How often do surface mount caps fail the one in photo

[sam_sam_sam]

If they fail what type do you use
The ones marked 10-35 are those caps as well ?

[mariushm]

Re: How often do surface mount caps fail the one in photo

The yellow ones are tantalum capacitors, which tend to have the habit of exploding or igniting, with flames.
The black ones with a white bar seem to be polymer capacitors and as far as i know they don't fail more often than others, or well.. don't have a reputation for failures.

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

The yellow ones are tantalum capacitors, which tend to have the habit of exploding or igniting, with flames.

How do I know the value of them and how do I find them on Digikey ?

Is this the right choice for this cap

476
20k
0*15
Here is the link

http://www.digikey.com/product-detai...4-1-ND/3759311

Or is there a better one

Also the black ones as well thanks

[stj]

Re: How often do surface mount caps fail the one in photo

black & yellow are both tant's

value is on them.
black 10-35 = 10uf 35v
yellow 476 20 = 47uf 20v

[Th3_uN1Qu3]

Re: How often do surface mount caps fail the one in photo

476 is 470uF not 47.

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

476 is 470uF not 47.

OK thanks

[Khron]

Re: How often do surface mount caps fail the one in photo

Isn't it "47" and 6 zeros after it? That's 47,000,000pF, which is indeed 47uF

476 is 470uF not 47.

[Th3_uN1Qu3]

Re: How often do surface mount caps fail the one in photo

Haha, my bad. I knew 105 meant 1uF but somehow my mind jumped over 106 = 10uF and went straight to 100.

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

Haha, my bad. I knew 105 meant 1uF but somehow my mind jumped over 106 = 10uF and went straight to 100.

OK which is it 47 or 470uf --->which one you have now confused me

[Khron]

Re: How often do surface mount caps fail the one in photo

47uF - see the reasoning i mentioned.

http://www.csgnetwork.com/capcodeinfo.html

Last line of the 2nd table - 474 = 0.47uF; add two more zeroes at the end to make it 476, and when you move the decimal point two digits to the right, you get 47uF

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

47uF - see the reasoning i mentioned.

http://www.csgnetwork.com/capcodeinfo.html

Last line of the 2nd table - 474 = 0.47uF; add two more zeroes at the end to make it 476, and when you move the decimal point two digits to the right, you get 47uF

OK thanks that what I though that it was 47uf

[ChaosLegionnaire]

Re: How often do surface mount caps fail the one in photo

like mariushm said, the only way tantalum caps fail is by violently bursting into flames. other than that tant caps last a very long time. asus at one point designed a mobo with tant caps for the cpu vrm instead of electrolytics/polymers at the height of the bad caps scandal. the tant caps physically look ok so i highly doubt they failed. look elsewhere for the problem.

[mariushm]

Re: How often do surface mount caps fail the one in photo

I was wrong about the black capacitors because there are black capacitors that are actually polymer or hybrid polymer capacitors, just not in the typical round cases.

Here's for example a Panasonic HX series : https://cdn.badcaps-static.com/pdfs/...057483404e.pdf

Looks like the difference is more or less in the shape of the metal leads.

Here's polymer tantalum hybrids (from what I remember, they have the "advantage" of being safer than simple tantalum capacitors but naturally there are some tradeoffs) : https://na.industrial.panasonic.com/...ntalum-polymer


Also worth mentioning that unlike regular electrolytic and round polymer capacitors, on these capacitors the bar tells you it's the POSITIVE side of the capacitor. Some capacitors even have the + printed on the bar.

and ps. Remember that some companies (Asus and Asrock are the major ones in IT) which like to print the silkscreen on their printed circuit boards in such a way that the black bar on the pcb is actually for the POSITIVE side of capacitors, exactly the opposite from the common way.
It's OK though, they're consistent, if you see basically all capacitors apparently inserted the wrong way, you know it's the stupid silkscreen.

If you see just a few, it could be that someone already tried to fix the board before you and put the capacitors the wrong way so it would be worth investigating (ex. determine the negative and positive side by following the traces, often negative side goes to ground so that's easy to determine with a multimeter set on continuity)

also a note... polymer capacitors with voltage rating over 25v are kind of rare in this surface mount format, and typically the capacitances are small - see the datasheet above where they stop at 33uF 25v.

// and last edit: on the board on the first post, I would definitely rework some of the joints which seem to have little solder to me. Basically, if you don't have one already, you should buy a flux pen or separate liquid flux. wet the joints with some flux and then add a bit of leaded solder to various terminals on the pcb, all those surface mounted diodes and mosfets/transistors on the left side seem a bit dodgy to me.

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

mariushm
Thank you for those links I have not done much these types of caps before
What would the ESR reading on these type of caps as a general rule
I learned something new today thanks

[mariushm]

Re: How often do surface mount caps fail the one in photo

I'm not sure, tantalum capacitors should probably be between 0.1 ohm and around 0.5-1 ohm, depending on capacitance, voltage rating etc

Tantalum capacitors in lower capacitance values (like 100uF 25v) were usually used in circuits because the ESR was usually around 0.2-0.3 ohm in contrast with electrolytic capacitors of those times which at that capacitance had up to 1-2 ohms ESR.

These days electrolytic capacitors have improved quite a lot but at those low capacitance values electrolytics still have higher ESR values.

Polymer capacitors... there's several kinds, the most common polymer capacitors (the round silvery ones you see on motherboards) will typically have esr values lower than 0.1 ohm, around 0.03 - 0.05 ohm
There's also polymer capacitors that are made differently and they have higher esr values.

My advice would be to go on online stores like Digikey.com or Mouser.com and pick a bunch of capacitance values ( 100uF , 330uF, 470uF, 680uF) and then pick various capacitor series from various manufacturers.
For each of those, on the page of that particular capacitor, you'll see a link to the datasheet of that capacitor.
In the datasheet, you'll see the ESR values (well, you'll look at a column titled impedance at 100 khz and a positive temperature, which is technically not ESR but for our needs it's pretty much the same thing as ESR) for ALL capacitors in that series (all capacitance + voltage rating combos).

So once you compare several datasheets, you kind of get a picture of what ESR values you should expect from such capacitors.

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

I was looking at some data sheets and I have a question what is the life span of these type of cap i am not clear on how long they should last
The data sheet talks about 2000 hours at 85*C

There is a cap that has date code of 0*15 if understand the data sheet it was made 2000
The one in the photo is the one that I got of eBay which look like it is older

If this is the case it is over 15 years old

Keep in mind that the machine that I am talking about was made in 8/2008
Also this machine has an air condition unit for two of the BK2000 units
However the other two BK2000 units are just in metal boxes with no AC units
and in summer time it can be as hot as 100*F also in this box are two servo drive these two box get a little warm

[mariushm]

Re: How often do surface mount caps fail the one in photo

2000 h at 85c means that the capacitor is guaranteed to stay within the specifications in the datasheet or within some specified percentages up to 2000 h at the specified temperature (85c) and with the "stress" specified (current ripple), or something close to this.
In real world, your capacitor won't be working at 85c, won't be subjected to those peak conditions... there's a general convention that for each 10c of lower temperature, you double the lifetime rating.
So your 2000h @ 85c is 4000h @ 75c, 8000h @ 65c and so on.. for 35c average, that capacitor could be rated for 64000h or at least about 8 years of continuous usage.

But, keep in mind that circuits often work even when capacitors go slightly out of specs. For example, your capacitor may be specified to have an ESR of 0.2 ohm in the datasheet, but maybe the engineer designed the circuit to work with capacitors having ESR between 0.1 ohm and 1 ohm, to account for capacitor aging and also to make it easier to switch between series of capacitors if the preferred one suddenly is on backorder (requiring weeks to be made and delivered) or simply to make it possible to use better or cheaper series of capacitors for some territories (for example you may need higher quality caps for countries with heat and humidity, but you could do with lower grade capacitors in temperate climates or in regions where customers don't base their purchase decisions on warranty, so you could sell your product with less warranty and not get complaints).

Even if the ambient temperature is let's say 30c, the capacitor will heat due to that internal resistance (equivalent series resistance = ESR) and other losses, and it can also potentially warm up through the traces in the pcb and the leads, if the capacitor is close to other components connected to same traces (for example a bridge rectifier or a linear regulator could radiate heat through the pcb traces and then through the capacitor leads.
Point is, even after all those hours, it just means the capacitor may be outside the specifications, more than some degree.. but who knows by how much.

So the general consensus is double lifetime for each 10c for electrolytic capacitors. Some companies say 4 times the lifetime for each 10c for polymer capacitors.

If you want to be more anal about it, here's some formulas from Illinois Capacitors:

Page (also have calculators on page) : http://www.illinoiscapacitor.com/tec...lculators.aspx

Electrolytic radial:



Electrolytic axial :


Polymer:

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

mariushm
Thanks for the info

[stj]

Re: How often do surface mount caps fail the one in photo

tants last for decades, dont worry about it.
the only bad tants i'v seen came bad from the factory.
they had bad leakage but uf looked fine.

[sam_sam_sam]

Re: How often do surface mount caps fail the one in photo

A Beckhoff KL2488 8 Channel Output Module ?

Here is the problem I am having with is machine is that I am getting an light-buss error undefined meaning that unknown problem but if you power down the machine and power it back up It might work for awhile and then the same thing may happen two or three time that day or the next day

Also this same machine also has a servo error undefined meaning unknown fault but on this one I am thinking that the
The light-buss error might be related to each other because you get light-buss error or the servo drive error the first error then you might get other error a few minutes or hours Later and they not always in the same order

This happen when you first turn on the machine and may happen again a few minutes or few hours later and tech support is not a lot of help with this issue
Then they tell you maybe you need to have a tech come out to service your machine and it cost $2000.00 to $4000.00 and may not be able find out what your problem is any way because it is not broken to point that you can not use it

We another machine that had a problem that tech support could not really help me with they tell you to check slot of things
I found out what was wrong with it after checking a lot of things that they did not even talk about
Tech support told me the same thing that we need come out and service your machine for same kind a money

Now would the LED in the TOTX170A fiber optic transmitter be going ( bad) meaning that they are getting week

What would be the best way to test there output any ideas

Also how would you test the TORX170 receiver to see if the transistor or LED is getting week

Now I understand how to check fiber optic cable to see if they are good or not