Embarking on an Altair 8800C build

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

Re: Embarking on an Altair 8800C build

Postby TheoAU » September 13th, 2022, 4:34 am

BillO wrote:It's really hard to see the timing between the two traces. It would be great if you could turn up the graticule (all the little squares) so we could see more accurately.

In any case, it looks like the time between the fall of the yellow trace and the rise of the blue trace is less than 50ns. It should be 70ns or more. You might be able to adjust this by decreasing the duty cycle of the yellow trace. Try to make it more like 50% rather than 56%. You can do this by decreasing R42. 5.6K might work. And you might try reducing the duty cycle of the blue pulse. Right now you are well over 100ns but only need ~70ns-80ns or so. Try 10K or so for R41. That should give you ~100ns. Maybe Mike has a better suggestion, so wait until he responds.

Also, the voltages seem way too low. They need to be 9V or higher. Yours look like they just over 4V. Not sure what's causing that. Can you make sure IC N is a 7406? Also check to make sure resistors R29, R30, R32 and R33 measure to be 470 ohms? I had to decrease those resistors on mine a bit, but even so I was not off by that much. Also check that you are getting 12V across D1, you might have put a 5.1V Zener in there by mistake.

Edit: the implication of putting a 5.1V in for D1 means you might have put a 12V in for D2 which would mean the -V rail is getting ~-8V (or whatever you set your PS to). Better check that first.


I checked IC-N and it is a 7406. I also checked D1and I'm getting over 12v on D1. I recall checking both D1 and D2 as part of my initial checks after the build and they both got the correct voltages. I'm starting to see what you mean about the low voltage though. I just uploaded another screen shot from the clock test I did on the 8800b CPU board. And the voltages that board is showing are more like what you're saying. Ironically that CPU board doesn't work and my reproduction CPU above seems to work fine. I'm not sure about the resistors. Is there a way they can be tested on the board or do they need to be removed?
Last edited by TheoAU on September 13th, 2022, 8:05 pm, edited 1 time in total.
TheoAU
 
Posts: 50
Joined: July 21st, 2022, 2:59 am

Re: Embarking on an Altair 8800C build

Postby BillO » September 13th, 2022, 8:45 am

If you have 12Vdc across D1 and 5.1Vdc across D2 then they are likely fine. There should be no ripple as you are using regulated supplies ahead of them.

You can see if they have the correct forward voltage (usually about 1V) by taking the board out of the backplane and testing with a decent DMM. I've heard of DMMs that will check the Zener voltage but have never actually seen one. PEak also make a device called the ZEN50 that will test them, but it's not very cheap. Digi Key have it for about $170.

There seem to be quite a few videos on U-Tube about it.
BillO
 
Posts: 136
Joined: November 11th, 2020, 6:29 am

Re: Embarking on an Altair 8800C build

Postby BillO » September 13th, 2022, 3:04 pm

Oh, just thought of something. Where did you take those scope readings, specifically?
BillO
 
Posts: 136
Joined: November 11th, 2020, 6:29 am

Re: Embarking on an Altair 8800C build

Postby TheoAU » September 13th, 2022, 8:13 pm

BillO wrote:Oh, just thought of something. Where did you take those scope readings, specifically?


The readings were taken on pins 5 and 13 of IC-Q, as per Mike's instructions above in this thread. Pin 5 is phi 2 and pin 13 is phi 1.

Also, I meant to ask about checking resitors. Can they be checked on the board or do they need to be removed?
TheoAU
 
Posts: 50
Joined: July 21st, 2022, 2:59 am

Re: Embarking on an Altair 8800C build

Postby AltairClone » September 13th, 2022, 11:14 pm

As is very common on the 8800 CPU board, you need to at least shorten the phi2 high time to increase the time from its falling edge to the phi1 rising edge. However, before making any changes, switch to 1v/div so the signals aren’t so short and repost pics. The rise and fall time will look more “normal” then.

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

Re: Embarking on an Altair 8800C build

Postby BillO » September 14th, 2022, 10:09 am

TheoAU wrote:
BillO wrote:Oh, just thought of something. Where did you take those scope readings, specifically?


The readings were taken on pins 5 and 13 of IC-Q, as per Mike's instructions above in this thread. Pin 5 is phi 2 and pin 13 is phi 1.

Also, I meant to ask about checking resitors. Can they be checked on the board or do they need to be removed?


Ah, okay. That makes sense. Those voltages shown are fine given those locations. It might be an idea to check at Pins 8 and 2 of IC N as well to see if the voltages are correct. Then you don't have to check the resistors if the voltages are over 9V.

Also as Mike and I suggested shortening phi2 would improve things. As mentioned previously replacing R42 with 5.6K would be a good first try.

(plus brightening up the graticule would be awesome to help see the trace to trace timing better)
BillO
 
Posts: 136
Joined: November 11th, 2020, 6:29 am

Re: Embarking on an Altair 8800C build

Postby TheoAU » September 15th, 2022, 7:25 am

I think I'm starting to understand this. Here are some more captures of the clock. The only difference between these three captures is the time scale. The first in 10ns/div, second is 25ns/div and the third is 50ns/div. As Mike suggested, all captures are at 1v/div. I tried to get it to zoom in more with the 10ns/div screen as BillO suggested. I also used the T cursor to get an idea of the time between the falling edge of CH2 and the rising edge of CH1. It seems it's at around 35ns. Which is half what it should be going by what you guys have said. It seems to me that the width of CH2 crest is 288ns. If this was shorter by around 30ns to 40ns, it would increase the gap between the edges to what's needed.

I hope I've got some of that right.
Attachments
3.png
1v/div and 50ns/div
2.png
1v/div and 25ns/div
1.png
1v/div and 10ns/div
TheoAU
 
Posts: 50
Joined: July 21st, 2022, 2:59 am

Re: Embarking on an Altair 8800C build

Postby BillO » September 15th, 2022, 9:18 am

You're getting the idea. R42 is the one that needs changing. The next common value down would 5.6K That may indeed not be enough and you may need to go to5.1K or even 4.7K but that timing needs to be increased. 70ns is published minimum so aim for 80ns or 90ns. You have plenty of pulse width on phi2 to spare. Another way to look at it is you want the make the phi2 duty cycle 50%.

Also, don't forget to check the clocks at pins 8 and 2 of IC N as well to see if the voltages are correct (9V or more). Mine were too low and the rise times were shoddy.
BillO
 
Posts: 136
Joined: November 11th, 2020, 6:29 am

Re: Embarking on an Altair 8800C build

Postby AltairClone » September 15th, 2022, 11:49 am

Yes - good looking scope shots and it sounds like you have your head wrapped around the issue.

The minimum phi2 high time is 220ns, so I don't like to take all the time we need from phi2 - that leaves us too close to the 220ns parameter.

The minimum phi1 high time 70ns, so we can take some time from there as well, which adds to the time between phi2 fall and phi1 rise that we need.

Assuming R42 for phi2 is presently 6.2K, then 5.6K will take you from 280ns on phi2 down to around 250ns. That gets us about 30ns

Assuming R41 for phi1 is presently 13K, then 10K will take you from 125ns on phi1 down to around 95ns. This gets us another 30ns. This give us 60ns total additional time between phi2 fall and phi1 rise.

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

Re: Embarking on an Altair 8800C build

Postby TheoAU » September 17th, 2022, 4:35 am

BillO wrote:You're getting the idea. R42 is the one that needs changing. The next common value down would 5.6K That may indeed not be enough and you may need to go to5.1K or even 4.7K but that timing needs to be increased. 70ns is published minimum so aim for 80ns or 90ns. You have plenty of pulse width on phi2 to spare. Another way to look at it is you want the make the phi2 duty cycle 50%.

Also, don't forget to check the clocks at pins 8 and 2 of IC N as well to see if the voltages are correct (9V or more). Mine were too low and the rise times were shoddy.


Hi BillO,

Here's a capture of IC-N, pins 2 and 8. CH1 is connected to pin 2 and CH2 to 8. Voltages seem good, 13v on both.
Attachments
IC-N_pins_CH1_pin_2_and_CH2_pin_8.png
Last edited by TheoAU on September 17th, 2022, 5:36 am, edited 3 times in total.
TheoAU
 
Posts: 50
Joined: July 21st, 2022, 2:59 am

PreviousNext

Return to Altair 8800c

Who is online

Users browsing this forum: No registered users and 5 guests