Power Profiler kit disconnects and the NRF52DK

[agrelogix]

Hello, 

First time posting, Here are the details of my setup, I would greatly appreciate any help from anyone at this point.
I have purchased 2 nrf52 development kits with 2 power profiler kits to go with them. I am trying to measure the current / power consumption of a lithium ion battery CR2477 in a small test circuit.

1 NRF52DK with 1 PPK plugged together / stacked.

1 external circuit / hardware in breadboard.

1 CR2477 coin cell lithium ion battery (3.0 V).

1 Computer with NRF Connect wand power consumption app installed. (V. 2.3.2 for NRF connect, and 2.0 for the power) 

The NRF52DK has jumper SB9 cut, and SB12 joined (as SB12 is for external source)

I have 2 jumper cables running from the battery holder, to the PPK External supply. 

I also have 2 jumpers running from External DUT to power the breadboarded circuit.

After I've connected the DK to the computer (PC) it asks for a firmware update, I agree and continue. I select the COM port to use and select start.

It seems to work from here, and we'll get to that in a bit. But after pressing start, the system disconnects or drops into an error for no apparent reason. Please see image below:

I'm not sure if the quality is good enough, however the text under the red highlight it says: "Corrupt data detected, please check connection to PPK"

At this point nothing has changed, no cable has moved, and I checked various micro usb cables connected to the DK. 

The issue persists. 

Another screenshot of the logged info can be found below: 

I'm not sure if this is a known issue, but we're in a tight deadline to test our circuits for power consumption / battery life and it's a bit troublesome that it doesn't work as expected. 

I would greatly appreciate anyone's help. I have this setup up and ready to run if anyone has any suggestions or need more information from the device / me.

Thank you in advance

[agrelogix]

Hi,

I thought my response was posted yesterday, I apologize for that. Yes I do get data using the DK as a DUT. I have deleted the hex image in the DK using NRF Connect's programmer app, as well as flashed the ble_blinky program in the DK successfully. And I do get what the video I referred to previously got as a result. 

However after a few minutes (between 1 - 5) the program crashes with error:

Corrupt data detected, please check connection to PPK. 

PPK closed

PPK closed

I have used 4 different USB cables and do not touch the board while the acquisition is on. 

I don't have any external debugger available... If you have a product number or link to it I may be able to get it depending on the cost and lead time.

Thank you

[wlgrd]

Are you using any hubs? Is it the same thing with both ppk's? 
How do you program your external DUT then? You can use a DK as an external debugger, with just an extra cable (can find links later). 

[agrelogix]

I've been running both OS X and Windows. 

OS X specs:

- 3.1 GHz Intel Core i7

- 16 GB 2133 MHz DDR3 Ram

- 1536 MB Graphics

- macOS High Sierra 10.13.4

Windows specs:

- 1.50 GHz Celeron J3455

- 8 GB LDDR3

- 64bit OS

- Windows 10 Pro

Same errors on both, no abnormal resource usage on either.

Thank you

[wlgrd]

I'm at a loss right now. Just cannot see what could be different with your setup. For the external debugger cable, you can use these https://www.adafruit.com/product/1675 , but it shouldn't make a difference. I'm starting to wonder if some SEGGER OBD firmware update has made some changes that I haven't encountered here (nor heard of from anyone else). I will do some testing over the weekend and see if I can replicate. If nothing comes out of that, I guess you need to send us the boards, and we'll ship you a couple of new ones. 

[wlgrd]

I just opened three brand new packages with nRF52 DK and PPK. Just plugged in without any modifications (PPK on top of DK), plugged in my external DUT, set power select to external and provided 3V on the External supply pins of the PPK. Everything works on 3 out of 3 boards, both with nRF52DK and nRF52840 as segger host. 

I also note that we have the exact same Segger version, so that shouldn't matter.

I have experienced USB ports that hasn't behaved, but since you are using two different computers, that shouldn't be the case either. What the USB cable, can you try to swap that with another one from a different vendor? We have seen really weird things when it comes to cable quality some times. Also, try the setup with absolutely no modifications, just to make sure that nothing has happened here. I then upgraded the debugger to v6.20i, and things still work. I have also tried using regulator and DK as power source for the external DUT: Both works.

If this is not solved any time soon, I suggest you do an RMA and also send the defective units so I can debug it further. 

[agrelogix]

Hi,

Thank you for your response, my duties have me else where. However I did manage to try different cables in previous tests and obtained the same results. 

I do have a question in the meantime, as soon as I power the ppk board from an external battery (coin cell using an attachment) the PPK board does fire up (turn on) is this supposed to happen? I do understand that in order to measure consumption of the DUT, it has to go through the PPK, however if I don't plug anything to the DK or PPK except for that coin cell battery through the external supply, should LEDs turn on?

Another question, and this would be more of the connectivity to the DUT, I believe the documentation states it considerably obvious and directly but just want to be 100% sure.

Connecting the PPK to the DUT through P16, would connect to the DUT as if it was a battery, am I correct? as in positive to positive, and gnd to gnd? 

Reason why I ask, is because when trying to measure current with an amp-meter (multi-meter) the connection has to be done in series to the DUT. 

Also, I can simply have any program on the NRF52-DK like blinky or anything else and it will not affect the PPK's firmware / app to measure the power consumption on my DUT? 

I don't believe this last question is covered clearly in the documentation, or I may have missed it somewhere. But would greatly appreciate your response! 

Thank you in advance 

[wlgrd]

Should the battery power the PPK? 

It does, but quite frankly, it shouldn't. Right now, the external power supply (to the DUT) powers the PPK. The same power is used for the External DUT, so I understand your confusion. This means that you cannot do a longevity test of your battery by connecting it through the PPK, but the power consumption numbers are not affected by the PPK also being powered by the external supply.

Connecting the external DUT:

Yes, positive to positive, negative to negative. The PPK needs to power the DUT in order for our measurement circuitry to work. We are thinking of adding the "amp-meter method" as well, i.e in series with the DUT. This is more useful, since we can more easily adjust the voltage to the DUT through the PPK's regulator.

Will any firmware on the DK affect the PPK?

No. You can have any firmware you like on the DK, it wouldn't matter. It can even be empty. The only thing the PPK uses the DK for, is the on-board debugger circuit that the DK has. You are right, this is not covered in the docs. We might add a note for it.

Really sorry to hear that you haven't succeeded with making any of your PPK's work, and I would love to have some hands on to try and figure out why these fail. 

[agrelogix]

Thank you. I will reconvene with my team leader and see what our next course of action will be. Any recommendations on what else I could try for longevity test of my battery? 

Thank you again

[Bjørn Kvaale]

Will you need simultaneous current & voltage measurements from the battery? From what I've understood from this conversation, it seems that you need to power the PPK via the nrf52 dk, via a usb cable to the PPK or via an external power supply, but not a coin cell battery.

An "easy" solution would be to run an example on the DK that has an almost constant current draw from the battery. I have written a blog post on this here. It is absolutely not an ideal way of doing this, as I am using the multimeter to measure the current. 

Instead, you could consider running a nordic example that has a "repeatable" current profile. That way, you should be able to connect up the ppk to the board & take current measurements without using a battery. Then, you could hook up the battery to the dev kit, run the same example & measure the coin cell voltage via the ADC. 

You would need some kind of timer that starts the application after a given number of seconds to be able to synchronize the current measurements with the voltage measurements afterwards. 

This is all just conjecture & I am not sure it will work, but it could be worth a shot!