TECHNICAL

50 W power amplifier by Peter, SP2DMB
Published on 28 January 2014 by Bo OZ2M
Peter, SP2DMB, has made a new power amplifier with 50 W amplifier for 4 m. It uses a Mitsubishi mosfet - RD70HVF1. Power supply is 13.8 V.

FT-847 modifications by SP2DMB and SP3RNZ
Published on 11 October 2013 by Greg SP3RNZ
Part 1 - description of modules, diagrams, mods and some measurements
Part 2 - "howto" do it in practice.

Diamond D-130N on 4 m
Published on 20 May 2012 by Marc PA1O
Some time ago I have investigated the frequency adjustment of my Diamond discone antenna with the results in the below picture.
Frequency [MHz]3550707580859095100105110
Length [mm]127082038031025721518516312210082

Standard there is a top rod with dimensions for 50 MHz operation, but it can easily be tuned up or down across a wide range.

There is a lot of D-130N discones around, and I think the information will be helpful for others who want to use this popular antenna on 4 m!

Diamond D-130 datasheet.

Microwave Modules MML 144/100 mods for 4 m
Published on 6 February 2012 by Paul PA0OCD
Modification at your own risk (yes!) of the Microwave Modules MML144/100 Linear to 70 MHz.

First I tested this (second hand) PA on 144 MHz and on 13.8 V it produced max about 60 W! Perhaps the SRF 1397 is not (and never was) up to 100 Watts hi. This PA 144/100/1/2 is old and has no Preamp, so it can be that later produced PA's have a different PCB Layout....

I modified this 144 MHz PA to 70 MHz using from the “4 Meter Website” both the diagrams and remarks / description (Tnx!) given by SV1DH and GI0GDP. This modification has now become a kind of mixture....

The result is a a 70 MHz PA which has about 10 dB gain and is up to 50 W reasonable Linear and can produce about max 70 W and 10 A at 13.8 V (saturation). The idle current for the PA transistor on TX is about 250- 300 mA (see potentiometer A). Using PSK31 to measure the third order IMD I found IMD-3 at -30 dB (good) at about 15 W output on a normal (slow) Watt-meter and about -20 dB at 20 W output.

The SWR at the input is about 1 to 1 when carefully adjusting the 60 pF Trimmers at No. 2 and No. 3.

Use at the output the 3 coil/2 Doorknob Capacitor Low-pass Filter described by OZ2M or the 3 Coil/ 2 PCB Capacitor Low-pass filter of IK0VAQ. The Doorknob Filter I made here is up to 100 W easily. I saw other filter on the 4 Meter website  from G4CJZ and GI0GDP as well.

I made two photos of the modified PA with and without numbers at the places of the modifications:There is plenty of room in this PA!

By the way: Before you start take good pictures of the original PA, I did not hi. And while most of the radio-amateurs are human, like me, we can make mistakes so following this description of the modification hereunder is fully at your own risk.



A lot of the original parts can stay in place (hurrah!) , but I removed:
  • At 2: remove the 470 pF ceramic capacitor.
  • At7 : choice: or remove the original 60 pF trimmer or replace it with a low-loss Teflon trimmer of 60 pF.
  • At 14/14A: remove the silver-mica from PCB track to Ground.
  • At 17B: remove the 470 pF ceramic capacitor and put a piece of wire in place or leave the 470 pF here.
Now a few changes:
>No. 1A: cut the PCB track located under Coil 1.

>No. 1 = simple Low-pass input filter 50-50 Ohm (Tnx Elsie) with about 20 dB attenuation of 140 MHz.

4 Turns 1.3 mm CuAg 8 mm diameter, long about 17 mm  and at both sides 100 pF Silver-mica form the coil to ground. Perhaps SMD capacitors size 1206 and 50 V can also be used here.

>No. 2: 60 pF low Loss trimmer Capacitor,  probably Teflon, in series with the track at the place of the original 470 pF capacitor. Remove this one.

>No. 3: 60 pF Low loss trimmer Capacitor from the PCB track to ground.

>No. 4 : 100 nF SMD 1206, 50 volts or better, between ground and the Bias- track to the new 10 uH coil going to the base of the PA transistor.

>No. 5A and 5B: cut the track of the original PCB etched Bias- coil going from the Bias Regulation to the base of the PA transistor.

>No  5: new 10 uH coil, replaces the original PCB coil, made from an old  ferrite choke of about 50 uH 1.5 amp, by peeling 2/3 windings of.
 
>No. 6 A: cut the PCB track under the coil at No. 6.

>No 6:  Coil 2 turn 1.3 mm CuAg 8 mm Diameter, long 9 mm

>No. 7: 60 pF Trimmer Capacitor from PCB track to ground

>No. 8: just under and beside (left) of No. 7,  a 470 pF SMD 1206 from PCB track to ground or use the removed 470 pF ceramic C.

Leave about 25mm PCB track between Capacitors No.7 / 8 and the base of the PA Transistor .

>No. 9: New collector coil 8 Turns 1.3 mm CuAg 6.5 mm diameter, long 18 mm

>No. 10: 2 x 100 nF SMD 1206 and 2x 4n7 SMD 1206, 50 volts Capacitors from collector-coil No. 9 to ground.

>No 11a: Cut the PCB track between the two sets of the original silver-mica capacitors (at the collector side of the transistor) and place here coil no 11.

>No. 11: Coil 1 Turn 1.3 mm CuAg 1.3mm, about 5 mm long

>Leave between the end of Coil 11 and the capacitors at no 12 about 25 mm PCB track and cut it there at 13 A.

>No. 12:  2 x 180 pF brown Silver-mica 500 volt from PCB track to ground.

>No. 13 A: Cut PCB track under coil No. 13

>No. 13: Coil of 4 turns 1.3 mm CuAg diameter 6.5 mm, long about 10 mm.

>No. 14/14 A: Remove

>No. 14: Small low loss 10 pF Teflon trimmer from coil 13 to ground.

>No. 15 A: cut PCB track under Coil 15.

>No.15 : Coil 4 turns 1.3 mm CuAg Diameter 6.5 mm, long 10 mm

>No. 16 : Small low loss 10 pF Teflon trimmer from coil No. 15 to ground

>No. 17a: cut PCB track under 60 pF air-trimmer.

>No. 17 : 60 pF air-trimmer.

> No. 17 B: Originally there is after this coil No. 17 an 470 pF capacitor place in series with the PCB track. I removed this one and connected both sides of the PCB track with a piece of wire. Perhaps this (original) capacitor can stay.

>No. 18: Not (yet ?) solved, is the VSWR shut-down now, never tested it when the PA was still original though for the 2M Band. Did a short test without a dummy connected at the output, but the shut-down did not work, but the PA is still ok though hi.

>at A: Resistor trimmer for the Idle Bias current. Fully Anti-clockwise first (lowest current without any RF drive) then move the trimmer a bit up clockwise so that there is an increase in current of 250-300 mA (at 13.8 V).

>at B: Resistor trimmer for the Vox-delay. Put it at about 2/3 clockwise, Delay is ok then. When put to much anticlockwise, so with very short delay, I had problems that it would fall off
sometimes....



Success, it looks a lot of work but it is not!

Comparison of commercial 4 m verticals
Published on 26 May 2011 by Ian Hogan G6TGO
A comparison of commercial 4m verticals
Read More...

70 MHz low pass filter
Published on 26 May 2011 by Alex IK0VAQ
A low pass filter for 70MHz by Alex IK0VAQ
Read More...

Improvements to the YG5-4 antenna
Published on 25 May 2011 by Marcel Pertus F5DQK
Improvements to the YG5-4 antenna by Marcel Pertus F5DQK
Read More...

70 MHz Big Wheel
Published on 25 May 2011 by Peter OE5MPL and Rudi OE5VRL
70MHz Big Wheel antenna by Peter OE5MPL and Rudi OE5VRL
Read More...

MMANA analysis of the Trident TA4M4L antenna
Published on 25 May 2011 by Marcel F5DQK
MMANA analysis of the Trident TA4M4L antenna by Marcel F5DQK
Read More...

Interfacing Icom HF sets with the Spectrum transverter
Published on 25 May 2011 by Alex GD3UMW
Interfacing Icom HF sets with the Spectrum transverter by Alex GD3UMW
Read More...