search

revaldinho / cpc_ram_expansion

A series of Amstrad CPC PCBs including a backplane, ROM and 512K and 1MByte RAM expansions.
GNU General Public License v3.0
56 stars 13 forks source link

464 RAM Card: Unreliable operation of Card/Reading of DIP Switches #4

Closed revaldinho closed 4 years ago

revaldinho commented 5 years ago

LambdaMikel reports some issues with his (early build) RAM card where the cursor in Future OS disappears or acts like an erase. This occurs frequently but not always, can be fixed with a power cycle on some machines and others are dependent on the type of ROM card being used.

This issue is most likely due to the CPLD incorrectly reading the value of DIP3. On the original boards the SIL pull-up/down values were 47K/10K respectively. This was found to lead to unreliable reading of the DIP switches, possibly due to pull-ups internally on the CPLD pins when tristated . ( There are different versions of the datasheet for example and CPLDs have been obtained from several batches.)

This issue has already been address on the other board builds, either by adding a 10K SIL in parallel with the 47K pulldown (on the first few built cards) or by switching to much lower resistances for pull-up and pull-down at the expense of a few mA of static current.

Action is currently with LambdaMikel to add a 10K pullup on the back of his early card between DIP3 and ground and report if this resolves the issue.

revaldinho commented 4 years ago

Closing this issue. No problems with the firmware RC-6-Fighter.