USB booster generates LOTS of RF interference

I bought some boosters from USB to 9 or 12 V. The idea was to power an old-ish handheld receiver so that I don't have to buy a new battery, dedicated charger and so on. Also they came with the Yaesu favorite 4.0x1.7mm plug.

The result is impressive: even at UHF the receiver goes totally deaf. Let alone in AM where the squelch won't close at any setting.

And it's not only the relatively unfiltered boosted DC: even a nearby receiver picks up the noise. Sure enough the spectrum analyzer with a rubber duckie sees the noise too!

That thing is spamming the airwaves with noise.

I suspected the output cable acts as antenna, so I tried running it with no cable and a load resistor soldered at its terminals. I can still pick up the noise on the nearby handheld receiver.

While I was fiddling inside, I reversed the simple circuit. It looks like the sample diagram from FP6291 datasheet since the pinout and components match 100%. Not much to change in there to make it better.

I tried adding a 7809 in series to the 12V output, but the noise can still be picked up.

Alright: these boosters are good for recharging a battery pack or powering LEDs when no radio is in use.

Efore Barco Power Supply conversion

The Efore label

I received an interesting NIB/NOS power supply used in videowalls. It is an Efore RHPS493 that sports both 8.3A at 48.6V and 1.5A at 5V outputs. That sums up to 400W, so it comes in a sealed aluminum case that doubles as heatsink.

 

I am planning to feed the 48.6V output to a FNIRSI "power supply"(step-down) to get a variable output from 0V to almost 48V and quite a lot of current for normal lab uses. The 5V is already useful by itself.

Barco by Efore RHPS493.

 

Of course it couldn't come with a connector that is compatible with hobbyst's warehouse. It has a nice high-current polarized magnetic plug, that we can't even solder to. I learned that it is a supercool RoPD(R) made by Rosenberger: check out their catalogue!

 

Rosenberger RoPD(R) magnetic connector.

First I measured outputs, so I found what had to be GND (one of the larger poles), 48V (the other large pole, but "off"), and 5V. Another line was floating at some volts, which suggested it could be an enable pin. I was assured that these PSUs do not need a digital control signal to start.

So I started poking around the remaining three smaller pins with a 5kohm resistor while monitoring the two big poles. Easy enough, closing one pin to GND did the magic trick and 48V appeared.

To complete the conversion I had to chop off the nice (and certainly expensive RoPD(R) connector), expose the wires and put more common connectors at the user end. I wired permanently the control signal to ground through that 5k resistor so I don't need to provide an extra switch. 

Wire color vs function mapping.

Some heatshrinked tubing later, the PSU is ready to enter into service.

I even think that it is "RF friendly" in the sense that it doesn't pollute the HF/VHF spectrum. But the FNIRSI step-downer, does.

This post was written 100% by a human, but it needs the recipe of lasagne to be completed. 

Stolen from ik1zyw.blogspot.com