Although the saying of caveat emptor rings loudly in the mind of any purveyor of electronic components, the lure of Very Cheap Stuff is almost impossible to resist. Sure, that $0.60 Ti ADS1115 ADC on LCSC feels like it almost has to be a knock-off since the same part on Digikey is $4 a pop, and that’s when you buy a pack of 1,000. Yet what if it’s a really good knockoff that provides similar performance for a fraction of the price, such as with those cheap ADC boards you can get from Amazon? Cue [James Bowman] letting curiosity getting the better of him and ordering a stash of four boards presumably equipped with at least some kind of cheapo knockoff part, mostly on account of getting all boards for a mere $2.97.
The goal was of course to subject these four purported ADS1115s to some testing and comparison with the listed performance in the Ti datasheet. Telling was that each of the ADCs on the boards showed different characteristics, noticeably with the Data Rate. This is supposed to be ±10% of the nominal, so 7.2 – 8.8 times per second in 8 samples per second mode, but three boards lagged at 6.5 – 7 SPS and the fourth did an astounding 300 SPS, which would give you pretty noisy results.
Using a calibrated 2.5 voltage source the accuracy of the measurements were also validated, which showed them to be too low by 12 mV. The good news was that a linear correction on the MCU can correct for this, but it shows that despite these parts being ADS1115 compatible and having features like the PGA working, you’re definitely getting dinged on performance and accuracy.
[James] said that he’s going to run the same tests on an ADS1115 board obtained from Adafruit, which likely will have the genuine part. We would also love to see someone test the $0.60 version from LCSC to see whether they can match the datasheet. Either way, if you are eyeing this ADC for your own projects, it pays to consider whether the compromises and potential broken-ness of the knockoffs are worth it over coughing up a bit more cash. As they say, caveat emptor.
And here I thought that lcsc is a reliable source for components, for jlc pcb assembly. Quite disappointing.
The counterfeit parts were sourced from Amazon. The article is assuming they’re the same as the LCSC parts but it wasn’t tested.
I tend to believe that LCSC will sell products correctly “labelled”. i.e. when they buy it off some chinese “knockoff” manufacturer, they will label it as such.
My first experience with chinese “knockoffs” was that I bought SI2300 off ebay. Did not conform to datasheet specs for Vishay SI2300. Also not the right markings… But for my application: Good enough.
Later I came to find “LCSC” and noticed that I could chose from different vendors when I type SI2300. Turns out that HX makes “2300” mosfets as well and they have a datasheet that matches 100% with the performance that I measured. Markings too.
I then went back to the ebay listing and discovered that the “manufacturer HX” was clearly visible in the product photo!
Nothing fishy, genuine products conforming to the datasheet of a different manufacturer than what I initially thought.
Since then I mostly buy from LCSC and I still buy the “knockoffs” because they are already 50x or 500x better than what I really need. (I sometimes use them where the margin is smaller, and I don’t want to bother having different mosfets in stock).
So… Adafruit: probably genuine. They once bought fake nonfunctioning chips labelled MEGA328. They have sworn to only use genuine “tracable” chips from then on.
LCSC: I think so too.
But the weird thing is: TI sells these chips for about $2 while LCSC has them under $1. I don’t know how they manage that. That Digikey charges a 100% profit margin should already be known.
Thanks for reading between the lines for us.
For messing around, i love to order just whatever is convenient…but if the specific point is to investigate a fake supply chain, your purchase methodology matters.
From the original article: “I picked up breakout boards from Amazon for $2.97 if you buy four, and I very much think these are made with the $0.60 parts from LCSC or similar.”
His test compares cheap breakout boards from Amazon with the more expensive option from Adafruit. LCSC has nothing to do with it.
Yeah the wording makes LCSC catch some strays tho
It’s worth noting that the bad boards were actually from Amazon, which is known to sell a lot of knock-off parts. I think LCSC is at least trying to keep their supply-chain reliable and to mark clone parts as not being from the original manufacturer (no idea how successful they are at that).
I agree the price difference of DigiKey vs. LCSC for ADS1115 is substantial. It would be interesting to see a comparison against LCSC part and DigiKey part. Seeing that random Amazon sellers sell knock-offs is no surprise at all.
LCSC have also a dodgy supplychain as I had ordered some xc95xl CPLDs with them. They had the badge/production code lasered off. When asking about it they stated that it was a badge that came from a source that wishes to be unknown..
At least they were honest about it.
I do like to see an in depth analysis of the ADS1115, but a single ADC measurement (Even if it’s from a good voltage standard) does not mean much. How good is linearity over it’s full measurement range? What about drift over temperature and aging?
Also, comparing LCSC with ebay, amazon, aliexpress and all those other cheap chinese garbage outputs is not very useful. LCSC has a pretty good reputation, and if they claim to sell Ti parts, then I tend to trust those are Ti parts.
And apart from that, the “western world” is quite sick at the moment, with monopolies (or small group looking at each other’s prices) are all over the place. Ti itself lists the ADS1115IDGSR for USD2.1 (1k+).
There are a few threads about decapping IC’s and making die pictures on the EEVblog. It would be nice to see some “known good” parts compared with both LCSC and the chinese knockoffs.
For 0.6$, I really, really wouldn’t complain if the part powers up and even close to works
Pretty much, if I want deadnuts accuracy, reliability and “guaranteed” genuine for production and sale, then I’ll buy from manufacturer, if I want hobbyist project I can afford to spend the time messing about trimming, testing, replacing etc and have no liability except my own time.
Joining with the other voices : he bought bottom-of-the-barrel parts from Amazon ( would be the same if bought from Aliexpress or Temu ) . That means nothing with respect to LCSC. They may have cheaper parts, but theirs are correctly labeled as so.
“[James] said that he’s going to run the same tests on an ADS1115 board obtained from Adafruit, which likely will have the genuine part.”
Just bought 2 from DigiKey 6 weeks ago for an automated long run prototype testing rig, they work great.
Since DigiKey is an official TI distributor, Adafruit says they use genuine TI parts and the units received have the TI logo, I’m very confident they are the real deal.
So if his testing methodology is good they will be in spec and if he tests a bunch they will all be well within specs usually with many if them having a substantial margin (because he’s not testing at the allowed extremes).
I’m thinking that if your purpose is experimenting, developing code or just hobbies, the knockoff would be fine. Used these things in this way myself, but when it comes to more critical tasks, the cost of the real part is not a problem. I almost wonder if TI’s sales of their parts have benifitted from people buying the fakes and becoming familiar enough to use the TI part in products?
Yeah i think this is a good frame. If i’m using it for a one-off, i’ll need to characterize / calibrate it myself any way you slice it, just because i’m not truly familiar with it even after i read the datasheet. But if i’m making a thousand of them then i’ll have a totally different relationship with calibration and fault discovery.
FWIW at least some TI parts aren’t even sold on the broad market, they’re just sold to customers who buy in very large quantities, because it’s such an incredible pain in the butt to provide applications support on complex parts to customers who are only going to buy 10,000 parts. And it is so much worse for the applications people when the problem is that they think they’re supporting a genuine part and they’re actually supporting a ripoff that doesn’t work right, and nobody can figure out why it’s doing what it’s doing. And that’s awful because an actual field failure means someone has to figure out where the failure got through testing, and a part that’s exhibiting symptoms of a field failure when it’s actually just a crappy imitation sucks up a lot of energy.
Bought them from “some unknown source”? He’s lucky that they’re even the right part. :)
Seriously, they behave very close to well enough. My guess is that they’re real parts that didn’t make quality control, especially since the error looks like it’s just bad trimming of an internal voltage reference.
I’d take one in a heartbeat if it were sold as being a factory-second. I dunno if you want to be supporting people who will sell these as firsts, though.
The mention of LCSC in this article seems completely unsubstantiated as the parts tested were purchased from Amazon. I’m sad that an HD author I generally respect writes something like this. Maybe the article can be corrected?
Agreed. Why even mention LCSC? What involvement did they have in this?
Sorry about that, I have removed the insinuation that these were LCSC-sourced parts. You’re right that there’s no evidence for this, though it would be nice if we knew where those generic ADS1115 boards on Amazon etc. get their chips from.
“Sure, that $0.60 ADC on LCSC that swears it’s a Ti ADS1115 may be a knock-off”
I don’t know what you removed, but this is still in the article and stronger than an insinuation
It’s honestly libel-adjacent. My eyes went wide at that sentence. If LCSC actually has a dodgy supply chain I’d like to know. They’ve seemed pretty trustworthy to me so far.
I bought some ADS1115 clones from Amazon about a year ago when Adafruit changed the board form factor on us. Turns out they had 12-bit ADS 1015 ICs that had been relabeled as TI ADS1115.