Fake Audiophile Opamps Revealed


The OPA627 is an old, popular, and very high-end opamp found in gear cherished by the most discerning audiophiles. This chip usually sells for at least $15, but when [Zeptobars] found a few of these expensive chips on ebay going for $2, his curiosity was piqued. Something just isn’t right here.

[Zeptobars] is well known for his decapsulating and high-resolution photography skills, so he cut the can off a real OPA627, and dissolved one of the improbably cheap ebay chips to reveal the die. Under the microscope, he found an amazing piece of engineering in the real chip – laser trimmed resistors, and even a nice bit of die art.

The ebay chip, if it were real, would look the same. It did not. The ebay chip only contained one laser trimmed resistor and looks to be a much simpler circuit. After a bit of research, [Zeptobars] found it was actually an AD774 opamp. The difference is small, but the AD774 still has much higher noise – something audiophiles could easily differentiate with their $300 oxygen-free volume knobs.

This isn’t the first instance of component counterfeiting [Zeptobars] has come across. He’s found fake FTDI chips before, and we’re counting the days until he gets around to putting a few obviously fake ebay 6581 SID chips under the microscope.

64 thoughts on “Fake Audiophile Opamps Revealed

      1. The bad filter SIDs aren’t fake. They’re defective chips from a massive chip pulling operation. The leads quite visibly show that they aren’t new chips. The good ones were sold off long ago but whoever had the bad chips realised that you can sell them as well.

        The actual fake SIDs aren’t even SIDs. They’re just some other chip that has a similar package that gets ground and then relabeled. Some of them can’t even be bothered to laser etch the new markings and they’re just printed on.

  1. What’s worse is when you open up the unit you just bought and reveal that the innards are really gizzards and you just paid $900 for a headphone amplifier that’s all of about $10 in parts. All “precision” aside, as an audiophile, there are some serious differences in what qualifies as great (devialet, linn, B&O…) what’s just marketing. Have a look at the Grado headphone amp teardown. http://diyaudioprojects.blogspot.com/2007/08/grado-ra1-headphone-amplifier.html?m=1

    I love the cmoy knock-off in that article with the TI Opamp. Some nice logarithmic pots and a few precision discretes and this can sound great.

    1. The audiophile stuff is mostly marketing and slightly more expensive and nicer materials used for the packaging. You rarely see “high-end” audiophile stuff in cheap a looking enclosure, not because it wouldn’t work just as well but, because it couldn’t command the crazy price

      1. Sometimes. There is a shitload worthless marketing but there’s also some very good design. The Grado headphone amp is in the worthless category. On the flip side I’ve pulled apart some equipment and found some real high end and carefully thought through circuits and components plus assembly that’s to die for.

        In that case the price is the result of economies of scale. I’m sure the top of the line McIntosh or Electromechanical would be 1/3rd of the price if they sold more than 5 of them a year or something like that. The same could not be said for the Grado headphone amp where selling a single one would have recovered all their R&D costs 3 times over.

  2. This sort of fakery is infuriating to those of us who spend hours chasing down noise or bugs in our code (referring to his earlier article). So I wonder who supplies chips for Lenovo? (not really)

  3. It’s certainly appropriate to cover fake chips on HaD as these fakes are interesting hacks themselves. The economics behind them must also be interesting and it’d be neat if someone could do an “investigative” story into the business behind fake chips.

    1. It’s not just fake chips. The Chinese are known to capitalise on brands by making fake anything. Hell they faked an entire company. NEC had a multi story building in Beijing and a factory where they manufactured everything from fake fridges to DVD players. No one found out about it for years until the real NEC got some support calls for the electronics which were dying, except they never made any of the products in question.

      Comparatively this is quite a small scale forgery.

    1. If you can manufacture a limited number of fakes cheaply ($.10 a piece say) you sell them cheap, but still at a good profit, so you can blow them out and move onto the next project. Better to sell 10,000 fake chips at $2 a piece in a month than sell them at $10 a piece and have to keep them in the warehouse for a year or more. Take the money and run.

          1. That’s what I meant, why sell them so _suspiciously_ cheap? $10 down from $15 is a believable price, I think. Anyone who knows anything about electronics would realize that’s an outright scam. Then again, audiophools don’t know anything, so I guess they’re selling to the right people.

          2. Enough people won’t think at all about it and just buy it. You see it all the time on eBay. A secondhand item goes for not much under the retail price. They don’t think hang on… I could spend a tiny bit more and have a brand new item with full manufacturer warranty.

          3. Oh yea, it’s a red flag for a lot of people, but others think “Yea, I just got a smoking deal!” And on EBay you can get more than enough suckers to make your money fast, why do you think people still buy speakers out of vans in parking lots?

          4. Well, on ebay the price range is usually this big. E.g. 10 pieces of ne555 will cost something between 0,79-18eur. Can you tell, which price is the right one for these?

          5. Personally, I suspect it’s like how some 419 scammers will make deliberate obvious grammar/spelling mistakes to scare away anyone but the very dumbest. That way they minimize the chance of selling to someone dumb enough to buy but smart enough to know who to complain to when the product inevitably fails.

  4. Counterfeit components are a plague that should be eradicated from Ebay. Reporting the vendors and leaving them negative feedback can work, but often they backfire by refusing any business in the future with the buyer.
    Cloning *out of production* ICs is another matter and I’m all for supporting that, so if someone made for example working SID chips clones (not that hard being a 30+ old design) he would get my blessing.

      1. Because some of them don’t actually know what they’re selling. Along with counterfeit chips they often have perfectly legit parts at convenient prices. Nearly every chinese seller on ebay has counterefeit chips, transistors, mosfets, capacitors and even resistors (metal resistor especially) plus other genuine parts, it is hard to hit the fakes market without hurting ourselves.
        A viable way of hurting the sellers who don’t act against fakes, and consequently the makers too, hopefully, would be buying small samples of random parts, then after some time expose them anonymously on a site made for this purpose.

          1. This is a typical case where something that is illegal could actually be right. Textool ZIF sockets prices are clearly too high considering they changed nothing in decades, therefore the price is that high “because they can” (read: lots of lawyers ready to strangle any emulation attempt). In this case too I’m all for cloning, provided an adequate quality level is offered.
            Chinese ZIF sockets clones are surely of inferior quality compared to original Textool ones but I still have to see one fail in a measurable lifetime; small differences, if any, IMO don’t justify an order of magnitude in price difference.

  5. Ebay is a magnet for scammers. That is why I have not done business there for quite some time now. Sure I’m missing out on a lot of great deals but at the same time I’m avoiding a lot of rip offs too.

        1. Then he shouldn’t have bought copper cable in the first place. Everyone and their mum knows that those are better suited for trumpets and other brass instruments.

          Foor woodwinds, he’d be better off with a balsa wood-sleeved cable with a maple core.

    1. I have a buddy who swears by analog synths and audiophile gear. I recently showed him my new bass line synthesizer and he was blow away by the rich sound and deep bass response. When he asked what circuit I used I opened the case and showed him the lone Arduino Uno mounted inside. Ha!

        1. oh youbetcha there is… a couple of them even. typical small tube has a socket diameter of a dime, typical medium tube has a socket diameter of a quarter, typical large tube has a socket diameter midway between a Franklin and an Ike. You could fit a boatload of micros under most tube sockets and if you did it right no one would notice. Soldering might be challenging, but then that’s half the fun.

          Heck, throw in an LED for some lighting effects up through the base (a number of designs do this now, and no one has yet pointed out that the non-visible light output from the LED might be effecting the spectral purity of the anode current, but what the hey…), and everyone will be so wakka wakka over the cool lit-up tube, no one will bother to check to see if the LED has any adjacent friends hiding under the base.

    1. “So he had to open them up to know that they were fake?” –> Yes. He opened it up to determine without a doubt if it was fake. It was.
      “What’s the point in buying the original part when your ears can’t tell the difference to a fake?” –> Where in this article or his website does it say he ever even hooked them up? He saw a deal that was too good to be true, then set out to see if his suspisions were right. They were.

    2. He did list several of the differences between the two chips from a pure datasheet perspective, the key two being more noise and something about different voltages.

    3. Audiopfiles are able to tell the difference _most of the time_, BUT they cant tell which one is which :D They can tell you “those two sound different”, but picking the “good one” will be as likely as a coin toss. No matter if you test speakers, headphones, CD decks.

      Of course then we get to speaker cables, power cables (lol), and volume knobs where audiophiles SWEAR they hear the difference, but fail at ABX testing

  6. I’ve run into fakes before at a manufacturer of high-end weather instruments. We received a shipment of ADCs that were supposed to be MIL spec, temperature range -40 to +60C, supposedly from TI. They crapped out at -10C. Called TI to complain, and was informed that the lot number was not theirs…they were Chinese counterfeits.

  7. bought clones labeled with an in-house partcode
    had half the alloweable power dissapation in datasheet :(

    this was a part with CURRENT regulation as a feature…
    so much for using the feature, had to use resistors,
    it turened out to half-regulate

  8. Even worse, to me, than counterfeit components – which at least actually perform some function (the examples above notwithstanding) – are the obviously useless cables and adapters plastered all over eBay, DX, and similar places. For example: sub-$5 cables with an HDMI connector on one end and an HD15 and two RCA plugs at the other. Not only are they taking advantage of the uninformed, they’re also wasting perfectly good connectors and wire (not to mention the fuel used to ship them overseas), which drives up the cost of the cables that perform an actual function!

    1. These cables are often used for specific applications, for example I had an old nvidia graphics card that had a port which looked like s-video but by changing driver settings you could set s-video, composite, or component. Sadly, most sellers don’t tell you which cards the cable is compatible with so its a bit of a crapshoot if they got the wiring right for you device or not though! And really quite shady in any case

    2. That cable is used for projectors and such, it might have other uses too. Just because you dont know what it is doesn’t mean it’s useless. Like … said, that kind of stuff is specifically targeted at people who think throwing 2 connectors on a cord means they’ll magically work how they think it will.

  9. Many years ago (it’s all in the past yer ‘onour) I worked for a firm that could “find” almost any component regardless of age or type.
    It was occasionally convenient to “make” a particularly elusive part by re-marking another (sometimes just making the same part appear to be made by a different manufacturer) or remarking a superior part to appear to be an inferior one. This was usually to satisfy some arcane QC procedure or other on the part of the client. The customer was sometimes aware, sometimes not but we never had any complaints! Ahhh, the good old days….

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.