : Case completed including hinge and spring tube holders. He is going to start with the front panel (Aluminium). Tuning disk and slow motion original. He is collecting the electrical components.
Some questions:
1) Tubes socket type ?
2) Panel and case color  ?
3) Case has some handle ?
4) Some inside photo
5) Key drawing not clear

I7KVG : No info

Before summer vacatios I sent to Jo - ON9CFJ the FUZ and other questions


e-mail from ON9CFJ :

1) Tube sockets are in 1 mm  Pertinax
2) Panel and case color is silver-grey
3) No handle
4) Will follow by mail new inside photos.

I0BR : Noted some errors on RX section layout.
The power for 1.5 k and 10 k resistors on supply line are not correct.
Home made the coil support for TX gluing several carton sheets on a 35 mm core. After the tube has been fixed and hardned with transformer compound. Final result very similar to the original.
He started with electrical assembling. Wires cotton covered. Noted small space available.

IK5FUZ : Panel completed. He is proceding with the electrical assembling. Tube sockets used in pertinax.  TX coil different diameter but with the same inductance then original. Lamps used 3.5 V 200 mA because the original are not available Problem on the key and asks for a better drawing and if the case have some lock system

IK0MOZ : Front panel paintend and silkscreened. The lettering has been made using 3.7 mm Letraset


Mechanical assembling done. Found all parts (Modern and Surplus) less the RF chokes and tubes. The AF impedance made using a  3.5 VA 220/6 Vca transformer. The inductance is 18 H instead the 36 H required. I find also an old TV audio output transformer with 33 H inductance and R 900 Ohm, but being the fixing holes in different position, I will start with the first solution. if necessary I will change the item in a second step.


On middle of september arrives a new envelope from  ON9CFJ including :

1) A detailed drawing of the key (Originator ON9CFJ)

2) Additional  photos of his Paraset .



IK0MOZ :  Completely assembled the TX section. Found new error on the TX layout (on the supply line).  HF chokes problem momentary solved using RS coils cat. N° .........
Found tubes  from surplus dealer ESCO -  6V6  is glass type.

TX section completed..  In the center the RX coil is missing.

TX section different view. Is possible see the  "Made I0BR" key, and the 0.1 MF 750 V Sprague capacitor.


ON9CFJ via E-mail confirms the errors in the schematics and confirm that in the case there are not locking system.

I0BR set completed less case. Fixed a sked for on air tests

br set

Front panel of the I0BR set

br interno

the same from the inside

The Paraset has been built using all surplus parts. The final result is very good.
Pout on 80 m. around  7 W, POut on 40 m around 5 W. TX tuning very easy.  Receiver quite broad reaction control very critical. Roberto shortened the variable capacitors because the max capacity were around 120 pF instead of the required 100 pF. The adjustments has been easily done taking away the last 4 blades of the capacitors.
For the tests has been used some FT243 series crystals.  It has been also tested some modern TV crystals with the max Power of  1 W. The possibility to use this kind of very cheap and common crystals has been rejected.
On late afternoon we make the on air test. Distance betwen our QTH around 60 Km.
incontro br/moz

I0BR/Roberto (on the left) and IK0MOZ/Mario. Subject : Paraset to do list

I  use my SEG100 and send a long serie of V to allow  the receiver tuning. When I pass on receive mode I can ear very clearly the I0BR sigs. RST around 579 absolutely no chirp. We made a short QSO without problem.

IK5FUZ : Also Alberto finished the set it is  missing  only the lettering on the panel. He made al the test and noted the same Roberto results. Confirm the criticity of the receiving section, but he found very stable. He send to me the first photosask to me .

set di fuz

IK5FUZ / Alberto complete set

fuz interno

and seen from the inside

IK0QDQ/Edoardo  ask to me all the Paraset documentation, after analisys he will decide if start  with the project.

IK0MOZ : Set completed. Still problem on slow motion drive.  Built a PSU using 2 tranformers : 1) 220/6 Vca for filaments. 2) 220/220 Vca for HT. Rectifing (4x1N4007) and filtering (130 MF 450 V) the secondary 220 V the DC is around 300 V.  In this condition the TX  Pout is around 4 W. Also the receiver is OK and noted the same tuning difficulty already mentioned from I0BR and IK5FUZ.  The reaction controll is quite difficult and need practice.

IK0MOZ complete Paraset , but not yet case

Built the case using aluminium. Sides fixed with screws.  Missing  the lid

I0BR : Buil the case including tube holders and tuning table.  First PARASET Replica made in Italy.

br cover

I0BR set - The cover with tubes and tuning table

br completo

I0BR Paraset in working order

IK5FUZ : Lettering on the panel done. PARASET replica n° 2 finished.
He will send as soon as possible his crystals availability to establish a radio sked. Evaluated the possibility of a split frequency QSO.

The IK5FUZ paraset in working order

IK0MOZ / I0BR : On air test for IK0MOZ set.
Time = 17:30 UTC ; QRG = 3600 KHz
I will use the  Paraset
Roberto will use his VRQ102 (P out around 100 W)
I will send a series of V letter Roberto will answer to me in voice (USB). In that QRG the Pout in antenna  of my TX is around  3 W.
Results : I0BR lissen my signal but he is very QRMed from a strong RTTY signal.
I receive 100 % the answer (around 56) and any difficulty to center his USB emission
We can say Positive the results of the test. I have to find a crystal working on a lower QRG after we will try for a  2xPARASET tests.

IK0MOZ :  Built the lid complete with the tuning table and spring tube holder. Botton case riveted and painted. Set N. 3 complete.

IK0MOZ Paraset - Replica n.3 combat ready


The 3 sets are completed and working.  We decide to make a meeting to match the experiences, see from the live the different sets, test the sets. The meeting will be held on IK5FUZ's  QRA. Comparing the sets they are very similar also because the "Paraset project" has been the first subject on our regular radio sked , and then several solutions has been found during these QSO.  We never seen an original Paraset but comparing with the photos in our hands the sets more similar to the original seems to be the I0BR Paraset.  Infact Roberto make particular care in the contruction of some parts like the key, the coils, and the LF impedance, tring to replicate  maximum as possible the real set.
Common comments is the good reliability of the design infact the 3 sets also if made using surplus and home made parts worked immediately without any problem. All the components has been tested before the assembling.

The Paraset production line.....

From the left : I0BR/Roberto, IK0MOZ/Mario, IK5FUZ/Alberto
--.-   .-.  -..-     --.- .-. -..-


To be continued.......   Go to next page....


                                               Updated on Feb/25/2002

Vintage Military Radio SiteRing (WS#19 Shown) The Vintage Military Radio

This site owned by
38303 Mario - IK0MOZ
Previous Site List Sites Random Site Join Ring Next Site
SiteRing by