Charger let the smoke out..

Ypedal

Ypedal

100 TW
Joined
Dec 27, 2006
Messages
12,536
Location
Moncton NB, Canada
Ok.. so after about 1 year of faithfull service, my ebay TP1010 knock off charger smoked..

I was attempting to revive some Ridgid 18650 cells that were below 3v per cell, the charger was in " recovery mode " and trickle charging, one of my aligator clips came loose and fell off ( without shorting out, i was watching it whe it happened, it simply came off cleanly ) and at the same time i heard a " Pop " followed by that ever familiar smell.. :evil:

the display then said something like " Incorrect Input Voltage " and started to beep..
 

Attachments

  • charger1.JPG
    charger1.JPG
    71.5 KB · Views: 1,103
I followed my nose and found this :

Labelled " J226 " .... my limited knowledge of electrical components tell me this is a resistor.. ?? :lol:

I tried googling J226 and surface mount resistor but no joy.. any chance i can swap this with a regular round type resistor if i can figure out the value somehow ?
 

Attachments

  • J226.JPG
    J226.JPG
    42.9 KB · Views: 1,099
Y:

there is a designation "c11" close to your smoked component, what is that about? "c"'s are usually capacitors.

dick
 
Ypedal said:
http://en.wikipedia.org/wiki/Resistor

Found the J code designation : + or - 5 %

:?
Click to expand...


I don't know what the value of it is but I suspect it's part of a voltage divider. Can you find the resistor it likely connects to and give the code? I suspect the last number will be a 5, 6 or 7.

Typically, the first two digits is the first two digits of the number and the last number indicates the amount of zeros following. So, 226 would mean 22 followed by 6 zeroes or 22,000,000 or 22 M-ohm. That seems like a godly high number for a voltage divider, though...

What does the silk-screen say? Does the first letter say R?

I think it states C. That would indicate 22,000,000 pF which is 22 uF. From the size of that thing, I'm going to guess that's a tantalum capacitor which the darkened color suggests. Can you get a measurement on it? Is it 3.5 mm x 1.2 mm or thereabouts?

It looks like the line indicates it's polarized, so it's most likely a tantalum capacitor. You may be able to replace it with a 22 uF ceramic cap, if you can mount it correctly and all. I'm going to guess that somehow the voltage was reversed or the voltage limit was possibly exceeded, which that big inductor seems to have the potential to do so if the connection is suddenly disrupted. You might want to replace the capacitor with a higher voltage rated one. A 50V ceramic cap is a good start if you can get it to fit.
 
can you measure the length? size code A is 3.2mm L, 1.6 W,1.6 H

page 862 of mouser, 581-TPSA226K006R0500 is 500mohm ESR

i think that could be your part.

edit, p 864: 581-TCJA226M6R300 is 300mohm ESR the conductive polymer electrode

and a high temp series p865: 581-THJB226K006RJN is 2.5R ESR but the case is 3.5mmL 2.8mmW

there are specs for the width of the metal strip that folds under the cap too.

also discovered the surface mount aluminum organic caps 7.3 mm long 4.3mm W which have aluminum anode and aluminum oxide, but a conductive organic polymer film to replace the liquid electrolyte.

p 727: 80-A700V226M006ATE028 is 28 ohm for the ESR but has 7.3mm L

very educational.
 
Definitely a capacitor, 226 = 22uF as others have said. I don't think it's a tantalum, though. The markings aren't right, there should be a polarity band and probably more lettering, and it would be bigger. Probably just a garden-variety ceramic capacitor. Eyeballing, it looks like maybe 1210 or 1812 surface-mount size, that would be either .12x.10" or .18x.12." If you can figure out the right size, just buy whatever the highest voltage rating available for that size/value combination is and you should be good to go.
 
rhitee05 said:
Definitely a capacitor, 226 = 22uF as others have said. I don't think it's a tantalum, though. The markings aren't right, there should be a polarity band and probably more lettering, and it would be bigger. Probably just a garden-variety ceramic capacitor. Eyeballing, it looks like maybe 1210 or 1812 surface-mount size, that would be either .12x.10" or .18x.12." If you can figure out the right size, just buy whatever the highest voltage rating available for that size/value combination is and you should be good to go.
Click to expand...

:shock:

We must be living in reversed universes.

There should be a polarity band on the capacitor, however.

It looks like it might've been partially covered from the charred remains as I think I see a little white peeking through the left side of the capacitor.
 
did you buy the desoldering station we were talking about a couple of weeks ago?

just wondering.

now we are going to find out just how steady your hands are. magnifying glass, tweezers, good lighting, teeny tiny tip on the old soldering iron.

rick
 
In a situation like this, I would try to upgrade the part. For the kind of size you'll get with that upgrade, you'll probably go best leaded.

After removing the original component (A regular soldering iron can be used along with tweezers)...

If you have the appropriate drill size (A tiny .8 mm), drill a hole in each pad. Thread the through hole component, solder and you're done. If that's not an option, solder small sized wires to the via and the "right side"(As opposed to wrong side) of R39 and attach the leaded component to the other side of the wires. Observe polarity if you have a polarized part.

If you can't find 22 uF ceramic capacitors, you may be able to use MLCC or possibly leaded tantalum if you're not concerned by polarity reversal. By going with a technology that inherently has a higher voltage rating and/or polarity protection, you can decrease the likelihood something like this would happen.

There's no guarantee that there aren't other damaged components, though, so it still might not work if that's replaced.
 
it did not look like a normal ceramic surface mount cap. you can see one to the left of the one that blew. it looks 3.5mm long, type A.

that's why i said you should measure it first, seems like it is large for the type A part.

from it's apparent size, i wondered if it could actually be one of the aluminum organic types, 7.3mm long.

i think john was right about the J being for 6.3V spec, 22x10^6 picofarad and polarized so tantulum or niobium oxide or maybe the aluminum organic. measure the length first.
 
yes, identical part. i can see it is larger than the normal ceramic caps, if it measures 7.3mm, you may want to use the aluminum organic polymer cap. also polarized the same way. that may be one there.

maybe you can reach the original manufacturer and get some technical advice. they may have the parts list somewhere and can give you a part number for c11. i am now thinking it is not a tantulum cap either.
 
Good call on finding another part to compare. Looks like I was wrong before, it is polarized and bigger than I thought. That's what I get for estimating size from a picture with no reference...

Did a quick search for reference. For Kemet parts at least, looks like you can't get 22uF tantalum in that package with anything higher than 10V rating. That would make sense if the "J" is for 6.3V rating as others have said (not sure myself). The organic polymer caps also come in the same package with ratings up to 10V. The polymer caps have lower ESR than the tantalum, but either would probably work. Depending on cost, it wouldn't hurt to go with the highest voltage/lowest ESR that will fit, though! Not sure what might be available from other manufacturers, but now you at least know enough to do a parameter search on Digikey and see what's out there.
 
Hey Ypedal,

Maybe I could help.

I repaired my Megapower 960SR 3 times from now after I tortured it with several reverse polarity and ALSO a CONNECTION THAT FELT OFF.

I replaced some fets, caps and resistor on it. it was easier than i expected for me to repair it each time. I also have some modification on it to have the 4 wire cucrrent sense measurement to get perfect RI measurement.

I think that to troubleshoot something you must know what happened and why. this is the key.

On your situation, I think I know wat may happened. I remember you said thatyour charger was in "recovery" mode. These charger are CC-CV so they are constant current as well. and constant current should be independent to teh wire lenght or the total resistance in your wire.

so let say what happen when your wires are longer between the battery abnd the cells... The voltage at the charger output will be adjusted to compensate the resistance loss and let the current to not deviate from the original setpoint.

so by increasing resisatance voltage will increase to preserve the current stable.

so ... if the resistance goes INSTANTANUOUSLY INFINITE ( wire disconnected situation..).... The voltage will rise very fast to try to restabilise the current... but as we know it's impossible cause the resistance is now infinite!!

So what happened is that your charger constant current circuit tried to rise the voltage very fast to compensate the open circuit... and reached too high voltage!.. ( Overshoot).. and the (commonly called CROWBAR circuit) to damp that peak seems not very evoluated.. and the little capacitor ( 226) that is probably in parallel with the output or somewhere, just received overvolt and blown.. but it is also possible that other component might be affected.. pay attention to that :wink:

This is current in RC application that connection come off between cell and charger and when people reconnect it, there is a large spark due to the high voltage rise ( 50-60V in some case) on the charger.. so when you reconnect 60V to a 8.4V battery, the voltage difference and the stored energy in the output capacitor that diacharge suddenly may be impressive!

On my megapower it happen often and it blown one time because of to that.

While charging a single A123 using that charger, if i disconnect it while charging it at let say 10A, I can easily hear the PWM circuit in the charger modulating like "ssss zzzi i i i i i i i i" to boost suddenly the voltage to try to recover the 10A current..... with ... infinite resistance as well....

:shock: ... Dude!! .. so many word to explain something simple!!... :oops: .. well....

that's me... ... lol.. Hemett Brown aparently was like that too... :mrgreen:


Doc
 
What ever you do, do NOT drill that board anywhere for any reason. That is definatinately a multi-layer board, you have no idea what you would be drilling into and shorting or breaking that you can't see. Trying to drill a smt pad trace nearly always results in lifting the pad anyways.

You can pickup a through hole 22uf cap from anywhere, i would try to get one rated for 50v. Don't solder through hole legs directly to surface mount pads. Solder a little runner of the most flexible little wire you can find to the pads, then connect that flexible wire to the legs of the component. Mount the component with silicone to something near, and then put a bit of silicone over each runner wire where it connects to the pads.

However... you don't know if some other failure is what caused the cap to pop, or if the cap was the failure. But for $0.50 to give replacing the cap a shot, it's a good first start.
 
liveforphysics said:
What ever you do, do NOT drill that board anywhere for any reason. That is definatinately a multi-layer board, you have no idea what you would be drilling into and shorting or breaking that you can't see. Trying to drill a smt pad trace nearly always results in lifting the pad anyways.
Click to expand...

If the solder is cleaned up, then my experience suggests that doesn't happen (I use a .8 mm drill bit at 30,000 RPM). If it is a multi-layer board, however, then it's pretty wise to not drill it. It seems difficult to believe a cheap generic eBay charger would use more than two layers and if it is a double layer board, then checking what you're drilling through the other side is wise if not common sense. You'd have to take the pcb out anyways to do the drilling... so... it'd be kind of obvious, hopefully.

But attaching wires is going to have a much higher chance of success if you're inexperienced with board repair (which seems to be the case) so I'd definitely support that.
 
I managed to change a surface mount resistor on the DrainBrain a long long time ago with a round type http://www.ypedal.com/Chaos3.htm

, that was the smallest thing i've ever soldered .. this should not be so bad once i locate a cap to use.. i hope it's all that blew ... that charger was 200$.. :|
 
If you take the board out to repair, hold it up to the light. If it's totally opaque (power plane) or you can see traces that aren't visible from the front or back, it's multilayer. If you can see straight through the area around the pads, you might be clear to drill (with caution!). Replacing with either surface-mount or a dead-bugged thru-lead component would be advisable, though. At least it's pretty big, as surface mount goes. Soldering something in 0402 size (40x20 mil) is a little more sporting...
 
i still think you should replace with the same part. it's not 50 cents, more like $2.75 as i recall from mouser so digikey or newark would be more.

the much larger size over a standard ceramic cap implies it is doing more at that voltage than the ceramic cap does or it would not have been designed in, imo. a large fat electrolytic can would not fit and the heat would boil the electrolyte out anyway, so aluminum organic makes sense to me now, just did not know about them before. that's why i said this was educational.

it's a tight fit for the soldering iron down between those 2 big shunt resistors, but if you put a tiny bead of solder on the one end, you can hold it in place with your other hand using a long thin screwdriver while the iron goes in from the other side to tack it on the right end it will settle in the solder on the trace now, then turn it around, tack the other end and you are done.

i liked doc's analysis.
 
The cap protects another circuit, better find/trace that one. I doubt replacing it will fix your problem. :(
 
You must log in or register to reply here.
Back
Top