I keep frying RFD868's, what is going on?

Hardware Radios
,
1

After frying my HolyBro SiK radio, I decided to go for the RFD900 family. Since I am in EU, I bought the RFD868x.

For a short while it worked, I was able to confirm connection with GCS. But shortly after, it failed. Later I reasoned that I must have accidentally connected it to a 5V level UART, which was supposed to be plugged into another cable on my system.

So I bought 2 more RFD868x’s. Today I tested everything and it was working, but very unreliable. AT commands, including remote, seemed to be reliable, but connection with FC had problems.

Somewhere during testing I may have shorted something out while plugging in the connectors. But I was very surprised to see that I just bricked my second RFD868x. Most likely I must have miswired something, but I am not sure.

Voltage is 5.2-5.3.

When measuring resistance on the supply pins (2 & 4) when fully disconnected, the still working unit gives me .5M, which increased over time (I assume a capacitor being charged?).

Failed unit #1 (which I thought might haved been exposed to a 5v level UART, but not sure) has a voltage of 1.3 and dropping. This is after being fully powered off for a day and disconnected. Must also be a charged capacitor?

Failed unit #2, the resistance is 150K and climbing.

I have two questions:

  • What are common causes for frying RFD900-series units? Any particular sensitivities or common user errors?
  • What methods can I use to find out how I fried it. I’m thinking about removing the shield and probing around in the unit. If the error was due to accidental wrongful insertion of power connector, it would be good to confirm this. If the problem was caused by accidental connection to to a 5V UART, that would also be good to confirm.

Both dead units draw 0A at 5V.

2

Long shot - I assume you had antennas attached while testing? No antenna while powered will kill them…

3

During the first frying session I later noticed that one of the SMA connectors was a bit loose, the nut got slightly loosened. The second attempts, one of the SMAs may have been slightly on the loose side as well (barely handtight).

4

That fries the RF endstage. The antennas must be securely connected when in operation

5

that being said, you have to power them up quite a long time without antennas.
they won’t fry in seconds. Mine even survived 2-3 minutes.

so that combined with your suggestion that you “might, probably, maybe, must have, most likely miswired them” leads to the conclusion that it’s probably a blunt user error.

not to talk you down, we all have miswired stuff and had things die on us. But when you install a module with this price tag, you might want to triple check all the connections.

besides that: i’m confused why your uart pins have 5 volt, or did i understand that wrong? normally these GPIO pins run on 3.3 volt. the rfd868 does not like 5 volt on the gpio pins. max is 3.5 V
the max input also is 5.5. so your 5.3 are closer to the max voltage then the nominal input.

but i would indeed guess the same like the others did: were the antennas (both!) securely connected?

best of luck

6

They may not fry in seconds as you indicate, but the PA can be damaged immediately and function at reduced power. There is no foldback protection.

7

Sure, that’s all true. I just noticed that they are more resilient in practice than One would think.