We are using some nRF52 DKs for debugging and programming prototype boards. Everything went fine until we wanted to expand our testing setup and got some new nRF52 DKs. These new nRF52 DKs were hardware version v2.1.0, whereas previously we were working with v1.2.4. The new nRF52 DKs gave error messages when running nrfjprog commands, whereas this didn't happen when using a previous version nrf52 DK. We just switched the USB between the two nRF52 DKs, so everything else (supply voltage, jig connection, ...) was the same.
Through further testing, we noticed that raising the supply voltage of the external nRF52xxx resolved the issue and we could use nrfjprog as before. The threshold seems to be around 1.92V, everything above works fine, everything below gives the same message. Is this a known (perhaps intentional) thing that has changed compared to v1.X.X versions of the nRF52 DK?
This seems strange seeing that the nRF52 family devices can go up to 1.7V according to the datasheet. We actually have one of our products run on 1.8V supply and that's how we came across this thing. So right now it seems that it will no longer be possible to program/debug these devices using the latest version nRF52 DK?
Having seen that warning, I didn't even try to use the debug out on a 1.8V project - just went straight to a proper J-Link.
AIUI, the nRF52-DK senses the target Vdd at Debgug Out, and uses that to switch…
The debug output works with 3V logic so I think you must have been lucky with the old DK you had. I have not been able to program targets at 1.8V with the few v1.x.x boards I have tested. The standalone…
AFAIK, the nRF52 DK has never supported debug out at other than 3V
I think the nRF52840-DK might?
Thanks for showing this information. I'm definitely talking about the nRF52 DK (with nRF52832). I must have not checked this information because I wasn't aware that this was only supported for 3V external supplied MCUs. I also never needed to check this, because it just worked on our devices with the nRF52832 supplied by 1.8V. It's only now that we switched to this newer hardware version of nRF52 DK that this issue became apparent.
I guess since the 3V support limit is stated in the documentation it's not a bug or anything like that, but I'm interested to find out what changed between v1.x.x and v2.x.x. that could lead to this behavior.
AIUI, the nRF52-DK senses the target Vdd at Debgug Out, and uses that to switch the SWD connection from the onboard nRF52 to the external target.
So I guess it must be the thresholds in the sensing stuff that have changed ... ?
other debug probes are available.
We also have a J-Link lying around which we use for the higher volume, more formal stuff. The nRF52-DK as debugger was mainly to debottleneck things and try out some new stuff and it stuck around because it just works. At least it did up till the new hardware version. I guess it's time to invest in some additional J-Links then.
The debug output works with 3V logic so I think you must have been lucky with the old DK you had. I have not been able to program targets at 1.8V with the few v1.x.x boards I have tested. The standalone J-links come with a built-in level shifter to support different target voltages.