Analog discovery Oscilloscope not working, AWG works fine. Help!

[OscarBarajas]

I recently re-soldered the USB connector on my legacy analog discovery, since the d+ and d- pads got lifted, I also had to scrape the silkscreen on the tiny the traces that go to the FTDI chip and painstakingly bridged  them using tiny isolated wire I got from an old data cable.

The AD was working just fine before the micro USB was damaged, but now I'm running into some issues:

When using short cables I get a DptilO failed ERC:7 Device configuration failed (PLL 3). I solved it by using use a LONG cable and now it to be recognized and programed correctly, which is counter intuitive as supposedly a PLL error means the AD needs more power, so a longer cable would make that worse. 

Now, when testing it with the latest version of waveforms, it seems like everything works BUT the oscilloscope, I can only see the noise floor when selecting a small enough vertical range, I have tested both channels of the AWG with a benchtop oscilloscope and they work just fine, the power supply works too, the audio out works as well.

It seems something happened to the Oscilloscope input channel stages, but I don't know where to look on the board or what to test for, can someone help me with the schematics/ layout or expected test values of the relevant part of the board? I am a graduate student and I can't spend another 180 dollars in a new AD, I'd rather give it a shot at replacing some components and fixing it myself, I do have a very cheap usb microcope and a rather decent SMT soldering station.

[attila]

Hi @OscarBarajas

The USB socket on AD1 can be broken with bad quality cables or by forcing it. Luckily this is no more an issue with AD2.
In case someone needs to repair AD1 usb, below can find a post with details/PCB.

Not only the cable length, but also the wire gauge and contact quality/material counts in the cable resistance.
Some microUSB cables might work well with one device but not with the other device, have contact issues due to pin alignment, wear, rust...

About the scope issue... the noise indicates that the ADC is working.
Make sure you have connected the wires correctly and started the AWG.
In case you are using the BNC adapter verify the AC/DC coupling, specially when measuring DC voltages.

 

[OscarBarajas]

Quote

The USB socket on AD1 can be broken with bad quality cables or by forcing it. Luckily this is no more an issue with AD2.
In case someone needs to repair AD1 usb, below can find a post with details/PCB.

Hi attila, I was hoping you answered, I have already replaced the USB port, fortunately only the data pins lifted so i didnt have to make any bridges under the connector. I also already found a couple long cables that work.

I have tripple checked the connections and coupling with and without the BNC adapter, but I get the following oscilloscope calibration results *AWG is spot on after calibration. I also make sure the AWG channels are running when trying to measure them.

Please find the calibration results and error prompt atached.

 

 

 

 

[attila]

Hi @OscarBarajas

The AWG 1 might be calibrated wrongly... I would suggest to try to reset factory calibration.

Connect AWG1 to Scope1+, AWG2 to Scope2+, and 1-/2- to GND,
Run a test using Scope and Wavegen interfaces, similar to the earlier screenshot.

[OscarBarajas]

Dear attila, I have ran that test with and without the bnc adapter multiple times with no luck. The AWG is correclty calibrated as confirmed in the screenshots. Please find a screenshot of waveforms so you trust me the AWG is running, the connections are fine too.

 

It seems like no input is getting into the scope channels, so I'm guessing the issue is somewhere on the input circuitry, but I need help troubleshooting and testing. I read some people have replaced the Multiplexer to get the scope working again but I dont know if that's what I have to do or not, I'd need more info such as layout pics or general direction.

 

[attila]

Hi @OscarBarajas

The 5.5V, 3.3V rails are probably fine (since the AWG also uses them) but I would verify these after filter, next to the ICs on:
- C2/C11   AVCC_+/-5V5_SC
- C30/C39 AVCC_+/-3V3_SC
Apply an signal to Scope 1+ input and follow it toward the ADC:
- R20 is after ADG612, signal attenuated
- R16 is after AD8066, signal attenuated
- R13 is after ADA4940, signal further attenuated with 0.9V offset

Here you can find the reference manual, schematic:
https://reference.digilentinc.com/reference/instrumentation/analog-discovery/reference-manual#scope_input_divider_and_gain_selection

Top:

Bottom:

 

[OscarBarajas]

Hi @Attila sorry about the delay, I had no access to an oscilloscope until today, so I checked and the Minus 5.5V is reading 1.07V, the PLUS 5.5V is working fine,

both 3.3V and -3.3 are working. 

How can I make the -5.5V work again? where is the negative power supply IC on the board?

 

Also, when I trace the input waveform past the first variable cap (on pin 3 of the gain selector) all I get is a 600mV DC signal, and I dont understand why. But first things first, lets fix the -5.5V rail.

[attila]

Hi @OscarBarajas

I forgot to mention that you should try to connect with the WaveForms application to the device to enable the analog circuitry, IC20:
https://forum.digilentinc.com/topic/9063-analog-discovery-2-with-error-pll-not-locked/?do=findComment&comment=29345

On your device it is probably something damaged and takes too much power from the AVCC-5V5_SC, like: ADG612, AD8066 or ADA4940 from one (or both) of the two scope inputs.

You could verify the -5.5V on C217,  AVCC-5V5 and AVCC-5V5_SC on the FBs. The shielding metal piece is also ground.
In case the voltage is dropping toward the AVCC-5V5_SC there is too much consumption.
See: https://reference.digilentinc.com/analog_discovery_2/refmanual#figure_30

[OscarBarajas]

Hi, I can not find FB16 and C217 (see attached picture), can you help me?.

When I probe one side of FB6 I get -5.3V, but when I probe the other side I get 1V, so yeah I think that means something is damaged and using too much power, any pointers on finding out which one it may be? ADG612, AD8066 or ADA4940?

Thanks for your help so far.

 

 

[attila]

Hi @OscarBarajas

I have added to the previous post a device photo highlighting the PCB area.

I just notice that you have Analog Discovery (1). The earlier captures are from AD2.

Is any of the ADG612, AD8066, ADA4940 components heating up?
The power/heat might be limited by the FB6...

I have seen some post about the ADG612 getting damaged. This usually results in constant high or low out of scale readings.
The scope inputs are rated to +/-50V, but higher 100-1000Vs are likely to damage this first IC, ADG612.
https://forum.digilentinc.com/topic/18761-i-think-its-broken/

On AD1 FB16 is on the bottom, upper-right:

 

The C217 on the top, upper-middle:

[attila]

Hi @OscarBarajas

In case you have voltage drop only on FB6, but not on FB16. Disconnect the USB and verify the continuity on FB6, should be less than 1 Ohm. This could have been also damaged by short in the above mentioned ICs.

[OscarBarajas]

I think ADG612 is the culprit, I just ordered a replacement and I´m waiting for it, I´ll update you as soon as I replace it.