• You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.

Connex 3300HP / Galaxy DX55HP - setting up the RF output section.

BayouRadioAmplifier

Active Member
Jul 5, 2022
89
82
28
65
Connex 3300HP Galaxy DX55HP circuit EPT360015C

- I got my hands on a Connex 3300HP to repair.
I see that many have problems with this radio, and all those other
IRF520 MOSFET RF output section radios in general.

- I only have worked on a few of these radios, to get them to work in SSB mode.
But I am very familiar with these IRF510 - 540 devices, using them in
shortwave band pirate radio class-E AM transmitters. Letting them transmit
for hours and hours, without burning up.

- I got this 3300HP to work, with 7 watts carrier, with good modulation.
It can be overmodulated and all. Turn down the carrier power to 3 watts,
turn up the mic audio. and it is a "swing machine." Modulation
transistor TR51 does not blow up. I have been doing many "CB whistles"
into it and all. But TR51 gets warmer that it should due to the design.

Before I did this modification, made adjustments, the input current was
about 5 A, swinging up to 7 A or so with modulation.
Then transistor TR51 got so hot, smoked and blew up on me!
(the "smoke test") (and it worked!) (fun!)

------

But, it is, what it is, so I made it work.

on the top of the board, cut the bare jumper wire from TP7 to TP8.
On bottom side of board, add a wire from TP8 to the emitter trace of
TR51. This puts the the IRF520 driver TR44 drain circuit at 13.8 V.
Modulating this driver is not needed. It causes problems.

Looking down at the board, turn VR10/20 fully ccw, and VR11 fully cw
for no gate bias voltage, no bias.

I soldered a .001 uF disc capacitor across TR45 collector to ground.
Easy to do on bottom side of board. This kills the RF from going to the
IRF520's, so bias adjustments can be made.

Put a current meter in line with the 13.8 V going to the radio,
that can measure small current adjustments.
Turn the microphone all the way down, set the voltage at
TP7 to 5.5 V with VR13. (with this radio here, the Receive RF Gain
was modified to set this level.)

Key up radio, notice the input current (maybe around .44 A)
and turn VR11 ccw for an additional .05 A,
then each VR10/20 cw for and additional .05 A each.
I had .58 A or so total.

Turn 13.8 V off, then unsolder the capacitor on bottom
of board that kills the RF.

Power up radio again. Key up in AM mode.
Check your carrier power. This radio here had very low power.
I adjusted driver bias pot VR11 a little ccw and RF output
increased. I got it up to 14 watts. So turn it down for
7 watts of carrier. And check or adjust again for around 5.5 V at TP7.

turn up the mic gain. modulate it. always monitoring the heat at
the side of the radio at the modulation/regulator transistor.

while modulating this radio here, with 7 watts of carrier,
the input current stays steady at around 4 amps,
until you crank it up and whistle, then the current will drop
at peaks.

This all works, and the best that can be done with this design.
"it is, what it is" as they say.
 
Last edited:

it is a "swing machine."

that is because the meter on the radio is reacting
to the positive going signal. so it shows swinging
upward. but the overmodulated signal is also going
to zero in the other direction. cutting off. going to
no RF output. as in -
"crank it up and whistle, then the current will drop
at peaks."

this can be solved with an added "keep alive voltage"
circuit at TP7. to keep it from going below .5 V or so.
 
another thing on this radio set up. I got it after it was worked on
before. It was blowing fuses. Regulator/mod transistor
was all shorted out, and also found the short of a solder blob
between solder joints at one of the IRF520's from drain to ground.
I removed them 5.6 V zener diodes soldered onto the board
to clean up at this solder and flux mess. It looks like it was
the original factory work, since there was a mark on
the heat compound. I left these diodes off.
they are not needed. they will stop the bias voltage from going
up to 5.6 V, but it cannot go that high anyways.
 
  • Like
Reactions: Shadetree Mechanic

dxChat
Help Users
  • No one is chatting at the moment.
  • @ Bandit1:
    Hey guys I been out a while
  • @ Naysayer:
    ARRL paid ransom for their files back from s hack attack? Todays Amateur thread Reddit
  • @ Naysayer:
    In middle of an antenna job. Misplaced paperwork on purchased balun. It's a 1:1 current balun for a dipole that I disconnected a while ago. I purchased the balun from Canadian company where one input connection has a Red Dot. That for the center of coax, right?