Forums

Hi, firstly sorry for my first post here being one asking for help but I've not had a Master System for many years until a few days ago.

I have a Master System II I bought as a job lot bit of a gamble, I just tried it out but it is being very strange.

I'm testing the built-in Alex Kidd game as I don't have any cartridges right now. I've taken a video of what I'm seeing as I can't quite describe it, it is more than video corruption, the audio sounds strange and it seems the game isn't doing what I'd expect.

It looks like sprites are loading in the wrong places and corrupted, after watching the intro screen, it goes right to the map screen without me doing anything and stays there. If I press the "Pause" button on the console then it appears to go to the motorbike demo sequence but it shows Alex running instead of on a bike! If I press Pause again then it goes back to the map. I don't think this should happen.

I have removed the PCB from the unit and visually checked for dry solder joints and broken tracks, I've just done this by eye for now but it looks clean. I've also checked the voltage in and out of the regulator and it is 9.5v in, 5.06v out which seems within tolerance? Those remain stable.

Am I looking at a corrupted CPU or bios/rom chip here?

Here's a video, it's really weird:

https://photos.app.goo.gl/rQ48YWWPXwkpy2As8

EDIT: worth pointing out that I have nothing connected to the console at this point, not even a control pad, just RF cable and power supply. It's not the original PSU, it's actually from a Sinclair Spectrum 128 +2 but this is a centre negative, 9v 1,25A power supply. I vaguely remember using the same PSU back in the early 90s as I had both an SMS and 128 +2 at the time.

cheers,
Paul.

Wow!

Something is really wrong with the console or the bios. It could be RAM, VRAM or maybe some broken trace so addresses are slightly wrong...

hayesey wrote

Am I looking at a corrupted CPU or bios/rom chip here?
Paul.

more likely the RAM. If you want to exclude the possibility of the bios rom being faulty, try inserting any cart. If the cart behaves strangely as well, it's the RAM. It's not VRAM only, because otherwise you wouldn't have corruption of the game logic.

kamillebidan wrote

hayesey wrote

Am I looking at a corrupted CPU or bios/rom chip here?
Paul.

more likely the RAM. If you want to exclude the possibility of the bios rom being faulty, try inserting any cart. If the cart behaves strangely as well, it's the RAM. It's not VRAM only, because otherwise you wouldn't have corruption of the game logic.

great, thanks. I'll get a scope on the ram chips data lines and look for clues. I'll also check the pcb traces in more detail.

I wonder if replacement RAM chips are available? I'll do some searching...

I will update here when I've hopefully got somewhere.

well I've done some testing but haven't made any progress.

I've scoped the data lines on both the ROM and the RAM chips, I've attached an image, all 8 lines on both chips gave a similar signal. It looks OK, it's fuzzy because the system is constantly accessing memory so makes sense.

Also scoped the clock to the CPU which looks correct too, ~279nS period.

So no signs of a fault yet but not to say that either the ROM or RAM chip are actually working correctly but no obvious faults like a dead data line.

I think the next step is going to be to get hold of a cartridge to see if that works OK. But I'm not sure I want to spend more money on this really or just try to get back what this machine owes me, it's in good condition aside from whatever this fault is obviously!

It has a sanyo lc3664-lc ram chip which I see are on ebay (from chinese sellers) for not a lot of money but would rather try and get hold of a cart to try first.

p.s. apologies for the state of the oscilloscope, it's not actually mine I borrowed it and it was in someone's loft for a while :)

IMG_4117.png (1.03 MB)

ram chip data line signal

IMG_4111.png (1.12 MB)

cpu clock trace

+1 for this looking likely as a RAM-related issue.

It could very well be a faulty LC3664 SRAM chip but it's worth making sure it's not a mainboard issue first.

I would do a continuity test for all the Address lines and Data lines between the CPU and the RAM chip, i.e., test that A0 on the CPU has continuity with A0 on the RAM and no connection with any other Address line, then repeat the test for A1, then A2, and so on.

5.06 Volts on the Regulator output is OK. It would be a good idea to check the power rails on the LC3664 RAM chip too to make sure it's getting a solid 5 volts. Also have a close look at the ceramic capacitor and electrolytic capacitor for the SRAM chip to make sure nothings bent or broken, if you haven't already.

I've just tested continuity on all address and data lines between the cpu, ram chip and rom chip. No issues found and no shorts with any other pins on those chips.

Checked voltage from output of regulator, power to the pins on ram & rom chips and getting 5.07v stable on all.

Really think there's little left to try but changing the ram chip.

I would check VRAM too anyway, it looks like it could be missing continuity (or a short) on an address pin.

Based on what you've said it looks like the SRAM chip itself is faulty.

The SMS board and traces are quite delicate and it's easy to damage traces in the process of de-soldering a chip.

If your purchasing SRAM off ebay I'd suggest soldering in a DIP-14 socket as fake chips are everywhere on ebay and a socket would make it easy to swap out a fake or defective replacement.

If a genuine LC3664 becomes too hard to source, the Samsung KM6264 was also used in SMS boards in place of the LC3664. The LC3664 (surface mount version though) was also one of the apparently many possible SRAM chips found in the Megadrive 2. So you may have some options for replacements.

Good luck.

MD2 SRAM.PNG (125.09 KB)

MD2 SRAM

thanks for the info, I'll check out the KM6264 chip too as I am aware that there are many counterfeit chips and am dubious about buying these from a seller based in china.

How about the NEC UPD4168C-20? This is what is listed in the Master system II service manual on this site, seems to be an equivalent of the sanyo chip from the datasheets (although I can only find a chinese datasheet for the sanyo chip but the pin out is identical).

There's a seller in the UK selling the NEC chip for not silly amounts of money so might try that.

I'll pin out the VRAM chips too, no harm in doing. I have to say though I've inspected the PCB closely and it looks new, I can't see any breaks or any dry joints, all solder joints are shiny and new looking.

Any 28-pin SRAM that is fast enough (but maybe not too fast?) should be ok. They're supposed to be interchangeable.

Author	Message
hayesey Joined: 04 Nov 2021 Posts: 5	My Master System II has gone crazy Posted: Thu Nov 04, 2021 1:52 pm
hayesey Joined: 04 Nov 2021 Posts: 5	Hi, firstly sorry for my first post here being one asking for help but I've not had a Master System for many years until a few days ago. I have a Master System II I bought as a job lot bit of a gamble, I just tried it out but it is being very strange. I'm testing the built-in Alex Kidd game as I don't have any cartridges right now. I've taken a video of what I'm seeing as I can't quite describe it, it is more than video corruption, the audio sounds strange and it seems the game isn't doing what I'd expect. It looks like sprites are loading in the wrong places and corrupted, after watching the intro screen, it goes right to the map screen without me doing anything and stays there. If I press the "Pause" button on the console then it appears to go to the motorbike demo sequence but it shows Alex running instead of on a bike! If I press Pause again then it goes back to the map. I don't think this should happen. I have removed the PCB from the unit and visually checked for dry solder joints and broken tracks, I've just done this by eye for now but it looks clean. I've also checked the voltage in and out of the regulator and it is 9.5v in, 5.06v out which seems within tolerance? Those remain stable. Am I looking at a corrupted CPU or bios/rom chip here? Here's a video, it's really weird: https://photos.app.goo.gl/rQ48YWWPXwkpy2As8 EDIT: worth pointing out that I have nothing connected to the console at this point, not even a control pad, just RF cable and power supply. It's not the original PSU, it's actually from a Sinclair Spectrum 128 +2 but this is a centre negative, 9v 1,25A power supply. I vaguely remember using the same PSU back in the early 90s as I had both an SMS and 128 +2 at the time. cheers, Paul.

kusfo Joined: 29 Mar 2012 Posts: 886 Location: Spain	Posted: Fri Nov 05, 2021 8:53 am
kusfo Joined: 29 Mar 2012 Posts: 886 Location: Spain	Wow! Something is really wrong with the console or the bios. It could be RAM, VRAM or maybe some broken trace so addresses are slightly wrong...

kamillebidan Joined: 25 Feb 2013 Posts: 384 Location: Osaka	Posted: Fri Nov 05, 2021 10:29 am
	hayesey wrote Am I looking at a corrupted CPU or bios/rom chip here? Paul. more likely the RAM. If you want to exclude the possibility of the bios rom being faulty, try inserting any cart. If the cart behaves strangely as well, it's the RAM. It's not VRAM only, because otherwise you wouldn't have corruption of the game logic.

hayesey Joined: 04 Nov 2021 Posts: 5	Posted: Fri Nov 05, 2021 12:18 pm
hayesey Joined: 04 Nov 2021 Posts: 5	kamillebidan wrote hayesey wrote Am I looking at a corrupted CPU or bios/rom chip here? Paul. more likely the RAM. If you want to exclude the possibility of the bios rom being faulty, try inserting any cart. If the cart behaves strangely as well, it's the RAM. It's not VRAM only, because otherwise you wouldn't have corruption of the game logic. great, thanks. I'll get a scope on the ram chips data lines and look for clues. I'll also check the pcb traces in more detail. I wonder if replacement RAM chips are available? I'll do some searching... I will update here when I've hopefully got somewhere.

hayesey Joined: 04 Nov 2021 Posts: 5	Posted: Mon Nov 22, 2021 12:04 pm Last edited by hayesey on Mon Nov 22, 2021 3:20 pm; edited 1 time in total
hayesey Joined: 04 Nov 2021 Posts: 5	well I've done some testing but haven't made any progress. I've scoped the data lines on both the ROM and the RAM chips, I've attached an image, all 8 lines on both chips gave a similar signal. It looks OK, it's fuzzy because the system is constantly accessing memory so makes sense. Also scoped the clock to the CPU which looks correct too, ~279nS period. So no signs of a fault yet but not to say that either the ROM or RAM chip are actually working correctly but no obvious faults like a dead data line. I think the next step is going to be to get hold of a cartridge to see if that works OK. But I'm not sure I want to spend more money on this really or just try to get back what this machine owes me, it's in good condition aside from whatever this fault is obviously! It has a sanyo lc3664-lc ram chip which I see are on ebay (from chinese sellers) for not a lot of money but would rather try and get hold of a cart to try first. p.s. apologies for the state of the oscilloscope, it's not actually mine I borrowed it and it was in someone's loft for a while :) IMG_4117.png (1.03 MB) ram chip data line signal IMG_4111.png (1.12 MB) cpu clock trace

asynchronous Joined: 14 Aug 2000 Posts: 742 Location: Adelaide, Australia	My Master System II has gone crazy Posted: Mon Nov 22, 2021 3:17 pm
	+1 for this looking likely as a RAM-related issue. It could very well be a faulty LC3664 SRAM chip but it's worth making sure it's not a mainboard issue first. I would do a continuity test for all the Address lines and Data lines between the CPU and the RAM chip, i.e., test that A0 on the CPU has continuity with A0 on the RAM and no connection with any other Address line, then repeat the test for A1, then A2, and so on. 5.06 Volts on the Regulator output is OK. It would be a good idea to check the power rails on the LC3664 RAM chip too to make sure it's getting a solid 5 volts. Also have a close look at the ceramic capacitor and electrolytic capacitor for the SRAM chip to make sure nothings bent or broken, if you haven't already.

hayesey Joined: 04 Nov 2021 Posts: 5	Posted: Mon Nov 22, 2021 9:31 pm
hayesey Joined: 04 Nov 2021 Posts: 5	I've just tested continuity on all address and data lines between the cpu, ram chip and rom chip. No issues found and no shorts with any other pins on those chips. Checked voltage from output of regulator, power to the pins on ram & rom chips and getting 5.07v stable on all. Really think there's little left to try but changing the ram chip.

sverx Joined: 05 Sep 2013 Posts: 3823 Location: Stockholm, Sweden	Posted: Tue Nov 23, 2021 11:45 am
	I would check VRAM too anyway, it looks like it could be missing continuity (or a short) on an address pin.

asynchronous Joined: 14 Aug 2000 Posts: 742 Location: Adelaide, Australia	My Master System II has gone crazy Posted: Tue Nov 23, 2021 1:17 pm
	Based on what you've said it looks like the SRAM chip itself is faulty. The SMS board and traces are quite delicate and it's easy to damage traces in the process of de-soldering a chip. If your purchasing SRAM off ebay I'd suggest soldering in a DIP-14 socket as fake chips are everywhere on ebay and a socket would make it easy to swap out a fake or defective replacement. If a genuine LC3664 becomes too hard to source, the Samsung KM6264 was also used in SMS boards in place of the LC3664. The LC3664 (surface mount version though) was also one of the apparently many possible SRAM chips found in the Megadrive 2. So you may have some options for replacements. Good luck. MD2 SRAM.PNG (125.09 KB) MD2 SRAM

hayesey Joined: 04 Nov 2021 Posts: 5	Posted: Tue Nov 23, 2021 1:57 pm
hayesey Joined: 04 Nov 2021 Posts: 5	thanks for the info, I'll check out the KM6264 chip too as I am aware that there are many counterfeit chips and am dubious about buying these from a seller based in china. How about the NEC UPD4168C-20? This is what is listed in the Master system II service manual on this site, seems to be an equivalent of the sanyo chip from the datasheets (although I can only find a chinese datasheet for the sanyo chip but the pin out is identical). There's a seller in the UK selling the NEC chip for not silly amounts of money so might try that. I'll pin out the VRAM chips too, no harm in doing. I have to say though I've inspected the PCB closely and it looks new, I can't see any breaks or any dry joints, all solder joints are shiny and new looking.

lidnariq Joined: 24 Mar 2021 Posts: 120	Posted: Tue Nov 23, 2021 6:08 pm
lidnariq Joined: 24 Mar 2021 Posts: 120	Any 28-pin SRAM that is fast enough (but maybe not too fast?) should be ok. They're supposed to be interchangeable.

Forums

View topic - My Master System II has gone crazy