Site icon PhilWIP

PCB Repair: NBA Hangtime

Problem

Barely discernible sound.

Diagnosis

As with other Midway games of the time, the audio section of the Wolf Unit PCB requires -5V. Before testing the PCB, my -5V supply was confirmed working.

I first verified the digital section of the audio system to ensure that the audio produced by the ADSP-2105 DSP was both present and correct. I captured the serial data fed to the AD1851 DAC by the DSP with my logic analyzer and converted it to an audible format:

https://philwip.com/wp-content/uploads/2018/05/chicago1.wav?_=1

This confirmed that the fault lay either at or beyond the analog output stage of the DAC.

As I had an UMK3 PCB to hand, I decided to make my life easier by checking various signals around the audio section so I would know what to expect from a working PCB. Somewhat mysteriously however, UMK3 had suddenly developed the same sound fault as NBA. It quickly dawned on me that NBA had killed the -5V output of my JAMMA SuperGun. Arse! I then did what I should have done right at the start: measure the resistance between -5V and ground on the PCB. The reading was 0Ω. No wonder…

In the meantime, a short funeral service was held for the dead DC-DC converter on the SuperGun that was responsible for producing the -5V supply, and a new replacement was hastily ordered from eBay.

RIP

We still needed to find the short on the PCB. There was what appeared to be a 10µF filter capacitor for the -5V supply at C10. The resistance across it measured 0Ω, so I removed it to test out of circuit and got the same result. The next morning I took a trip to my local electronics surplus store and found a suitable replacement. After installing it, I confirmed that the audio was fixed using my old JAMMA test rig, which has a working -5V supply but no video output…

C10 and friends

Fix

Replace 10µF 10V B3528 SMD tantalum capacitor at C10.

Exit mobile version