Hi everyone,I hope you're all doing well?
This is my first post since I started building my Apple-1 replica.
I’ve read a lot of the resources on this forum, but today I’ve hit a wall (probably because I’ve also reached the limit of my electronics knowledge...).Over the past few months, I’ve bought (or been given) most of the components to build my Apple-1 replica.
All voltages are okay.
I have a real 2513, 2519.
Here's a glimpse of my fully populated PCB:
Today, I’m finally seeing some results (i.e., I have a display!), but there’s still a bit of a way to go before I have a working replica...Here’s where I’m at:
I’ve populated all the video components (and more).It seems to me that I have a “flickering”/oscillating signal (with the DS0026). When I don’t install an AM1404, I get a string of @_@_@_@_, where the @s are blinking. That seems a normal situation from what I've read so far.
When I install the AM1404s—using an initial batch from China (I read post from Uncle Bernie explaining those where counterfeits...) —I get a completely static screen of @_@_@_@_, except for the first column of @s, which flash.
See below:
So I figured my AM1404s were fakes (and probably ICs of a different type actually...).So I bought some other AM1404s that were supposed to be good. But I’m still skeptical. When I installed them this time, I got a screen filled with @@@@@@ (so 0s, if I understand correctly) that are all blinking.
So to this point, I unsure when I am...
I’m inclined to think that none of my AM1404s are working. Real or working ones are so hard to find in 2026...But I’m not sure.Maybe my problem lies somewhere else? Are there other components I should test before?
I detail that seems important to mention: When I try to send a CLEAR signal (via the keyboard pins), my screen goes blank briefly, but as soon as I let go, it returns to its @_@_@_@ or @@@@@@ state. I never get a single blinking @. Same for the RESET signal.
So I need your advices...
I have an oscilloscope and a multimeter, and I’m ready to test anything you might suggest.
Thank you for your help guys! Much appreciated.
Love from France,
Jonathan.
Hi Jonathan
Great job on your build so far!
If your AM1404 are from AliExpress and/or are laser engraved rather than paint then most likely are counterfeit.
There is a 1404 / 2504 test board you can build to test your ICs or I could post you one from the UK or if you wanted to wait a month post one from France?
Thank you @Sarnian!
Let me send you a PM on this. Sure I'm interested on having a tester for these AM1404s.
Still I’m puzzled by the fact that:
So that makes me think that the problem might actually lie upstream of the AM1404 shift registers (which, of course, need to be tested).Could it be something with the 74166 or 74161 chips, aside from the AM1404s?
Thank you guys.
If you have a T48 or similar ROM burner, the software has an IC test suite that may support the AM1404. That has been my goto for suspect chips. It does not have a pattern for everything, but might get lucky. If you don't already have one, they can be found on the jungle site fairly cheap.
Thank you for your feedbacks, appreciated!
Okay I will get a tester for these chips. And also contact the seller if they don't work.
I noticed that the 74166 tends to heat a bit. Could it be useful to consider having spare parts for it?
Sorry, but as far as I know, XGecu programmers don't have test vectors for the 1024-bit shift registers 1404/2504/2804.
In my opinion, the easiest way to test these chips (as a preliminary check) is actually that 2504 tester from Digiteka/Claudio (P-Lab).
It has always tested correctly for me without exception.
1024 bit shift register IC tester
IMG_4800.JPG
holy cow! I fogot the early chips came in cans!
Not that I have an escuse to forget, I have a few hundred canned op-amps somewhere in my lab. You are welcome to some, if you need.
Just got an AM1404 tester and checked all those chips I received and guess what... ;-) All wrong!I need now to (try to) get refunded and find out some working ones (probably with TO-5 package...since real DIP-8 are so expensive or hard to find nowadays...).
IMG_1035.JPG
Just by the date code, I would never expect those chips could be genuine. The type of shift registers used in the Apple 1 were made from the early 1970s until the mid 1980s†; no way they were still being made in 2009. The shape of the packages' index notch is also wrong.
† The 1404A is mentioned in an application note from AMD published July, 1973, "Applications of Dynamic Shift Registers". The last AMD data book to mention them was issued in 1981.
When chips this old and rare are for sale, they most often are not coming from new-old-stock in sealed tubes: so different chips in a lot will not carry the same date codes. Instead, they will be a mix of date codes across the years of production.
You could have saved all the effort of testing by just thinking harder about how these chips were marked.
AM2804PC is the drop-in substitute and they still are available:
https://www.digipart.com/part/AM2804PC
... but beware of Chinese sellers, these most likely have only counterfeits.
The AM2804 shift register has a NSN (5962-01-306-3796) and is used in many US weapons systems that are still actively maintained / in use, and from time to time stocks of overaged ICs from the National Stockpile are sold off to IC brokers, this is how they get at them, so even if briefly being sold out at brokers, check a few years later and they have reappeared.
It also may be possible to re-wire those ICs from the same family which have dual 512 bit shift registers. I have a few of them but never got around to try it out.
I don't see any reason to buy Apple-1 ICs from Chinese counterfeiters. You just need to know which drop-in replacements exits and where there are no replacements. Such as for the 2519 - but Manoshevitz Electronics in Israel has ~5000 of them in stock, so if you pay their moon price, you can get 'em. This stock (the only known stock of 2519 left in the world) will last Apple-1 builders for the next few Centuries (at the current build rate).
- Uncle Bernie
Thank you for those feedbacks and insights robespierre and UncleBernie, appreciated.
The seller agreed to refund me and for me to return the counterfeit items to him.To be on the safe side (am I?), I decided to buy some AM1404 in TO-5 package on eBay, like the following ones:
s-l1600 (2).webp
Once I get them, I’ll follow the method described here: https://www.applefritter.com/content/another-apple-1-lives
I'll lose the original look, but I can live with that.
Hopefully it will work this time ;-)
I recently ordered a small batch of 1404A chips in TO-99 cans from the Chinese supplier UTSource. First 10 units, then another 20, at a price of under $4 each.
I fully agree with Uncle Bernie’s assessment regarding the reliability of Chinese sources; my own experiences have been similar.
However, in this case, I was foolish enough to think that, if things went wrong, I could simply reclaim my money via PayPal. So, I decided to take a chance, even though I was pretty sure nothing useful would come of it.
To my surprise, all the 1404As I received work perfectly, even passing a continuous (almost) two-week test in my Apple-1.
Of course, the situation with UTSource could be completely different with the next order, dealing with them is always a gamble, and you never know what you’re going to get.
It turned out that those AM1404A were the right solution.It allowed me to end up my whole Apple-1 repro project ;-)
IMG_1085.JPG
IMG_1083.JPG
IMG_1075.JPG
IMG_1074.JPG
IMG_1084.JPG
Thanks again for your help!
Jonathan
Looks great, congratulations!
A very nice looking Apple-1 build which succeeded despite not using my IC kit (long sold out) !
I wonder where you got that white ceramic 6502 from and if it had a reasonable price.
"Reasonable" price for me means less than what it would cost me to transfer a 6502 die from a plastic package into a donor white ceramic package with the right type of leads, as was used by MOS Technology in Y1976. We have a local company here which can do that but it's a defense contractor and hence, pricey. So far I was reluctant to give them the go (and my money).
Yet another question - did you use the six damping resistors from my "reliability mods" ? If not, is your DRAM known to work 100% error free ? Just asking because the feedback I got from builders of my kits was that none of them was lucky without the six resistors added - the diagnostics page I provided in my PROMs always found DRAM errors without these resistors.
- Uncle Bernie
Thank you very much!
Yeah that wasn't easy to gather the parts without your IC kit...
I would love to have a real ceramic MOS 6502 but unfortunately that's not the case... way too expensive currently when you find one.
The one you see on the picture is only a plastic 6502 thoroughly painted in white and with a sticker on top to look like a MOS MCS from 75.
I've used only part of your reliability mods so far, but not all yet (still need to apply the video fixes when I get the necessary hardware). See the picture below:
IMG_1098 2.JPG
So far my RAM is working correctly. I tested it with a RAM utlity from this RAM/ROM card from Jurassic Computing: https://www.tindie.com/products/jurassicomp/apple-1-64kb-romram-card/
The tests passed OK many times.
(Obviously I did used the oncard RAM, not my onboard RAM)
Yesterday however, during a program execution, my screen got progressively filled up with full @@@@. Maybe it can be due to RAM issues?
Jonathan.
I also tried to get my ACI working.I have all the components for now.
I connected C to R on the plug pad.
However, when I perform a reset (after cleared the screen at first), it doesn't return control to me with the "\" prompt.
On the other hand, if I remove the two PROMs (3 & 4) from the ACI (as seen in the photo), I *do* get the "\" prompt.
I assume the CPU is being interfered with when the PROMs are installed.
It would seem strange for those two PROMs to be incorrectly programmed, given that PROMs 1 and 2—from the same batch and the same source—work fine.
My question is: Could I have missed something in the wiring? On the PCB? Or on the ACI?
Thank you for your help.
Jonathan
IMG_1100.JPG
IMG_1099.JPG
I think my 7402 is busted.
Pin 2: 3.52 V
Pin 1: 3.89 V
PROMs pin 13 (empty socket): 1.48 V
PROMs pin 14 (empty socket): 1.48 V
So pin 2 is high (3.52 V) and pon 1 is high at the same time (3.89 V). This looks impossible for a functioning gate. Therefore, it’s ignoring the block selection.
This is probably what’s causing the crash on reset, and explains the average 1.48 V on pins 13 and 14: the selection is constantly switching.
I think I will need to replace the 7402.