Smallest Discrete Transistor 555 Timer

Over at Tiny Transistor labs, [Robo] took it upon himself to reproduce the classic 555 timer in discrete transistor form. For bonus points, he also managed to put it in a package that’s the same basic size, pin compatible with, and a plug-in replacement for the original. The first task was deciding which 555 circuit to implement. He examined a handful of different implementations — and by examined, we mean dissected them and studied the die circuitry under a microscope. In the end, he went with Hans Camenzind’s original circuit, both as a tribute and because it used the fewest transistors — a point which helped manage the final size, which is only a little bit bigger than the IC!

Speaking of sizes, have you ever soldered an EIA 01005 resistor? We agree with [mbedded.ninja] who wrote on a post about standard chip resistor sizes, the 01005 is a “ridiculously small chip package that can barely be seen by the naked eye.”  It is 16 thou x 8 thou (0.4 mm x 0.2 mm) in size, and despite its name and placement in the Imperial series, it is not half the size of an 0201. The transistors are your standard 2N3904 / 2N3906, but purchased in a not-so-standard DFN (Dual Flat Pack, No Leads). We might think a 1.0 x 0.6 mm component as small, but compared to its neighboring resistors in this circuit, it’s huge.

[Robo] has done this kind of project before, most recently making a discrete recreation of of the classic 741 op-amp. We covered a similar, but larger, discrete 555 timer project back in 2011. If you want to go really big-scale with your own reproduction project, check out the MOnSter 6502 from five years ago for further inspiration. Thanks to [Lucas] for the tip.

28 thoughts on “Smallest Discrete Transistor 555 Timer

  1. 01005 is close to the size of a grain of table salt. No way I can hand solder those! The smallest I can do is 0603 and I prefer to use 0805 for hand soldering.

      1. There’s confidence, but there’s also equipment and surface tension. Due to surface tension with an iron, 0603 is realistically as small as many people can do. Most hot air stations blow smaller parts off the board. When a reflow oven is doing the soldering, it doesn’t matter the size once the parts are placed.

        1. People who have problem with hot air often just set it too high, with hot air I can go down to 0402 easily with the naked eye (I am very short sighted, that helps).
          Using magnification makes anything smaller possible with tweezers and patience, but not that hard, the smaller it is, the less heat you need, so the less airflow you need.

          Using an oven for small parts without pick and place is infuriating, the smaller they get, the higher chance they have of tombstoning…

          In general, soldering fine things is as much a matter of experience as it is a matter of having good tools (Mainly a quality hot air station, fine leaded solder, and fine tweezers)

  2. Could this be the perfect place for “could have just used a 555”?

    Whilst I’m all for doing something for the sake of it, there’s something odd about making the smallest version of a large version of small thing made out of large things.

  3. I appreciate the effort this takes, but you have to be a masochist to try to pull it off. Just thinking about hand soldering it gives me a headache. He could put together a whole guide on soldering techniques for these microscopic parts. I see he uses a hand soldering iron and no reflow.

  4. [Could have done this with a 555 -> not]

    Many say to use a 555 as WDT because the internal micro-controller WDTs are not considered fail-safe as software is blah blah. I have used a 555 solution where power consumption is not a design issue and where cannot be dependent on the controller’s internal timer; but use of specialty ICs as an external reset source is indicated and probably less design effort and more reliable.

    And ‘low-current’ 555 solutions using external power-steering transistors is not a good fail-safe solution because you are banging input power on and off which can never be considered to be reliably mono-tonic and you are reducing reliability by adding even more components.

  5. He could have made it smaller. That PCB substrate is pretty darned thick. If I were doing this, I would have routed out some channels to install parts inside the board for a more 3D (as opposed to his single-sided) approach. Get it smaller than a SOIC.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.