Jump to content
Due to a large amount of spamers, accounts will now have to be approved by the Admins so please be patient. ×
IGNORED

Turbo Outrun Repair Log


Recommended Posts

Been getting my Outrun fix lately, with this stood in the garage.

 

fMbK1wqh.jpg

 

nTJ6RD7h.jpg

 

It was imported as faulty and as the cab mains wiring was horrible and the board was in a box in the bottom of the cab I passed on powering it up and focused on the PCB.

 

r2ETX4Fh.jpg

 

This board was born an Outrun, then upgraded with a Sega kit to convert it to Turbo, presumably in 1989.

 

nohMZ74h.jpg

 

Visually it was in very good condition, clean and no signs of hackery, but it was riddled with Toshiba glossy topped TM2063 and TM2016 SRAMs.

 

These are pretty unreliable in their old age, especially the 2063s, if this board had a full working set it would be surprising.

 

It also had an FD1094 suicide CPU, which contains a CR2032 cell powering a drop of SRAM that holds the decryption key to match the encrypted program ROMs. The 317-0118 is the Sega part number for the FD1094 with the encryption key loaded, the 0118 denoting that it contains the key for the Turbo Outrun upgrade kit.

 

vyLfplBh.jpg

 

The battery has been in there since 1989 so there was a fair chance that was dead, or perhaps not, either way it was another unknown.

 

Powering up these boards on the bend is a pain in the arse, as much as I hate doing it I had to solder on some power wires to the underside of the strange lever lock 10 pin connectors. It was either that or pull the wiring out of the cabinet, and as that was soldered to a tags in the bottom of the board I took the easy option. When plugged into the loom I had made for the last Outrun board I got a black screen and nothing else. With the scope I could see that /RESET and /HALT on the pair of 68000 CPUs were pulsing high and low about twice a second. Basically the board was watch-dogging, ie locking up, failing to respond to the watchdog timer circuit, which was detecting that as a crash and kicking the CPU to try again.

 

This version of Turbo Outrun (Sega 315-0118) is the upgrade kit supplied to convert existing Outrun Cabinets, and the decrypted version is not in the usual sets, but there was some old chat on newsgroups that said that a certain bootleg version was a working replacement, so burning 4 EPROMs with that code and putting in a standard 10MHz 68K CPU got me... nothing, no change.

 

At this point I didn't know if the decrypts were correct, if there was a board fault involved. There is a RAM test set of ROMs for this board set that someone awesome has provided, so with two of the original ROMs back on the board, and the four test ROMs burnt (a pair for the main CPU, and a pair that basically shut down the Aux CPU so it doesn't get in the way of tests) the board would seemingly boot, but I could only tell that the crashing had stopped, as /RESET and /HALT stayed high on the main 68000 CPU.

 

LzlVxaMh.jpg

 

So it looks like the board has a video fault in the mix and as I really needed to see what the test output was trying to tell me that was the next port of call. At the RGBS output I could see a healthy sync signal on the scope, but all the colour lines were held low. A common question about troubleshooting is "where do you start", basically at the fault and move out, or in this case in.

 

The RGB lines track back to three resistor arrays hanging across the outputs of two 74HC273 at IC94 and IC97, both had silent outputs, but also silent inputs, so working upstream further you get to a pair of 74LS245s at IC93 and IC96. These are the bus transceivers that handle data going into the two video RAM ICs at IC92 and IC95.

 

These two LS245s were doing nothing, despite having active data on the A side pins, because the shared enable signal to both on their 1G pins and the LS241 at IC 98 (which is involved in the video SRAM control, and the /DIR line on the 254s) was held high. As /G means enabled when low, these three pins being high was shutting down the video output. These three enables all connected back on a line on the schematic called CS2, which went to pin 35 on the CN2 board interconnect, and then all the way to IC123 (a 74ls244) on the CPU board. No surprise this is right next to the main video controller customer IC. On the scope its outputs were half OK but a load were stuck, and piggybacking a working 74LS244 on top lit the video output up.

 

So IC123 came off, and ...

 

x5pl0Iuh.jpg

 

...faulty, so a salvaged one went on to the board.

 

So I could now see what was going on, an IC 73 issue and some faults with the road section MSBs were getting flagged.

 

PX1gwGyh.jpg

 

So, off came IC73 and yep...

 

m60MZmAh.jpg

 

... it's dead (purple means a pin is not responding).

 

With that replaced, I could swap back in the FD1094 and the original ROMs and it would boot into Turbo finally, the FD1094 battery is still alive - impressive!!!

 

Hmm, yes, road issues.

 

pMn8mNKh.jpg

 

rTZsM0bh.jpg

 

Based on the error messages I had to do some digging to work out which chips were implicated. The errors were all on the MSB lines (most significant bit I assume)

 

0 1

MSB IC21 IC39 Reported as fail - issue is /WR never moves on these two. Stuck high on 39 and low on 21. IC39 all outputs seem floating

LSB IC20 IC38

 

These are /WR2(IC21) and /WR3 (IC39), track to LS155 at IC35 pins 9 and 11 respectively, at the bottom right of this picture, with the 4 TMM2015s SRAMs that it controls above.

 

SfQON1oh.jpg

 

Outputs 9 and 11 are on second gate, which is set by 2G (pin 14) and 2C (pin 15) (tied to ground), so there was no reason these outputs would be silent.

 

The 74LS155 is an unusual chip in arcadia, every time I track down a datasheet for a chip I save it in my archive, and it is getting rarer and rarer that I need to go looking for a datasheet I haven't needed before. Being unusual meant I didn't have one on any scrap boards, except one that was a Fujitsu and faulty.

 

After some extremely careful but rapid desoldering (as it's legs were right up against the board interconnect socket which would not stand the desoldering gun heat for long) it was out, and into my tester.

 

U9fnnIlh.jpg

 

Faulty! Thankfully this is one of the few 74 series ICs that Jaycar do still stock, not cheap for one (they are about 30 cents online), but certainly two weeks quicker than buying from the usual sources online.

 

BTIBYmeh.jpg

 

CYDCPTDh.jpg

 

With that installed the board now passed all the tests.

 

rC1QTPMh.jpg

 

and, with the game ROMs refitted, booted into Turbo Outrun fine.

 

G0D1rEGh.jpg

 

tabvknWh.jpg

 

Board fixed!

 

The FD1094 needs a new battery (keepin' it original folks) which will take it up to 2048 if it lasts as long as the old one did.

 

On to the cabinet, the mains wiring was a mess so had to go. A lot of houses in the US don't have 3 pin sockets (pre-1970 it was not the standard) so when arcade cabinets retired from arcades to peoples homes or garages they had no way to plug them in, without snapping off the earth pin, which is what they did. I have no idea if this causes compliance issues over there but it certainly would not be considered safe in Australia. All exposed metal parts of a cabinet need to be earthed (connected back to that pin) if they connect to anything that could end up connected to the mains through an internal fault. So coin doors, metal CPs and control pedals on this cabinet.

 

Also the "snapped off earth pin" mains cable wasn't original, it comprised of wires twisted together with the internal cab wiring and wrapped in electrical tape, which had half fallen off with age.

 

GI4H3xYh.jpg

 

Nastiness!

 

Whoever did this got live (or "hot" in the US) and neutral flipped, so the main fuse and input switch were on the wrong side. So two things that could save your life in a major fault scenario, disabled.

 

As this cabinet was going to run via a step down I needed to replace the US mains lead, which thankfully some vendors keeps delivering in large numbers to the labs at work, just so we can throw them away.

 

ZOkRXi7h.jpg

 

So one got the snip, and a connector from Jaycar added, plus a lug to go to the earthing mount on the (cleaned) mains input plate...

 

DPdsk9gh.jpg

 

With that reinstalled, and the missing earth connection to the pedal assembly sorted via a new bolt, the board was installed on the back door of the cabinet and hooked up.

 

Impt2gJh.jpg

 

Power up time...

 

QgpZnoph.jpg

 

... it's alive!

 

Fan was bloody noisy, so that was disconnected, not really needed unless the cabinet is going to be run for long periods of time.

 

Screen did look a bit dim, but not surprising really, as the cabinet glass and tube were totally caked in dust and shite.

 

AFTL1yGh.jpg

 

A quarter century of fluff...

 

kHn1NWhh.jpg

 

...and a quarter!

 

PiYX2Rph.jpg

 

With that cleaned off things looked much better...

 

WgzwqEth.jpg

 

... but the monitor probably still isn't a keeper. There's a smear of burn-in that makes the middle section of the screen look muddy, and the reds are a bit weak, leaning towards orange. Also no amount of degausing wand action could fully cure the edge issues. Plus there is a scratch in the tube glass that is more noticeable without the thick layer of dirt.

 

All up that's about it, the cabinet needs some a good clean and polish but that's not my department (not my cabinet :cool:). Oh and the steering shaker mech needs a strip, degrease and re-lube, am pretty sure that shaker motor is working it's guts out.

 

vuc3FPch.jpg

 

It could also use a new set of steering wheel centering springs, I found the snapped bits of the left one scattered in the bottom of the cabinet.

 

0zQRRqch.jpg

 

So just these three, causing all the trouble.

 

ZD56X1Lh.jpg?1

Edited by Womble
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...