1 post / 0 new
Last seen: 3 hours 12 min ago
Joined: Apr 1 2020 - 16:46
Posts: 306

I often get asked about hooking up transformers for the Apple-1 PSU and sometimes about fuses, despite everything can be found in the datasheets of the transformers. I'm still reluctant to give any detailed wiring instructions because it's a test: if you can read and understand the transformer datasheets, then it is likely, but not guaranteed, that you have the level of electrical expertise to hook them up correctly to the Apple-1 and to the line voltage ("mains") without killing yourself or your loved ones or setting the house on fire. See, I do not have the "License To Kill" from the Government like those manufacturers of the "vaccine" have got (yesterday I heard about 1400 victims in Colorado Springs alone, from a gentleman who got a blood clot in his leg, and then a piece of the clot broke off and caused a mild stroke, and his doctor told him the number of victims, and that he can't sue, and that he may not survive, nice deal !), and unlike these big pharma crooks, I have no blanket immunity against any liability either. I would LOVE to have a "License To Kill" and that blanket immunity against prosecution and getting sued. This world would be a much better place if I had that. But since I don't have that, I can't bite off a huge piece of liability by telling you things you should better not know to try out. But as far as the transformers/fuses issue goes, I can regurgitate some of the info in the datasheets of the transformer and fuse manufacturers and add a little bit higher safety margins. Still, use all this information in this post at your own risk only ! I assume no liability, for any incidental or consequential damages arising for the use or misuse of this information.




There are a lot of myths and misconceptions about fuses out there, and some relate to fuses on transformers. Let me debunk a few of them:


1. Myth #1: A 110V fuse won't protect a 12V circuit. You have to use 12V automotive fuses for that.


Untrue. The voltage rating of a fuse specifies the maximum AC voltage it can safely disconnect when open. This boils down to arcing, isolation and spacings of its connectors. A 110V rated fuse can protect a 12V circuit, but not the other way around.


2. Myth #2: A 110V (or 220V) fuse will open slower when running from a lower voltage, such as the secondary of a transformer.


Untrue. A fuse blows by melting of its filament. The melting of the filament is done by the current alone, and not by the voltage. The filament has no idea about the voltage in the circuit until it opens ... see Myth #1.


3. Myth #3: A transformer can be protected by a single fuse on its primary winding.


Untrue for transformers with multiple secondary windings (including center tapped ("CT") windings). But it is true for a transformer with a single secondary winding. After all, a transformer transforms the currents by the winding ratio for each pair of windings. If you draw 1A out of a 12Vac secondary winding, the primary winding running at 120Vac (not a typo) will draw 0.1A, and so a single fuse on the primary winding can protect both the primary winding and the secondary winding of the transformer from overload conditions.If the transformer has multiple secondary windings, a fuse on the primary winding will protect the primary winding, but may not protect the secondary winding if that secondary winding is overloaded.


4. Myth #4: One fuse in the primary feed can protect a multitude of transformers connected to it.


Untrue with a few exceptions. If you have two transformers of the same power rating and don't want to run them at their limits, one fuse in the primary feed may be able to protect both (but observe the explanation at Myth #3). This may even be used as an advantage as one overload condition would disconnect both transformers from the mains, and so all voltages into the load(s) will be gone, which may be beneficial for some loads which don't like some voltages being present and others being gone. So by using this concept, you can save a lot of components for additional protection circuits. But in the general case, you can't protect two transformers with a single fuse, especially if one of them has a lower power rating than the other. The fuse will not blow when the little guy is overloaded and starts smoking and dies. This is the case with the Apple-1 PSU in its original form, as shown in the Apple-1 manual.




1. Use a fuse with appropriate voltage rating for the highest voltage you will ever use it. Automotive fuses are perfectly fine for the secondary side of transformers if their voltages are below 30V or so but you will find out that despite automotive fuses are dirt cheap, the fuse holders are expensive. The most economical solution is to use 220Vac rated, 5x20mm fuses, which you can find, buy and use all over the world.


2. How quickly a fuse opens depends on its speed class. To protect transformers from thermal overload, "slow blow" or "time-lag" fuses are the best choice. They allow you to use a lower current rating for the fuse which helps protect the transformer better. Quick acting fuses are not recommended as they tend to blow during the power-on surge unless they have a current rating that may be too high to protect the transformer from long but weak overload conditions.


3. Use a fuse of the appropriate current rating on each secondary winding on each transformer. Typically, modern transformer manufactureres specify the recommended fuse(s) in their transformer datasheets. For instance, the datasheet for the TRIAD VPP16-1900 transformer I recommend in my "Tips & Tricks", they specify a Littelfuse p/n 313 5.0HXP, 5A 250V, slow blow fuse, or the equivalent from Cooper Bussman, if the secondary windings are connected in parallel, which is the case for the Apple-1. These are 5A fuses for their 3.8A windings (if in parallel, each winding can take 1.9A). Hmm. OK, they supposedly have tested their transformer with that 5A fuse not to start smoking or catching fire. Fine. Remember, this is about UL certification. UL does not care whether the transformer is damaged or dead afterwards. All they want to see is no smoke and no fire before the fuse blows, thus ending the threat.


I am always a bit on the safe side when it comes to my designs, so I did not become a rocket scientist / designer as my dad. He told me: "if you design a rocket according to the safety legislation for steam boilers, it will never leave the lauch pad." Hence, you need the lauch crew be safe in thick concrete bunkers. And God bless those daredevils who ride on the machine. Allegedly, one of the early U.S. astronauts asked Wernher von Braun if he would swap the place in the capsule with him, and take the ride to fame. Von Braun shook his head and said: "No, no way. You fly." The astronaut asked" "Why ?". Von Braun answered: "Well, if it explodes, among us two I'm the one who can go back to the drawing board and improve the design." Oh, and the astronaut did fly. I don't remember which one, but it may have been John Glenn. The exact wording may have been different, too. It's from my memory.  


OK. Nice anecdote. But I think you get my point: different applications need different safety margins. While the rocket designer wants a machine that can lift off and gets a good payload into orbit, the transformer company wants a competitive power figure while still dodging the wrath of the UL, but we, as hobbyists, want to tinker with our Apple-1 and not fry transformers in the process. So I would recommend a 3A slow blow fuse, and not a 5A slow blow fuse  as recommended by the transformer datasheet. 3A is plenty for the Apple-1.Also keep in my that the LM323K regulator has in internal current limit and thermal protection. So it will limit the current on its output side to levels which the TRIAD VPP16-1900 can easily sustain forever. But what if the LM323K itself fails ? Or one of the rectifier diodes fails (they typically fail as a short circuit) ? Or something else happens on the input side of the LM323K ?

It can happen - I will tell you a story later in this post.


As for using a lower rated fuse which would blow at lower currents, there is some deeper engineering reason behind my caution. In the datasheets of the fuse you will find a table which typically goes like that:


1.5  In   MIN 1 hr.

2.1  In   MAX 2 min.

2.75 In   MIN 600ms MAX 10 sec



This was taken from the datasheet of the BELFUSE 5ET series (www.belfuse.com)


"In" is the rated current of the fuse. So for the 5A fuse mentioned above, 1.5 x 5A = 7.5A would be needed to blow it, but how long would that take ? The "MIN" value means it won't blow earlier than 1 hour under these 7.5A (Oops !)This is roughly twice the rated current for the transformer secondary (2 x 3.8A = 7.6A) and I don't think the transformer will like that current for such a long time. And note that the fuse does not necessarily blow after 1 hour. It's a "MIN" spec. There is no "MAX" spec. Sneaky, isn't it ? So at 7.5A it may blow never.


Whereas, for the 3A fuse, 2.1 x 3A = 6.3A and at that current it will blow in no less than 2 minutes (it's a MAX spec, see). I'm quite sure the transformer will not get damaged if it has to source 6.3A for a mere 2 minutes. It may get hot, though.


Along the same lines of thought, for the VPP24-1250 which is rated at 1.25A for each of its secondary windings, you may use a 1A fuse in each of its secondary windings (no fuse in the CT leg, please). Or use 0.75A fuses or even lower. The regulators on the +/-12V also have self protection circuitry, but the same rationale applies as for the LM323K.


So with three cheap fuses, 1 x 3A and 2 x 1A, you can mitigate disaster and protect your transformers. An appropriate fuse holder looks like this:






Well, this is probably OK if you keep your Apple-1 under adult supervision all the time when it is powered on, no exception, not even a bathroom break. See, if anything bad happens and an IC dies, or a short circuit occurs, the Apple-1 will not just continue to run happily, but it will crash the running program or show some other symptoms of distress, such as smoke or stink. Now, if you see a program crash and the usual RESET/CLR SCREEN does not bring remedy, or you see or smell smoke, SIMPLY TURN OFF THE POWER IMMEDIATELY ! And do not touch the possibly hot parts. You may use your contactless IR thermometer as discussed in post #5 of this thread:




to try find out which IC ran too hot and died, before everything has cooled down again. Note that most dying ICs will not get hot and smoke. These are more difficult to find.


Now, suppose you have plugged in your Apple-1, and it powers up correctly, and you have loaded a program, and then take a bathroom break. A few minutes later, approaching the Apple-1 again, you can already smell it (and I don't mean the bathroom stink you left behind, I mean a BURNING smell, as in "FIRE" !). Then you find your Apple-1 does not run the program anymore, and the transformer looks like this:



How did it happen ? Well, it was late in the evening and the light was poor, and when I plugged in the J1 connector I plugged it in by one pin offset. So the poor VPP28-1060 transformer had to source the full current load of both the +5V side and the +/-12V side. It did its best, a glorious struggle, until it gave up the holy smoke. Dead. Another hero which died on Uncle Bernie's electronic battlefield. No medal granted, though. Thanks to the good transformer designers at TRIAD there was no open fire (according to their datasheet they use bobbin materials with a flame retardent that exceeds UL requirements). But there was an awful stink in the living room which lingered for days despite I had immediately ventilated the whole house. This was the first power supply I had built for the Apple-1 and of course it had no fuses because I would never let it run unsupervised. Except for that one incident of plugging it in wrongly and that one bathroom break immediately afterwards, so I could not see the catastrophy unfolding before it was too late. Like with airliner crashes, most catastrophies are not being caused by one sole point of failure, as the designers have foreseen them. But any combination of two or three failures may be too much, and then no trained procedure may be able to save man and machine.


In the case of the transformer, three cheap fuses would have prevented this dire outcome. It's not so much the $15 or so for the transformer, it's feeling stupid and then on top of this psychological punishment having to spend hours to build a new power supply. This time with fuses, of course, I promise !


Comments invited !