Membership upgrades and revisions

Last updated Aug 18 2010. Edited by Herb Johnson, (c) Herb Johnson, except for content written by Lee Hart and others. Contact Herb at, an email address is on that page..


See the linked Web page about Lee Hart's 1802-based "membership" card and other 1802 activities and history. Lee produced a kit from this design: see the linked Web page for my construction of the kit and the notes and comments from Lee. This page is about revisions to the board or docs, and some possible upgrades, for the Membership Card. It's in a little bit of "flux" due to the Rev B release.

- Herb Johnson

Membership card Rev B release, Aug 18 2010:

Check the Membership Card home page for Rev B links.

Membership card revisions, Rev A, to Aug 2010:

See this linked note, incorporating most of the revisions found in Rev A of the Membership Card kit as discussed below. - Herb Johnson

Q: Where's the most current REv A schematic and notes?

A: This Web page will reference Rev A changes and link to Web pages about them. Check the Rev A kit pages and other pages on the Membership card home page.

Q: The IN or LOAD or RUN toggle switch operations are not working.

A: The 100K resistors for U8 are too low.See this discussion for details, but change R8, R9, R10 from 100K to 470K resistors.

Q: Some CMOS RAM chips go into low power mode when deselected. The Membership card design does not do this in all cases when it is not powered.

A: Lee came up with a lower power mod for the Membership card. See my tech note on this linked Web page.

Q: There's an ambiguity on the *schematic*: "all switches are shown DOWN" it says, but the pushbutton switch is apparently drawn UNPRESSED.

A: I changed note 6 on the schematic to read:

6. Toggle switches S0-S10 shown in "down" position. Pushbutton S11 shown in "unpressed" position.

Q: The latch U7, a 74HC373, does not "latch" the data from the OUT 5 or OUT 7 instruction

A: As of July 2010, the "fix" is to replace the '373 with a 74HC374. This will capture the data earlier. See this linked Web page for discussion. This fix may not be compatible with the "lower power mod" noted above.

Q: On the original Cosmac ELF, OUT 4 operates the LED's. That does not work on the Membership card.

A: The Membership card uses OUT 5 or OUT 7. Change the opcode from 64 (hex) on the Elf 2K software to 65 or 67 on the Membership Card. (Thanks to Todd Decker, Eric Tchong)

RAM power use depends on chip

March 26 2010:One other nagging question (for the next board):

The present board has a minimal address decoding scheme. Since there's only one memory chip, A15 is used as its chip select. This puts the single RAM or EPROM at all addresses from 0000 to 7FFF (hex).

Old 1802-vintage CMOS RAMs are true CMOS; they have a 6-transistor cell, and like other CMOS chips, they draw essentially zero power when all inputs are stable (at either 5v or gnd).

But the RAMs we can get today are all pseudo-CMOS. They draw essentially zero power when chip select is high; but considerable power when chip select is low (typically 5-50ma).

This means that if the 1802 is running a program in low memory, A15 is almost always low, so power consumption is basically set by that RAM.

Does anyone have:

a) A source of old bytewide RAMs? Parts like RCA CDM6116, CDM6264, MW6116, MW6264, Harris or Intersil HM6516, 6564, etc?

b) Ideas for modifying the design to keep the RAM chip select high as much as possible? If I had room for another IC, I could make RAM chip select only go low during TPB high (1/8th the time).

c) Are there operational tricks? For example, power consumption usually doesn't matter when running a program; it only needs to be "zero" when it is stopped (IDL instruction, reset, wait, load, etc.) Is there a way to make sure A15 high before stopping?

It's easy to test [current draw]. Stick one in a solderless breadboard socket. Wire all the address inputs, output enable, and write enable pins high. Measure the VCC current at 5v, first with the chip select pin low, and then with chip select pin high. All CMOS parts will have a VCC supply current of less than 1 milliamp with chip select high. But only the true CMOS parts will less than 1 milliamp supply current with chip select low.

--- Lee Hart


Rick Cortes notes "Reading the HM6264P and HM6264LP specs, the only difference I can see is standby current is a bit lower for the LP type. I looked at two different datasheets and got operating current at 7-15 mA in one and 60-110 mA in the other, but both had a current vs. frequency graph at < 1 mA."

Contact information:
Herb Johnson
New Jersey, USA
To email @ me, see
see my home Web page.

This page and edited content is copyright Herb Johnson (c) 2010. Contents written by Lee Hart, are copyright Lee Hart (c) 2010. Copyright of other contents beyond brief quotes, is held by those authors. Contact Herb at, an email address is available on that page..