• About EEVblog

    Check Also

    EEVblog #1047 – Solar Roadways FINALLY BUSTED! Colas Wattway

    Dave finally puts an end to the idea of Solar Roadways. We have the test ...

    • addidis

      You are usually right on with your rants. This one is embarrassing I think. You assumed the resistors were good. You only had to deal with X of them. Digi-key deals with billions of parts. Measure twice solder once? You basically proved that it wasnt digi-keys fault with the microscope that clearly indicates the manufacturer is to blame. Its clearly manufacturer marked and you go as far as to praise the manufacturer, and rant about digikey , who I will add posted this this morning on twitter.

      @eevblog Our apologies for any inconvenience. Email [email protected] with the issue & your contact info so we can correct your order.

      • The photos don’t really prove anything. Digikey could very well have mixed up the parts with lesser tolerance ones, we don’t know for sure.
        It’s best to start with Digikey and work backwards to the manufacturer. Digikey are in the best position to investigate.

        Yes, I assumed the resistors were good because they are the exact same ones I have ordered and used before with no problems.
        Yes, I should have done more exhaustive testing. I was lulled into a false sense of security based on all my previous batches being spot on.
        At least I admit it.

        • addidis


          • Jason

            Surely you can’t criticize Dave for not checking the resistance of the part before soldering it it and not criticize DigiKey for not doing their own quality assurance?

            Even if it is the manufacturer’s fault, DigiKey should still have a system in place to test their components to verify that they do have the right labels on the parts that they are on-selling to the public.

            The fact is that small companies, and hobbyist designers do rely on the quality assurance systems of the larger companies when they can’t afford to do it themselves (or hire someone else).

            • Even if it is the manufacturer

            • addidis

              I agree, it is as reasonable for dave to expect to get parts within spec as it is for digikey to expect to from suppliers.
              I dont check every resistor I get , but then again when a design hinged around a tight spec on a penny part perhaps I would check those.
              My point is They reached out to him before he even bothered. Then posts this , then damns the person who is like what are you doin man THEY came to you to resolve it.

              Ive had one problem with one part from them. I went online talked to a live chat rep explained what my meter showed *He understood the part was bad from me telling him the meter readings* 2 days later I had a new part free in hand and that includes the day I contacted them late night. Old one went in the trash. Thats the best service ive ever had.
              I just know how far they will go to fix something and guessed (correctly by daves admission) that he didnt even give them a chance, or see them reaching out to him 10 hours before my first post. How often do you see companies go out of their way for you.

              • Oh FFS, don’t be so serious.
                I run a video blog where people like to see me share my experiences. This was a good story too good to pass up sharing.
                What was I supposed to do, contact Digikey and then wait (maybe weeks) and HOPE that they actually a) investigate thoroughly, and b) come back and explain to me exactly what happened so I can then put together a fully polished and thoroughly accurate blog about what happened?
                No thanks.

                I have NO DOUBT Digikey offer excellent customer service and will fix my problem ASAP with no fuss. But that is an entirely separate issue to the story about what happened.

                I just know how far they will go to fix something and guessed (correctly by daves admission) that he didnt even give them a chance, or see them reaching out to him 10 hours before my first post. How often do you see companies go out of their way for you.

                Do you think I watch Twitter 24/7?
                Do you know about time zones?
                Do you understand my previous post about it making Digikeys response making no difference what so ever to my blog about it?

            • addidis

              No, thats why I posted. Because your a silly aussie and were probably sleeping when it was posted. So I posted it here for you. Your welcome.

          • Thats what Im saying. But before you publicly slam a company thats already reached out to you to help, on a problem that was most likely at the manufacturer level before you know any thing except that you didnt do your job as a manufacturer any better then theirs did , I would have made an effort to have answers.

            It’s not your blog.
            I shoot from the hip, that’s my style.
            I was not going to wait weeks to (possibly) get a response from Digikey.
            Publish and be damned.
            I didn’t really slam Digikey, I simply presented what’s happened and asked for answers.

            FYI, I didn’t see the Digikey tweet before I posted the video, not that it would have made any difference.
            I expect them to simply apologise and send me new ones. I’d be surprised if they actually let me know the exact cause of the problem. But I’d be happy to be surprised.

            Dont slam companies publicly before you know what who did . When the facts say its identical to the right part which probably means the manufacturer mislabeled it.

            Look at the response below with someone else with a similar issue with Digikey, and this isn’t the first time I’ve heard of it either, I’ve been hearing a lot lately. But that’s beside the point.
            I know shit happens, so I don’t really blame Digikey (or Bourns), but when shit does I like to talk about it on my blog.
            What difference does it really make who’s fault it is anyway, fact is it happened, and it’s a useful story for others.
            The blog is about telling stories, and I think most people appreciate me telling the story as it happens.

            You say you dont know , what does the twitter say , what does the title say. Whats really dodgy here?

            What’s really dodgy here is you telling me how I should run my blog.

            • addidis

              I just shoot from the hip ….. works for u works for me…

              • I just shoot from the hip

            • addidis

              I never expected to win , but i did expect you to not get all this is mine i win hardy har ahr on me .
              I expected that you would reply : “oh kraky I didnt see that addidis when does a company ever do that ?!? Kudos to them lets see what happens! ”

              Not im the dictator of this show stfu. If i hadnt respected you , I wouldnt have said any thing as this makes you look bad. Thats my opinion, and when doing a show to the public their opinions count. Real companies pay for opinions.

              • I get people like you on a daily basis basis telling me what I should and should not do on my blog, what I should and should not say, what I should and should not have done etc. Most are well intentioned, some are just being dicks. But overall I think I do an ok job of being tolerant of the dicks, and listening to the well intentioned and well reasoned opinions.
                Do you know how many comments and opinions on my show I get each week? (hint, it’s a lot) with almost every single person having a different opinion or viewpoint, I cant possibly please all of them, ever.
                Ultimately it’s my show and I’ll do what I want.

                My reply was intended to be a lighthearted response to your (maybe?) light hearted comment.
                If you have a problem with it, that ain’t my problem.

            • addidis

              As previously stated I started out with respect for you, and looking forward to rants. I didnt realize I had to have the same opinion as you to post. My bad.
              You keep saying “i didnt slam digikey”

              Blog #133 rendering now. @Digikey & @Bourns had better look out… about 18 hours ago via web

              @digikey and/or @Bourns will get hammered in my next blog. Someone has screwed up parts in my

              • *yawn*
                I hardly hammered them, I was just telling the story and asking for answers.
                Get over it.

            • addidis

              I know i know its your twitter feed , you win.

            • addidis

              response to being shown that what you say now directly conflicts with what you said hours ago:

              Get over it.

              a true wanker response imho.

            • Gonnafail

              According to my twitter feed Dave Posted “Blog #133 rendering now. @Digikey & @Bourns had better look out…” at 7:26am EST. the Reply from digikey came at 10:43 EST so by the time Digikey replied the video had already been shot and submitted to youtube. One can hardly fault Dave for not commenting on Digikey’s response in the video considering it hadn’t happened yet.

            • Rob

              To addidis:

              I think it is well established that when this self-important guy is shown that he is wrong, he just says this is my blog and I do what I want. so, really you are wasting your time.

              Talk about irony. He is blaming companies for not testing their parts before sending them to the customer, but yet that is exactly what he did himself! Shipped parts without testing them.


              What a douche.

              • How am I wrong?
                If you mean that I tweeted that Digikey and Bourns might get hammered and then I actually went very soft on them, then yeah, I’m wrong.
                If you mean that I simply pointed out some bad parts I got as part of a story to teach people that these sorts of things can happen, then yeah, I’m wrong.
                If you mean that I’m honest with my customers, and thorough providing info to suppliers when something goes wrong, then yeah, I’m wrong.
                What’s going on? All I hear is the usual rubbish, a couple of people bitching about what I have said or how I said it etc, and most of the other sensible people (including Digikey) actually understanding what and how I said something and supporting me.
                If you think I’m such a douche, why do you spend your time watching?

                • Nick

                  Methinks Rob and addidis doth protest too much :>

        • Rubi

          I also had troubles with Digikey, f.e. they sent a wrong coil I used to build a N2PK vna.
          Or they send wrong samples in behalv of Ti.

          But usually you tell them and they correct their faults as fast a possible.

          Considering their hough component throughput,and their super fast delivery it is impossible to avoid errors.

          It is not fair to blame Digikey imho.


    • George Graves

      Wow Dave. I think it’s great that you’ve made this post and didn’t just sweep it under the rug.

      “I don’t know who’s actually at fault here”

      I kinda cringed when you said that, because we all know at the end of the day, you can’t pass the blame to anyone else but yourself. I don’t think you are or are trying to. Not at all! And I commend you getting out in front of something like this.

      What I’m trying to say is – that people may have missed that point, so they might give you some grief about that!!!

      And I guess I kinda do expect test gear such as this to be checked/calibrated before it’s shipped – but maybe I’m naive about that? I take it that you weren’t checking for accuracy? Only working/non-working function?

      Having a problem after shipping a bunch of units is a nightmare scenario that keeps me awake at night.

      Hope to learn something from seeing how you handle the situation!

      George Graves

      • Wow Dave. I think it

      • addidis

        Didnt miss that he did catch it, he did admit it. I caught every thing, Including Digi-key reaching out to him. It wasnt a lack of information or understanding that led to my comment. It was indeed information that clearly the author didnt have by his own admission.

    • JC

      Digikey has recently had some problems with their stock. Just a week ago, I bought some 2.2k 0603 resistors, and got what looks like 0603 capacitors, in a bag labeled with the 2.2k 0603 resistors I ordered. Same sort of thing happened a few months back, and our solders tell me they’ve had a few of those problems with other customers too.

      I figure with so much stuff in stock, they are bound, sooner or later, to start mixing up parts. Fortunately, the Digikey customer service folks are pretty good about it.

    • Dave-

      I’m going to side with you on this one. You were following the commonly accepted principle of skip-lot QA, where after a number of examples of acceptable supplier quality, you reduced the frequency of incoming inspections to something like one in ever 5 or 10 lots.

      This is an entirely reasonable position.

      With suppliers like these in the chain, quality at the highest level is a reasonable expectation. It’s also not so subtly implied by the ad copy of both. I maintain you did absolutely nothing wrong by not testing the lot as it came in. Based on your example variation, it’s entirely possible you could have grabbed a conforming resistor to test and called it good.

      You’re not a large manufacturer. You rely on your partners to play a role in the QA and QC process. They dropped the ball.

      • You are pretty much spot-on to what happened.
        When I first released the

    • Brian


      This is the reason companies who want happy customers, test their units.

      Note that every semiconductor data sheet has an Electrical Characteristics table. Those are there to tell customers what they can expect from the part, but they are also the list of tested parameters for the part.

      Actually, you’ve been lucky. Have you ever tested your circuit over a temperature range to see if you still get acceptable performance? Resistors will drift with temp, but the input offset and gain of the Op-Amp will probably drift a fair amount with temp.

      These kinds of things are what seperate hobbyists from succesful manufacturers. You can’t test good quality in, but you can test bad quality out!

      • Karl (not that Karl, the other Karl)

        Have you ever tested your circuit over a temperature range to see if you still get acceptable performance?

        Didn’t Dave build his very own temperature chamber in one of his video blogs to do such things?

        BTW, one buys from a reputable supplier to not have to test each and every detail of each and every component of each and every unit build (e.g. think type approval). If the supplier lets you down he has to share the blame.

      • Have you ever tested your circuit over a temperature range to see if you still get acceptable performance?

        Yes I have, and it’s not hugely significant considering it’s designed to be used in a lab type environment as much test gear is.
        IIRC it’s typically under 0.01%/degC

    • BrizzyMike

      Well at least you didn’t ship 100’s or thousands with defective parts and chances are not all are faulty. Thanks for another interesting blog. At least we may learn from this.

    • PedroV


      Dave thumbs up to publish the story as it was/is.

      I bought one of the uCurrent, and let me tell you that i appreciate your honesty with this.

      Also i got a lesson here. I’m myself building my stuff, most of the time alone and i rely on quality. So through this story i got knowledge of simple thing that one doesn’t expect that can go wrong.

      Would i expect my 0.1% resistors to be grossly out of spec? NOOOO, and not specially after getting confidence in the product/lots as you did.


    • Coder


      I agree with your frustration and sentiments here.

      A serious quality control lapse has occurred either at Digikey or at Bourns. In your case, there is an obvious and readily measurable difference between the right and wrong components, as you have shown. What if the pulse rating of a resistor, or the voltage rating of a chip capacitor was mixed up? Hands up if anyone routinely tests these parameters?! Months down the track, the capacitors may leak, the resistors may go open-circuit and you could be staring down the barrel of an expensive recall. For technical or economic reasons, some parameters just can’t be tested. We must inevitably trust our suppliers to deliver what they promise.

      I hope Digikey gets to the bottom of this. I’m sure it is a rare event, but it would be good to know that they take it seriously.

    • Tim K

      I love how Dave goes off on a rant, but still manages to teach people something.

    • Steve

      Wow, a whole lot of hot air in some these posts, if you know what I mean!

      Dave, a spot on intelligent, albeit emotionally metered response to a problem that you are reporting, basically to your customers. At the same time you are also teaching a lesson to all current and future developers. Bravo!

      Excellent, I understand you frustration as a consumer. You buy a product and you expect the dam thing to work, wants is so wrong with that!!!


    • Bob Tipton

      Anyone who sells a product should be happy whenever a customer brings a problem to their attention, and most are. Whatever negative publicity Digi-Key suffers is nothing compared to what could have happened had they continued to ship out-of-spec parts. In the past, companies had no way to learn about these sorts of problems unless the customer contacted them directly. Now companies can protect their reputation by actively looking online for complaints about their service, and dealing with them quickly and publicly – turning a negative experience into an opportunity to demonstrate the quality of their customer support. That’s exactly what has happened here, and not only has it helped Digi-Key, but it’s helped their customers as well by reducing the chance that future orders will have similar problems.

      • It has indeed helped Digikey and Bourns.
        The Digikey service has been excellent and they been very prompt to investigate as you’d expect. They have found other reels that measure incorrectly. They will investigate further and talk to Bourns.
        Those reels likely would have likely caused issues with other companies too, as you don’t use a 0.1% resistor unless you actually need that tolerance for some reason.
        Will keep everyone updated.

        • Any update after 3 months?

    • walter delbono



    • Fraser

      An interesting blog and resultant comments. I note that some believe Dave should have tested each unit before despatch for tolerance. Whilst this is possibly fair comment where precision test equipment is involved, I feel I must make comment that in this case we are talking about the uCurrent. IMHO Dave has sold these units at a very reasonable cost and whilst this component fault is annoying it should be remembered that time is money and if people want individual unit testing rather than batch testing you can expect the cost to increase accordingly. uCurrent for $100 anyone ? I feel for Dave on this one. Modern components are often high quality and close tolerance and Dave has just been very unlucky. The cost to rectify this situation is being born by Dave which has to be painful. Has this put Dave off doing these small production runs as a favour to the community ? I hope not. All in all a very interesting Blog. I liked the explanation of 4 wire resistance testing… I have it on several bench meters and now fully appreciate it’s usefullness. Thanks Dave.

    • Erik W

      Wrongly labeled and fake components may be due to the component shortage that is taking place right now. Factories were shut down during the financial crisis and now the industry have problems delivering components.

      Distributors are searching everywhere to find components for their customers and fake components exist.

      • BT

        Any reputable ‘franchised’ distributor has a lot of systems in place to keep counterfeit components out of the supply chain. Fake components are a non issue with the likes of Digi-Key, Mouser, Newark and the other bigger guys.

        Having worked at a distributor, parts that are incorrectly taken from bins in a warehouse are usually grossly incorrect because of a deliberate warehouse configuration and an error of an employee. A lot of warehouses are set up to minimize similar parts being stored next to each other. A problem such as this is almost certainly on the manufacturer of the part.

    • f4eru

      Hello Dave

      If you want to fix it quickly withut waiting for correct replacements, you could solder a high value trim R in paralell


    • Michael Thompson

      I agree: The great thing for us is that we can learn something from Dave’s trials and tribulations.

      Dave, I’m sorry about that stuff going down.
      I hope it all gets sorted out for everyone.

    • Dave,

      You did exactly what your audience has come to expect and grown very fond of over the years. If you didn

      • These 0.1% resistors are going to get a ton of scrutiny from Digikey and the full distribution chain.

        That’s what I figure too.
        If I didn’t do the blog and had simply emailed them like a regular customer with yet another “I got the wrong parts” complaint, they may not have investigated as thoroughly.
        I’d like to think some good might come from this. Who knows, it could even potentially fix some other issues unrelated to my order, but might help others.

    • When testing large batches of consumer products we use a final random inspecting to test the End Product. Based on AQL sampling plan… Randomly test a batches – Qty of test based on Qty of MFG and what quality level is needed…. It’s too expensive to test all but a random inspection gives the best value for money based on the quality level required.



    • Tony

      I have to side with Dave too. When design is known-good, just spot checking should be necessary. Now, if there is problem – like this – testing should be changed to cover the problem area. I guess in the future the spot check will verify proper operation in spot current ranges, for at least ten or so randomly picked modules of the batch, if not for all.

      That being said, I’ve been in similar situation, though I got lucky – I expected chip in SOT6 (so some other similar, don’t remember exactly) and got LM7805 in TO-252 instead. Package labeling matched the component I expected, just the components itself was wrong. Funnily enough they (and mind you, this was/is *very* reputable component dealer) sent me *three* times batch of same LM7805s in bag with correct component’s label before they finally managed to deliver correct components. Had I been in rush to to deliver I’d be pissed, but fortunately this was non-critical prototyping part …

      So shit happens.

    • Pellemannen

      10ohm is small value, maybe somebody couldnt hold their breath, oxidation or something… test set 2 20ohm in parallel. That should zero -down the error.

    • John

      It’s hard to pinpoint fault, really. I mean, it could’ve been a manufacturing malfunction that wasn’t caught at the factory, I’m sure they make millions of resistors. Digi-Key isn’t fault-free either, I’ve gotten a 1Mohm SMT pot when the package said it was a 470 ohm 0805 SMT resistor.

      I don’t know about anyone else, but testing every resistor personally seems to be a little ridiculous. I know it’s a small run, but saying Dave is at fault for not measuring each resistor seems a little…silly. In a real production environment, you expect to be able to get a cut tape or a reel of components and send them off to the fabrication house without having to check every component yourself. Isn’t that the factory’s job?

      Anyways, never seen a 4-wire resistance measurement, Dave. Pretty cool, maybe I should pick up one of those meters. 🙂

    • Me

      Thanks, Dave for telling the truth. Just don’t go to Sweden.

    • Ling SM


      You are doing the community a service. If I were you, I would be very pissed also.
      Digikey response is not bad so far, bad batch happens sometime, your closing remark reflected that too.

      But with your blog, it shall wake Digikey, Mouser, element14, etc a little so they shall at least do some sample testings on those items they can. If they have done that, than many of the customers can be saved of this task, and can confidently keep buying from them than going for the cheaper route of ebay way.

      Excellent!!! Keep up your true reaction.

    • I personnaly like the blog and see dave’s rants on any subject.

      I understand these wrong parts must have annoyed him greatly , sending out units that are out of spec…

      So i have no problem of dave expressing this in his blog.

      It would be nice if Digikey resolves this in a quick fashion he also mentions this on his blog.

      I’m also interested in what the origin of this particular problem was.

    • Dave Wise

      Either Bourns had a glitch or – I doubt it – DigiKey bought fakes. How much incoming QA does DK do?

      If you do four-wire by measuring amps and volts, you don’t even need a current source, since you’re measuring it anyway. Just grab a battery or power supply, and a series resistor.

    • Drone

      Any product like this that relies on high tolerance components and/or adjustments requires test before ship. Especially for the first units of a production run and a statistically significant sample of all others that follow. Real-life examples of how you can put yourself out of business if you don’t do this were used when I was taught probability & statistics in EE school.

      You can wag your Blog at the manufacturing & supply chain all you want, but in the end… ‘nuf said.

      • Any product like this that relies on high tolerance components and/or adjustments requires test before ship. Especially for the first units of a production run

        I did, it just so happened to pass.

        You can wag your Blog at the manufacturing & supply chain all you want, but in the end

    • Laurie

      Stubmled on your problem by chance, but I’m old enough to rememeber a training video 30 years ago when high-reliability was becoming fashionable.

      TRW was supplying space mil-spec which were perfect % at manufacture and even into the distribution centres, but when packaged in smaller quantity packages for shipping to clients, they suffered from poor (read non-existent) static handling procedures. Microscopic analysis revealed ‘puncture’ marks in the resistive coating.

      Your resistors are almost certainly suffering from electrostatic discharge craters. Before you die laughing, consider the fact that the restance VALUE doesnt make the difference. It is the percentage of the ‘active’ area and all SMD active areas are very similar.

      From memory the guy was Dan Andersen – CEO of Andersen Electronics in the space tech arena.

      His company worked with another to combine detergent and a pink colour to give the very first anti-static bags – the pink ones – later used to drape spacecraft during assembly to avoid another squib accident! The conductive skin made all surfaces equal voltage and avoided the ZAP that has caused your problem.

      Bourns are TRW quality. I strongly suggest the
      packagers were NOT static aware.

      Hope this has helped.

    • Laurie


      The company was Richmond Industries and the CEO was Dan Anderson.

      Oh, and I’m five years older than I thought – the 1960’s video was called ZAP and well worth looking at even nowdays.

      Google “Thermoformed Containers for Electrostatic Sensitive Devices” to see reference to the process.

    • Well Digikey should make clarifications on this issue. I have one that works perfectly and I think they are using it for some reason.

    • Longhair


      When something like this does happen – faulty product sent to consumers due to bad part – who is held financially responsible in the end?

      * Postage Costs: End User, Product Manufacturer, Part Distributor, Parts Supplier, etc.

      * Rework Costs: Time, labor, supplies and additional postage cost if not done in house.

    • Ryan

      The commenters claiming that Digikey is going to “make this right” are delusional. Shipping a handful of new resistors is absolutely irrelevant. None of the involved parties

    • Earlier this month, I had a bad batch of resistors from Digikey, but it was clearly their fault rather than Bourns’s. The parts were supposed to be 1.00k 1%, and the bag was so labelled. Inside, however, the parts were labelled 10R2, i.e. 10.2 ohms, and that’s what they were.

      We called up Digikey, who were very responsive, and the following day we had a second batch of resistors with exactly the same problem! They obviously hadn’t checked the stock bin before resending the parts.


      Phil Hobbs

    • Great video and proof that something is wrong. I think the comments were just as enjoyable. Nice that Digikey is vocal about the issue but I would have expected them to be less abrasive.

    • Roeb

      Great video Dave, love the four wire resistance measurement technique.

      • eStu

        Is there any drawback (beyond effort) to rolling ones own .1% resistors by combining common 1% types in series or parallel, confirming the value with a 4-wire ohmmeter? ie, are .1% resistors much better in stability somehow?

        • 0.1% resistors are generally available in lower ppm drift versions than 1% resistors.

    • ben

      So which is the better portable microscope value?

      Extech’s MC108: Digital Mini Microscope
      (7x-27x optical zoom)

      Microscope Wand
      (10x-200x optical zoom)

      USB Pen Microscope

      I’ve read for smt soldering, its best to get one that can switch between 10x and 20x at a distance of 6″ (most likely requiring a .5x barlow lens, so I guess you need 20x and 40x magnification to start)

      • No idea, I don’t have the others to try them. I’ve found that x6 and x8 are just right for SMD soldering.

    The EEVblog Store generally ships twice a week, on Tuesdays & Fridays, Sydney time. Dismiss