Project Logs

Collapse

• Troubleshooting

  guilldeas • 11/30/2022 at 11:22 • 0 comments

  After disassembling the device a visual inspection of both components and traces was conducted, however there were no evident signs of failure (i.e. scorching marks on the PCB, bloated capacitors, charred resistors, etc).

  Because the machine wouldn't turn on the first thing to be more properly examined was the power supply. The machine could be powered either by an external standard 5V AC/DC converter or a set of AA batteries, selecting which to connect to the 5V rail by switching an SPDT. After corroborating the external PSU was in working condition the internal power schematic was traced [Fig. 1]  

                                                          Fig. 1: Simplified internal power schematic.

  The external power input was examined which was composed of a MicroUSB connector immediately followed by a DC to DC switching converter [Fig. 2], which was found to be rated for a 5V nominal input voltage and 5V nominal output. 

                                                      Fig. 2: DC to DC converter NXE2S0505MC.

  The converter voltages were first measured with the component isolated from the load (as shown in Fig. 1). Both levels read out healthy values, however with the converter tied to the load the voltage dropped well below minimal nominal values [Table 1].    

  Voltage	Isolated from load [V]	Tied to load [V]	Min. [V]
  +VIN	4.83	4.82	4.5
  +VOUT	6.22	3.31	4.5

                                                      Table 1: Measured values on converter.

                                                    Fig. 3: Dropping 5v supply rail at converter output.

  This could indicate the converter was demanded too much current from a short further down the line, thus dropping the 5V rail to a point where the microcontroller [Fig. 2] could not switch on.

  To prove this wasn't instead a problem with the converter being faulty the following test was performed. A regulated DC power supply was connected to the battery terminals thus powering the synthesizer while bypassing the converter. The power supply's voltage was set to the equivalent of the synthesizer's 3 AA batteries in series (4.5 V). The supply was then current limited to the synthesizer maximum current draw. Nonetheless this information was not stated on the owners manual so the maximum current drawn was estimated to be the maximum current delivered by the DC/DC converter. According to the NXE2 datasheet the maximum power delivered is 2W, which means the maximum current at 5V is:

  Thus it was assumed a healthy converter would be expected to supply at maximum 400 mA without dropping it's output voltage. The DC power supply was then set to 4.5V and current limited to 400 mA. Upon powering the synthesizer from the external power supply the over current protection was immediately tripped which proved the problem was indeed a short and not the converter.

  To find a shorted component without visual cues the device was run for a while (current limited by the regulated DC supply to avoid further damage). This allowed the failing component to heat up to a point that it could be identified by touch. This component was found to be IC11 [Fig. 4], an audio codec (SGTL5000).

                       Fig. 4: Overheating audio codec (IC11) and both output jacks LINEOUT and HP.

  This component implements various functions like switching inputs into...

  Read more »

View project log