You know, I've always wondered why surface mount ceramic capacitors are not marked. On more than one occasion I've had to remove a capacitor from a board just so I could do an out-of-circuit test to verify that it was the correct part.
I for one would be willing to pay double to get ceramic capacitors that are clearly marked.
Will IC's be next?
Is anyone buying the justification that the reason for removing the markings is for environmental reasons?
Hi Rick -- the hot-off-the-press news is that, almost as soon as I'd posted this blog, I heard that the rumor in the street is that Yageo has agreed to continue supplying marked SMD resistors to Digi-Key and other Distributors.
@Max: the hot-off-the-press news is that, almost as soon as I'd posted this blog, I heard that the rumor in the street is that Yageo has agreed to continue supplying marked SMD resistors to Digi-Key and other Distributors.
I just got confirmation of that from Yageo. My two questions are:
1. Since they didn't change the part number, how can I be assured that I will not receive any of the millions of unmarked ones they just produced?
2. Will ALL of their new resistors be marked now, or is it just by request? Aside from looking at each reel, how will the distributors be able to tell the difference?
I sent an inquiry to Yageo a few minutes ago. I'll post the response here.
I just received this from the Managing Director at Yageo:
We are in process with Digieky to make all their inventory in "Marking Product". Then you just order as usual. However this process is still running and we can not make a digital swtich. Please understand.
I think I'll steer clear of their product for a while until you folks use up all of the unmarked stuff. Whoever uses the last unmarked resistor, please give me a heads-up!
@Etmax: Often I get parts that are not what's on the label and so I now check all parts received before they are loaded on the board.
It just blows my mind that every part isn;t identified in some way -- even the smallest parts would support colored dots -- also it wouldn't be beyond the bounds of possibility to have different colored packaging materials -- one for resistors, one for capacitors, and one for ICs...
Absolutely, and while we're at it why don't high pin count IC's have a mark in the molding every 5 or 10 pins? Amortised across the No. of chips that come out of a mold die it would be near zero additional cost
These days, component packaging is often based on high volume, mobile devices. The money simply may not be there for multiple form factors (e.g. DIPs and larger SMT packages). The same goes for cost structures. At the high volumes, any fraction of cent can make a difference.
The specific chip may not be a high volume chip, but that manufacturer (or the package manufacturer) likely sells enough of that package type to be concerned about micro costs. Personally, though, I think the mark every five pins is an awesome idea.
I suspect that cost is really the driving factor behind the removal of markings on SMT components. It's not just the ink. It's a completely separate manufacturing step. That extra step can be a significant part of the component cost.
@Zeeglen: You CAN get SMT capacitors marked with value codes - http://www.vishay.com/doc?45028
You had me all excited for a minute there....until I looked on Digi-key. The only ones I can find listed are more than 10X what I had been paying for the same values! It's hard to find them, since there's no parametric search option that I can find for markings.
Do you know of any available marked caps that won't break the bank?
@Rick: Like I said, I would GLADLY pay extra for that.
Actually, I think you said "I would gladly pay double to get capacitors with markings on them." ... So, if I bought a reel of capacitors for $x and market them up, you'd pay me $2x ... maybe I should go into business doing this LOL
@Wnderer: The very least they could do is color code them within a factor of ten.
That would really help, but for 1206 and larger they could just as easily write the value.
On second thought, perhaps they could write the value and then use a color code for the voltage or dielectric type.
Now that we're talking about this, perhaps the reason they don't mark them is that for a resistor you can tell the approximate wattage just from looking at it, and you can guesstimate the maximum allowable voltage too, but for capacitors, it would be good to know the value, the voltage, and the tempco at least.
I was thinking about ceramic 0603s where I standardize tempco and voltage. Below 1nF they should all be 50V NPO and between 1nF and 100nF they should all be 50V X7R. Above 100nF things start getting complicated. And yes in higher voltage circuits we have some higher voltage parts. I guess I see your point.
@Max "the colored bands on the resistors served to distinguish their values,"
Speak for yourself!
By that I mean what if you can;t distinguish resistor colors? In particular ,red, green, brown. I can't and haven't for many years. Even at my first job, I always used 1% resistors because the values were printed with numbers.
I once write a piece called "Help me read this resistor." Perahps it's time to dig it up and post it here.
Exactly -some of the paints are so unclear -red, orange brown can be hard to distinguish. Sometimes the brown band to designate 1% is the same width as all the other bands - so which end do you read from? And if the body colour is other than neutral, be careful - the body colour can mix with the bands to make them undecipherable.
Another PO (no, not Purchase Order) is those thru-hole resistors printed with text - when reverse engineering a board, the damn value is always facing DOWN against the pcb. Same with thruhole caps and diodes - ARRRRGGGHH!
Picking resistors with readable color codes was part of speccing the component back in the thru-holes days. Yes there were these cheap cylindrical brown ones that were almost impossible to read. We used the peanut shaped beige ones which were readable and the 1% resistors had color codes, too. The 1% had an extra color line. I use to train assemblers to read the color codes and use them to inspect boards. Color codes were very useful. Even if they didn't read the codes, comparing to a known good board, a wrong color component was visibly different. Try that with tiny printing.
It's a conspiracy, probably started by Analog Devices and Analogic when they both made ADC, ADC, and PGA modules. They copied each other all the time. I know because I was once given the job of tracing out a schematic of one onf "Bernie's" modules when I worked at ADI right out of college. We copied them.
@Dom Lodge: I wonder what future impact this may have on counterfeiting if it was to be applied to all components. One fo the first things to check is whether the markings are correct / have been changed
Good point! Without markings we certainly lose a lot of traceability.
I think SMD MOVs have never had any marking on them. I have a product that uses a 5V5 and a 33V MOV and you cannot tell the difference. Of course you know what happened- they got interchanged. We got some very strange results reported from the test department of our sub.
The only way to test an MOV is to force it into clamping using a current limiting resistor- It is not a simple exercise, especially when you are using 1206 components since you first have to remove the part from the design- connect a resistor and then apply a voltage.
Even more concerning is the fact that if they used a 33V in place of a 5V, test would not normally detect it, but the product would be unprotected.
The production house could even place a capacitor there and we would be none the wiser.
It would be so much easier if they were marked!!!!
One product I ran into used MELF diodes with a single blue color band. The circuit they were used in was very strange, appeared that it would draw excessive current. Fortunately when studying an earlier revision of the product that used thru-hole the number 1N5305 was marked on the diodes, turned out to be not a true diode at all but a 2 mA current regulator made from a FET with gate and source tied together through an internal resistor.
SOT23 packages with their 2 or 3 letter codes are not quite as bad, there are several code-lookup sites online. Where it does get tough is when the code is not listed on these sites.
I hear and understand about an undocumented change and I have had even Vishay pull something like this on Digi-Key stocked parts, but changing lead diameters on through-hole inductors (they made 'em bigger, great for board designers when the data sheet and the acutal part don't match PHYSICALLY).
I would simply change the supplier of the resistors. It seems that Visahy Dale, Rohm, Panasonic, and Stackpole are likely to have a substitute for your Yageo part, and the 0603s and larger are going to be marked if my experience is any indicator.
Regarding marking capacitors, tantalums in molded housings are generally marked. The conformally coated Vishay (orange) tantalums are often not marked, and most ceramics are not marked. It is my understanding that part of this is the design of the ceramic capacitors and a history with the military from many years ago, when marking the parts actually degraded the reliability of the capacitors because the ceramic was fragile and thin on outer layers. Today, coating one surface of the ceramic chip and laser marking would seem quite reasonable, but until accepted as the "norm" would likely bring a cost penalty. For me, the cost penalty does not seem like much since most of what I design only makes at most 10K-20K boards of a type per year, and the costs associated with assembly are often much greater than the cost of purchase for chip passives. I agree that having all components marked would be a great help in inspection and trouble-shooting.
The notion that the blind shall lead the blind is simply unacceptable to me as a customer. It's time to take a stand and tightenen up our contract/order wording to explicitly require labeling. This is an example of penny wise for the component manufacturer and dollar foolish for the customer. Once orders to the guilty manufacturers dry up, they'll get the message or rightly go belly up. We must enforce, "The Customer is Always Right."
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.