Path: ...!news-out.netnews.com!postmaster.netnews.com!us7.netnews.com!not-for-mail X-Trace: DXC=D1ZePNOSHgY2BFZQ=P]QnYU5[F2hIijD_7J470dMQQ7[J4R`5ADBYnR^blOYK0N2kR\lfPTGbeKk[Nfhi@Ij_EG]Uh:ZAVm]2iT1NI1l8fb?f^XlnNGcjcaFT X-Complaints-To: support@frugalusenet.com Date: Sat, 25 Jan 2025 20:16:07 -0500 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: Intermittent fault on Korg SDD 3300 Newsgroups: sci.electronics.repair References: <67951f89$1$3620713$882e4bbb@reader.netnews.com> Content-Language: en-US From: bitrex In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-Antivirus: AVG (VPS 250125-4, 1/25/2025), Outbound message X-Antivirus-Status: Clean Lines: 47 Message-ID: <67958cd6$0$654094$882e4bbb@reader.netnews.com> NNTP-Posting-Host: 127.0.0.1 X-Trace: 1737854166 reader.netnews.com 654094 127.0.0.1:60375 Bytes: 3466 On 1/25/2025 2:58 PM, Dave Platt wrote: > In article <67951f89$1$3620713$882e4bbb@reader.netnews.com>, > bitrex wrote: > >> It's annoyingly unpredictable and seems somewhat related to how it's >> mounted? I can power it up 25 times on the bench and it works fine >> making me think I solved the problem, then slide it back in the rack and >> it starts acting up again. I've tried a number of things like adding >> bypass capacitance on the display board which is separated from the CPU >> by good distance, disconnecting the backlight thinking it might be >> interference from the inverter, moving the wiring around, also recapping >> the PSU which I was planning on doing anyway. Doesn't seem to help. >> >> Here is the upper digital board (the analog board is on the lower >> level), the lines to the LCD from the NEC Z80 variant are on the far right: > > I'd be looking for things which could cause an intermittent connection > on that cable and its connectors... maybe a bad or dirty pin or > socket, maybe a hairline crack where one of the pins is soldered to > the PCB. A bad bus-driver on the main PCB (either a separate chip, or > dedicated pins on the microprocessor) might have a similar effect. > > If the characters being shown were characteristically off by one bit, > it'd point to one of the data lines. As it is, they seem to be > rather unpredictably garbled, which suggests to me that one of the > clock or handshaking lines might be bad. Glitchy rising or falling > edges on the "latch your data" signal might result in data being > latched at the wrong time, while the data bus was in transition, > and this could lead to all sorts of nonsense being displayed. > > Might be worth pulling the cables, fluxing and re-flowing the > connecting pins on the PCB, cleaning everything thoroughly. > > If you have a DSO or logic analyzer which has a "look for glitch and > runt pulses" acquisition feature, scoping the data and handshaking > lines at the LCD while actively driving the display, and tapping on > cables and the PCBs, might prove instructive. > > Thanks for getting back, I'll follow up on those suggestions thank you! No logic analyzer available at home unfortunately but I have one I can probably use for a time if nothing else works.. -- This email has been checked for viruses by AVG antivirus software. www.avg.com