I'm wondering whether anyone can tell me how the Apple 1 startup screen gets populated with alternating @ and _ characters? I just finished a second board and instead of alternating characters I'm getting an entire screen of flashing @ symbols, and the board doesn't run reliably. This board is populated with a lot of period correct Fairchild and Signetics ICs from the mid-late 70s, so suspecting that one of these was the culprit, I swapped all the ICs from my other 100% working Apple 1 board, but the problem remains. I'm now suspecting one of the passive components - perhaps a vintage Philips 22uF cap, but knowing how the startup screen gets populated with alternating characters would help narrow it down.
Anonymous
User login
Please support the defense of Ukraine.
Direct or via Unclutter App
Active forum topics
Recent content
Navigation
No Ads.
No Trackers.
No Social Media.
All Content Locally Hosted.
Built on Free Software.
We have complied with zero government requests for information.
Turns out the problem was indeed bad vintage Philips 22uF electrolytic caps. I guess this messed up the timing of the terminal section. I did notice that the flashing was quite a bit faster than normal, and one of these caps is involved with setting the timing of the 555, so that makes sense.
Hi,
You searched well. The "startup board" is generated using the 555 timer, corresponding to, inter alia, for the blinking of the cursor. The signal along with the pulses from the multiplexer output 74157 (C14) goes to the input 6 shift register 2519. Before the screen is cleared, inputs I1-I5 of chip 2519 are low, because - input 15 (E - CLR) of 74157 chip (C4) activates them when it is in the low state. In the high state, input 15 causes the outputs 4,7,9,12 (Za-Zd) to be low.
Input I6 changes its state to the clock cycle of the chip 555. Now look at the contents of the chip 2513 (character generator). The address 000000 corresponds to the "@" character and 111110 to the "_" character.
Hope I was right. If I was wrong, correct it.
Regards,
Mateusz / SQ9PXB
Thanks for the explanation, that's super helpful! I couldn't find this information anywhere and I really enjoy learning more about how the circuit works.
I'm surprised that the faster timing of the 555 caused only @ symbols though, and that they all flashed. But I estimate that it was running around double the normal rate, so could it just miss the _ characters in this case?
I wonder if anyone has any insight into another problem I'm seeing with this board? All the ICs on it were working perfectly in another PCB, but I recently completed a Newton-NTI board and moved them across to that. The problem I'm seeing is that the terminal section works fine, but as soon as I clear then reset the processor, I get a backslash, but usually the @ symbol appears right after it on the same line and it doesn't accept keyboard input. Sometimes I'll get a random character and then a backslash, and every now and then it does drop down a line before the @, but not every time. I've checked the address lines of the 6502 with a scope and it seems that after a reset all activity stops within a fraction of a second. Occasionally after many resets, the processor will keep working, but when I type just one key, it displays the character on screen and the processor stops again.
I've tried a different 6502 and 6820 from a good working board, also tried a different 74123 but the symptoms are the same. I have added a variable resistor and trimmed the timing of the 74123 to exactly 480ns, but that didn't make a difference either. One difference between this board and others I've built that work properly, is that I used 100nF film caps rather than ceramics to mimic the look of the NTI boards - could that cause a problem???
One of the many problems/quirks of the Apple-1 is the inadequate power supply bypass capacitors. On the prototype they used (at the time) super modern and high performance multilayer ceramic capacitors which - I think - came out of Japan, because at the time being only the Japanese could make these. Proof: photos of the prototype used in the ads.
For some unknown reason, the "BYTE SHOP" machines sold by Steve Jobs used the much inferior ceramic disc type capacitors of BEL brand which at the time were ubiquitous in Silicon Valley and were found in many of the arcade video games of the time being. On the 2nd production run aka "NTI" version they used the multilayer capacitors again, but unfortuantely I lack the insight if these were the same as seen in the prototypes from the ads.
This problem is how my "reliability mods" came about. If you add these additional bypass capacitors, the Apple-1 will work. If you further add the six damping resistors, it will even work better.
Other than that, the Apple-1 was a sound design. I have analyzed every nook and cranny of that circuit and it's sound. A testimony of Woz' skills as a digital designer. The infamous ACI, not so much. It's a piece of sh*t. Woz ain't no analog / mixed signal designer. As much as I'm no RF designer. If I would try to design a microwave circuit, it would be a piece of sh*t, too. But at least I can design digital and mixed signal circuits which are below RF.
We should be not too harsh on Woz. When he designed the Apple-1 he was young. With no experience. The fact that his designs of the Apple-1 and Apple ][ overall were sound, despite of the lack of experience, is just a testimony for his genius. It is easy for battle hardened verterans of the digital/analog/mixed signal craft to nit-pick on his designs but if you have 40 years in the industry it's easy to do that. And it would be unfair. I audited many job applicants (I had to) but only few would pass. Most lack even the basic understanding of Physics. Hey, if you can't write down the differential equations for the circuits you intend to design in that job, please don't apply and don't waste our time.
Woz never got this level of education and still he designed awesome machines. Only great minds can to that. And then, lacking the gurus, they fall on simple things like the RF performance of their bypass capacitors. I'm sure that they never put their capacitors on a Hewlett Packard network analyzer. Otherwise, the Apple-1 would have worked robust and reliably. All the other quirks it has are not fatal.
- Uncle Bernie
I had a similar problem, /@ in one line. Turned out it was a 47pF capacitor, I bought a dozen at Unicorn and they all behaved the same, and I didn't use a 27k resistor, just put 22k to be closer to the 480 ns value. Probably a bad batch... Later the other guy had the same thing, I suggested he change that capacitor and it worked. You can read more here in the Facebook group.
"I'm getting an entire screen of flashing @ symbols, and the board doesn't run reliably."
I just happen to had the very same effect today and it was because of a defect 0025. You might want to look into this.
EDIT: Saw that you solved it already.
I found the problem. It turned out to be a bad memory chip in bank X. I'm not sure how, but this one memory IC will prevent the board from running properly even when it's in one of the sockets of bank W. Once I narrowed down the problem chip and swapped it out for a good one, both banks are passing memory tests and the board is stable.
Thanks for the suggestions by the way!