Reminds me of when some legacy control equipment stated to have problems. Turned out to be a case of die shrink. The SRAM chips were being made with newer processes and the speed was way faster then advertised. So the SRAM would clock in glitches that were never seen by the original SRAM parts. Pretty wild.

When will we see a 160 meter band transceiver made with ONLY chinese 555's and no other semiconductors? 😄

They are taking advantage of special mailing rates for developing countries. It's the way Chinese companies game the shipping. That box probably never made it to that island and instead someone just slipped it into mail coming from Tuvalu...

The other thing with Tuvalu, is that their domain names are "dot TV" and they "export" domain names to TV companies.

I read once Tavalu was a sinking atol and UN was trying to help them by free/low cast shipping in and out of this tiny country. Maybe those Chinese sellers are taking advantage of the free shipping.

I also stumbled upon Tuvalu just a few weeks ago. I have a little maritime interest and Tuvalu came up there. The maritime business is a little special. The have often very complex ownership and the shipping companies often flag their ships in tax paradises to avoid taxes. Tuvalu is most certainly one of those tax paradises. Very interesting video btw.

Did you test the presence of ESD protection diodes at input pins? That would be a clue for CMOS.

They don’t need to meet any data sheet spec. They need to be able to run the circuits people put them in and last long enough. Suggest you connect pins 2, 3 and 6 and see how fast that goes.

Interesting - The datasheet I have for a TI TLC555 CMOS chip (dated July 2019) gives a max freq up around(ish) 2 MHz.

I like videos where you poke around in these basic parts. You asked questions and do things that are similar to what I might do that I don't find anywhere else. I like these videos with basic components

I worked with a guy from Tuvalu, (TOO-VA-LOO), back in the 80's. We got on really well.

There are also originals that are better than the originals. From wikipedia, it appears CMOS parts have been made by TI (and others, intersil etc) since the 1980s. TI's LMC555 created in 2000 thinks it can go to 3Mhz. There are some that will work with high voltages, and some that will work down to low voltages (1-3V). This became relevant for some in the maker community when a popular design of cheap chinese capacitive moisture sensors used a 555 in the circuit, and many times people were powering it with 3.3V or it would have a little regulator on-board. Some of them came with any ol' 555 SOIC package, where if it was an old one only specified working at around the min 4.5V level you'd be lucky if your unit worked. You could test what power consumptions they use as well. God those HP part numbers make it annoying to repair some of their equipment 😫 It'd be real nice if someone leaked their catalogue of numbers for all equipment pre-90's (at least for off-the-shelf parts) 😅

Worked the T2C DXpedition in Tuvalu on 4 bands over the last week. They are there from 10 to 30 October.

A very fun and interesting video. Thank you. We could see the little bumps on the scope in the speed test. This makes a great story, and it looks like you now have a lifetime supply of better 555s. PS - I have a story of clones being better than the original. When I worked in an office we all got new computers. My boss wanted the alsolute best for word processing and spreadsheets so he got a 6MHz IBM AT with 10M drive (I think) and green monochrome monitor. It was built like a tank, the keyboard was like a rock, and the letters on the monitor were crisp and clear. The rest of us were to be getting IBM XTs, so we talked to the salesman and he said he could do a deal on clones, as long as we all got the same kind. These were 16MHz 386SXs with 20M Seagate drives, Hercules compatible video cards and orange monitors. Very nice for the time, faster than the IBM AT, and perfect for staying behind after work to play games with the Hercules graphics. Eventually my boss saw the light and he appropriated one of the clones. His poor old IBM AT sat unused in the supply cupboard for the next couple of years :-)

The 555 is such an ancient part it would be an effort to make a modern clone of it preform as badly.

Interesting test. I guess I'll put some 555 timers on my next AliExpress order. A million uses for "the most useful versatile chip ever made."

Its where the ".tv" TLD comes from.

Tuvalu is a coral atoll. Much of the land area is low-lying and has been disappearing for a long time, well before global warming and climate change were common phrases. I know about it because my missus is from a neighbouring island, Niue. They have been accepting people migrating from Tuvalu for many decades, and the process has been made easier for them due to these circumstances.

A sharper edge is not always good. There are electronics that need a minimum rise time. Too fast, and certain circuits fail. The sharper the edge, the more bandwidth is put out, and sometimes the harmonics cause other stuff in your electronics to fail.

Ever seen a .tv domain? That's Tuvalu.

New project idea: 160 meter band CW transmitter using a 555.

There is an active DXpedition there right now. Two active hams there

Max sink and source output current capability would provide additional clues.

Tuvalu is not far from Australia where I live. It's a poor country from what I just read. They use the Australian dollar, and we do give them aid. It does sound like a nice place. "Tuvalu, in the South Pacific, is an independent island nation within the British Commonwealth. Its 9 islands comprise small, thinly populated atolls and reef islands with palm-fringed beaches and WWII sites. Off Funafuti, the capital, the Funafuti Conservation Area offers calm waters for diving and snorkelling among sea turtles and tropical fish, plus several uninhabited islets sheltering sea birds."

The CMOS-BiPolar question can easily be answered by measuring the current consumptions. Ron W4BIN

You have to do better than the horrible connection of power to the breadboard. Separate power leads have a lot of inductance than is more likely to resonate with ceramic decoupling capacitors than be properly "cancelled" by them. Get the inductance down by twisting the wires together. Use a good bulk capacitor of 10 µF or more. The original 555 timer had large, slow output transistors. They were large so they could handle the output currents specified. The fact they were slow lead to high shoot-through current where both source and sink transistors were partially on. This shoot-through current is what makes proper power supply and decoupling a necessity. The "bumps" on the output are probably due to excessive power supply impedance at the IC.

At least they work, I have had Op Amps that are hollow inside 😂

A theory of mine regarding fake chips is that semiconductor fabs do test runs of their production lines making 555's, 358's, 741's classic cheap parts to tune up their parameters.

Would be interested in some more tests. Things like going to max. output current and maybe maximum power supply voltage. The other thing is that even "original" Components may be improved. I tested some BC547 Transistors, new-old-stock ones form early 80s, newer ones from around 2000 and more recent ones from 2020. All of them (kind of) meet the datasheets. but the newer ones are faster and have somewhat higher gain. Of course, 40 years of storage in the craft cellar can also change values...

How bizarre, I would love you to dig a bit deeper too ! I am doing tests on opamp's (based of some of your work) to the best I can with my skillset and kit ! and it's helping me learn as well as being fascinating. You make a great point that the people who do this cloning probably have better kit nowadays and make a better product. I think your the first on the internet to make this point so I'm calling it !! you heard the theory here first folk's Hahaha ! and I believe my comment is the first also so I am the second person who would throw his hat in on this idea ! and I love the story of the Island and that in of itself is worth exploring...cheers !

would the current draw be a hint as to whether it was CMOS? Or perhaps the minimum supply voltage?

Those are cheap enough to cleave the encapsulation off and see the chip. Ken Shirriff's blog has a genuine 555 chip photo. On the other hand the TI LMC555 is a genuine CMOS part that this may be a clone of.

Interesting, never seen the bump or wiggle but then never really looked for either. To determine whether CMOS you could check the supply current (conventional parts around 2-3mA, CMOS around 50-110uA, with same supply and external components).

Look at the output (pin 3) how close it gets to the rail, a bipolar transistor keeps it's saturation voltage while a mos transistor makes an resistive connection which gets real close to the rail without load.

Ive had this with things from aliexpress coming from or via Tuvalu. I knew of the place because the machines I work with at work are designated with names of countries (easier to remember than numbers) and one of them is Tuvalu.

I tested my own ne555 v.s. some banggood ones I got a few years ago and the banggood ones behave exactly as the ones you are measuring. They have faster edges probably about the same. I think these are just faster bipolar versions. I checked the resistance between the control and vcc pin and the original ne555 is about 4.8k and the banggood one is about 5.6K. The original draws about 5ma with a high output (not oscillating) and the banggood ones were about 3ma. I don't think mine is a CMOS part as typically the resistance ladder is higher and the quiescent current is much lower. It must just be a bipolar 555 that is redesigned to be compatible with a newer process. I don't know about yours, but I think this is a safe bet for mine.

You've violated Bettridge's law of headlines here.

In the early 2000s, I used to get some medicine online from a sketchy website and it came from Vanuatu, a similar tiny South Pacific nation.

Well, the NE555 comes from 1972. At this time the world was still not very familiar with really good transistor designs. So this device comes with some very strange properties. First of all: the output stage is not shoot through protected. So it uses up to 200 mA or even more during the switching phase. It is necessary to operate with at least 220 nF immediately between the power rails to compensate for power line fluctuations. Furthermore the inputs of the comparaters have low impedance due to the voltage devider consisting of three 5 kOhm resistors. Higher resistors would be preferred if you use the voltage control pin for any reason. And the output stage is not really symmetric and it doesn't go from rail to rail. So you cannot control precisely the frequency and the duty cycle by direct use of the output pin. You have to use the (hopefully stable) operating voltage and the discharge pin to ground. All these issues have been solved in the low power CMOS versions like the TLC555. And it can be operated furthermore at least with twice of the frequency of the NE555. So operation up to 2 MHz is generally not the problem and the switching signals are steeper anc cleaner at the same frequency. That's why I also prefer the use of the CMOS versions over the bipolar version of the 555 timer.

Tuvalu also exports excellent coffee beans.

Texas Instruments make SE555, LM555, TLC555, NA555, NE555, SA555, LMC555, and they vary in speed, voltage, and other parameters.... 🤷‍♂ Maybe the clones took one that has the 555 number and just passed it to the fabs and they keep on making these...

I've only heard of Tuvalu because there was recently a DXpedition there that I made contact with.

I bought some cheap irl540 mosfets a while ago. They turned out to be better than originals considering Rds on resistance. Lucky this time, you cant know how they will turnout to be.

In the time it took to watch the video, I found as many different block diagrams as there are 555 variants. The block diagram of the signetics shows one transistor while other show as many as four. That’s one example. Another is how the comparator polarities are different in some variants. Would it be possible to set up the diagrams on bread boards to determine what makes one better of will that introduce to many more variables? Or run a spice simulation of the original vs a current block diagram ? I’m truly curious.

The fakes probably have a microcontroller inside running a 555 emulator. :-)

I'm going to dig through my 555s now and see if I can replicate this, the rise time looked quite a bit faster.

You cold also check the Vih Vil input transition voltages, likely different if CMOS.

Can you really call a Chinese version of a jellybean part a "fake"? Dozens of companies have made 555s down the years and I wouldn't have thought there would be a patent on them anymore. The only fakery is the passing off of the logo, and the ones I've seen had some Chinese identifier on them.

How much current can they source and sink? Speed isn't everything there's power too.

First thing i tested on these when i received them is the maximum speed in astable mode, and i noticed it could easily reach double the frequency stated in the datasheet. Compared it to an 'old' 555 and it performed a lot faster than the 'original'. Wish the same was true for opamps.

Tuvalu is clearly in the drop ship and repost business as well.

I enjoy watching the Olympics, specially the openings, so Tuvalu is pretty much in my memory.

Did you check the current consumption as the old 555 was not at all very efficient.

I think the 555 is such a robust circuit that even faking it is going to be fine. How those compare to ICM7555?

Can you compare it to a genuine CMOS 7555 part? Might be interesting.

Well, it would naturally hold that both the die size being smaller would increase speed and raise profits. But, if they shrank the die, did you lose output current capacity? Nice video & 73…

I think what's going on with the Tuvalu thing is a trick you see the Chinese sellers using lately to keep the slow cheap shipping slow and cheap. IIRC poor countries get subsidized international mail and China got kicked off that list a year or two ago, or something like that. So now it seems they're forwarding the cheap little "e packet" type parcels through countries that still get reduced rates. The last cheapo forget about it order I put in came from some other out of the way country with the usual chinese label under that one.

Im curious if the internal voltage divider still burns as much current as the original parts. The fake is definitely a lot faster on the rise time.

Common for items from China to have the return address as some small island in the Pacific, I always assume it is to do with post, like it is not going back where it came from, or the postal service is allowed to destroy it.

those HP part numbers always piss me off

Is it possible that they are a legitimate Chinese equivalent to the 555 which normally has a different part number never seen outside China? Then someone arranged to mark a bunch of them as 555 to sell to the foreign gweilo who wouldn't accept the original part number?

Interesting, it depends. You looked at just a single parameter. Maybe for simple applications their better, you might need to look at others requirements, EMI, temperature, applied voltage, and load conditions. The question is what defines a cloned part? Is it the die process? Would a TI or Fairchild parts be clones of the original Signetics? Maybe it just the copy right.

The problem with the clones, not whether they are better or worst than the original, but that it is always a lottery.

Time is just different in Tuvalu . The clone kind of fall in middle when it comes to data factor i.e faster than real old but fall short of the really good chips . The low noise high gain are failures but the old slow ones are fine for it purpose . The cost makes the jelly bean parts so to speak .

If you look pure speed of clines then there is gd32 microcontrollers that are "clones" (they are not clones in strict sense they are as i know 100% comparibile, but use differnt type of flesh memory) and they are running on about 30% higher clock speeds.

7:50 First Ali "bump"was way "slower"

If CMOS, surely the power consumption will be much lower. Decap it and have a good look at the die.

I've had them not withstand the full max voltage.

Immitation would be the highest form of flattery... if not for improvement!

The triple 5kohm resistor network is a triple 100K network . So if measure between pin 8 and 5 you should measure 5k for the old or close to 100K if it’s a cmos. this was the reason for the 555 name the cmos should be called 100100100 :-)

I have some XR2206 chips from AliX and I assume they are clones. Just using for Sine Waves I can get reasonable stable up to 3MHz. The datasheet implies up to 1MHz. Ok its not crystal stable, but still not too bad, and shape is also fairly good. So not all clones from China cheap and nasty.

Yeah, I noticed before, that all 555 isn't identical. One dead circuit, after replacing 555 with OEM from different manufacturer and slightly newer, didn't work as older one.. External values needed to be tweak. Same thing with parts made in USSR.. Moder replacements need to be tweak.. 😂 First time I was shocked, when I sell modern replacement part for USSR part to customer, back in days, when worked at Elfa Distrelec, and customer said to me, that, this time, replacement didn't work as intended.. First thought was about faulty chip, but then I did similar comparison in my lab on breadboard.. Yeah, resistance and capacitance tweaking is needed.. 😬 But then I didn't test further, why that is happening.. I did it later and noticed difference, even, between manufacturers..

I wonder if that would apply to the 556 dual 555

Not really surprised. After all there are more than 50 years between the original and the modern parts. China is silly, as sometimes they make better products than the original (e.g. fake Ferrari bag, had much higher manufavturing quality than original), but they do not build a brand and try to sell it under a fake. As to faster parts, smaller structure size requires less silicon die area. That makes the parts cheaper AND faster! That is the unexpected side effects of greed. 😂

Uh there are CMOS 555s designed and made by TI for a long time. You need to compare it to one of those. It's probably a clone of that.

1:45 that is a RETURN address in Tuvalu........Seriously?? Never heard from Tuvalu? Just like the Fiji Islands...It is a "caldera"ring(volcano) in the water. ........Say NOTHING about where it came from....If I order from Ali, most of the time I get a return address in a town 15 KM from my home here in Holland....I can guarantee that this is NOT the address the stuff was sent from. Stuff gets sent back to there , gathered and auctioned off...

555 comes in a faster CMOS version, ICL7555/LMC555 etc. One way to tell the CMOS parts is that the voltage divider on the comparators is 5k/5k/5k on the bipolar parts, 100k/100k/100k or so on the CMOS ones. Try feeding a small current into pin 5 and see how it responds.

I have 2Mhz 555's from TI, maybe those clone one's are just TI parts not clones..

Not "fake" but at the most "counterfeit" and otherwisw copies.....

What is a fake 555 timer?

the Empire say's ''clones'' R better! & Darth Vader prefers ''Clones''! ok tudaloo! imagine having Fed-Ex on this island! I've read some article headlines, that claim China has lithography tech!

LMC555

If it is better than the original is it fair to call it fake? OTOH if a circuit is designed with a 'real' 555 as an astable multivibrator at 500 khz and you repair it with a clone that runs at 800 khz... the circuit is not repaired.