2006-01-06

70 MHz Transverter Components

Revision 1 - Kits sold in 2005

Resistors

Resistor Nominal value Delivered Colour code
R1, R22 27 kΩ 27,4 kΩ Red Violet Yellow Red Brown
R2, R23 10 kΩ 10,0 kΩ Brown Black Black Red Brown
R3, R14, R18 180 Ω 182 Ω Brown Grey Red Black Brown
R4, R7, R25, R29 22 Ω 22,1 Ω Red Red Brown Gold Brown
R5, R19 47 Ω 47,5 Ω Yellow Violet Green Gold Brown Red
R6, R17 150 Ω 150 Ω Brown Green Black Black Brown (Red)
R8 8,2 kΩ 8,25 kΩ Grey Red Green Brown Brown Red
R9 2,2 kΩ 2,21 kΩ Red Red Brown Brown Brown
R10 1,2 Ω 1,21 Ω Brown Red Brown Silver Brown Brown
R11 5,6 kΩ 5,62 kΩ
5,49 kΩ
Green Blue Red Brown Brown Red
Green Yellow White Brown Brown Red
R12 15 kΩ 15,0 kΩ Brown Green Black Red Brown
R13, R26 1 kΩ 1,00 kΩ Brown Black Black Brown Brown
R15, R16, R24 39 Ω 39,2 Ω Orange White Red Gold Brown Red
R20 22 kΩ 22,1 kΩ Red Red Brown Red Brown Red
R21 3,3 kΩ 3,32 kΩ Orange Orange Red Brown Brown
R27 4,7 kΩ 4,75 kΩ Yellow Violet Green Brown Brown Red
R28 12 Ω 12,1 Ω Brown Red Brown Gold Brown Red
R30, R31 82 Ω 82,5 Ω Grey Red Green Gold Brown
VR 250 Ω 220 Ω  

All resistors are 0,25 W/0,4 W and metal film type.

Capacitors

Capasitor Nominal value Marking
C1, C7, C14, C24, C26, C32 120 pF n12
C2, C8, C28 27 pF 27
C3, C4, C30, C31 22 pF 22
C5, C6, C27, C29 1 pF 1
C9 270 pF 271
C10, C11, C25 33 pF 33
C12 3,3 pF 3.3
C13, C34, C35 39 pF 39J 100V
C15 100 nF polyester, 2-modules µ1K63, yellow
C16 1 µF, 100 V, electrolytic, axial 1µ0/100V
C17, C18 22 µF, 25 V, electrolytic, axial 22µ/25V
C19, C21 82 pF 82
C20 150 pF n15
C22 10 pF 10
C23 47 pF 47
C33 1 nF 1n0, small yellow
C36, C37 12 pF 12p, grey-green black top
C38-C58 10 nF 10n, yellow green top/brown 103

All capacitors are 1 module, 2,54 mm, ceramic types unless otherwise stated.

Semiconductors

Component Possible component Delivered
D1, D2 5,6 V zener diode, 0,5 W Blue
D3, D4 1N4004 or equivalent 1N4006, black
D5, D6, D7 1N4148 or equivalent 1N4148, red
IC1 78L08 78L08
Q1, Q8 BF988, BF981, BF900, BF960 BF988, text facing down, D long terminal
Q2, Q7 J310, U310 J310
Q3, Q4 BD442, BD434, BD438, BD136, BD140 BD442, Q3 metal facing west, Q4 metal facing east
Q5 BC337 BC337
Q6 BFW93, BFW92 BFW93, text facing up, C long terminal
Q9 BFR96 BFR96, text facing up, C long terminal

Inductors

Inductor Type Comments
DR1, DR2, DR6, DR8 10 µH Neosid, blue plastic cap, 10
DR2, DR6 18 µH Neosid, blue plastic cap, 18
DR3, DR4, DR5 Ferrite bead Run a piece of wire through the ferrite beads, big beads
DR7 15 µH RF-choke Axial, Brown Green Black Silver
DR9 (Drain-choke) Ferrite bead To be mounted over Q1 drain terminal, small bead
L1, L2, L3, L4, L12, L13, L14, L15 Neosid 00 5231 03 Grey metal housing
L5, L16 Neosid 00 5061 00 Marked 5061
L6, L7, L8 Neosid 00 5048 00 The top is marked with yellow and grey dots
L9, L10, L11 Neosid 00 5049 00 Marked 5049

L1-L16: In case the inductor has a fifth terminal, one that is not located in a corner, cut it away.

Miscellaneous

MX1, MX2 SBL-1, IE500, IE800, HPF-505
Crystal 42 MHz, third overtone, HC49/U
Metal sheet box 148 x 74 x 30 mm3
BNC-connectors Four BNC-connectors. Remove the four terminals on each connectors.
Feed through capacitors 1 nF, solderable into a 3,2 mm hole
Relay Omron G5V-2
10 cm coax Used in case of separate 28 MHz RX

Documentation

Changes

 

Improvements

42 MHz LO

In some transverters the LO signal at 42 MHz can be found on the output connector. This residual can be reduced by grounding the crystal and solder a screen from L11 to L16 and ground the screen also between the two inductors. This fix has been seen to reduce the 42 MHz signal by more than 15 dB.

I have tested the same fix on another transverter and it had the opposite effect increasing the LO from -60 dBc to -50 dBc. Here are some successful LO screen ideas mounted on component side: Paul, GW8IZR, Tim, OZ4GA.

According to Paul, GW8IZR, a screen on the track side has even greater effect than on the top side. See an example here.
Paul, GW8IZR, did some more research on the spurious and what seems to be the solution is a screen on the trackside covering the oscillator circuit. First he installed two ground rails, then added the screen plate and finally did this measurement showing no LO or associated harmonics, i.e. all below -65 dBc and with out the topside screen. However, the topscreen does not harm so if already installed leave it.

TX instability

Some transverters show a tendency to oscillate on the output. A way to cure this is by reducing the gain in Q8 by removing C52 results. Doing so reduces the TX gain about 6 dB. Paul, GW8IZR, has also tried to vary the supply voltage between 10 and 18 V and there is no sign of oscillation when the TX BPF has been sweeped.

Uffe, PA5DD, writes:
"By searching the Internet I found a reference to an old article in the German magazine Funktelegramm. The article was written by Mike DL1GNM and should contain some changes to improve the stability of the original OE9PMJ transverter. I E-mailed Mike, and the change was to place a 10 nF capacitor parallel to C33 (1 nF) - the emitter de-coupling of the final transistor Q9 (BFR96).
To my surprise, this changed the tuning of Q9 quite a bit - an indication, that the 10nF capacitor had quite an effect. At the same time I have no more oscillations, even in the original configuration  and "max smoke" tuning method. I did solder my extra 1 kΩ resistor at G1 of Q8 back in to decrease the gain a bit.
I reckon the 1 nF capacitor was chosen with regard to BFR96 being a microwave transistor, but it seems that it does not provide proper decoupling at the working frequency due to the low impedance, at the emitter.
My conclusion is to add a 10 nF SMD on the backside of C33 if you experience instability in the transmitter."


Bo, OZ2M, www.rudius.net/oz2m