Some 555-Timer-Dies

[Noopy]

Absoluty right. 
I recently got a document showing the differences between some more versions of the 555.
Perhaps I´ll draw some schematics of the 555-variants as soon as I find free time.  Perhaps I find some parts that are not in the datasheets.

[magic]

You are sure that the Fairchild/TI chips that you decap are genuine?

Zeptobars has already posted pictures of some chips they thought were real, but it later turned out they were fakes.

[Noopy]

You are sure that the Fairchild/TI chips that you decap are genuine?

No, not 100% sure. 
You can only conclude what is likely in view of the appearence of the die.

[magic]

I don't know what genuine Fairchild chips are supposed to look like, but that style of DIP8 package with grainy top and two perfectly flat bands along the edges is something I have only seen on fake chips so far. Specifically, on fake NE5532 (I still don't know what they really are, but certainly not recycled 5532s and not the common RC4558-style counterfeits identified by Zeptobals) and on an NE555 which was marked "TI", but IMO it isn't real TI because all TI packages from legitimate sources I've seen so far don't have those flat bands, only a dot in the corner.

The notch on the UTC package looks very similar to a lot of fake chips these days, maybe those chips are rebranded Youwang or maybe somebody else just uses similar molds as Youwang or maybe the chip you decapped is a fake Youwang

[Noopy]

In my view the outer appearance can be misleading especially in case of the 555 which has been produced for nearly 50 years by many companies.

For example this one looks very strange to me:



But the die is nearly the same as in this one which looks also unusual but not really fake:



In fact the history speaks for an original:
The lower is a older Sescosem which changed to Thomson (the upper one). Inside are the same dies but slightly revised.
You can´t be 100% sure but thats convincing for me.


Also this one looks kind of strange:



But the die seems to be the right one:





Let´s discuss what a (555-)fake is:

1. There is no die or a completely different die
You will find out very quickly. 

2. It was some old 555 which was relabeled to make it look new or more expensive (special company)
If the original die has a marking or a structure which is known you can recognize the fake.
But especially with newer or "no-name" products you can´t be sure whether it´s a relabeled fake or they have bought the design.
With every 555-die you see your success rate in recognizing relabeled fakes gets higher as you know a lot designs and see different kind of for example a TI NE555.

3. A new 555-design is labeled with a "famous" 555-name
I´m not sure whether that´s a good business model anyway. You have to design a new chip which you can sell only very cheap.
Such "die-fakes" are a good opportunity to earn some money if you can put a cheap existing die (for example a medium power NPN) in a more costly package (2N3055 or some newer design).
I found this strange TI-NE555-dies which I posted further up and were also posted by zeptobars. They could fit in catogery 3 but I´m not sure.
See also: https://www.richis-lab.de/555_10.htm
The marking is sometimes very strange but wouldn´t it be possible that TI has a chinese company building the last few cheap NE555-DIP and they are no longer interested in the perfect quality? The dies of "these modern" TI-NE555 seem to be all the same. There was only one die shrink. Could be a chinese fake-company. Could be a cheap TI-outsourcing... 

[Noopy]

Today I have a TDB0555 produced by Siemens:



It seems that they had an own design.
It has interesting resistors with round edges.

Apparently the 555 had seen corrosive chemicals. There is severe damage around some bondpads.

Whole story here (in german):

https://www.richis-lab.de/555_24.htm

[floobydust]

I have heard of failures due to aggressive flux getting into plastic DIP packages. Looked at some IC failures at https://www.semlab.com/examples/deviceexamples/ but nothing sorta matched what you have.
SEMLab Example 4 corrosion "... In addition, fractures were found in the plastic molding compound suggesting that the parts were damaged by “popcorning” during solder reflow."
Some SEMLAB pictures can help identify corrosion, ESD or overcurrent damage on dies. The Siemens die looked very nice.

Looking in my junk bin, the oldest 555's I have - Fairchild 7728, Exar 8142 and CA555E RCA141. PM if these are interesting, I can part with them.

[Noopy]

Interesting link! Thanks for that one! 

I am interested in your 555-variants!  You got PM!

[Noopy]

A new variant of the famous NE555:

A ROHM BA222
https://www.richis-lab.de/555_25.htm


Quite interesting (for people who are interested in the NE555-history). 


Besides the study of the die there is a nice cluster of three different resistor types:



- Normal resistors build with the base-material.
- Low-resistance resistor build with the emitter-material.
- High-resistance resistor build with base-material and an overlay of the emitter-material so that the cross section of the resistor is reduced.




Some squares and you get:
- Green resistors: 4,7kOhm, 4,7kOhm, 200Ohm /// 40,5sq, 41sq, 1,5sq
=> 115 - 147Ohm/sq
- Yellow resistor: 100Ohm /// 2,5sq
=> 40Ohm/sq
- Blue resistor: 100kOhm /// 25sq
=> 4,2kOhm/sq

[aheid]

Awesome work, thanks!

[Noopy]

Thanks for the feedback! Keeps me doing what I do. 

[Noopy]

Today a CMOS-version:



Nothing very special but nice to see a CMOS-NE555 with smaler structures and protection circuits at the bondpads...

[Noopy]

Damn, the link is missing! 

https://www.richis-lab.de/555_26.htm

 

[Noopy]

Today I have four Intersil ICM7555 for you:

https://www.richis-lab.de/555_27.htm

I´m not sure which one is the newer design or whether there is a newer design anyway. Hm...

[Noopy]

Some TLC555:


https://www.richis-lab.de/555_7.htm





CMOS-LTP 

[Noopy]

Raytheon RC555:
https://www.richis-lab.de/555_28.htm

And Raytheon RM555:
https://www.richis-lab.de/555_29.htm

[Noopy]

I got a new Siemens TDB0555:

https://www.richis-lab.de/555_24.htm#Update

It looks different than the first one.
Sometime I´ll have to create a NE555-bloodline... 

[texaspyro]

If you are in the Dallas, Texas area Tanner Electronics has some 555's from the first year of manufacturing (1972).   They are $3 each.  Tanner does not do mail order.

[Noopy]

If you are in the Dallas, Texas area Tanner Electronics has some 555's from the first year of manufacturing (1972).   They are $3 each.  Tanner does not do mail order.

Not today and not tomorrow but who knows perhaps I have to take a vacation in Texas. 
Thanks for the hint!

[texaspyro]

If you are in the Dallas, Texas area Tanner Electronics has some 555's from the first year of manufacturing (1972).   They are $3 each.  Tanner does not do mail order.

Not today and not tomorrow but who knows perhaps I have to take a vacation in Texas. 
Thanks for the hint!

Unless Tanner can find a new, affordable store location by June... they will be closing.  That would be a devastating loss to humanity.

[Noopy]

Thanks to floobydust et al. here more Fairchild 555-dies:

https://www.richis-lab.de/555_30.htm

It seems that Fairchild used different designs in the LM555 and in the µA555.

Does anybody know why they had two different 555-variants?

[magic]

Fairchild acquired Raytheon's and later Samsung's semiconductor divisions in the 90s. That's how they ended up with two different 5532s, for example. And they had their own designs to begin with.

In particular, looking for Fairchild LM555 datasheets I found a Samsung-style datasheet but with Fairchild logo.
http://www.elenota.pl/datasheet-pdf/158320/Fairchild-Semiconductor/LM555

The LM555 could also be fake.

[Noopy]

Fairchild acquired Raytheon's and later Samsung's semiconductor divisions in the 90s. That's how they ended up with two different 5532s, for example. And they had their own designs to begin with.

 

A view minutes ago I was told that Fairchild belonged to National from 1987 to 1997.

The LM555 could also be fake.

I always said that you can´t be 100% sure whether a part is a fake-part. But somehow I don´t think the LM555 is a fake. We will see as soon as I have mor LM555... 

[Wolfgang]

Thanks to floobydust et al. here more Fairchild 555-dies:

https://www.richis-lab.de/555_30.htm

It seems that Fairchild used different designs in the LM555 and in the µA555.

Does anybody know why they had two different 555-variants?

Hi Richi, 

thanks again for your interesting die analysis work. One comment, however:
Your setup for measuring 555 risetimes and waveforms I dont like too much. I think it should be
made on a groundplane, and better supply decoupling as well as probing via 450R resistors directly into the (50R)
scope inputs where possible. To me the waveforms look like deformed by excessive lead length and some strays.

[Noopy]

Thanks to floobydust et al. here more Fairchild 555-dies:

https://www.richis-lab.de/555_30.htm

It seems that Fairchild used different designs in the LM555 and in the µA555.

Does anybody know why they had two different 555-variants?

Hi Richi, 

thanks again for your interesting die analysis work. One comment, however:
Your setup for measuring 555 risetimes and waveforms I dont like too much. I think it should be
made on a groundplane, and better supply decoupling as well as probing via 450R resistors directly into the (50R)
scope inputs where possible. To me the waveforms look like deformed by excessive lead length and some strays.

Hello Wolfgang,

I know the measurement circuit isn´t perfect but looking at the performance of the CMOS-555 it is good enough:



All artefacts on top of this square are caused by the particular bipolar 555.

With this in mind that should be ok?

Greetings,

Richard