I am happy to say that an "Apple 1" was born tonight. I spent the last week troubleshooting. Many thanks to Corey for graciously letting me bounce questions off him. As it turns out Unicorn supplied a DS0026 in my parts kit. Corey pointed me to a bunch of references dealing with the 0026 Vs the 0025. After contacting Unicorn on Friday Rob quickly shipped out an order I placed for a DS0025. I got it today. Once the chip was installed I still had a problem, but I had eliminated the possibility of the 0026 making the rails too noisy.
I think the thing that helped was understanding the actual circuitry. SB-Projects was a great resource, thanks San!) Once I could tell that the shift registers weren’t clearing it pointed to the clock driver again. It may have very well ran on an 0026, but I never tested with it once I got my 0025. Once the 0025 was in I scoped the actual two phase clock. One “leg” wasn’t hitting all the way down to the negative rail. After a while of making sure the inputs were equal to the clock driver I assumed that I may have had a 2504 loading down the clock. So I pulled three of the 2504’s and the clock signal looked good on the scope with both legs hitting the rail. I put the pulled chips back in one at a time until I found the bad chip. luckily I had bought a complete set of spares from Unicorn. Plopped one in, when I tested the clear screen everything worked perfectly.
As part of my trouble shooting I had swapped out a lot of the TTL with newer IC's. Interestingly enough when I went back and repopulated the older TTL the screen clears and the computer runs. (The character set print routine). Although the computer terminal boots with random characters in the shift registers, not the uniform @_ character filled screen.
I'd also like to thanks Mike W. and many others who I have gained a lot of knowledge from to help me get this done.
CONGRATULATION ! and best wishes for evenings with discovery of early computing experiences...
including the challanges of realizing own ideas at the machine by hard- and software....
Lol, I'm just playing catchup with the Apple 1. My first experiences date back to around 1974/75. I designed and built my own 8008. Unfortunately my original computer went to the good will. You can check out some of my stuff at 8008chron
For all of those wanting a closer look here are some higher res pictures of my board:
Len's Apple 1
Just gave a quick once over on your board using my tiny iPhone screen.
You will have problems with the 74LS257 instead of the 74S257. It causes issues with the ACI.
I'll take a look over later on a big screen.
Ah crap... I missed that. I'll swap those out. The ACI was the next thing I was going to play with.
Very nice job, I'm glad to see another one up and running.
I'll just add a observation to share with others that may be contemplating building a Mimeo or other clone.
I would say that Len's bring up experience is about typical, sometimes you are lucky, and they work the first time and other times, you'll have multiple problems to overcome.
Troubleshooting skills are helpful to have when executing these projects. Even if you do a careful job with soldering, there is always the chance of a bad component, a bad socket, a bent under leg or a PCB manufacturing fault. On the Apple 1, except for intermittant DRAM issues due to the Apple 1's noisy power rails, problems are a lot more common with the video circuit than the microprocessor section. There have been a boat load of bad 2504's floating around the last couple of years. Some of the bad ones get real hot, so you can sometimes debug just by touching them.
That said, through email, phone and instant messaging, I've helped a number of people learn the basics of digital circuit troubleshooting while bringing up their Apple 1 clones. Most people can learn how to do it, but it takes a great attitude, time, persistence, and a decent oscilloscope.
I used the info from your build guide as a reference, as well as some postings on your site. Very helpful. When I build I assemble based on altitude of the parts. Diodes, resistors, IC sockets, capacitors, etc. It has been a pretty safe way to go. I also tend to just go one part at a time. I'm not to worried about building fast, I'm more concerned about quality. You guide made a great check list.
I just plugged in the chips to my cassette board/kit that I got from you. I think the first thing that I will load is your memory test. Between you and Corey I get to stand on some pretty tall shoulders.
I've been wanting to build an Apple 1 for a very long time. Ironically I emailed Woz back in early 2002 looking for artwork to build my own board. I was really impressed to get an email back.
May I also suggest you implement Mike's suggestions in resolving the DRAM noise and video noise (perhaps after your build is fully debugged and working). It has made my Apple 1 (Obtronix) solid. Before these fixes, I had many memory issues and the video quality was poor. You can make the fixes on the bottom of the board where they aren't visible.
What kinds of problems were you seeing? Mine seems to be very stable. I got Mikes cassette interface running tonight. I loaded micro chess, Mikes memory test program, and basic. I let the memory test program run for an hour without seeing any errors. Basic also came right up and I wrote a program that runs fine.
When I built my board I bought a ton of .1uf bypass caps. I went through the entire lot and selected the highest values of the batch. All were over .1uf. And used those on my system. I swapped the ls257 chips this afternoon as Corey suggested.
So far I'm not seeing any quirks. If there is something I can stress test the board with, let me know.
I also have a fan running on the 323k heat sink to keep things cool. So far I'm very pleased!!!
it´s the same thing like with the Apple II.....
the base system is rather stable..... even more stable with the fan.....
in general the problem start raising ahead when you start expanding the system beyond the ACI or the
CFFA1. Some of the Noise gets lurking tricky around the corner when expanding the system beyond the
one expansioncard at the system....
at the Apple II it starts when expanding beyond 4 cards that some systems start getting instable....
So it seems at least partially related to the limits of the powersource, and when expanding the
adressinglines and datalines beyond the limits of the mainboard due to problems related with
the capacity of that lines when expanding the system....
Regarding the memory problems with the Apple 1, it's a marginal problem and some systems are perfectly fine. Others will experience random application crashes, and the memory test will fail after a few passes. I've found that if the memory test passes against both banks for 5 or 10 minutes, everything will be fine with applications. I've been able to get systems working by cherry picking caps and sometimes DRAM, without resorting to adding extra capacitors to the -5 volt rail, but I think it depends upon the particular parts you end up using in your build.
I've never had a need to run a fan, but I've never run one in a full enclosure, either. That includes multiple weekend shows, where I ran the system all weekend. I may consider putting a small fan in the replica wooden enclosure that I will eventually build.
I also now build based on altitude of parts. I haven't updated my manual since changing my methods, and probably will not, since the sockets first approach also works pretty well. These days I usually populate resistors and similar small axial leaded parts first. After those are in, sockets and then taller parts like ceramic capacitors. This altitude approach works better on the SCELBI boards, but there are dozens of resistors on most of the SCELBI boards.
It makes sense that the problems come with the loading of the PS.
I got my ACI going wonderfully last night! I'm going to try and load up my CFFA later today. I'm deciding what to do next on the expansion. It's an internal debate I'm having in my mind on how to deal with the case and the possibility of having space for an appendage sticking out the side. I am leaning towards a case built of wood.
I think possibly an extra slot or two would be all I would ever use, maybe just for a serial port. Since I've just got things running I'm still thinking about what would be fun to add.
A few years back I built a Transistor Clock. There was something like 2700 solder joints for over a 1000 parts. Working up by size was the only practical way to get a uniform board. My work area is my glass office desk, so I have a really flat surface to work on. (Albeit very cluttered. I think everyone finds what works well for them...
Now I do a lot of surface mount. In a way I find it easier than through the hole. You don't have to trim up the leads after. For a long time I avoided going SMD, until I realized how much of the electronics world I was missing out on. Although I try to not mix vintage and SMD unless I need to so something special, like a USB memory loader, etc.. My technique with surface mount is to usually mount the highest lead count devices first. That way if something goes bad I didn't spend a a lot of time populating all the small parts.
I thought your documentation was done really well. It helped me considerably.
When you use the memory test to run against bank 0 do you alter the code to look at the beginning of E000 for the configuration bytes? i.e. DO you load the code up at E280?
Which memory test program are you using?
As Speedy mentioned, I would see problems when running programs on my CFFA1 that spanned the memory range. My CFFA1 port of figForth, in particular, caused problems. Oddly, the memory test ran fine. I also had the horizontal spans in the characters much brighter than the vertical spans. With Mike's fixes, figForth runs forever and the video looks like an Apple II. Of course, you may have no issues with your build. I'm sure cherry picking caps is quite beneficial, but I like the extra margin offered with a few simple and transparent updates.
The one that Mike did: http://www.willegal.net/appleii/6502mem.htm
I got my CFFA1 up and running tonight. I'll have to load up figForth tomorrow and play a bit and see if I can see any issues.
I went through several of my CF cards before I could get one to format. The transend cards had issues. The sandisk 4 gig worked fine, although a bit of overkill.
SO far so good!
What a blast. I spent an hour or so tonight playing Corey's lunar lander. It was like traveling back in time. It is amazing how satisfying it is to play a simple program on a home built computer. I not sure most people would appreciate it as much as I did. But after collecting all the parts, soldering up the board, trouble shooting it to get it running, it brought a lot of satisfaction to play a game.
I'm running on a little 5.5 inch green CRT monochrome monitor I bought off eBay. Makes it feel a bit more authentic, along with my home built keyboard. Now I just need to build a case.
Thanks Corey for the memories.
I did have to chuckle to see how long it took to "list" out the entire program.
All in all everything is very stable. No crashes or errors.
Are you aware of Corey's version of lunar lander of the Apple 1 that has sound?
Hook up a small amp with speaker to the ACI out. Lunar Lander was an experiment for VCF a few years ago to demonstrate ascii game graphics and audio capabilities of the Apple-1 with ACI.
I have an little LM386 based 9 volt battery powered, amplified, speaker that I built into a small project box. I made it originally for another project. It was dirt simple to hook it up to the ACI output with alligator clips in order to generate sound output on the Apple 1 with Corey's modified lander program.
There are LM386 amplifier plans all over the internet.
I'll plug in a speaker tonight.
be aware that if you hang up a speaker :
1. you will hear nearly nothing...
2. if you do this for time more than 5 minutes you may damage your ACIcard !
the Output of the ACI has about 5 kiloOhm impendance ( or resistance if you like this more )
and the spearker in General has only 4 Ohms to 16 Ohms resistance....
this causes the Output of the ACI to permit output at more current than it is designed for....
It´s therefor recomended to use a amplified speaker-system....
I have 2 different very simple and cheap application pages for simple active speaker-systems
from RC-pages related to audio for shipmodels running by battery power:
and some additional ones listed in the left side at top of the page at:
depending to the kind of chip that you have availiable at your place....
though the pages are in German language the circuitplans and pictures can be
understood by everybody without knowledge to German language
Speaker was a general term... I actually play an engineer in real life.
My last audio project is shown here:
Click on the pictures link and scroll down.
i just read the general term and wanted to protect your ACI from possible damage... nothing else...
glad to hear that you know about the issue...
just followed up your given link...
reminds me to similar projects issued in early 80´s at the elektor ( formant ) and
similar projeckts published in the Germen ELRAD magazin....
maybe you´ll find some interesting stuff at the bottom of my main Apple Box pages
below the Term "Synthisizer projects for Musicians"...
at the bottom i placed a bunch of links to that various synthisizer projects from the early 80´s
containing also a bunch of circuitplans and pictures :
Oh Cool SpeedyG,
Those look great. I'll have have to see if I can run some of the PDF's through Google Translate.
i quess that won´t work, because they are generated from picture scans....
but i still have the original scans and OCR Software.....
and some parts of the projects i even have as issues from Great Britain too... !
so maybe i can even mail you from some projects the related pages from original english
issue.... best would be to make a list from which parts you want translated text
and send the list here to me by pm....
then i can check if original english text ist also availible in my archive....
and then i might be able to mail you pdf´s from that files....
i believe this would be first and best option and then you might perform attempt
to google translation as second option...
that parts have been made when the appleII-box has been only in German language....
when i started later to switch to english for international public i have not spent
the time yet to create the according english pages yet due to other tasks....
that´s the reason that pages haven´t been translated yet...