Updated power supply and the RTC card
 

The Altair Real Time Clock card allows for two independent real time clocks for interrupt generation. One clock is crystal controlled while the other is based on a presumed 60 Hz line frequency. This second clock is generated by a circuit that watches one of the S-100 unregulated voltages for a residual sinusoidal wave.

I was wondering: is it possible that the updated power supply is too good to have a sufficient residual 60 Hz signal for the original RTC circuit to work?

The Altairkit power supplies are switching units instead of the original linear supply, and you're right, they do completely block any 60Hz ripple.

It looks to me, though, like the RTC board could only work off of the system clock in a standard 8800 anyway. The RTC manual says it can get its timing signal from the 2Mhz clock or by picking up a 60Hz signal on the unrectified voltage on bus pin 55, but according to the 8800 manual, pin 55 isn't connected to anything. It is labeled as RTC in the 8800b manual and flagged as a new signal for the 8800b.

Geoff.

My interest in the RTC card stems from a need to have a reliable, high resolution periodic interrupt in one of my probable applications. It doesn't matter if it's from the system crystal or from the line frequency as long as it's there. I don't recall if there were other cards that generated periodic interrupts; perhaps the disk controller?

I'd like to hear from anyone who has gotten one of the new, super combo S-100 cards (64 KB RAM, flash, disk emulator, I/O). Do these actually exist yet? If so, do they work? If such are available, then that would strongly influence the ordering and use of other cards. And if they can generate a high resolution periodic interrupt, then so much the better.

Alternative 60 Hz input
 

For only a few dollars, a single chip all-in-one microcontroller could be programmed to generate a 60 Hz sine wave and feed it to the S-100 RTC pin. Alternatively, the output could be a square wave and be sent straight into the RTC card's counter, bypassing the Schmitt trigger input.

But the main advantage of choosing the 60Hz signal is how precise it is, at least over long time spans. A microcontroller generated signal would be no more accurate than the system clock.

If you need the precision of the line frequency, couldn't you hook up a small transformer and feed an unrectified 5v to bus pin 55 and let the VI/RTC card work like it would in an 8800b?

Geoff.

I'd guess that the 60 Hz signal, or any other signal source, should have the same electrical ground as the circuit that's receiving it. There has to be a return path for the electron flow. Without this, it would be like trying to take a measurement with an oscilloscope probe without connecting its ground clip to the test circuit's ground.

Re: Updated power supply and the RTC card
 

I have an RTC board and have it jumpered to use the 2 MHz crystal for a time base. Since I'm only going to use it to generate regular interrupts for MITS Time Share BASIC and not for an accurate clock, I don't care if it drifts a little as long as I keep getting interupts. One question I do have, though is that the manual for the RTC specifies that:

"If you are constructing the 88-RTC, a hardwire connection must be made between the Mother board and the power supply board. This connection brings the unrectified +16 volts to buss #55 for use as the line frequency input for the 88-RTC."

Is this nessa* even when the board will be using the 2 MHz crystal? If so, then I won't be able to use the 88-RTC without the mods listed in the thread. If it's not nessa*, then I'm all set. I can't wait to get my Altair replica!

* A humorous substitute for "necessary" from Mel Brooks' film, "High Anxiety"

No, you don't need to connect buss line 55 if you are using the system clock (pad S jumpered to CF). You need unrectified voltage on line 55 only if you are using line frequency to supply the clock source (pad S jumpered to LF).