MOS 8501 CPU replacement

A 65C02 is not recommended because it has more outputs to which the adapter applies 5V, this won't be good for heat production. The 65C02 also won't support the undocumented opcodes, which you need for compatibility.Considering how easy it is to find NMOS 6502s, you should go for NMOS.

Sorry! It has been fixed.

Thank you very much for your quick response - and also for that great project! It helped me to revive two C16 and one Plus/4. There are just a view things to the BOM: In the schematics you've used Zener-Diodes. But in the discussion below you are mentioning 4148. I used 4148 and they work perfectly fine. You just need to make sure they have the right size. It turned out, that 0805 is the perfect size. I also tried 0603 - which work fine - but since they are so small, they are a pain to use. My diodes have the part number TS4148RBGCT-ND The "bigger" ICs also kept me busy a lot, since there are many variants of the same IC. I ordered SOIC. My exact part numbers are SN74HCT573DWR and SN74HCT541DWR. For the resistors and capacitors I would also recommend 0805. The bigger 1206 are already too big and therefor very difficult to solder. There is a little problem with the smallest IC, the 74HCT245. The chip - although TSSOP - is a little too big for the pads on the PCB. My parts number is SN74HCT245PWR. Maybe if you redesign the pcb once, could you enlarge the distance between the pads of this IC by 2mm or so?

Thanks for your feedback! Diodes are typically not measured in units like 0805, the BOM therefore says "SOD323" as the size you need to use. However... this indeed is quite similar to 0805 and your TS4148 is indeed "officially" classified as 0805. "SOD523", which is similar to 0603 works in my experience, but it not a great choice. I have used 1206 resistors, this did work quite well, but indeed 0805 is the preferred size. Your part numbers look correct to me and are Texas Instruments parts. I am using Philips 74HCT541D and 74HC573D. For the 245 I am using Nexperia 74HCT245PW. I'll keep your feedback in mind, there will be an updated PCB at some point.

The schematic shows Schottky-diodes, but Schottky-diodes are not a requirement, normal switching diodes work fine.

Schottky... of course! Zener would not be a good idea ;-) By the way - today I received W65C02-CPUs from China. I ordered 10. 7 did work. These were pretty cheap, but I don't know, why 2 did not work, although my BackBit ChipTester Pro tested them fine. 1 CPU failed completely. I'll try them also in a VIC 20.

You should not use 65C02, but the NMOS 6502! Using 65C02 is a bad idea because some pins that are NC on the NMOS 6502 are in use on the 65C02 and the adapter connects some of those. Check if your CPUs have not been relabeled.

Maybe you're right. I ran a burn-in test on my ChipTester Pro. It worked perfectly fine. I tried to remove the label with acetone, but this did not reveal something else. It's strange. Maybe these are rebadged 6502.... which would be great for me :D By the way - I also tried them in a VIC-20 and they worked fine there too.

The four resistors on the top of the PCB are 47kOhm, the two 5.6kOhm resistors belong on the bottom of the PCB. The BOM has been adjusted.

Thanks a lot! In the meantime I checked the schematics (on freepascal dot org slash tilde daniel page) and I got the same results... I managed to put one together, but it gives black screen. - I used HCT logic ICs instead of LS's. (HCT245, HCT541 and HCT 573). Do you think it could cause any problem? - I have only -25 rated GAL chips. I hope it should not cause any troubles.. - I have used 4148 diodes. Are they fast enough? (in fact LL4148 which is not the best candidate..) Thanks for your suggestions (I start checking my solderings). Uh-oh. Could you close the vias with mask? Sometimes it is really worrying to do some misconnection... And one more question (last one, I promise): why is the P1 pad exposed right beside R4?

HCT chips work fine. I do not see any problems with LL4148 as well, the diodes do not need to be very fast because the address bus gets the right value near the start of the clock cycle (when PHI2 is low), while the right value value needs to be at the input of the GAL at the end of the clock cycle, so there is quite a bit of time for the diodes to do their work. The P1 pad contains the unused but of the CPU port, bit 5, that is not available on the pins. I think the best explanation is still a construction error. It can be tricky in my experience and I have been stuck with a black screen quite a few times, finding the error after quite a of troubleshooting. Remove the 6502, connect R/W to ground via a resistor (1000 ohm should work). Connect A15 to Vcc with a resistor. Connect an oscilloscope to pin A0 of a ROM on the mainboard. Now connect A0 of the 6502 to either GND or Vcc via a resistor. You should be able to see the switching between GND and Vcc on an oscilloscope. Now change from A0 to A1, then to A2, until you have tested the entire address bus. Then also test D0..D7. In my experience this is an effective troubleshooting method.

Thanks a lot for your support, it works now. Unfortunately I received 65c02 chips from China.. (wrong date code on chip and JMP ($31FF) jumps to address of 31FF-3200). Otherwise C16 prompt is present! LL4148 (not a good choice) and with HCT logic chips.

Great to hear that! Just curious: Why do you think LL4148 is a poor choice? I am using 1SS355 myself and those work fine. I have also used BAT54C (a Schottky diode) and technically they were working great, but they did really suck during soldering, so I did stop with those. As for China, MOS is the brand with the most value, makes most sense to counterfeit chips as MOS. I have had better succes when buying Rockwell or UMC chips. Best is to use a professional IC trader, they are often in the far east as well, but they know how to avoid fake stuff.

Hello Daniel, the LL4148 is too long and hard to solder as it is round and hard to hold it with a tweezer... But it you use enough flux and solder it will work - as it does in my case. I will try R6502. Best wishes, S