Danexmurder
Student
Awesome! I have a set of their screw drivers and love them. In fact I used a Wiha flat-head to pull the chips on the board. This looks even better. May have to scoop that up.
- Thread starter Danexmurder
- Start date
It's supposed to be like that, mine is the same.
It's just an inverter. Since it's connected to the crystal oscillator it's just there to provide a better clock signal. Crystals have poor driving ability, so this is pretty common.
They cut the legs as a factory fix probably. I can't think of any other reason... Your PCB might be a revision where they fixed that.
They cut the legs as a factory fix probably. I can't think of any other reason... Your PCB might be a revision where they fixed that.
Danexmurder
Student
Oh man! Wish I'd heard about that earlier! I ended up sending it back. For what it is worth W4050 agreed to an exchange. Things went from bad to worse with the board and I decided I didn't pay the money to spend all that time troubleshooting it. Last I heard (A week ago) The board was still languishing in customs. Really hoping the dude comes through for me because I'm past the period now where ebay will step in.
People are usually reasonable with this type of thing. He'll probably fix it and send it back. Give him a tip about the crystal in any case.
For what it's worth, I spent way to much time pinpointing the fault and only figured it out because I didn't have activity in the inverter when it didn't boot. I actually thought the crystal was bad.
For future reference, I also resoldered the switch and changed the cap next to it otherwise it would only boot after pressing the switch (it's a reset button).
I believe the capacitor is part of the reset circuit and it just went bad.
For what it's worth, I spent way to much time pinpointing the fault and only figured it out because I didn't have activity in the inverter when it didn't boot. I actually thought the crystal was bad.
For future reference, I also resoldered the switch and changed the cap next to it otherwise it would only boot after pressing the switch (it's a reset button).
I believe the capacitor is part of the reset circuit and it just went bad.
ninjaz0mbie
Student
@ic3b4ll I have a Raiden Fighters Jet board (same one as OP, I think) which is exhibiting a similar problem. I will turn it on and it will operate fine for about a minute and then freeze. (sound is also missing, but I’m just focusing on one problem at a time)
I’ve cleaned and re-seated the chips. I’ve replaced the capacitor near the oscillator. I’ve re-flowed a few things, but no luck.
By “re-solder the main oscillator legs” do you just mean reflowing those four joints on the underside? Any other ideas or tips to diagnose this issue? Anything I can prove while it’s in this frozen state to learn more?
Curiously it seems to pretty consistently freeze at the same place each time if I let the attract mode play. Would that indicate it’s likely a RAM or ROM issue rather than the oscillator?
Yes, just the 4 pins underneath. My board had very little solder on those pins.
Try applying pressure or slightly flexing the board to see if it behaves differently -- freezing sooner or later, glitching, etc.
This would indicate cold solder joints on one or more SMD chips -- usually this affects those 100+ pins chips.
ROM issues could be the culprit, you can try dumping them to see if they are corrupted in some way, it's quick and easy.
It can also be that the RAM is bad, but I think it would crash the board almost immediately. It is unlikely that the game program didn't use a specific area of memory that would cause it to crash or freeze sooner, but it's a possibility.
Don't discount the sound issues. They may be related to the freezing. The board uses a Z80 and they do tend to die. If it's dead and the main CPU tries to communicate with it, it may result in the game locking up waiting for a reply.
ninjaz0mbie
Student
Thank you, this gives me a lot to go on. I think I may have a spare Z80 sitting around from my Donkey Kong’s high score save kit. I may swap that in just to see what happens.
Another thing that may be noteworthy is that “reset button” near the oscillator does absolutely nothing. I used my multimeter to check the legs of the momentary switch and it seems to not have any effect when pushed. What is it “supposed” to do exactly?
I really with this board had some sort of power on self-test for the RAM and ROMs :/
Make sure the Z80 is fast enough. I think it's a 6 or 8 MHz part.
The reset button should (gasp!) reset the board. Now that I think of it, I had to resolder it too, it had the same behaviour as yours.
Start with the oscillator and reset switch and see if the problem goes away.
Grab a logic probe or scope they'll come in handy if things become more complicated.
Good luck!
The reset button should (gasp!) reset the board. Now that I think of it, I had to resolder it too, it had the same behaviour as yours.
Start with the oscillator and reset switch and see if the problem goes away.
Grab a logic probe or scope they'll come in handy if things become more complicated.
Good luck!
ninjaz0mbie
Student
I tried replacing the Z80 with a Zilog Z84C0020PEC that I had lying around. I looked up the data sheet and (if I’m reading it correctly) it should be compatible. The MAME notes that it should run at roughly 4.9mhz
https://github.com/mamedev/mame/blo...f4ae28b410/src/mame/drivers/seibuspi.cpp#L704
Nothing changed unfortunately. Still freezes at the same place.
Interestingly, if I turn the game on without the Z80 in the socket at all it behaves identically to how it behaves with the Z80 in place. Maybe something upstream from the Z80 is at fault? I’ve never used a logic probe before, nor do I have an oscilloscope. All I’ve got to work with at the moment is my multimeter
. May be time to invest.
First I’ll try messing with the momentary switch by the crystal and see what’s going on there.
Whatever failed regarding the sound must have failed recently because the previous owner sent me a video of the game playing with sound prior to shipping it to me last fall.
https://github.com/mamedev/mame/blo...f4ae28b410/src/mame/drivers/seibuspi.cpp#L704
Nothing changed unfortunately. Still freezes at the same place.
Interestingly, if I turn the game on without the Z80 in the socket at all it behaves identically to how it behaves with the Z80 in place. Maybe something upstream from the Z80 is at fault? I’ve never used a logic probe before, nor do I have an oscilloscope. All I’ve got to work with at the moment is my multimeter
. May be time to invest.First I’ll try messing with the momentary switch by the crystal and see what’s going on there.
Whatever failed regarding the sound must have failed recently because the previous owner sent me a video of the game playing with sound prior to shipping it to me last fall.
ninjaz0mbie
Student
If anybody else has a RFJ board and could try running it without the Z80, I’d be interested to know whether it freezes for you at this exact spot as it does for me. Just turn it on and let the attract mode run for a minute without pressing anything. For me this is consistent.
Attachments
The Z80 you used is fine.
My RFJ is stored away otherwise I'd check it for you.
The fact it always hangs in the exact same spot suggests that whatever is causing the freeze is probably related to a memory access.
If you have the means to dump the program ROMs (main and audio CPU) I'd try that first.
Oh, and make sure the voltages are good. It's easy to forget that.
A logic probe is cheap and easy to use and should help with basic troubleshooting. Get that first if you don't want to spend much.
My RFJ is stored away otherwise I'd check it for you.
The fact it always hangs in the exact same spot suggests that whatever is causing the freeze is probably related to a memory access.
If you have the means to dump the program ROMs (main and audio CPU) I'd try that first.
Oh, and make sure the voltages are good. It's easy to forget that.
A logic probe is cheap and easy to use and should help with basic troubleshooting. Get that first if you don't want to spend much.
