A14f / nte5800 / mr5000 cannot be found. What other models can be used instead

At least this is what I found with MR501 I tried. The plastic body of these was as long as the spacing between the holes in the PCB, so there is so way to bend the diode leads to fit into the holes without looking unprofessional and ugly (you would need three bends on each lead to snake the lead under the body of the diode OR put them in "standing", perpendicular to the PCB).

I wonder if this is a problem with the cloned motherboards or  if the package of the MR500 series changed in the past 40 years, or whatever.

I prefer glass frit diodes which look like glass beads with leads.

The look is not 100% authentic, but they will fit if bent "carefully". Note: I won't be using them on my next build.

IMG_3564.jpeg

In post #5, flatsixracer wrote:

Will fit if bent "carefully"

Uncle Bernie warns:

... except that your example is not "careful" bending of component leads. If you bend the leads that close to the plastic body, the stress on the plastic is high, and may crack the plastic. Even if the crack has not yet manifested itself, over time and over thermal cycles it will.

Proper bending of radial component leads goes like that:

Method A): use a 'lead bender' tool and seat the component into a place where it fits easily (not too tight). Then hold the tool in one hand and with the thumb of that hand hold down the component body snugly into the tool. Use fingers of other hand to bend the leads.

The key observation here is that after the procedure the lead still comes out of the component straight and not bent at all, until the bend starts at a distance from the body. This is the proper way that has been taught in electronics trade schools since these existed. It is prohibited to bend any lead "free form" just using fingers. Because it tends to start the bend at the component body and hereby cause undue stress there which will lead to reliability issues. This effect has been discovered with the first point contact transistors which had metal can packages with glass sealed leads. Bend the lead at the glass, initiate a micro crack, which will grow over thermal cycles and then the lead gets loose and the 'cat whisker' attached to the lead wiggles around, making the transistor fail. Similar issue with carbon composition resistors: bend at the body and see the resistance change. If it gets bad, the lead will break loose. But most often, this happens only during the soldering process, because is more work to fix it.

Method B): use needle nose (or skinny nose) pliers to hold the lead near the component body, but the tool must not touch the body itself ! Then use your fingers of the other hand to bend the lead as you want. I prefer pliers which have a flat and smooth grabbing surface and a round shape outwards. The round shape makes nice round bends. Needle nose pliers are round all around so the grabbing surface is round, too, which means the lead is squeezed too much if you apply some force.

I know it may sound funny that there is a lesson 101 of how to bend leads with the proper tools and the proper technique, but actually it is sad that this knowledge is long forgotten in most places. This loss of knowledge on how to do it the right way leads to a bad outcome, in this case, lack of reliability due to creation of a potential failure point. If you use these wrong methods often enough to build any machine of any significant complexity (number of parts is meant here) it will never work or it will cease working soon, and be a maintenance nightmare.  I think this way to dubious work is only appropriate if you work in a slave labor camp or in a communist country, which essentially is the same. If you want to build things that last, and work reliably, you have to use the proper tools and the proper techniques. Alas, all these skills get lost and those who learned them from the ground up are all retired before they trained their replacements. Which often are half a world away and don't speak your language and on top of that are slave laborers and so the moronic managers did not even ask to train them. For managers, the quarterly results matter. Nothing else matters as long as the bonus payments flow into their pockets. Apres nous la deluge.

This is how you bring a civilisation based on craftsmanship, technology and science on its knees. Forget how it is done right and corrupt the editors of the technical journals and corrupt the science by controlling the grant money. Hire H1B "professors" from India who have a worthless Ph.D. they got by bribing their equally incompetent professors in India. You will harvest junk graduates, junk science, junk scientific papers, junk products, products for which you pay for with unsound junk money, and which won't last. You will get junk medical care which will kill or cripple you, based on junk or outright fraudulent "science", too. The airbag gas generator in your car will not inflate the bag, it will detonate instead, and take your head right off or send shrapnel to your heart. I could go on and on. But there is a way to stop this ... by telling people how to do it right ! (this is part of my mission).

In post #7, macintosh_nik wrote: 

"Not everyone who builds Apple-1s is a professional in this field. I am one of them and I have my legs bent just like the flatsixracer. There is no other way to do it."

Exactly this is the problem for hobbyists: no professional training or professional experience in the field of the hobby. A true professional gets this training in trade school or in college, but alas, the scope and quality of this training went down as the prices (tuition fees) went up. When I went to the Technical University to get my M.S.E.E., I had to take a mandatory 6 week crash course in a trade school which outside the summer break would train real craftsmen. There I learned the basics of how to file, drill, hone, cut threads, rivet, solder, weld (electrical stick and oxyacetylene), forge nails and clamps, turn parts on a lathe, or machine them on a mill, all just at a basic level, but the intent is that the finished engineer would not draw funny blueprints that can't be manufactured. All these skills, although only at basic level, were invaluable for me during my whole life. Despite there are no rivets in ICs !

The best advice I can give to any serious hobbyist is to seek professional training from a true master craftsman. Most retired master craftsmen are fed up and don't want to have anything to do with their profession anymore, but some would love to show a few of their tricks of the trade to hobbyists. Not necessarily for free, and with no compensation, but maybe for a few nice dinners or a crate of good wine ? Or doing some work on their property they don't want to do anymore due to age ? Or you just pay a bit of GTP which will soon be worthless for a commercial course offered by places who want to sell their equipment ? (Alas, we have to wait until the pandemic is over).

If you don't seek such professional advice, you will make mistakes. Which can be costly. Or compromise the reliability or quality of your work. Or waste valuable materials. The few hints and "Tips & Tricks" I can give to you sure can help to avoid the most common mistakes when it comes to Apple-1 builds, but I can't cover the whole territory of electronics.

Anyways, here is an example on how to bend the diode leads correctly (using method B of above): 

PDRM2057.JPG

I used the tool to the left first to make a round 180 degree bend after the straight part of the lead (which was grabbed with the flat surface inside of the pliers). Then I used the tool to the right to grab the inside of the bend and I used my fingers to make the final 90 degree bend on each lead. The tool on top was used to straighten and align the leads. Method a) with the lead bender tool would need three 90 degree  bends on each lead and the bends would not be round like above but would look more rectangular. The example is not perfect, but good enough. If you want perfection, buy 100 of those diodes and practice, practice, practice ;-)

This is how the diode fits into the PCB:

PDRM2061.JPG

... and here is the top view:

PDRM2063.JPG

It looks ugly, because it exposes the mistake of inadequate hole spacing of the PCB (or a wrong part). But it will be reliable if soldered properly.  Note that the clearance to the J1 connector is tight. Do not solder the diode in before you have checked the J1 connector (including the plug) will still fit and not touch the diode lead. If you really think about it this situation has an advantage, too: it will be more difficult to plug in the J1 connector offset by one pin, which happened to me and fried a transformer (dead !). So sometimes, uglyness avoids trouble.

The 'glass bead' diodes look better. The BY series avalance diode you suggested sure could be used, and you can forget the avalanche function as the Apple-1 has no 1000V inside. These diodes are meant to be used with flyback transformers and the like such as found in horizontal deflection circuits of CRT based monitors or TVs. Inductors have the useful but dangerous feature to increase their voltage until the same inductor current continues to flow, which happens when the current path is interrupted, and that voltage increase has no limits: the current will continue to flow no matter what gets into its way: the inductor will simply increase the voltage high enough until some isolator breaks down. A normal diode getting into the way will break down if reverse biased and seeing a high voltage exceeding its breakdown voltage, and will get damaged due to 'hot spots' in the crystal. Avalanche diodes also break down but do not get damaged because they are designed to break down in a controlled manner which avoids 'hot spots'. The limit is the energy they can handle during such an event, which in case of an inductor is limited by the energy stored in the magnetic field.  All this needs careful calculations and choice of components. This may be one reason why most electronics engineers leave design of power converters using inductors (= switchmode power supplies) to the few specialists who know all the pitfalls and all the tricks. And these specialists get fewer and fewer. 

One problem with high voltage diodes like that in the Apple-1 is their higher forward voltage drop (in this case 1.5V) so they will get hotter and the input voltage of the LM323K will be lower. It may work, though. I'd rather look for glass frit diodes that are not of the high voltage type (look for low forward voltage at the 3A current).

But our choices are dwindling: more and more leaded components are declared "obsolete" by their manufacturers and are taken off the "released product list". And then all what is left disappears one by one. Another trick they use is to increase minimum order size to > 10000 pieces, so if you order them, they will make  a batch just for you. This often is done with wafers that have been stored just for that purpose. The wafer fab to make these wafers may not exist anymore. So eventually even these sources will run dry, too.

So get your vintage electronic parts as long as you can !

Thanks UncleBernie. Yeah, the picture made the bends look worse than they are in person. I will try to bend the spares nicer and swap them out, just because I don't want them to crack.

I don't mind the look and actually prefer them over ceramic or glass components other than for the authentic look.

But I want to do it right  with  the NTI look on my next board.  I've learned a lot building my first Apple-1 build and will keep it forever!