FLIR S40 Repair

[Chanc3]

So we were given an old S45 by one of our customers, which apparently no longer powered up. Naturally, I decided to investigate by stripping it down to see if I could locate any obvious signs of damage. Upon initial inspection, it looks like one of the components has cooked! I've seen this before and read that on older devices that these degrade with age and use and are relatively easy to replace. My problem is, I cannot remember the name of the component nor can I trace where to find them (I'm based in the UK). Can anyone assist?

Please see images attached.

[frenky]

I am pretty sure it is a big smd tantalum capacitor. Similar to the yellow one in front of it.

Try to find it's siblings on the board...

[frenky]

I did not know that before:

Is yellow tantalum capacitor better than the black one?

Actually yellow tantalum capacitor and black ones have the same structure and their main material is almost the same. The difference between them is the encapsulation resin because different manufacturers have their own marking technology and they select resin with suitable color to get obvious mark on the body. Manufacturers in USA or Europe produce yellow tantalum capacitors and Manufacturers in China and Japan produce black ones.
Now, the performances of Chinese tantalum capacitors have reached the world’s leading level. However, there were some inferior goods circulating in the markets at shortage period. Unfortunately, these goods with poor quality were black which makes people lose confidence in black tantalum capacitors. We are sure no matter which color tantalum capacitor is, you will get good quality products from reliable manufacturers.
http://www.sunlordinc.com/FAQ.aspx?cat=12

[Fraser]

Yes, it is a Tantalum capacitor.

It is the same value and rating as the yellow one that is in line with it. Part of the power supply Buck converters if I remember correctly.

KEMET 47uF 20V rated

http://www.kemet.com/Lists/ProductCatalog/Attachments/253/KEM_TC101_STD.pdf

Page 21 refers.

Inspect all Tantalum capacitors for case cracking or evidence of self heating. They can fail in a pretty spectacular fashion and burn a hole in the PCB!  I suggest you remove the capacitor with either 'hot soldering tweezers' or a pair of soldering irons. Trying to lift each end alternately with one soldering iron risks losing a PCB pad as the thermal stress from the capacitor failure can damage the pad Bond strength. carefully remove any residual solder from the pads with solder wick to allow the new capacitors to sit flat. Apply flux to both the pads and new capacitor solder points. When soldering these into place be aware that one end is often tied to the PCB power plane that acts as a heatsink. Quite a lot of thermal capacity or a direct driven soldering tip is often needed.

These capacitors often fail short circuit or low resistance. This shuts down the supply rail that it sits across. Hopefully fitting a new 47uF 20V Tantalum capacitor will allow the power rail to recover and no further damage has been done. If it will fit, I would replace both of those Tantalum capacitors with ones rated at 35V. Do not be tempted to fit common wet electrolytic capacitors in place of theses Tantalum types.

Fraser

[Chanc3]

That's the name! I think the sleep deprivation has gotten the better of me. Thanks frenky and Fraser. Last question, do you know where I can source a couple of this from in the UK that's reliable?

[Fraser]

Farnell have them and they will be good quality parts

Radio Spares is another option

Fraser

[Bill W]

A T491 in an SMPS or worse still on the input of an SMPS exposed to input power application is rather asking for trouble unless massively underrated.  A T495 would be better.

To be clear this is not a Kemet problem, as the same applies to AVX (TAJ vs TPS series) and Siemens / EPCOS.  Different grades for different uses, and you don't get something for nothing.
The black ones tend to be Panasonic IIRC.

Raytheon BST cores had a tant with habit of going POP as well for similar reasons, sometimes associated with folks wiring them into vehicles without due protection.

Bill

[Fraser]

Ooooops.... I did not mean to infer that the Tantalum was a particular model from KEMET. The datasheet was really just to explain the markings etc. Apologies for any confusion caused.

Bill is far more knowledgeable about Tantalum capacitors than me.

The failed Tantalum capacitor in the provided pictures has a type marking of "RL" ? It would be worth replacing it with the same if that is a more robust type.

Fraser

[Fraser]

I found this KEMET series markings guide.

http://alasir.com/reference/kemet_tantalum_capacitors/

It has the R+ but not the RL.

The R appears to indicate it is "Surge Robust" which ties in with what Bill stated. The T495 series is rated as such.

L normally indicates "Low DC leakage"

It would be worth looking for a Tantalum series that provides the robust plus low DC leakage spec. I think the RL is now obsolete though as I could not find it on the KEMET site.

[Fraser]

I have learnt a lot about Tantalum capacitors thanks to this thread.

I was unaware of the large range that are available for different tasks. In my younger days, a Tant cap was "just" a Tant cap !

I was interested to see that there are fused versions that will fail open if they break down. I suspect they are not used in many designs due to higher unit cost. Using a Tant Cap in a satellite is a different matter. A shorted Tant could cause havoc on power rails so I was not surprised to see Space Rated versions.

It is interesting that, as a beginner in electronics at a young age, Tantalum capacitors were hailed as the luxury end of the capacitor market, far more reliable than wet Electrolytics of the time. Wet Tantalum capacitors were even better than standard Tantalum's. They cost more so it was assumed they were the deluxe option in a design. Racal used them throughout their high end equipment lines.

Move forward a few decades and we read of Tantalum failures destroying PCB's and having major burn-up events, going short circuit and generally being a tad unreliable in old age. But that is the key factor.... age. Most manufacturers expect their quality kit to be serviceable for at least 10 years. Advantest even state the life expectancy of different parts of their Spectrum Analysers...... PSU : 5 years !

I suppose we must consider Tantalum capacitors to still be a high grade component, it just has some unfortunate failure modes as it reaches old age, or if it is unsuited to the task that it is expecting to perform.

My thanks to Bill W for his insight into the appropriate use of Tantalum capacitors.

Fraser

[frenky]

I have learnt a lot about Tantalum capacitors thanks to this thread.

I was unaware of the large range that are available for different tasks. In my younger days, a Tant cap was "just" a Tant cap !

I was interested to see that there are fused versions that will fail open if they break down. I suspect they are not used in many designs due to higher unit cost. Using a Tant Cap in a satellite is a different matter. A shorted Tant could cause havoc on power rails so I was not surprised to see Space Rated versions.

It is interesting that, as a beginner in electronics at a young age, Tantalum capacitors were hailed as the luxury end of the capacitor market, far more reliable than wet Electrolytics of the time. Wet Tantalum capacitors were even better than standard Tantalum's. They cost more so it was assumed they were the deluxe option in a design. Racal used them throughout their high end equipment lines.

Move forward a few decades and we read of Tantalum failures destroying PCB's and having major burn-up events, going short circuit and generally being a tad unreliable in old age. But that is the key factor.... age. Most manufacturers expect their quality kit to be serviceable for at least 10 years. Advantest even state the life expectancy of different parts of their Spectrum Analysers...... PSU : 5 years !

I suppose we must consider Tantalum capacitors to still be a high grade component, it just has some unfortunate failure modes as it reaches old age, or if it is unsuited to the task that it is expecting to perform.

My thanks to Bill W for his insight into the appropriate use of Tantalum capacitors.

Fraser

The marking on the burned cap is R+ or RZ imho... The yellow one is definitly R+.

Sent from my LG-D855 using Tapatalk

[Fraser]

Frenky,

Yes the symbol is indistinct in the photos and at first looked like an L. Another look suggests to me that it is definitely an R but the second character appears to have both horizontal and vertical elements. It could be a + or even a t. Chanc3 will have a better view of the part so maybe he can confirm the markings ? The two pictures at different angles show different features of the possible markings. The resolution is just a bit too low to be sure of what is there.

A T495 type (R+) fits the application though.

Fraser

[Fraser]

Purely for those interested in how a Solid Tantalum capacitor is constructed......

[SeanB]

Wet Tantalum is incredibly reliable, unless you bought the cheap version that is sealed with a rubber bung. this suffers from the same evaporation as a regular electrolytic, and gradually the ESR rises with age while the capacitance slowly drops. Replaced a whole lot of them with either regular electrolytics ( which were as reliable anyway, plus also available as well) or with the glass sealed versions. Aside from being not at all happy with ESD events that cause oxide failure long term, or absolutely not at all happy with any sort of reverse bias for any reason ( a standard electrolytic will survive being reverse biased to a volt or two with only long term degradation, but any tantalum will be quite happy to turn into a small squib nearly instantly) they are otherwise very nice capacitors, and have a very good life so long as they are used correctly, which means no overvoltage or ESD, no reverse voltage ( even a Shottky diode as reverse protection is too high a reverse voltage) and keeping the ripple current in specs, though the wet ones have a pretty good short term ripple rating in use, eventually getting hot enough to melt open the case solder seam.

[Fraser]

More trivia from me....

KEMET appear to be changing their standard case colour of yellow, to black for some Tantalum series. They state that it is to align with an industry standard colour scheme.

http://www.kemet.com/Tantalum-Polymer-SMD-Black-Epoxy-Conversion-Final-Stage---ICT-Impacted-Series

Fraser

[Fraser]

Further trivia.....

KEMET also produce Polymer capacitors that look the same as standard Tantalum types. Marked KO series.

http://www.kemet.com/KO-CAP

Fraser

[Fraser]

I have not previously needed to study the different series of Tantalum capacitors before. I always replaced like with like but did not pay much attention to the differences in the series. I now know better !

Tantalum Capacitors are so common in equipment that they deserve their own discussion thread that details user experiences and the various issue with their use. The KEMET web site certainly provides plenty of detail on the actual components.

It is life experiences as detailed by Bill W and SeanB that are most helpful. Using the wrong Tantalum capacitor series in a design or repair could end in tears ! I have read many posts where people have criticised Solid Tantalum capacitors for their less than circuit friendly failure modes. Maybe we should be considering whether the circuit designer got his selection of component right. Would it be better with a more robust series or even a different capacitor technology. Modern electrolytics can achieve excellent ESR values and we now have solid electrolytics to consider.

Chanc3, thanks for starting this thread, it has been most interesting looking into Tantalum capacitor technology. I am still a little puzzled why the KEMET capacitor that failed is Black instead of the usual Yellow of the era. FLIR would normally use the same manufacturer for the same value and series of capacitor. They have used a mix of KEMET and Panasonic on that board which isn't unusual, but the blown capacitor would normally be a yellow KEMET and not black.

For info, just like the dreaded issue with electrolytic capacitors of the 1990's, there was also a problem with black cased Tantalum capacitors. Some black cased Tantalums were apparently prone to failure. This appears to have been common enough for one manufacturer of Black cased Tantalum capacitors to highlight the issue and state that theirs are not the same and do not suffer from the problem.

Fraser

[Fraser]

Thread started to further discuss Solid Tantalum capacitors use in new designs and in repairs. I am also hoping forum members will contrubute their life experiences of this component technology.

https://www.eevblog.com/forum/projects/tantalum-capacitor-types-in-new-designs-and-repairs-discuss/

Fraser

[Chanc3]

Fraser, once upon a time it was yellow! It just cooked for a little too long!

Funnily enough, I replaced one of these of a ti45 from fluke. Exactly the same problem!

[Fraser]

Ahhhhhh that explains it. Thanks Chanc3. Having looked again at the pictures I can see a yellow/brown area at the end. She sure did cook well  It is fortunate that the PCB has not suffered too badly from the heat. You may wish to closely inspect any PCB tracks that pass under the failed capacitor though.

My original thought that the two 'in-line' capacitors in your pictures will be the same value and type is correct then. That is to say they will both be the R (Robust) version The + is just the polarity confirmation. As Bill stated, you need to source the T495 series Tantalum capacitors.

Fraser

[Chanc3]

Thanks all. I've ordered the higher rated capacitors, so should be arriving today/tomorrow. Will then take a few days to repair - will report back!

[Chanc3]

So I replaced the tantalum caps with the higher rated ones and voila! It fires up! Bad news is, the viewfinder is dead - no display whatsoever and connecting via component video gives a very glitchy display and no thermal feed! See the attached image.

Not sure what the problem is now! A visual inspection didn't result in any other physically damaged components and I definitely connected all the cables up correctly! Fraser - any ideas?

[Fraser]

Yikes! That does not look too good 

The fact that the FLIR logo is present in the top left corner proves that the video encoder chip and its associated input data bus is working OK. Video encoding on later PM series cameras is done by an Altera FPGA but I am not sure what was used in the P/S series.

The weird display suggests something nasty is going on in the video processing stage. No great surprises there.

Does the FFC shutter operate ? The FFC shutter drive is independent of the video processing stage.

Can the camera be switched off using its power button or do you have to eject the battery ?

Does the pressing of menu buttons change the pattern on the displayed image ?

OK, now the bad news......

These P/S series cameras are pigs to work on ! I do not have their schematics either.

I regret that I do not have any useful suggestions for you at the moment.

Fraser

[Chanc3]

FFC operates as does the rest of the camera - I can move around and select buttons etc, zoom, turn on/off with power button. Just the display! Thinking a chip has been wiped out when the other component went.

Fancy a trade? I can't waste much more time on it!

[Fraser]

I'll PM you.

Fraser