rtl-sdr.com

I must have had a ground loop or ground fault somewhere. Everything was supposed to be connected to the same ground. But when I went to connect my RTL-SDR to an antenna switch so I can use my nice dual band antenna, there was a spark between the grounds (outside of the SMA connectors). Oops.

The device still seems to work fine for 25+ mHz quadrature sampling. But direct sampling Q branch for HF reception is dead. I see a few spikes that move with the tuner so I know they're ghosts, but nothing else comes through. It worked immediately before I tried to let the magic smoke out.

Is there a most likely component that got fried that I should try replacing first? R1 seems to have a slightly melty look to the top but there's no obvious damage to any other components. I'm pretty good with a soldering iron... but 0201 SMDs will be especially fun.

Oh I should probably say that this is a v3, bearing what looks like a build date of "2018 4 18" (and doesn't exactly match the photo of the v3 on the "choose a genuine" image, but close enough that I think it's a minor revision).

It seems to me the most vulnerable thing is the input of the RTL chip.
I do not know the internals of the dongle, whether there is ESD on the direct sampling input or if it would fail short circuit.
Alan

Most likely component to die from ESD is the BGA2867 amplifier chip on the direct sampling circuit. It's in the same palce on all revisions, with the SMA on the left side, it's the 5 pin component on the bottom left of the PCB.

It's protected by some good ESD diodes, but a major ESD event could still kill it.

rtlsdrblog wrote: ↑

Wed Jan 20, 2021 5:40 am

Most likely component to die from ESD is the BGA2867 amplifier chip on the direct sampling circuit. It's in the same palce on all revisions, with the SMA on the left side, it's the 5 pin component on the bottom left of the PCB.

It's protected by some good ESD diodes, but a major ESD event could still kill it.

Sure it isn't six pins? That's at the bottom left, and the IC and connections match the datasheet from NXP (i.e. pins 2, 4, and 5 go to ground).

Also, further testing shows that reception in quadrature sampling mode is also much worse. So that would support a fried amp as the cause.

Thanks for the info. I thought I might have some around, but I should be able to get a couple pretty easily if not. I'll give it a shot at replacing it. It's a little under two years old, but this damage is definitely not covered by any reasonable warranty.

My bad, 6 pins indeed it is. If quadrature sampling is also bad after an ESD event then it must have been a pretty hefty event, as that usually survives most ESD.

rtlsdrblog wrote: ↑

Wed Jan 27, 2021 2:55 am

My bad, 6 pins indeed it is. If quadrature sampling is also bad after an ESD event then it must have been a pretty hefty event, as that usually survives most ESD.

It wasn't ESD. It was a power supply connected to other equipment on the same antenna pushing ground 12 volts high for several seconds or minutes. Datasheet says it's rated at max +7/-0.5, so it's not unreasonable for -12v to kill it. At least the rtl-sdr dongle kept my computer from taking a beating with bad voltage!