Impedance Analyzer Adapter has poor header connectivity to AD3

[irvmoy]

I received an Analog Discovery 3 Pro Bundle and an Impedance Analyzer Adapter hardware module this last Monday.  I connected the Impedance Analyzer Adapter to the AD3 via their headers, and noticed that it was a very loose fit, with no resistance from the Adapter's header.  It was difficult to keep the header properly seated, with the end for the digital I/O pins wanting to skew away from the AD3.  By comparison, the BNC Adapter's header fits snugly with the AD3's header, and the AD3 flywire assembly's header fits very, very snugly. 

As an example of how loose is the connection, I can easily move the end of the Impedance Analyzer Adapter's PCB that is opposite the header up and down after the Adapter has been attached to the AD3.  However, the Impedance Analyzer Adapter's header fits very snugly on the BNC Adapter's pass-through header.  It is rigidly attached and I can't move the PCB up or down.

As a test of the Impedance Analyzer Adapter, I put a 39 pF ceramic disc capacitor in its J2 socket and ran the WaveForms Impedance Analyzer instrument.

Capacitance and phase plots with the Adapter board level with the AD3, header fully seated.  The Capacitance plot looks like there is no capacitor present and the Phase plot is displaying noise.

 

Running the same plots with small shims under the Adapter board's rubber feet so that end of the board is tilted upwards, header fully seated.  The Capacitance plot shows the proper value of the capacitor being tested and the Phase plot is correct at a constant -90 degs.

 

Same plots after removing the capacitor, board still tilted, header fully seated.  They look similar to the plots with the capacitor installed on a level adapter board.

 

It seems like the Adapter's header does not make proper electrical contact with the AD3's header unless a small amount of force is applied to the connectors to get a solid connection.  Is this a correct assessment?

Edit: If it is, is the Adapter considered defective and should be replaced under the product warranty?  Or is this a larger problem with what seems to be differences in contact dimensions between the headers used for the adapter modules and the headers used on the AD3 and its flywire assembly?  The differences are tolerable in general but become too extreme in the specific case of the sockets on the Impedance Analyzer Adapter and the pins on the AD3?

Thanks!

Edited June 14 by irvmoy
Clarify purpose of question. Add information about fit with BNC Adapter pass-through header.

[JColvin]

No worries. You are neither the first nor last person to ask about this particular topic.

Mostly I need to find a visible way in the documentation to assure people that they will not break the any of the adapters, nor the 30 soldered pins on either the Analog Discovery or adapter by pushing too hard.

[JColvin]

Hi @irvmoy,

I might just have an old BNC adapter, but none of of my adapters (I tested all of them) fit snugly on my Analog Discovery 3 in such a way that I am not able to easily wiggle them with the barest effort on the opposite side. The same is true if I instead connect the adapters to the male end of the multiple BNC adapters that I have, which is not what you described.

The only secure connection I get with the Analog Discovery 3 is with the 2x15 flywire cable set, which I think is because the cable is able to insert into the AD3 enclosure about a millimeter or two farther than all of the other female housings on the various adapters.

To your point though, I also tested measuring a 39 pF ceramic capacitor with the AD3 and the Impedance Analyzer and experienced no connectivity issues, even when I was deliberately making a point of shaking, wiggling, and tilting the entire system during the measurements.
I only got disconnected plots like you showed in your 1st and 3rd screenshots after I had backed out the Impedance Analyzer so that the PCB was over a millimeter away from the AD3 housing rather than being flush against it.

Based on this and the fact that I have not seen other reports of this sort of connectivity issue, I am more inclined to believe your adapter might be defective.

Presuming you purchased the adapter directly from Digilent (if you instead purchased from a distributor such as Farnell, Mouser, or Amazon, you will need to work with them for an RMA), please send me a private message with the following information:

- Digilent Order number
- Purchase Date
- Adapter serial number (located on the underside of the board on the barcode)
- Preferred email contact

I will then get the information to our Sales team who will follow up with you for any additional information that they may need.

Thanks,
JColvin

[irvmoy]

Hi,

Thanks for the detailed breakdown of your testing!
I am a fool. 🙃  I looked again at how the BNC Adapter's header fits on the AD3 and realized the Impedance Analyzer Adapter's header was not seated to the same depth.
I gave the Impedance Analyzer board a really hard push and it seated further.  Good solid connection, not possible to easily wiggle the PCB up or down.
I guess I was initially too afraid to put that much insertion force on the header.
I repeated the test with the 39 pF capacitor and the plots are good.
Thanks for all the help!