No clock signals

Discuss construction, troubleshooting, and operation of the Altair 8800c computer

No clock signals

Postby TronDD » August 3rd, 2020, 4:27 pm

I'm stuck and don't have anything to compare to to know if what I"m looking at is good or not.

Got my 8800c all together, currently using the standard recommended boards. When I turn it on, I get all the data lights on, and some random amount (usually all or most) address lights on. None of the status lights are on. None of the front panel switches do anything.

Probing around the bus and CPU card, I don't get a square wave on the clock, Phi 1 or Phi 2 lines. No square wave at the CPU Phi inputs.

I understand you can't just measure the output of a crystal so I don't know what I should be seeing where. The crystal has about 1.25V going in and just millivolts coming out. That reads as a continuous low and propagates through the clock circuit as such so I get Phi 1 and Phi 2 always low and /CLOCK is about 3.6V on the bus.

I've tried 2 crystals. Looks the same. Tried another inverter which bumped the High output to 4.5V but still no square wave.

I'm assume absolutely nothing can work without a clock. Am I really 0-2 on crystals or can there be something else?
TronDD
 
Posts: 40
Joined: November 20th, 2018, 8:51 pm

Re: No clock signals

Postby AltairClone » August 4th, 2020, 10:13 am

The schematic for the oscillator is on the first page here: https://deramp.com/downloads/altair/har ... matics.pdf.

The clock was known for not always starting, and even when running, some resistor adjustment(s) in the one-shot timing that generates phi1 and phi2 was often required. There is a MITS recommended clock mod to improve oscillator start-up reliability in this Computer Notes newsletter https://altairclone.com/downloads/compu ... _02_03.pdf

But before performing that mod (it's fairly involved), I would double-check all resistor and capacitor values in the oscillator circuit. For the 7404, make sure you're using a straight 7404 and not an 74LS04. Is the crystal you chose designed for a series or parallel resonant circuit?

Mike
AltairClone
Site Admin
 
Posts: 632
Joined: April 5th, 2013, 10:55 am

Re: No clock signals

Postby TronDD » August 4th, 2020, 11:09 am

AltairClone wrote:For the 7404, make sure you're using a straight 7404 and not an 74LS04.


You're kidding me? That was it. IC I is also a 74xx04 and that's where my plain 7404 went. I had the 74LS04 in the clock circuit. Swapped them and there's my clock. 1.999999Mhz on the scope.

Still getting wonkiness on the front panel, but it's doing stuff now.
TronDD
 
Posts: 40
Joined: November 20th, 2018, 8:51 pm

Re: No clock signals

Postby AltairClone » August 4th, 2020, 1:32 pm

Next step will be to look at phi1 and phi2 overlaid on a scope at the same time. The early 8080's were very particular about the timing of phi1 and phi2 relative to each other. It's quite common to have to adjust R42 and sometimes R41 as well to bring the clock within spec.

Mike
AltairClone
Site Admin
 
Posts: 632
Joined: April 5th, 2013, 10:55 am

Re: No clock signals

Postby TronDD » August 4th, 2020, 3:18 pm

Yup, that's where I'm at. I'm using the timing graph from the newsletter you linked.

The stock setup had the end of Phi2 really close to the start of Phi1. Jerry from JMPrecision recommends a 4.3K resistor in R42 over MITS' 6.2K. That shortened Phi2 to get the right separation, but now Phi2 might be too short.

Code: Select all
Timing | Expected      | stock   | R42 = 4.3K
tCY      | .48 - 2     | .5      | .5
tr,tf    | 5 - 50      | ~5-10  | ~5-10
tPhi1    | 60 -        |130     | 130
tPhi2    | 220 -       |300     | 210
tD1      | 0 -         | 35     | 30
tD2      | 70 -        | 20     | 105
tD3      | 130 -       | 175    | 175


Do I need a little more resistance on R42 to get Phi2 back over 220?

I was about to say the computer still isn't working but I went to reproduce the behavior to document it and now it seems to be working correctly. Time to pop a program in and see if it runs.
TronDD
 
Posts: 40
Joined: November 20th, 2018, 8:51 pm

Re: No clock signals

Postby TronDD » August 4th, 2020, 3:36 pm

And now it's not working.

Examine always goes to 000070Q regardless of address switches.

Examine next executes a stack operation for 2 clicks then does whatever for a click before going to the stack again.

Deposit always deposits 377Q regardless of the switches.

Single step seems to do what's expected and executes whatever was randomly in memory. Run seems to execute random memory contents as well. I can't examine a PROM address to try to execute actual code.
TronDD
 
Posts: 40
Joined: November 20th, 2018, 8:51 pm

Re: No clock signals

Postby TronDD » August 4th, 2020, 7:29 pm

Sorry to keep talking to myself. It's working again. :D

I went to a 4.7K resistor and that brought all the numbers into spec.

It still had the same behavior as before. I fiddled with the FDC+, turning the RAM off and on again and it started working.

Entered Kill the Bit and killed some bits. Power cycled it and it was still in a functional state.

Maybe the FDC+ wasn't tight in the slot.
TronDD
 
Posts: 40
Joined: November 20th, 2018, 8:51 pm


Return to Altair 8800c

Who is online

Users browsing this forum: No registered users and 3 guests

cron