Briel Computers support forum

Hello,
A few days ago I received my long-awaited kits, and yesterday I assembled the Altair.
I have to say that the quality of the kits is great. Just holding the bare boards is a pleasure, they're really professional and well done.
Anyway, I assembled the boards in about 4 hours (overall they were easier than the Kim) and I assembled the switches and LEDs in another hour.
The video out connector was broken, and the power connector was missing; but for now I just soldered two wires to where the connector would go, and use those.
Assembly was uneventful, and the silkscreening was so good that I didn't have to look at the instructions at all.
When it came to assemble the switches, at first I inserted the lower row of "single-throw" ones - they had to be really forced in their holes, to make them go all the way down to the board; and once pushed in, they wouldn't fall. Then I put in the front panel, screwed it in, and put just a little solder on only one pin of each switch. Then holding the panel upright, I would melt the solder on a switch one by one and fine-adjust its position by hand. Then I would solder the second pins of the switches, and then make a second pass with more solder on the first pins. Then I simply repeated for the "temporary" switches and for the upper row.

Then for the LEDs, I decided not to have them stick all the way out of the panel. Being a small kit, I thought the LEDs looked too "big" on the panel. So I decided to solder them so that only the "dome" of each LED would stick out.
To do so, I simply inserted all the LEDs in the board; then I screwed the front panel back in again, and made all the LEDs go into their holes; then I put all the upper switches to the "UP" position and the lower switches to the "down" position, and put the panel on the bench "face down", and it would stay put without toppling; then I took a bunch of photos - I chose the right amount to make the right height - and put them under the LEDs, between the LEDs and the bench, so that the LED heads would stick out for just about 3mm. Then I soldered them, and that was it.


When it came to turn it on for the first time, no magic smoke came out... Before inserting the chips I measured the voltages and they were OK. So I put the chips in and started the computer, which booted at first try.
But I think there's something weird with it, or at least, it behaves differently from what I've seen on a few youtube videos.

When I turn it on, with all switches down (except RUN which is in STOP position), I correctly see the two "banks" of LEDs lighting for a moment, and then the W0 LED lights for about 6 seconds, and after that M1 and MEMR light too and I can use the computer.
In some videos I saw, the LED that lights first is not W0 but INT, and it takes much less time before MEMR and M1 also go on and the computer is usable. Mine seems to have a sort of "slow boot".

Then, if I put RESET up, most of the LEDs light up, but they don't do so instantly, but a few at a time, over the course of a couple of seconds. Some lights sometimes stay off. And sometimes some of the lights "flicker" a bit, as if there was being some "action" there. It's not a cold solder joint because during "normal" operation the lights are rock solid.
And when I put RESET down again, then the computer behaves as if I just turned it on, with the W0 and then M1 and MEMR lights, only that W0 stays on for a longer time before M1 and MEMR also come up, sometimes even 13-15 seconds.

Also, is it normal that if AUX is up, then all the address and data LEDs flicker on for a moment everytime that I press EXAMINE or EXAMINE NEXT?


I tried toggling the test addition program, and it worked "some of the times" - i.e. with some input data. Using 1 and 2 as input data was OK, using 7 and 15 was not. But I can't be sure it wasn't user error, I have to experiment some more.

Does any of this make any sense to you?

Cristian

I have exactly the same behaviour with my Altair 8800 Micro. When I first tested it, I thought it was broken: the lights were on all the time and when I waved my hand near the computer, the lights followed my hand! I had a look at the board and found that when the reset switch is up - as it was - the ATmega controller's reset line is high and its output lines (to the 74HC595s) go to a high-impedance state. So, because the cable from the panel to the main board is a bit long, it picks up a bit of capacitance. Once Reset is released, this doesn't seem to cause any problems.

When AUX is up, the LEDs are updated for each simulated 8080 cycle.

For more information, see the link on this thread:
viewtopic.php?f=23&t=1033

I have not tried the test program, so I don't know if my unit behaves the same as yours - sorry. It took me some time to get used to the various modes: the PROTECT switch, AUX switch and RUN/STOP switches are the most important ones.

Thank you for your reply.
I seem to remember that I had 40-pin hard disk cables somewhere, and some were very short, I could try replacing them.
So you're saying that your Altair too lights the W0 led instead of INT when switched on, before being ready?

I also noticed something weird when accessing addresses with A15 on (of course they're beyond the available memory). But when I try to examine those bytes up there, the data leds light up but not all together, they come up one by one in random order over the course of 3-4 seconds. A little similar to what happens when turning on the RESET switch. I suppose this too has something to do with the cable?
Is there something I can do to "fix" these "irregular" lightings of LEDs?

Cristian

Hi,

I've had a look at the power-on LED pattern of my unit - here's what I get:

For these tests, the RUN/STOP switch is on 'STOP'.

With AUX and PROT up, the right bank of LEDs (A8 to A0, D7 to D0) light for a second, and then these go out and the left bank (MEMR to INT, A15 to A9) light up. These go out and WO comes on.

With AUX down, the same as above, but WO goes out and MEMR, M1 and WO light up again.

Regarding replacing the cable: I don't think that this will work since the problem is not the cable, but the fact that the 74LS595s are operating while the lines are floating. It's nothing to worry about - it's just a bit irritating. A fix would be to tie the 595's _OE line to the reset, so that the outputs are disabled while reset is active. I don't currently have the schematic available, but perhaps your best bet would be to ask Vince or Garry for their advice.

I tried having a look at the addresses with A15 on, as you describe. I get exactly the same behaviour: the data LEDs light up and stay on, and sometimes go off - as you say, it appears that this is similar to the behaviour with the RESET switch. I wonder what is going on with that? After I examined a few addresses with A15 active, I switched A15 off and tried to look at address $0000 - but the A15 light remained active, as if 'stuck'.

Lex

lexthehex wrote:Hi,

I've had a look at the power-on LED pattern of my unit - here's what I get:

For these tests, the RUN/STOP switch is on 'STOP'.

With AUX and PROT up, the right bank of LEDs (A8 to A0, D7 to D0) light for a second, and then these go out and the left bank (MEMR to INT, A15 to A9) light up. These go out and WO comes on.

With AUX down, the same as above, but WO goes out and MEMR, M1 and WO light up again.

Same here!

I get exactly the same behavior. I guess the Setup & Users Manual that comes in the CD ROM is not updated on the INT light NOT coming ON. In fact I have not been able to see it ON at any moment, except for the brief blinking of LED banks at power-on. So, the INT LED is installed correctly but maybe it works differently on 5.1.

In fact mine behaves exactly the same as Quaxo76 describes.

I was breaking my head on this, and reopened the case in order to check for continuity and resolder all relevant points related to the INT LED. Now I can rest and enjoy using my AMicro #193.

Yeah, the manual has the outdated steps. On the Altair forums is a link to the Addendum page, go get that, sorry guys. I need to update the manual.

Yes, thank you. I already saved some shortcuts to the Addendum page and other pages on the Forums.