Go Down

Topic: Suggestions on a SMD transistor (hardwar) to use... (Read 18305 times) previous topic - next topic

dc42

#15
Jan 16, 2013, 07:22 pm Last Edit: Jan 16, 2013, 07:23 pm by dc42 Reason: 1
Quote
basically a transistor will allow a 'higher' voltage/current pass through it (to the target device)....(unaffected..ie: not regulated or anything)...  by allowing a 'smaller' voltage/current to 'toggle' the gate and allow said 'flow'...

correct?


Yes. Bear in mind that a BJT is a current-controlled device, unlike a mosfet which is voltage-controlled.

Quote
(Vcbo) Collector Base = 50v
(Vceo) Collector Emitter = 45v
(Vebo) Emitter Base = 5v

Does that mean it can have up to 50v connected (from battery/source)...
output up to 45v to the 'target' device..
and can take 'up to' 5v (max) to toggle/control the gate?


No. it can have up to 45V connected from the battery (Vceo). The Vebo rating means that if for some reason you apply a negative voltage to the base, it had better not go more negative than 5V.

Quote
Im also confused about the current ratings given.. it is 200mA available or 500mA available?  what are the difference between the two numbers given?

(Ic) Collector Current = 500mA
(Ibm) Peak Base Current = 200mA


The collector current rating is the one that matters, along with the power dissipation rating.

Quote
basically jumper/bridge from the battery +/- pads from the first pcb to the second.....  
but more importantly.. could I use the same PWM pin to control/fade BOTH pcb's at the same time??


Yes, provided that the current draw from the Arduino pin is below its 40mA limit. Currently you have 1K base resistors, so the current draw from the pin per board connected is only about 4.5mA.

Quote
my question is.. if the one resistor I use, limits the current to, say, ...40mA going to the 5050 led/unit...  how will the leds respond/behave?

Will all 3 leds split the '40mA' equally?  will one leds take more? or full mA leaving the others without full or equal current?


They will probably share the current fairly evenly but not perfectly. Best option would be to use a high enough voltage battery so that you can connect the 3 LEDs in series. Then the current draw remains 20mA per 5050 (assuming you want to run the individual LEDs at 20mA).

Quote
in the current example. if I changed to 5050 leds..  there are 5 leds.. going to all 5 x 5050 leds would be (5 x 60mA = 300mA) neded per pcb/board!!

Im not even sure the transistor can handle/provide that?.. however if @ 40mA each 5050 led looks and acts fine.. it might be a viable option... (or maybe just more 1206 leds!)  lol..  but I 'am' curious.


Connecting the LEDs in series and using a higher voltage supply addresses that issue. But 300mA may be just OK for a BC817 if you reduce the base resistor from 1K to 220 ohms
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

xl97

Hey guys-

being the weekend.. getting  few hours to work on things.


I want to wrap this one up and send out for some PCB's..  but I got to thinking..

I want to make these as 'universal' as I can....  meaning I am not always sure what battery source/pack these will be used with.. (I'd imagine +3.3v - 7.4v packs...and all in between)

that being said, I was wondering if maybe I should work a voltage regulator in there?

If yes.. I imagine it would go 'BEFORE' the transistor.. (between transistor and battery source)

or do you think using resistors would just be the best approach to being more 'flexible'?



If I use a +3.3v voltage regulator between transistor and battery source...  the battery source will ALWAYS have to be a +5v source or above.. right? (enough to power the +3.3v regulator...   and all SMD vRegulators I have seen only let 50mA through or so..??)

this also rules out the Arduino using the +3.3v pads to power anything..  (how much current does the +3.3v pads provide?)  (in case someone was using a +3.3v arduino or something)


using resistors matched to the battery source at the time of implementation might be best?  but man thats alot of waste isnt it?

if the battery pack is a +7.4v li-ion pack.. (powering the Arduino... 'as well as' going to the the led pcb above)..   using some leds with about 3.0-3.3 vF...   be wasting alot as heat?


just spit-balling for some good ideas.. and what would be the best approach.


thanks!


dc42

3.3V is too low a supply voltage to use with blue or green LEDs. If you will know what voltage you will be using when you populate the board, then there is no need for a voltage regulator, if you don't mind the LED brightness going down a little as the battery voltage decreases. If you do mind that, then you could use a 5V regulator.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

xl97

3.3v regulator is too low?   because of the drop across it?

I though ta 3.3v regulator put 'out' a 3.3v stream....no?

wouldnt that be enough to power a 3.0-3.3fV led?

most blue/green leds "I" have currently run at around there...  :)

your saying If I now the battery source/voltage when I populate I dont need a regulator.. because your saying.. just sue the correct resistor?  even its a 7.4v battery cutting it down to 3.3v/20mA... still 'ok' to do?






thanks!..


dc42


3.3v regulator is too low?   because of the drop across it?

I though ta 3.3v regulator put 'out' a 3.3v stream....no?

wouldnt that be enough to power a 3.0-3.3fV led?

most blue/green leds "I" have currently run at around there...  :)

your saying If I now the battery source/voltage when I populate I dont need a regulator.. because your saying.. just sue the correct resistor?  even its a 7.4v battery cutting it down to 3.3v/20mA... still 'ok' to do?


To drive an LED at about the right current, you need either a constant current driver, or a constant voltage and a series resistor such that the voltage drop across the series resistor is sufficient to define the current adequately despite variations in the forward voltage of the LED. For example, suppose your LED has a nominal 3.2V forward voltage at the current you choose, and you select a resistor so that a 3.3V supply gives that current. In practice, the forward voltage varies between different LEDs and with temperature. If your LED actually had 3.25V forward voltage, your resistor would be too large by a factor of 2. If it actually has 3.15V forward voltage, your resistor would be only 2/3 of the required value.

Therefore, when driving an LED from a constant voltage source through a series resistor, you should have sufficient headroom in the power supply voltage so that there is sufficient voltage drop across the resistor (e.g. 1V or more) that variations in the LED forward voltage won't alter the current much.

Also bear in mind that if you are driving an LED through a BJT, the BJT has a voltage drop (typically 0.1 to 0.3V at low to medium currents - look up Vce(sat) on the datasheet), so 3.3V isn't enough for a 3.2V LED driven through a BJT anyway.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

xl97

(Im going to have read it again for it sink in all the way)  :)

haha.. Im still confused on the best way then...


1.) to add or not to add a vReg..  (seems like maybe NOT?)

2.) I cant use a 3.3 (actually I think li-ion single cell are 3.7v...but whatever)..not enough to go through the transistor?   what is the minimum needed? 5v?


I guess we can start 'make' some requirements to make this easier..

lets 'mandate' (for example) it can only use with a pro-mini (or nano) running @ +5v/16MHz clock..

this kinda gives us a 'direction'  (since I think there will be a servo in the mix as well... those run @ 5v off the Arduino +5v line.. and the Arduino can give it enough current itself..form that control pin)...    So the Arduino 'will' be a +5v/16MHz type..

so we can NOT power that board with under +6v battery pack/source..

meaning we need at LEAST  +6v Alkaline back...or a 7.4v li-ion type pack..   (or similar)

this Arduino will connect to the PCB.. (in all reality there will probably be 2 pcbs',...another exact same pcb daisy chained to the first (bridging the battery +/- pads and the bridging the 'PWM' pad as well)... so that 1 PWM pin on the Arduino can pwm/control both pcbs at same time)


maybe Im over thinking it?  leave it alone.. just use a transistor.. and adjust the resistor(s) to reflect whatever battery pack is used?

I was just thinking there was a better, maybe more efficient way..  (but still keeping things small, and only adding in a regulator or something)

some of this all new..so Im trying to take it all in.. and work towards my goal.

thanks for the help/replies..etc  :)

dc42


1.) to add or not to add a vReg..  (seems like maybe NOT?)

2.) I cant use a 3.3 (actually I think li-ion single cell are 3.7v...but whatever)..not enough to go through the transistor?   what is the minimum needed? 5v?


3.7V would be just about OK if it doesn't vary too much, however it would be better to use a mosfet than a BJT in this case, because a well-chosen mosfet will have a lower voltage drop.


maybe Im over thinking it?  leave it alone.. just use a transistor.. and adjust the resistor(s) to reflect whatever battery pack is used?


That would be my recommendation.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

xl97

ABout to order some transistors for this..

http://www.digikey.com/scripts/dksearch/dksus.dll?FV=fff40015%2Cfff80078&k=BC817&vendor=0&mnonly=0&newproducts=0&ptm=0&fid=0&quantity=0&PV7=2

but alot of choices.. and not sure 'why'? or the difference between them?  (always one of the hardest parts for me is choosing the hardware...so many little details could be wrong.. and end up in a the wrong part ordered!)  :)


Here is the original post w/schematic and board ayout:

http://arduino.cc/forum/index.php/topic,142176.msg1074599.html#msg1074599


maybe this one:
http://www.digikey.com/product-detail/en/BC817-25-TP/BC817-25-TPMSCT-ND/1960198

or

http://www.digikey.com/product-detail/en/BC81725MTF/BC81725MTFCT-ND/1305367

perhaps??

would appreciate a second set of eyes.. so I order the right stuff..

(needs to be able to drive around 7-9 SMD leds)  :)

thanks!

dc42

Either of those should be OK (they are both BC817-25). However, if the current you will be switching is 300mA (as discussed in the other thread, and about the maximum I would use a BJT for), then I think BC817-40 would be a better choice, e.g. http://www.digikey.com/product-detail/en/BC817-40,215/568-1631-1-ND/763458. It has higher gain, so it should give a lower saturation voltage. Alternatively, use a small mosfet such as http://www.digikey.com/product-detail/en/ZXMN2A01FTA/ZXMN2A01FCT-ND/386044.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

xl97

#24
Mar 17, 2013, 07:25 pm Last Edit: Mar 17, 2013, 10:25 pm by xl97 Reason: 1
HI!..
thanks for the reply..

the battery pack (+7.4v li-ion pack) is what I'll be directly connecting to the transistor..

and the signal/control line to the Arduino 'pin' to PWM fade/flicker/blink the led array as a whole..

in total I believe the led array will need/pull not more than 200mA  (probably less.. but wanted some overhead)..  ther is 7 smd leds in total..  even @ 20mA each.. thats only 140mA total..


dc42

Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

Tumbleweed

"2.) I have seen some diagrams have the GND line from the Arduino be used and tied into the GND line from the battery source/pack?..
Why?  Im failin to see what part the Arduino circuit plays into this (powering the leds) if it can not give enough current.. 'and' a transistor is being used instead?   (maybe it was showing the Arduino GND to imply they needed to be joined together?)"

If they are using separate supplies grounds need to be connected, with the same battery supplying the daughter board(s) and the Arduino, they will be connected through the Arduino. As your circuit is drawn they have to be off the same supply for the transistor to have a circuit through the Arduino to the base to the emitter to the ground used by the Arduino .
Einstein once said you don't really understand anything until you can explain it to your Grandmother

MarkT


I'd go with FET with low Rds, otherwise have ~0.7V across the BJT transistor.



No, ~0.1V across a saturated BJT.  The collector falls well below the base voltage in saturation.
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Go Up