PoE PSU works but low noise linear PSU does not [fixed]

edited May 29 in Problems Now Fixed

I brought my KiwiSDR secondhand which included a fancy linear PSU designed for audio work. When I first set it up I simply used a PoE adaptor and everything work pretty well.

The noise floor sits at about -90dBm so I wanted to lower than. I moved it into a rack mount case and transferred the low noise PSU into there as well. However it will not work off that PSU, but does still work fine it the PoE PSU. In both cases the PSU outputs are floating with respect to the case chassis and only gets a ground reference when the case antenna connector is connected.

Looking at 5VE it is at 5V (+/- 100mV) in both cases and it looks stable on a scope, but the 5V from the the BBG is only about 0.7V when I use the low noise PSU. I realise this probably points to an issue U2 on the BBG but it defies logic as everything appears to be as you would expect.

Has anyone else seen an issue similar to this?


  • Is the single blue LED (not the four blue LEDs together) on the "right" side of the Ethernet RJ45 connector lit when it is in this condition?

    Does it work if you plug a "hot" cable from the linear supply to into the Kiwi DC jack? As opposed to turning on the linear supply and possibly running into the "slow ramp up time" issue where the Kiwi Beagle will refuse to turn on if the 5V ramps up slower than 50 msec?

  • Thanks for the suggestion. In the fault condition the blue light does not come on. I tried plugging in the DC jack after power on the PSU and the light came on, so slow ramp up does seem likely.

    I guess the easy fix is to add a front panel power switch in the DC feed.

  • Done that mod, all working good now. Same noise floor with 50R termination on so I guess that cheap PoE adaptor was not coupling any noise in directly.

    Thanks for the prompt suggestions.

  • Hello,

    an excellent read, a good article on the subject of noise rejection:

    Best regards, Philippe

  • @Tremolat Thanks for that. I have been looking at common noise filtering. I have modified the antenna cabling and I am waiting on some Ferrite Toroids to finish that process. I suspect the core problem is I have too much gear here that can radiate and the noise is being picked up at the antenna itself, not the feed lines.

    BTW, I think this thread can be moved to the "Problems Now Fixed" area but can't see how to do that.

Sign In or Register to comment.