wesnewell
100 GW
I've blown 2 transistors on my charger. Anyone know where I can get them? Part number?
Bleed transistors for Hyperion 1420i
- Thread starter wesnewell
- Start date
wesnewell
100 GW
After closer look I found the code 1Y on these transistors, which equates to these.
http://www.s-manuals.com/smd/1y
They are so fricking small I don't know if I can see good enough solder them on or not, but I'm going to try when I get some. I broke the 2 bad ones into pieces getting them off. If I could get a good one off I could move it to where I need it, but I don't think I can get one off without breaking it into pieces.
Cost a <dime each in quantity 1000 from digikey
Anyone got a few of these laying around they could mail to me?
http://www.s-manuals.com/smd/1y
They are so fricking small I don't know if I can see good enough solder them on or not, but I'm going to try when I get some. I broke the 2 bad ones into pieces getting them off. If I could get a good one off I could move it to where I need it, but I don't think I can get one off without breaking it into pieces.
Cost a <dime each in quantity 1000 from digikey
Anyone got a few of these laying around they could mail to me?
I can poke around in my stuff to see; got a lot of old boards of various kinds from all sorts of "dead" electronics. Dunno how long it will take to find anything though. 
Regarding removing the parts: If you use two soldering irons, it works pretty well most of the time. ON days my hands aren't shaking *and* my eyes work well enough (with LOTS of light, preferably direct sunlight) and some old bifocals a friend gave me, I can just about manage to move SMT resistors, caps, and transistors off one board and onto another. Haven't been able to do chips that way, though.
Regarding removing the parts: If you use two soldering irons, it works pretty well most of the time. ON days my hands aren't shaking *and* my eyes work well enough (with LOTS of light, preferably direct sunlight) and some old bifocals a friend gave me, I can just about manage to move SMT resistors, caps, and transistors off one board and onto another. Haven't been able to do chips that way, though.
wesnewell
100 GW
My normal reading glasses are 5.5x in one eye and 4.75x in the other. If I use another pair of 3.5x with them I can see pretty good close up within 3-5 inches. But move my head an inch or so and I'm blind again. Would love to have the eyesight I had 50 years ago. Guess I;ll go ahead and order 10 of these things and see if I can manage it.
dnmun
1 PW
those are most probably mosfets, maybe 20V, if you look at where the source leg and gate are connected you can guess if it is a n channel or p channel mosfet. i doubt if there is much current but they can handle hundreds of milliamps i suspect. maybe an amp.
dnmun
1 PW
amberwolf said:I can poke around in my stuff to see; got a lot of old boards of various kinds from all sorts of "dead" electronics. Dunno how long it will take to find anything though.
Regarding removing the parts: If you use two soldering irons, it works pretty well most of the time. ON days my hands aren't shaking *and* my eyes work well enough (with LOTS of light, preferably direct sunlight) and some old bifocals a friend gave me, I can just about manage to move SMT resistors, caps, and transistors off one board and onto another. Haven't been able to do chips that way, though.
i have replaced a few of those tiny 6 pin SOIC parts on the ping signalab.
i use solder wick, tiny tip on the soldering to push the solder wick into the edge of the leg's contact to the pad so i can soak up all the solder, do both sides, then with the solder wick still on the side with three legs, i push that sideways off the pads, and use the tip if another soldering iron to push the two leg sideways. so it rotates on the pad a little and slide it with the solder wick pulling it along until it is clear of the pads. if you don't clear the pads or other solder it will stick when it cools.
to place it, i use solder wick to remove any trace of solder from the pad. then slide the SOIC onto the pads, with the pick while holding down with a flat bladed jeweler's screwdriver. my hands shake a lot and the magnification only helps so much anymore. i try to carry the tiny balls of solder to the outer legs first and then the center from the edge. when they short i blot it up with solder wick, but not too much so there is still a little solder under each leg. not big blob of solder like with reflow but the electrical connection is made and i don't think it will fall off. blotting the pads of all solder after you remove the old chip is the most important. otherwise you can not get the chip to sit flat on the pcb when you push down and it moves when you touch it.
i actually use a barely dry iron and press each leg down onto the pad briefly to tack the legs in place, this is where it has to be square before you tack. the first leg is permanent. you have to unsolder it if it is not square and if it is not square and centered on the pads, it will short. the 6 pin SOIC that is.
the SOT-23 transistors are not too bad. solder wick is so cool.
wesnewell
100 GW
Just a follow up. Got 10 transistors from mouser for $3.80+$4.99 shipping +tax. I really hate having to pay more for shipping than the parts cost. Anyway, got them installed. It aint pretty, but it works. Man these really small sot-23 parts are a real pita to work with. Hope I never have to do it again.
bigmoose
1 MW
Liquid flux is the key in soldering these. I also use a 2.5x and a 3.5x magnifying visor. As said above I use solder wick spritzed with liquid flux to clean the pads, then I lightly tin the pads with new solder, just a little bump. Then a bit more liquid flux. Next secret is to grab the new part with right angle tweezers so you can damp the shake with your hand on the table. Place part on the solder bumps. Now with a fine soldering tip touch the pad/part leg. You will see an immediate flow of solder thanks to the flux. Repeat on the other legs and your done.
I am 60 and spent all day yesterday populating a surface mount prototype board. I have come to like surface mount over through hole. I can actually build it faster, the product is smaller, more rugged and the boards are cheaper. I did standardize with 1206 resistors however and not the smaller "specks"
I am surprised at the part you found, a 200 Volt PNP doesn't seem like the right part there. I've used 20 volt N channel and P channel FETs. Before you order I would check a good part, on the board to try to verify it being PNP with an ohm meter set to diode drop. Try to verify the 0.6 volt C-B and E-B junctions and make sure they are in the "right direction."
I am 60 and spent all day yesterday populating a surface mount prototype board. I have come to like surface mount over through hole. I can actually build it faster, the product is smaller, more rugged and the boards are cheaper. I did standardize with 1206 resistors however and not the smaller "specks"
I am surprised at the part you found, a 200 Volt PNP doesn't seem like the right part there. I've used 20 volt N channel and P channel FETs. Before you order I would check a good part, on the board to try to verify it being PNP with an ohm meter set to diode drop. Try to verify the 0.6 volt C-B and E-B junctions and make sure they are in the "right direction."
dnmun
1 PW
i know the BMSs use p channel FETs for the shunt transitor, that part has 30k on the gate resistor so i wonder if it really is a mosfet, even though that is what i would expect there. but would not expect there to be a current limiting resistor to the gate that large.
dnmun
1 PW
i looked at some pnp SOT-23 transistors in the mouser catalog to get an idea of what is possible. i ran across this 7 cent darlington that is just amazing at what it can do with no base current.
http://www.fairchildsemi.com/ds/MM/MMBTA63.pdf
but looking some more at his picture, i can see that each of the base resistors has that 30k resistor but the resistor placed in parallel is different for each of the three transistors that show up in his picture. so someone had to place the transistor and test the gain after it was placed, with the 30k maybe, but then after the gain was calculated, the second resistor was added so that the gain would keep the collector current within a specific range. and the second resistor is of the same order of magnitude so the equivalent resistance is gonna be around half of the 30k.
i also noticed in going through the catalog that the SOT-23 parts seem to be limited in power to around 1/2 watt. so i wonder if that is part of the design, to prevent the transistor from pulling more power than it can handle, by calculating the second resistor value, so that the gain does not allow it to flow too much current and therefore keep power under control so that they can use the SOT-23 parts for cheap and fast manufacturing.
not sure where they would probe to find the transistor gain. do the placement machines have the ability to actually probe the legs of the transistor as they remove it from the tape, and then the computer supervising the assembly would know what resistor to use so it would place all the resistors in that step, before the reflow or conductive thermal gas soldering?
if we could see more of the circuit maybe it would be more clear, but i do think it is a pnp transistor now, i should have noticed the resistors on the base then. there are a lot of pnp SOT-23 options in the 30V range too.
http://www.fairchildsemi.com/ds/MM/MMBTA63.pdf
but looking some more at his picture, i can see that each of the base resistors has that 30k resistor but the resistor placed in parallel is different for each of the three transistors that show up in his picture. so someone had to place the transistor and test the gain after it was placed, with the 30k maybe, but then after the gain was calculated, the second resistor was added so that the gain would keep the collector current within a specific range. and the second resistor is of the same order of magnitude so the equivalent resistance is gonna be around half of the 30k.
i also noticed in going through the catalog that the SOT-23 parts seem to be limited in power to around 1/2 watt. so i wonder if that is part of the design, to prevent the transistor from pulling more power than it can handle, by calculating the second resistor value, so that the gain does not allow it to flow too much current and therefore keep power under control so that they can use the SOT-23 parts for cheap and fast manufacturing.
not sure where they would probe to find the transistor gain. do the placement machines have the ability to actually probe the legs of the transistor as they remove it from the tape, and then the computer supervising the assembly would know what resistor to use so it would place all the resistors in that step, before the reflow or conductive thermal gas soldering?
if we could see more of the circuit maybe it would be more clear, but i do think it is a pnp transistor now, i should have noticed the resistors on the base then. there are a lot of pnp SOT-23 options in the 30V range too.
999zip999
100 TW
Bigmoose The trick of a light tinning of the pad is a great help as the old solder dose not stick well. So the cleaning and fluxing is so important to a easier solder job. But triing to solder a tiny thing is almost is just a hair puller. Good job wesnewell are you going to do it again or offer repairs ? Glad to hear the 1420 up and running again.
bigmoose
1 MW
zip you are speaking my language about difficult. I had ordered some boards and stuffed 90% of the components when the digikey order came in yesterday with the 3 parts I was missing. I laid out a LTC6915 using a SO16... when I took it out of the tube it was a SSOP16! The pitch of that SSOP16 is 0.025 inch! It took me 2 hours last night to solder 16 wires on it. Then I potted it in epoxy. I just finished soldering those 16 wires onto the SO16 pads ... what a nightmare. I'll wait to tomorrow to see if it power up with no shorts or opens.
dnmun
1 PW
my hands just shake too much. if i had to do much of these i would look into a mechanical placement device to hold them or use one of those rework tools with hot gas for soldering and unsoldering. real buckos.
999zip999
100 TW
Bigmoose I was triing to type Thank you for the tip. So again Thanks.
I do do jewerly repair silver and gold and with all the different types of gold and silver and triing to make some repairs to the types of silver from around the world is tuff. So much stuff you just look at it and should walk away. The shops ask me to do stuff they wouldn't touch. Some times you look at it and walk away.
A little sanding help also.
I do do jewerly repair silver and gold and with all the different types of gold and silver and triing to make some repairs to the types of silver from around the world is tuff. So much stuff you just look at it and should walk away. The shops ask me to do stuff they wouldn't touch. Some times you look at it and walk away.
A little sanding help also.
Similar threads
