Interestingly, the design wasn't patented and companies
like National Semiconductor and Fairchild soon released
their own versions. This helped to push the price down and
ensured that the chip would become ubiquitous in the decades
that followed.
Camenzind went on to have a long and successful career, but none of his subsequent designs would impact the world like his simple little timer. In fact, IEEE Spectrum named it one of 25 microchips that shook the world in their May 2009 issue.
-- Continued from above. Interesting, note the incentives vs the output.
I'm not sure what either of you are attempting to demonstrate with these excerpts.
I read the fact that he didn't patent the 555 with some disappointment. He had a successful career, good; and yet, he deserved to capture some of the marginal product of his labor, not simply to go on to be promoted super-Executive-Senior-Engineer-Level-XII. In this case, he did later start and sell a company after he left Signetics, and his reputation from the 555 timer surely helped him to success in that.
But to my mind not patenting the design clearly made it harder for him to partake in the direct rewards he deserved for creating such a useful circuit.
Those excerpts demonstrate that the 555 timer became significant in part because it was not patented. You could even argue that Camenzind wouldn't have been eulogized for his invention of the 555 timer if it wasn't for its success due to widespread copying.
That would be a specious and incredibly naive argument. First A (unpatented invention), then B (success), ergo A causes B -- is that the proposal here? Wouldn't I just have to show the existence of only one invention which became popular, despite being patented, to disprove this? (If I didn't reject it on the grounds that it is fallacious reasoning)
Yes, you could argue that he would not have been eulogized, but you would need to have a better argument and it would be speculative at best.
No, you're right, it's a specious argument. While digging for data to back it up, I found this interesting quote from Camenzind:
There are no patents on the 555. Signetics did not
want to apply for a patent. You see, the situation
with patents in Silicon Valley in 1970 was entirely
different than it is now. Everybody was stealing
from everybody else. I designed the 555 Signetics
produced it, and six months, or before a year later,
National had it, Fairchild had it, and nobody paid
any attention to patents. The people at Signetics
told me they didn’t want to apply for a patent,
because what would happen if they tried to enforce
that patent, is the people from Fairchild would come
back with a Manhattan-sized telephone book and say
“These are our patents, now let’s see what you’re
violating”. It was a house of cards – if you blew
on it, the whole thing collapsed.[1]
That's strong evidence that the lack of patent for the 555 was unrelated to its particular success, so I admit I was wrong about that. But if Camenzind is correct in his history, then it sounds like Silicon Valley had a solution 40 years ago to the patent mess: pretend they don't exist!
There was an interesting bit of industrial espionage intrigue around the 555, a story he told friends and family many times.
As he mentions in his book "Designing Analog Chips" he was at one point stuck on a design that required 9 pins, and that forced it into a 14 pin package. He initially submitted this design to Signetics and it was this design that someone in the company leaked to a competitor.
But the larger package bugged him and he kept at it until he was able to get it down to 8 pins which allowed an 8-pin package, much preferable over a 14 pin one. The competitors, unaware of the improved design, came out with the larger 14-pin package and this gave Signetics a significant edge when they started selling the 555.
>sounds like Silicon Valley had a solution 40 years ago to the patent mess: pretend they don't exist!
No they had a solution and it was MAD.
>The people at Signetics
told me they didn’t want to apply for a patent,
because what would happen if they tried to enforce
that patent, is the people from Fairchild would come
back with a Manhattan-sized telephone book and say
“These are our patents, now let’s see what you’re
violating”.
And it worked really well for a time. Unfortunately someone thought they could win and decided to "go nuclear" and now we're all stuck in the middle of a patent war.
Since when is non-existence a cause? I think you need to re-phrase this a bit. The simultaneity of non-existence is precisely not a cause. The fact that they are independent (statistically) is perhaps more interesting. Certainly from a public policy perspective.
The decision not to patent can be a cause. But other accounts in this thread suggests that patenting or not would have had little influence in the environment in the Silicon Valley.
Probably wouldn't have amounted to much, though. If Signetics had patented it, purchasing it from him, he would have gotten what, maybe a few hundred dollars? That's about what most engineers receive for their patents from the employing company.
Now, if he had left Signetics and founded another company to create the 555, then he might have been rolling in dough.
But instead he created a company that pioneered the concept of semi-custom integrated circuits, and rolled in the dough. :) I think the mentality that he was somehow cheated out of something would have been seriously lost on him. He did what he did because he loved it and the money was secondary.
Hans took on the 555 timer project after he had already started his own business and did it on contract. The contract helped him survive (with a wife and 4 young kids) while he built his business.
I never once heard him comment on disappointment about not getting royalties on the 555 timer, in fact, I never once heard him mention anything about royalties on any of the work he did for other companies during the 60's and early 70's. I don't think it was in the mentality of era. In the time since then, the success and popularity of the 555 timer was enough of a reward for him.
It definitely led to less rewards for him, but increased rewards for society. I wouldn't expect any for-profit business to be more concerned about society, though, strange that it wasn't patented.
I recommend Chapter 11 (starting at page 141 in the PDF file) for a detailed personal account of how he developed the 555 timer chip.
I'm also finding that the first 30 pages of the book is a very readable introduction to how semiconductors and transistors work -- it's written from an engineering mindset as opposed to the usual physics point of view.
I have many cherished memories of playing with the 555 timer. It was what drove me to learn how to hack PICs, and then the Arduino. After that came the mini2440, and the fun never really stopped. I built the siren described by a fellow hacker below, annoying my then girlfriend (and now wife) to no end with it. Then I built some fun strobe lights with LEDs. All for a couple of bucks, and available at the local Radio Shack. I still have a couple of them laying around, and some PIC 16F84 too. Fun times.
I was going to submit that but HN dupe check prevents it (it was originally submitted 1,001 days ago). Unfortunately it also appears that you can not discuss HN submissions that old, so there is no appropriate place to then discuss the other chips in that article.
Of particular interest to me is the DMD (Digital Micromirror Device). I had not really thought about how a digital projector might work, but if I had I would've said that it would be screen based. I wouldn't have guessed that the light would be projected at a microchip which would then reflect back only the pixels to be lit using incredibly small mirrors. And that this would happen so quick that the light would be sent through a gel of colour and would cycle through red, green, blue, and that the mirrors would reflect only a given colour at that moment for the pixel concerned.
To be fair the DMDs in projectors are a fairly recent innovation. I was interested in projectors in HS, and at that time all projectors were either CRT or LCD based. However, DMDs were an "exciting new technology" that had promises for cinema projection since they could be significantly brighter than the CRT and LCD projectors of the time. I certainly didn't think they would be available in home projectors in under a decade like they were (though the color wheel certainly helped with that)
I know, right? Someone recently contracted with me to build them a circuit to prevent accidental triggering of a pushbutton. So I did something where a double-click of the pushbutton results in a single contact closure output. Entire design was done in an 8-pin Tiny13 that cost about $0.50 @ qty 10 and the PC board is barely larger than the switch itself!
Still amazes me that I can buy an entire computer for fifty cents.
I was boggled by reading Bunnie Huang's diagnosis of some counterfeit SD cards. The SD card manufacturers have realized it's cheaper to put an entire ARM processor on each SD card to diagnose and bypass bad blocks of flash instead of using a very expensive machine to do the testing. Think about that for a sec.
Love the chip, the guys a hero up there with the also recently departed Bob Pease and Jim Williams.
I realize this is hacker news and pretty digital heavy, so I do feel the need to point out that "Incredibly, he created the chip alone, spending a year designing it by hand as a freelancer." and "His design incorporated 23 transistors, 15 resistors and a pair of diodes." is in the analog EE world not so much incredible as "probably about typical months design productivity". As usual in the analog world R+D and "productification stuff" and extensive testing burned up the other 11 months.
So Pease, Williams, and now Camenzind are gone. Hard to believe Widlar's been gone 20 years, now he was a guy who packed 2000 years equivalent of wild heroic living into, unfortunately, only about half a century. Losing a whole generation of analog arch-wizards. Sad.
My first IC - havent felt anywhere close to what I felt when I built my first 'siren' powered by a 555, a potentiometer and a cheap 4 ohms speaker.. RIP Hans.
That's one of the few chips that I still remember the pin-out for decades later. And a bunch of 74's. I don't know how many of those I consumed but they were a pretty steady item on the shopping list.
The 555 really was cool. I remember being in middle school way back in the day working with 555's. At that time, I had trouble reading resistance values without resorting to a color reference chart, but the 555 I knew inside and out.
When you think about it, a design would have to be 'Uber-Elegant' to have a middle school kid be able to understand it like that. For the time, I think it was.
(As an aside, I can remember going ballistic on one of my friends who soldered one to a board. I would only use them on breadboards because I thought they were too 'valuable'. LOL! Money in middle school terms!)
When those white breadboards came out that allowed you to wire entire circuits with no soldering at all it changed my life. Until then we would wire wrap, or solder to pieces of copper wire used as posts stuck in, erm, breadboard. chips always went into a socket holder whose leads were then soldered to the breadboard (or wire-wrapped).
My first was a HeathKit digital electronics course which contained power supplies, sources of +5/0 V, and leds to read output.
Aww man! I missed this when it happened. I can be dismissive of Arduino-based timing projects, driven in part by cherished memories of making LEDs blink and speakers chirp annoying through use of a 555.
Using an arduino for a timer is a spectacular waste of resources and very bad engineering (unless you already need the arduino), you might as well use a Core i7 to blink a led. The pricing will be about the same, you pay for the plastic package and shipping. Silicon price is very low.
Isn't having two factories, one to produce a 555 and the other to produce an ATtiny, the waste of resources? I personally prefer to only have to keep one chip in stock; the factory-calibrated RC oscillator is more accurate than what I would build out of discrete components anyway.
(Also, an ATtiny is not an Arduino. The Arduino chips cost like $5, an ATtiny is 50 cents. ATtinys have much less flash, EEPROM, and ROM. But can still produce a nice square wave :)
Ah, I loved the 555. I used it when it first came out to build blue boxes [1]. Of course, the frequency drifted all over the place and I was constantly having to retune the things. Even so, it was such a fun and easy chip to work with.
The first thing you usually do with the 555 is to blink a LED, that's the universal "hello world" of electronics. Then, you can connect a speaker and do funny noises. Can't compare it with a PIC or any uC, they are much more complex devices to operate. When you are 10 year old it blows your mind.
Not-entirely-related, but, RetroThing is a great site for all of you retro enthusiasts. They have a lot of actual original content, versus all the regurgitated stuff on all the other sites. (Not HN news though, this is all good regurgitated stuff!) ;D
Disclaimer: I'm friends with one of the writers/editors/major contributors(?) of RetroThing.
The 555 Timer was one of the fundamental IC's of my deeply rooted electronic fundamentals even though I am a modern day computer scientist. I remember using the 555 as a flip flop, a Schmitt trigger and to demonstrate a Monostable circuit.
It's basically a hobbyist's tool. If you wanted to make a simple circuit that would produce a "phaser" noise, it was great. If you were designing something that needed a stable oscillator you couldn't use a 555 because the oscillating frequency would vary tremendously with temperature. Same with the length of the pulse if you used it as a one-shot.
At the company I worked in the late '80s the frustrated management banned its use altogether after a series of failures in the field were traced to designs that relied on the 555 to be more stable than it really was.
If you were designing something that needed a stable oscillator you couldn't use a 555 because the oscillating frequency would vary tremendously with temperature. Same with the length of the pulse if you used it as a one-shot.
Those are all properties of the capacitors used, primarily, followed by Vcc stability. The 555 itself is a very long way down the list of Things That Make Oscillators Drift. Did you try polystyrene or polypropylene capacitors, or NPO ceramics? Supply pushing was also improved with the 7555 CMOS part, as I recall.
At the company I worked in the late '80s the frustrated management banned its use altogether after a series of failures in the field were traced to designs that relied on the 555 to be more stable than it really was.
>The 555 itself is a very long way down the list of Things That Make Oscillators Drift.
Well, this is true, but the chip put constraints on the types of circuits you ended up with.
>Did you try polystyrene or polypropylene capacitors, or NPO ceramics?
Eh, I don't recall, as this was 25 years ago. Probably. We were making one or two of each product, so the cost of the parts was a rounding error compared to development and test.
>Sounds enlightened.
They had been burned a few times in a sort of public way. Yeah, it was overkill. On the other hand by that time you could already buy phase locked loops on a chip, so there wasn't really a reason to keep using the 555 as an oscillator unless you were making something that had to use really cheap parts.
Indeed, but I only came across the obituary today; there seemed to be little discussion on HN at the time, and the retrothing writeup added some background to the story that I didn't see in the other obits.
