I'm looking to purchase the following period correct components as would have been found on an NTI board:
- Signetics N8T97B x 2
- Signetics N74S257B x 4
- National DS0025CN
- Signetics N74175B x 2
- Signetics N74157B
- Signetics N7432A
If anyone has these with date codes between 1974 and 1976, please get in touch!
I'm also looking for two "Circle D" 0.01µF ceramic disc capacitors like the one below:
IMG_4894.jpeg
If you find a source for them let me know. I have been looking for about 10 years now for one.
What component do you use instead of the original? I've found some that have a 50V max rating. Would that be OK to use instead of the original 0.01uF 25V part?
In post #5, fingerz asked:
"What component do you use instead of the original? I've found some that have a 50V max rating. Would that be OK to use instead of the original 0.01uF 25V part?"
Uncle Bernie answers:
In this case, yes. There is a common rule that you can always substitute a capacitor of the same kind and same capacitance having a higher voltage rating. Much as you can use higher wattage resistors. But there are possible pitfalls with that rule: a) the higher rated substitute may be mechanically larger and may not fit into the PCB footprint anymore or just look ugly, and b) it may have different parasitical effects which may degrade certain high performance circuits. But in the Apple-1 this pitfall lurks only with the 100nF power supply bypass capacitors. All other capacitors in the Apple-1 can be substituted according to the above rule with no fear.
Sometimes I admire the boldness of obvious electronics newbies who venture into to building the most quirky computer there is but don't even know the very basics of electronics. I even had a case of a builder who did not know yet how to solder because the Apple-1 was his very first electronics project.
How dare you !
But regardless, good luck ! (Oh, and even that soldering newbie succeeded using one of my famous kits. After his first success he ordered yet another kit and built a second Apple-1. Which probably looked much better than the first one, but both do work !)
Not only am I a relative newbie on electronics, I'm also restoring an original Apple-1 NTI board, not assembling a replica. Luckily I have some colleagues assisting me who are not newbies and can do repair work like this in their sleep.
The board is working, but has had many original components replaced over the years. Those are the components I'm looking for and what we are slowly. This will also be the second original Apple-1 we are restoring, and we assembled two Mimeo-1 boards together some years before.
The information you gave is helpful in case we are not able to find the exact replacement disc capacitors. My basic electornics knowledge led me to that assumption, but when dealing with an original Apple-1 I felt it was better to ask for advice than make a mistake I would later regret.
In Post #7, Fingerz wrote:
"I'm also restoring an original Apple-1 NTI board, not assembling a replica."
"The board is working, but has had many original components replaced over the years."
Uncle Bernie comments:
Aha ! So this is your story. It was not obvious to me that you restore an original. Btw, a "replica" ist not a "clone". What you did build with the Mimeo PCBs are clones. Please don't think this is nit-picking. Mixing up "replica" and "clone" has confused lots of people, me included, leading down a wrong path to a dead end. So we need to be more precise with our language. The fact that the "Open Source" Gerbers which came out of Russia (with love !) have the text "Apple-1 replica" on their solder side. Ouch ! For newcomers it's very confusing what kind of animal this is and I know several cases where this led to confusion with the "Replica-1" kit which uses more modern components.
I've explained this in post #3 of this thread, which - of course - started with the same confusion:
https://www.applefritter.com/content/looking-apple-i-replica-kit
Now a few more hints about restoring Apple-1: I think it's a sin to replace the original components unless absolutely necessary. Electrolytic capacitors, as long as their seal is intact, and they did not dry out, can be rejuvenated by a process called "reforming" which essentially is the same process their manufacturer uses to "form" the electrodes, but "reforming" is slower and gentler. I have successfully reformed Sprague39D capacitors with a mid 1979 date code and they work perfectly now. Alas, they are not 5300uF !
Important: after reforming, do a ripple test with the same rectifier topology and under full load current. Run this test for a few hours. If the ripple voltage amplitude is not too high and does not increase over time, and the capacitor does not get too warm, than it is fit for service for many further years. (Don't do this test in a real Apple-1. I built a lab test rig for that.)
About ceramic capacitors: it eludes me why somebody would have replaced them. Maybe they had issues with the crystal oscillator not starting up or not being stable, who knows (most of the time it's the crystal itself, as most primitive oscillator circuits overload it, and so it gets beaten up and degrades / fails). But there also is an effect in these ceramic capacitors which is nasty: some of the cheaper ceramic materials (the "Z" and "Y" families being the worst) are not very stable over time and they age and lose capacitance. It is not uncommon to find 40 years old ceramic capacitors of nominal value 100nF which only have 60nF left. If you have an oven with good temperature control, you can try to rejuvenate them by baking them for 24h at 150-160 deg Celsius. Desolder them, measure them, and place them on a wooden plank for the bake. Close the door and turn the oven on. Never open the door while hot. After the bake, let it cool down and measure them again. If their capacitance has recovered, they are fit for service again.
Resistors: the carbon composition kind used in the original Apple-1 is notorious for being unstable. On the Apple-1 motherboard, you can measure them all when the ICs connected to them are pulled (some instruments use low enough voltage to allow measurement of resistors in presence of pn junctions, buy this is not always the case). Those who are grossly off / open circuit should be replaced, but +/-20% from their nominal value is probably OK and won't cause the Apple-1 to stop working. Except for the one resistor which makes the 480ns timing pulse for the DRAM. Use an oscilloscope to measure where you are and then take corrective action. If the pulse is too wide, you can bring it back to spec by adding a parallel resistor on the solder side of the PCB. Remember that after any soldering on a carbon composition resistor, it may need many hours before its resistance has settled and stabilized. (Now you know why the industry threw them out after the carbon film and metal film resistors became available ... the only feature where carbon composition resistors excel is pulsed loads, so some values are still being made up to today).
Semiconductors: numerous failure mechanisms exist, none of which can be repaired. Tarnished leads can be fixed by tossing the IC into Tarn-X bath (although Corey disagrees, as he thinks it makes the leads brittle). The key to success is a thorough soak in deionized water after the Tarn-X bath. I use several soak stages to absolutely remove all the traces of Tarn-X. The dirty little secret is that there are two kinds of lead frame tarnish/corrosion: lead frames with high silver contents tend to get black silver sulfide tarnish, and Tarn-X strips that in a few seconds, and all is good. But there are some lead frames where they screwed up the galvanic process and there was residue of the process bath left, and these lead frames corrode from within. If you Tarn-X them, the outermost galvanic layer strips off and exposes a rotten lead, which may fall off just by looking at it (figuratively). Much the same damage can be caused if the circuit board in question was exposed to humidity (think humid attic, basement, or certain States like Florida).
IC sockets: IMHO the low profile TI sockets used in the original Apple-1 are the worst IC sockets ever conceived by the human mind, and whoever did that should be banned on a planet full of stone age savages so that designer never gets near any metals again. In other words, it's hopeless. There is a reason why these sockets are so rare nowadays: all the circuit boards using them failed soon, and ended up in a landfill (Let go of any hopes to dig them out). The industry quickly learned to avoid these TI sockets (and their knockoffs) when better socket designs appeared on the market.
All this is not from restoring Apple-1 originals, I never had the honor to have an original in my hands, which I compensate by having more than a dozen clones, but I collected and restored lots of vintage computers over the past 40 years, and the problems / damages found almost always are the same and can be remedied with the same methods.
I suppose that would be safe for ceramic package ICs as they are hermetically sealed, but plastic packages are not (this is why they require humidity control and pre-baking before reflow). Is there a danger that chemicals could creep into the package and not be able to be rinsed out?
Would dipping the leads in molten solder help with tarnish? Perhaps painting them in flux beforehand?
In post #9, Robespierre wrote:
"I suppose that would be safe for ceramic package ICs as they are hermetically sealed, but plastic packages are not (this is why they require humidity control and pre-baking before reflow). Is there a danger that chemicals could creep into the package and not be able to be rinsed out?"
Uncle Bernie answers:
The humidity control and pre-bake of the plastic packages has to do with the plastic's ability to soak up water (it's not hermetic and it's hygroscopic). Then when you reflow solder them, the water turns to steam and may pop the package ("Popcorn effect"). In many cases delamination of the package over the die (the "chip"). The older plastic packages were too thick for that to happen. The so-called "green" mold compound which was introduced a few years ago exacerbated the problem. These ICs have numerous reliability issues. Another nail in the coffin for modern electronics. The previous ban of lead in the solder was the worst sabotage, IMHO. This was not done for your health. Or do you eat printed circuit boards ? Consumers are not supposed to handle printed circuit boards anyways. The electronics industry itself bribed politicians to ban lead solder. Why ? First, some could leverage patents around lead free soldering. Second, electronics with lead free solder do not last. The lead free solder joints tend to crack, and that's the end of the gadget. You have to buy a new one. And when the cracks don't get it done, tin whiskers growing out of the lead free solder will cause shorts. Electronics built with leaded components (meaning: the wires) and soldered with lead based solder can last for decades and can be rebuilt indefinitely until you run out of parts. The modern, "green" electronics based on lead free SMD technology lives a few years at most and then needs be tossed into the trash. I refuse to buy any expensive modern electronics for that reason. I only buy disposable cheap items containing the modern crap electronics, and only if I can't avoid to have the piece. But I can't completely avoid to be ripped off by the lead free electronics junk ... I suspect my year 2012 car will die from rotten electronics which can't be repaired long before the mechanics are worn out or the body is rusted through. So far my take on that scam.
About the bath ... there is a small risk that a leadframe has delaminated from the plastic body on the occasional pin if the IC is old enough or if there were manufacturing issues with that lot. But if properly molded the adhesion between the plastic body and the lead frame is watertight. So nothing can sneak in. And if it does, there is hope that the much longer rinse bath in deionized water will be able to get the contaminants out, if any. Otherwise, the lead frame will rot, which, in a dry environment, not too much humidity, may take decades. And then the pin falls off. So what ? You replace the IC with a new-old-stock one. I have lots of computers where I tossed whole PCBs and wire wrapped backplanes into Tarn-X or equivalent bath because everything, including the wire wrap wires, were badly tarnished. You would suspect the insulation of the wire does not seal the gap between the insulation and the wire, and so chemicals from the bath would get in, and wreak havoc, but so far all these computers still work, after 20-40 years.
If you are dealing with a 1 Million US$ specimen like an original Apple-1, I would not use the "magic bath" method on any of the original ICs with the cherished and holy date codes. You can remove the tarnish with purely mechanical means, such as a rotary brush in a Dremel tool, or with a pencil eraser, like Corey recommends, and for $750 per hour even my lazy self certainly would never use a quick bath, but lovingly polish each pin, using mechanical tools only, and work very focused and slowly, pin after pin (you get the idea, here: time is money indeed).
For for quick restoration of vintage handware which only has scrap value I would not want to expend much time on tedious mechanical drudge work. The usual treatment for electronic scrap is to grind the ICs up and toss that powder into a cyanide bath to extract the gold, and then after it's depleted and no gold in it anymore, toss the cyanide bath into the next creek --- (NO, DON'T DO THAT) - I just wanted to show you what is done, pesky environmental swine, all of them... most of the pollution of the Oceans is done by third worlders. But the real criminals are those "recycling" companies who bring that electronic junk to these people.
And as for the use of a tin bath to remove the tarnish - I never tried that. But I know that tarnished leads can be very stubborn to solder, and out comes the "activated flux core" solder, nice halogens in there, you need to work under an exhaust vent / hood so as to avoid inhaling that nasty stuff. Now see the problem. How can you remove all the ionic contaminants brought in by the chemical magick in that activated flux ? So as to avoid future corrosion ? Just a few years ago my 1986 vintage Hitachi oscilloscope started to act up. But it turned out it was the ground clip wires on the probes. All the multistranded copper conductors in the wires were badly corroded and had fallen off at the solder joints. The culprit also was still there: activated flux, probably used to pre-dip the wires before they were pre-tinned in a solder bath. Still, this "Made in Japan" mess lasted for 33 years. Not bad. Even for activated flux.
In the end, we mere mortals have to accept that nothing lasts forever and that over time, every man-made object will return to the same sort of mineral oxides, sulfides, silicides or phosphates it was made from. There are so many nasty deterioration processes in all man-made objects that you can never, ever preserve them for eternity. Except if they are made out of solid, pure Gold or Platinum. But beware - the fine bond wires in our ICs, despite being pure, soft gold, where they are in contact with the aluminum bond pads, will, over the decades, form intermetallic compounds that are more brittle and higher ohmic than the original contact pairing. Another dead or limping IC ... so even Gold has its limits. Oh, and never try to fire a 100 years old gun. Intergranular corrosion is a topic here. You may end up with bits and pieces and a few lost fingers you can then pick off the ground. Maybe a good surgeon could stitch them back ?
Maybe this is naïve, but I would suggest that the ionic contaminants can be removed using the same cleaning processes as for PCB assembly and rework, where activated fluxes are also used. The halides should migrate to the surface of the solder as it crystallizes due to their lower melting point.
The once ubiquitous Kester 44 flux contains 0.44% halides, and is used for both assembly and rework. No doubt there are more aggressive fluxes, but residue removal should be a solved problem.