Th whole concept that just because a design is old that it must be useless is rather shortsighted and poor thinking. consider that the 4000 series CMOS line has been around even a bit longer than the 555 timer, and is still sold by most electronics distributors, IN QUANTITIES!!
And unlike so many of those programmable devices, the 555 can be put to work as soon as it arrives with no code development at all. The example of op-amps is a good one, they had been around long before the 555 was even considered, So while improvements can be made in almost every area, some tools never become obsolete.

Exactly.

• No programming required
• No software bugs
• No NRE (non-recurring engineering) costs for programming and debugging
• No manufacturing time loading code
• Wide supply voltage range
• No flash memory bit rot or soft errors from cosmic rays or voltage spikes
• No hung or stuck software states; no watchdog required

 

An extract from the Wikipedia link below reads:-

In 2017, it was said that over a billion 555 timers are produced annually by some estimates, and that the design was "probably the most popular integrated circuit ever made".

I might argue that op-amps have sold more – but as there are so many flavours, they don’t count.

https://en.wikipedia.org/wiki/555_timer_IC

 

As a frequency generator, it is rubbish compared to crystal oscillators, and a microprocessor with a crystal oscillator and a timer.
Where it really comes in handy is when two comparators and a set-reset latch are required. I use them as a zero-crossing detector on the full-wave rectified output of a transformer (and @Papabravo I use the SOIC-8 package!)

 

It can be used as the main control element in a switch mode converter - a billion sold in 2017 - I expect the numbers are similar now, it's use as a one shot, switch de-bouncer, and many other ckts make it the definition of ubiquitous, the smallest SMT version is about 3mm x 3mm ( VSSOP 8 )

 

Sorry - the DSBGA from 2016 is the smallest package at just under 1.5mm x 1.5mm

 

There is more to the world that the US, western Europe, and Japan. The vast majority of the world does not or can not program in C, Python, etc.

The last year I heard about was 2004, when the worldwide numbers were in the middle-billions. Today it might be less, but I'd be surprised of the total across all variations was under 1 billion per year.

I'm not a raving fan of the part, and in fact it contributed to my tagline. But it does some things *very* well.

ak

 

As a frequency generator, it is rubbish compared to crystal oscillators, and a microprocessor with a crystal oscillator and a timer.
Where it really comes in handy is when two comparators and a set-reset latch are required. I use them as a zero-crossing detector on the full-wave rectified output of a transformer (and @Papabravo I use the SOIC-8 package!)

Like a good shoe, one size does not fit everybody, nor every application. That is certainly true with the 555. But if it fits a few billion applications year after year, that is OK if it does not fit every application.

 

It can be used as the main control element in a switch mode converter - a billion sold in 2017 - I expect the numbers are similar now, it's use as a one shot, switch de-bouncer, and many other ckts make it the definition of ubiquitous, the smallest SMT version is about 3mm x 3mm ( VSSOP 8 )

A very good point. When I was learning the various control methods for switched modes, I designed each one around a 555, and it gave a very good insight into how they all work.
It's a great shame they don't make a 555 with a 1Amp output (although MCP1401 and all related devices can be used in most 555 circuits albeit with a wide tolerance on the thresholds and needing a diode to create "discharge")

 

easy enough to buffer the output with an emitter follower pair.

 

My wish-list 555 traded the Control input for a complimentary output, or an output polarity control pin, or an enable pin for a tristate output. I poked him about this, and he said that he had considered all of those options (and more), but at that time they added too much complexity to the circuit.

ak

 

complementary output in a 10 pin pacakge would have been very useful ! the size of the chip and the size of a useful npn/pnp emitter follower output - would have made the chip too big ( at the time ) for the std package.

 

It is a joke amongst the staff how much I like this old chip. I have several indexes of projects around this component.