Forums

Time ago I made a working SLOT 1/2 mapper using 16V8 PLDs and while it worked flawlessly the 16V8 PLDs are increasingly harder to come by and especially harder to flash due to 16V8 programmers being discontinued, so I want to replicate the PLD combinatory logic with off-the-shelf 74xx series.

So far I managed to replace the 2 16V8 PLDs with only 6 logic circuits, however I have a question about the ROM chip !CE line. While the original !CE used the following combinatorial equation:

!( !(A14 & A15) & CE & MREQ )

The current iteration uses the following one:

!( !(A14 & A15) & CE )

EDIT: (notice CE / MREQ are not negated, those signals come from inverters elsewhere in the circuit)

I was wondering if this is enough, since that would make the total bill of materials have just 6 logical circuits plus 1 or 2 latch ICs depending on whether the mapper would be SLOT2-only or SLOT1/2. This part is conveniently implemented using a single 74HC00 circuit, and I would want any changes (if needed) not to increase the IC count.

Once the doubts are resolved I'll finish the design and publish it here for all to use, like the previous design. Thanks in advance :)

LGTM

Mapper ready to be sent to fab. Will be doing some testing when it arrives. Meanwhile, I attach the schematic & gerbers to this post and the parts needed.

Qty Value Footprint
8 100nF 0805
1 Conn_02x20 Pin_Header_2x20_Pitch2.54mm
1 Conn_01x20 Pin_Header_1x20_Pitch2.54mm
5 10K 0805
2 74HC573 SOIC
2 74HC00 SOIC
3 74HC21 SOIC
1 74HC04 SOIC

mapper-module-v2-bw.pdf (83.83 KB)

Schematic

mapper-module-v2.2.zip (39.09 KB)

Gerber

Maybe the question is stupid, but for the #ROM_CE, why not directly using the #CE from the cartridge port?

NeonMan wrote

Time ago I made a working SLOT 1/2 mapper using 16V8 PLDs and while it worked flawlessly the 16V8 PLDs are increasingly harder to come by and especially harder to flash due to 16V8 programmers being discontinued, so I want to replicate the PLD combinatory logic with off-the-shelf 74xx series.

Hello,
Microchip (Atmel) still manufactures PLDs (both in THT and SMT) in 2022 like ATF16V8 and ATF22V10.
As for programmers, here are some of the options supporting those PLDs in 2022:
- TL866 ($50)
- TOP3000 ($80)
- Of course pro stuff like Dataman, Xeltek, but not cheap at all
- Ultimately you have the DIY solutions

I got around to actually doing testing of this device just today due to... Not having access to my consoles for different reasons. The result is a consistent "Software Error" so I guess it is time to find the errors in the implementation of the logical expressions.

It seems @doragasu is doing the same thing and I'm having the "almost there" feeling as well :)

I had a look to your schematic and everything seems correct, just two details that might be causing troubles:

* Your pull resistors might be too high. When I made my cart, I started with 47k and it wasn't working. I grabbed the oscilloscope and saw that when the latch output was deactivated, output capacitance discharge through the pull-down resistor was too slow and it took too much time to go to '0'. It didn't work reliably until I lowered the resistors to 4k7.
* I'm not sure this is really needed, but to be on the safe side, I would recommend using 74HCT chips instead of 74HC.

Hope you get it working soon!

Author	Message
NeonMan Joined: 24 Sep 2013 Posts: 141	Remaking the SEGA mapper clone out of discrete components Posted: Wed Jan 26, 2022 8:54 pm
NeonMan Joined: 24 Sep 2013 Posts: 141	Time ago I made a working SLOT 1/2 mapper using 16V8 PLDs and while it worked flawlessly the 16V8 PLDs are increasingly harder to come by and especially harder to flash due to 16V8 programmers being discontinued, so I want to replicate the PLD combinatory logic with off-the-shelf 74xx series. So far I managed to replace the 2 16V8 PLDs with only 6 logic circuits, however I have a question about the ROM chip !CE line. While the original !CE used the following combinatorial equation: !( !(A14 & A15) & CE & MREQ ) The current iteration uses the following one: !( !(A14 & A15) & CE ) EDIT: (notice CE / MREQ are not* negated, those signals come from inverters elsewhere in the circuit)* I was wondering if this is enough, since that would make the total bill of materials have just 6 logical circuits plus 1 or 2 latch ICs depending on whether the mapper would be SLOT2-only or SLOT1/2. This part is conveniently implemented using a single 74HC00 circuit, and I would want any changes (if needed) not to increase the IC count. Once the doubts are resolved I'll finish the design and publish it here for all to use, like the previous design. Thanks in advance :)

asynchronous Joined: 14 Aug 2000 Posts: 724 Location: Adelaide, Australia	Remaking the SEGA mapper clone out of discrete components Posted: Thu Jan 27, 2022 12:52 am
	LGTM

NeonMan Joined: 24 Sep 2013 Posts: 141	Posted: Thu Jan 27, 2022 9:11 pm
NeonMan Joined: 24 Sep 2013 Posts: 141	Mapper ready to be sent to fab. Will be doing some testing when it arrives. Meanwhile, I attach the schematic & gerbers to this post and the parts needed. Qty Value Footprint 8 100nF 0805 1 Conn_02x20 Pin_Header_2x20_Pitch2.54mm 1 Conn_01x20 Pin_Header_1x20_Pitch2.54mm 5 10K 0805 2 74HC573 SOIC 2 74HC00 SOIC 3 74HC21 SOIC 1 74HC04 SOIC mapper-module-v2-bw.pdf (83.83 KB) Schematic mapper-module-v2.2.zip (39.09 KB) Gerber

doragasu Joined: 03 Mar 2022 Posts: 55	Posted: Thu Mar 03, 2022 7:06 pm
doragasu Joined: 03 Mar 2022 Posts: 55	Maybe the question is stupid, but for the #ROM_CE, why not directly using the #CE from the cartridge port?

Apocalypse Joined: 26 Feb 2021 Posts: 155	Posted: Mon Mar 07, 2022 8:28 pm
Apocalypse Joined: 26 Feb 2021 Posts: 155	NeonMan wrote Time ago I made a working SLOT 1/2 mapper using 16V8 PLDs and while it worked flawlessly the 16V8 PLDs are increasingly harder to come by and especially harder to flash due to 16V8 programmers being discontinued, so I want to replicate the PLD combinatory logic with off-the-shelf 74xx series. Hello, Microchip (Atmel) still manufactures PLDs (both in THT and SMT) in 2022 like ATF16V8 and ATF22V10. As for programmers, here are some of the options supporting those PLDs in 2022: - TL866 ($50) - TOP3000 ($80) - Of course pro stuff like Dataman, Xeltek, but not cheap at all - Ultimately you have the DIY solutions

NeonMan Joined: 24 Sep 2013 Posts: 141	Posted: Fri Apr 21, 2023 4:57 pm
NeonMan Joined: 24 Sep 2013 Posts: 141	I got around to actually doing testing of this device just today due to... Not having access to my consoles for different reasons. The result is a consistent "Software Error" so I guess it is time to find the errors in the implementation of the logical expressions. It seems @doragasu is doing the same thing and I'm having the "almost there" feeling as well :)

doragasu Joined: 03 Mar 2022 Posts: 55	Posted: Tue Apr 25, 2023 6:28 am
doragasu Joined: 03 Mar 2022 Posts: 55	I had a look to your schematic and everything seems correct, just two details that might be causing troubles: * Your pull resistors might be too high. When I made my cart, I started with 47k and it wasn't working. I grabbed the oscilloscope and saw that when the latch output was deactivated, output capacitance discharge through the pull-down resistor was too slow and it took too much time to go to '0'. It didn't work reliably until I lowered the resistors to 4k7. * I'm not sure this is really needed, but to be on the safe side, I would recommend using 74HCT chips instead of 74HC. Hope you get it working soon!

Forums

View topic - Remaking the SEGA mapper clone out of discrete components