What's new
By:
Filters

Snow Bros PCB Fault

X

xodaraP

Legendary
Joined
Sep 21, 2016
Messages
4,751
Reaction score
3,794
Location
Gold Coast, Australia
Hi guys,

I'm looking at a Snow Bros PCB for a friend and I've run into a very weird fault. If ROM 2-A is installed but not 3A, it will run through various screens (squiggly lines, solid colours etc.) as you would expect since it's missing half the game code, if ROM 3-A is installed either by itself or together with 2-A I just get a perfectly synced black screen. To give you an idea of what I have done so far:

Removed, tested and socketed both 6264 RAMs above main program ROMs
Removed, tested and reinstalled LS245 @ IC16
Removed, tested and socketed 6116 RAM @ IC14 (I wanted to see where the Output Enable line for the LS245 was going and it was under the RAM IC)
Removed, tested and reinstalled LS157 @ IC10 (This was where the LS245 /OE was connected)

I have also tried replacing the 6264 RAMs and the EPROMs for the main program code, neither of these things got me any further. The sockets themselves look very clean, the pins are not corroded and both EPROMs verify perfectly off board in a prorammer. I know with these boards that the 41464 DRAMs tend to cause graphics glitches, but as far as I'm aware they are completely unrelated to main program code. There is also no sound, the game is not playing blind, it's not booting at all.
 
maybe both roms are being enabled ontop of each other.
trace the enable lines back, it's probably coming from an address decoder like an ls138
 
They are both connected to a 138 - both lines are connected together and then to pin 15 on the LS138
 
well they cant just be in parallel unless they are a pair of 8bit roms for a 16bit bus.
have you got a scope?
 
Sure do.

The chip enable lines are the same, tied together across both ROMs

Edit: So are the /OE lines - though I can't see what these are connected to.

On the scope both lines can be seen on the /OE, the /CE line is completely dead on both when both are installed, it pulses like mad on 2-A when it's installed by itself and stays dormant if 3-A is installed by itself.
 
Last edited:
maybe the pin on the eprom is bad and holding the line.
 
Good point - I might try burning a new set of ROMs

I also need to try and figure out where the /CE is coming from
 
xodaraP said:
I know with these boards that the 41464 DRAMs tend to cause graphics glitches, but as far as I'm aware they are completely unrelated to main program code. There is also no sound, the game is not playing blind, it's not booting at all.
Click to expand...
They are instead.Since all these 41464 DRAM are addressed by the Pandora custom chip which is addressable by main CPU.So if something goes wrong with it and the CPU can't reach the custom, the main code execution is halted (hence the bus arbiter circuit which the 74LS138 is part of is stuck).I have repaired many of this boards and problem was always the DRAMs or the Pandora custom which needed a reflow.
 
I've had a look at the pins on the custom and they appear to be fine - so I guess the next step is to remove the 4 DRAMs and replace them. Fortunately I do actually have 4 spare on a bootleg MK with battery damage
 
First probe the outputs of the DRAMs with your scope , if they are really bad you should see weak/unhealtly signals.
 
caius said:
First probe the outputs of the DRAMs with your scope , if they are really bad you should see weak/unhealtly signals.
Click to expand...
Which is exactly what I saw on at least 2 of them, the other 2 seem to be working but the signals are all over the shop. I'm going to install sockets in all 4 positions and then I can easily change them as required.
 
Got all 4 RAMs off the board, gouged a trace by accident too so I'll have to fix that.

The ground plane is so insanely thick the only way I was able to remove the ground pins was to heat the pin while removing the RAM
 
Yes, all Toaplan boards are very tough to rework due power supply planes.I usually use hot air at 300-320° to remove ICs.
 
I can only confirm what Caius said: I repaired two Snow Bros boards last year and each time DRAMs were the cause.
 
I've now socketed and replaced all 4 DRAMs - it's still not booting, but I do sometimes get differing levels of garbage like it's trying to boot but still getting stuck.

I'll try burning new ROMs to see if that helps, it may also still be the RAM, I pulled it from a dead bootleg MK board (battery damage) and they're certainly not a matched set, so it may well be that they don't want to work together on this board

If it's not that, the custom doesn't appear to have any loose pins, and I can't find whatever is sending the /CE signal to the main program ROMs
 
Last edited:
Hi guys, I have an update:

I replaced all 4 DRAMs with no change, so I replaced both main program ROMs with fresh EPROMs, which didn't help either

I moved my attention to the main program ROMs again as it doesn't make sense to me that it will boot garbage with 1 ROM installed but nothing with both.

Both ROMs have their /OE and /CE lines tied together, the chip controlling the /OE has been pulled and tested fine.

The /CE line is connected to the main CPU to an address line (pin 46) and with 2-A installed by itself it pulses low and high, with 3-A installed it is held low and with 3-A installed by itself it is stuck low.

It's not an issue with the ROM itself as replacing it didn't help, but the /HALT line isn't asserting either which suggests it's not crashing.

It feels like I'm getting closer, I'm just really hoping it's not a bad custom, but I feel like that's where I'm headed

Edit: Pushing the custom from various angles seems to give me results, so I'm going to attempt to reflow it with a hot air gun and see if that gets me anywhere

Edit 2: Poking a few of the customs legs with a scope probe to see if they move and..... quite a few do. It needs to be completely resoldered down, I don't have the skill for that so I may not be able to do much here
 
Last edited:
xodaraP said:
Edit: Pushing the custom from various angles seems to give me results, so I'm going to attempt to reflow it with a hot air gun and see if that gets me anywhere

Edit 2: Poking a few of the customs legs with a scope probe to see if they move and..... quite a few do. It needs to be completely resoldered down, I don't have the skill for that so I may not be able to do much here
Click to expand...
It seems I pointed you in this direction from the benginning but you didn't follow my advice :)
Any, let's explain you some tips.Use a needle to probe each pin of the Pandora custom chip, probably some of them are lifted or have bad contact.The you can choose if resolder the IC using the "drag soldering" technique /see on Youtube, no hot air is needed but a flat "chisel" iron tip) or solder the pins one by one by putting the iron tip on them and apply pressure.In both cases uses non-conductive flux or solder paste (this one is conductive, be careful!) and them clean perfectly the area with alochol or better wash it with warm water and dish soap (obviously dry the board under sun or in a preheated owen)
 
stj said:
invest in a $40 858D hot air unit.
Click to expand...
That's exactly what I have and I did use it but it couldn't get a lot of them to stay down, more solder needs to be added
caius said:
xodaraP said:
Edit: Pushing the custom from various angles seems to give me results, so I'm going to attempt to reflow it with a hot air gun and see if that gets me anywhere

Edit 2: Poking a few of the customs legs with a scope probe to see if they move and..... quite a few do. It needs to be completely resoldered down, I don't have the skill for that so I may not be able to do much here
Click to expand...
It seems I pointed you in this direction from the benginning but you didn't follow my advice :) Any, let's explain you some tips.Use a needle to probe each pin of the Pandora custom chip, probably some of them are lifted or have bad contact.The you can choose if resolder the IC using the "drag soldering" technique /see on Youtube, no hot air is needed but a flat "chisel" iron tip) or solder the pins one by one by putting the iron tip on them and apply pressure.In both cases uses non-conductive flux or solder paste (this one is conductive, be careful!) and them clean perfectly the area with alochol or better wash it with warm water and dish soap (obviously dry the board under sun or in a preheated owen)
Click to expand...
You did, yes :) but I was really hoping it wasn't the custom as this makes it an issue I really can't solve, I can do simple drag soldering but no way I'm doing it on a 160 pin QFP - I've messaged someone local to me who has done similar soldering for me in the past to see if he can help

I'm getting better with SMD, but I'm not anywhere near this level
 
You must log in or register to reply here.
Back
Top