Forums

Hey Mike, as you asked me about the watchdog, I got this circuit for you. I tested it and it works well enough to
keep the data alive in the SRAM. It's the same kind of watchdog circuit used in Ni****do cartridges.
But who cares.... the important part is that it keeps your data alive. The transistor I used for testing was a
2N5401, but a european BC558 should work as well...

I'll be without internet access until this sunday, so don't expect replies from me until sunday, if you mail me or post messages in the board.

See you guys sunday ... =)

Cya ...

Quote

> Hey Mike, as you asked me about the watchdog, I got this circuit for you. I tested it and it works well enough to
> keep the data alive in the SRAM. It's the same kind of watchdog circuit used in Nintendo cartridges.
> But who cares.... the important part is that it keeps your data alive. The transistor I used for testing was a
> 2N5401, but a european BC558 should work as well...

Thanks for that! I'll try it.

Before I saw your post I tried using a 62256 in conjunction with a Harris 7673, backed up by a CR2032 (I chose the 62256 because it was cheaper than the 6264, and avoids the noninverted CS pin problem). I've included a link to the data sheet for the Harris chip.

Well, it works up to a point... as it stands I can save games in Phantasy Star, switch off the SMS for a good while, turn it back on, and the data is still there... but as soon as I remove, then reinsert the cart, the data is lost! Infuriating.

Oh well, at least I know the static RAM works :-) I'll try replacing the Harris with the discrete circuit and see if that improves things.

Cheers,

Mike

Harris 7673 data sheet (PDF)

Quote

> Thanks for that! I'll try it.
> Before I saw your post I tried using a 62256 in conjunction with a Harris 7673, backed up by a CR2032 (I chose the 62256 because it was cheaper than the 6264, and avoids the noninverted CS pin problem). I've included a link to the data sheet for the Harris chip.
> Well, it works up to a point... as it stands I can save games in Phantasy Star, switch off the SMS for a good while, turn it back on, and the data is still there... but as soon as I remove, then reinsert the cart, the data is lost! Infuriating.
> Oh well, at least I know the static RAM works :-) I'll try replacing the Harris with the discrete circuit and see if that improves things.

Actually, I downloaded the datasheet and read it. The chip you choose can be used together with the 6264. I think all you need to
do is invert the status indicator output using a PNP transistor and connect the transistor output to the CS signal from the SRAM.
But I'm not sure, about that anyway.

I'm here, waiting for your success on this matter ... =)

Cya ...

Quote

> Actually, I downloaded the datasheet and read it. The chip you choose can be used together with the 6264. I think all you need to
> do is invert the status indicator output using a PNP transistor and connect the transistor output to the CS signal from the SRAM.
> But I'm not sure, about that anyway.

> I'm here, waiting for your success on this matter ... =)

Thanks Vic! To be honest, I think the problem is down to the battery itself - the voltage across it seems suspiciously low. (Around 2.5v, rather than the nominal 3v.) I'll try wiring up a standard battery pack, until I can get a replacement CR2032.

Mike

Quote

> > Actually, I downloaded the datasheet and read it. The chip you choose can be used together with the 6264. I think all you need to
> > do is invert the status indicator output using a PNP transistor and connect the transistor output to the CS signal from the SRAM.
> > But I'm not sure, about that anyway.

> > I'm here, waiting for your success on this matter ... =)

>
> Thanks Vic! To be honest, I think the problem is down to the battery itself - the voltage across it seems suspiciously low. (Around 2.5v, rather than the nominal 3v.) I'll try wiring up a standard battery pack, until I can get a replacement CR2032.

I measured the tension on both watchdog circuits I have here.
the one in the 315-5235 cartridge gave me 3.2v at the memory VCC pin. (Mitsumi IC watchdog)
The other made with discrete components gave me 2.45v. The battery is old, it's actual power is 2.90v and it's still enough
to keep the data.

A problem in using a SRAM with single chip enable line with watchdog circuits is that it need to have a pass thuru chip
enable signal. A real trouble to us. I gave up from using a 32kb SRAM because of that. I even thought about using a
621000 (128kb SRAM) but it's a sort of stupidity. If I'm going to stuck a such SRAM into a SMS cartridge It's better to use
it to load games and homemade demo programs.

Cya ...

(Latter I'll try to build a watchdog for single /CE line SRAMs, because I want to build a 512kb SRAM game cartridge for the
SMS, using the 315-5235)

Quote

> > Thanks Vic! To be honest, I think the problem is down to the battery itself - the voltage across it seems suspiciously low. (Around 2.5v, rather than the nominal 3v.) I'll try wiring up a standard battery pack, until I can get a replacement CR2032.

> I measured the tension on both watchdog circuits I have here.
> the one in the 315-5235 cartridge gave me 3.2v at the memory VCC pin. (Mitsumi IC watchdog)
> The other made with discrete components gave me 2.45v. The battery is old, it's actual power is 2.90v and it's still enough
> to keep the data.

I see... So that's probably not the issue then.

Quote

> A problem in using a SRAM with single chip enable line with watchdog circuits is that it need to have a pass thuru chip
> enable signal. A real trouble to us. I gave up from using a 32kb SRAM because of that.

Ah right! To be honest, I don't know a great deal about this. I assumed that using a 62256 would eliminate the chip enable issue, since the single, inverted chip enable pin could be connected directly to the 315-5235.

So what you're saying is that the SRAM chip has to be explicitly disabled when the cartridge isn't powered? I suppose that makes sense, I didn't really think it through.

I'll replace the 62256 with the 6264, and try what you suggest.

Mike

Quote

> I see... So that's probably not the issue then.
> > A problem in using a SRAM with single chip enable line with watchdog circuits is that it need to have a pass thuru chip
> > enable signal. A real trouble to us. I gave up from using a 32kb SRAM because of that.
> Ah right! To be honest, I don't know a great deal about this. I assumed that using a 62256 would eliminate the chip enable issue, since the single, inverted chip enable pin could be connected directly to the 315-5235.

Well ... Unfortunatelly, when you unplug the cartridge from the SMS and the /CS is grounded, the 62256 will receive write pulses, because /WR and /CS are both active-low input pins. The CS (non inverted Chip Enable)
are perfect for a such kind of control, because it's active high. It means that a SRAM chip with dual CS pins will never accept write or read pulses if one of the chip select signals are false. It's how the non inverted
CS acts when the system is powered down.

Quote

> So what you're saying is that the SRAM chip has to be explicitly disabled when the cartridge isn't powered? I suppose that makes sense, I didn't really think it through.

Actually, be sure that 315-5235 will keep SRAM /CS pin low when powered down, due to internal current drain. It means the SRAM will be enabled all the time (even in power-on/power-off cycles) making damage to the
stored data. Also, when the SRAM is in the active mode, it's power consuption is ingreased, because it will select the output buffers or the data write logic, which ingreases a lot the power consuption. Results: if the lithyum
battery is supposed to mantain the SRAM power for five years it will be drained in a week or two.

Quote

> I'll replace the 62256 with the 6264, and try what you suggest.

Again, I wish you luck with the SRAM circuit and I'm waiting for the results ...

Cya ...

Author	Message
Vic_Viper Joined: 12 Jul 2005 Posts: 55	SRAM 6264 watchdog circuit & my trip Posted: Wed Sep 06, 2000 1:00 pm
Vic_Viper Joined: 12 Jul 2005 Posts: 55	Hey Mike, as you asked me about the watchdog, I got this circuit for you. I tested it and it works well enough to keep the data alive in the SRAM. It's the same kind of watchdog circuit used in Ni****do cartridges. But who cares.... the important part is that it keeps your data alive. The transistor I used for testing was a 2N5401, but a european BC558 should work as well...

Vic_Viper Joined: 12 Jul 2005 Posts: 55	Forgot telling about the trip ... Posted: Wed Sep 06, 2000 1:04 pm
Vic_Viper Joined: 12 Jul 2005 Posts: 55	I'll be without internet access until this sunday, so don't expect replies from me until sunday, if you mail me or post messages in the board. See you guys sunday ... =) Cya ...

Mike G Joined: 21 Apr 2000 Posts: 598 Location: Newcastle upon Tyne, England	Posted: Wed Sep 06, 2000 10:00 pm
	Quote > Hey Mike, as you asked me about the watchdog, I got this circuit for you. I tested it and it works well enough to > keep the data alive in the SRAM. It's the same kind of watchdog circuit used in Nintendo cartridges. > But who cares.... the important part is that it keeps your data alive. The transistor I used for testing was a > 2N5401, but a european BC558 should work as well... Thanks for that! I'll try it. Before I saw your post I tried using a 62256 in conjunction with a Harris 7673, backed up by a CR2032 (I chose the 62256 because it was cheaper than the 6264, and avoids the noninverted CS pin problem). I've included a link to the data sheet for the Harris chip. Well, it works up to a point... as it stands I can save games in Phantasy Star, switch off the SMS for a good while, turn it back on, and the data is still there... but as soon as I remove, then reinsert the cart, the data is lost! Infuriating. Oh well, at least I know the static RAM works :-) I'll try replacing the Harris with the discrete circuit and see if that improves things. Cheers, Mike Harris 7673 data sheet (PDF)

Vic_Viper Joined: 12 Jul 2005 Posts: 55	Re: SRAM 6264 watchdog circuit Posted: Wed Sep 13, 2000 3:35 am
Vic_Viper Joined: 12 Jul 2005 Posts: 55	Quote > Thanks for that! I'll try it. > Before I saw your post I tried using a 62256 in conjunction with a Harris 7673, backed up by a CR2032 (I chose the 62256 because it was cheaper than the 6264, and avoids the noninverted CS pin problem). I've included a link to the data sheet for the Harris chip. > Well, it works up to a point... as it stands I can save games in Phantasy Star, switch off the SMS for a good while, turn it back on, and the data is still there... but as soon as I remove, then reinsert the cart, the data is lost! Infuriating. > Oh well, at least I know the static RAM works :-) I'll try replacing the Harris with the discrete circuit and see if that improves things. Actually, I downloaded the datasheet and read it. The chip you choose can be used together with the 6264. I think all you need to do is invert the status indicator output using a PNP transistor and connect the transistor output to the CS signal from the SRAM. But I'm not sure, about that anyway. I'm here, waiting for your success on this matter ... =) Cya ...

Mike G Joined: 21 Apr 2000 Posts: 598 Location: Newcastle upon Tyne, England	Re: SRAM 6264 watchdog circuit Posted: Thu Sep 14, 2000 11:12 am
	Quote > Actually, I downloaded the datasheet and read it. The chip you choose can be used together with the 6264. I think all you need to > do is invert the status indicator output using a PNP transistor and connect the transistor output to the CS signal from the SRAM. > But I'm not sure, about that anyway. > I'm here, waiting for your success on this matter ... =) Thanks Vic! To be honest, I think the problem is down to the battery itself - the voltage across it seems suspiciously low. (Around 2.5v, rather than the nominal 3v.) I'll try wiring up a standard battery pack, until I can get a replacement CR2032. Mike

Vic_Viper Joined: 12 Jul 2005 Posts: 55	Re: SRAM 6264 watchdog circuit Posted: Thu Sep 14, 2000 6:46 pm
Vic_Viper Joined: 12 Jul 2005 Posts: 55	Quote > > Actually, I downloaded the datasheet and read it. The chip you choose can be used together with the 6264. I think all you need to > > do is invert the status indicator output using a PNP transistor and connect the transistor output to the CS signal from the SRAM. > > But I'm not sure, about that anyway. > > I'm here, waiting for your success on this matter ... =) > > Thanks Vic! To be honest, I think the problem is down to the battery itself - the voltage across it seems suspiciously low. (Around 2.5v, rather than the nominal 3v.) I'll try wiring up a standard battery pack, until I can get a replacement CR2032. I measured the tension on both watchdog circuits I have here. the one in the 315-5235 cartridge gave me 3.2v at the memory VCC pin. (Mitsumi IC watchdog) The other made with discrete components gave me 2.45v. The battery is old, it's actual power is 2.90v and it's still enough to keep the data. A problem in using a SRAM with single chip enable line with watchdog circuits is that it need to have a pass thuru chip enable signal. A real trouble to us. I gave up from using a 32kb SRAM because of that. I even thought about using a 621000 (128kb SRAM) but it's a sort of stupidity. If I'm going to stuck a such SRAM into a SMS cartridge It's better to use it to load games and homemade demo programs. Cya ... (Latter I'll try to build a watchdog for single /CE line SRAMs, because I want to build a 512kb SRAM game cartridge for the SMS, using the 315-5235)

Mike G Joined: 21 Apr 2000 Posts: 598 Location: Newcastle upon Tyne, England	Re: SRAM 6264 watchdog circuit Posted: Thu Sep 14, 2000 7:53 pm
	Quote > > Thanks Vic! To be honest, I think the problem is down to the battery itself - the voltage across it seems suspiciously low. (Around 2.5v, rather than the nominal 3v.) I'll try wiring up a standard battery pack, until I can get a replacement CR2032. > I measured the tension on both watchdog circuits I have here. > the one in the 315-5235 cartridge gave me 3.2v at the memory VCC pin. (Mitsumi IC watchdog) > The other made with discrete components gave me 2.45v. The battery is old, it's actual power is 2.90v and it's still enough > to keep the data. I see... So that's probably not the issue then. Quote > A problem in using a SRAM with single chip enable line with watchdog circuits is that it need to have a pass thuru chip > enable signal. A real trouble to us. I gave up from using a 32kb SRAM because of that. Ah right! To be honest, I don't know a great deal about this. I assumed that using a 62256 would eliminate the chip enable issue, since the single, inverted chip enable pin could be connected directly to the 315-5235. So what you're saying is that the SRAM chip has to be explicitly disabled when the cartridge isn't powered? I suppose that makes sense, I didn't really think it through. I'll replace the 62256 with the 6264, and try what you suggest. Mike

Vic_Viper Joined: 12 Jul 2005 Posts: 55	Re: SRAM 6264 watchdog circuit Posted: Fri Sep 15, 2000 4:27 am
Vic_Viper Joined: 12 Jul 2005 Posts: 55	Quote > I see... So that's probably not the issue then. > > A problem in using a SRAM with single chip enable line with watchdog circuits is that it need to have a pass thuru chip > > enable signal. A real trouble to us. I gave up from using a 32kb SRAM because of that. > Ah right! To be honest, I don't know a great deal about this. I assumed that using a 62256 would eliminate the chip enable issue, since the single, inverted chip enable pin could be connected directly to the 315-5235. Well ... Unfortunatelly, when you unplug the cartridge from the SMS and the /CS is grounded, the 62256 will receive write pulses, because /WR and /CS are both active-low input pins. The CS (non inverted Chip Enable) are perfect for a such kind of control, because it's active high. It means that a SRAM chip with dual CS pins will never accept write or read pulses if one of the chip select signals are false. It's how the non inverted CS acts when the system is powered down. Quote > So what you're saying is that the SRAM chip has to be explicitly disabled when the cartridge isn't powered? I suppose that makes sense, I didn't really think it through. Actually, be sure that 315-5235 will keep SRAM /CS pin low when powered down, due to internal current drain. It means the SRAM will be enabled all the time (even in power-on/power-off cycles) making damage to the stored data. Also, when the SRAM is in the active mode, it's power consuption is ingreased, because it will select the output buffers or the data write logic, which ingreases a lot the power consuption. Results: if the lithyum battery is supposed to mantain the SRAM power for five years it will be drained in a week or two. Quote > I'll replace the 62256 with the 6264, and try what you suggest. Again, I wish you luck with the SRAM circuit and I'm waiting for the results ... Cya ...

Forums

View topic - SRAM 6264 watchdog circuit & my trip