Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

1 MCU open-drain pin to drive 4 NPN

Status
Not open for further replies.

banh

Advanced Member level 1
Joined
Dec 16, 2004
Messages
458
Helped
17
Reputation
34
Reaction score
5
Trophy points
1,298
Activity points
3,939
hi,

i intend to use 1 MCU open-drain pin to drive 4 npn's as in the figure.

IoL of the pin is 10mA.

each npn needs to be injected 1mA into the base for saturation.

So, this is what i expect:
- when the NMOS of the port is off -> all the NPN are turned on and the leds will be lighted up.
- when the NMOS of the port is on -> all the NPN are off, and the leds are off.

i need your comments on this circuit.

thanks.
 

artem

Advanced Member level 4
Joined
May 22, 2003
Messages
1,350
Helped
126
Reputation
252
Reaction score
32
Trophy points
1,328
Location
Turkey
Activity points
13,461
instead of 560 Ohm resistor put pnp transistor with collector connected to npn's base resistors, base connected through resistor to open-drain mcu output
and emitter to Vcc
 

    banh

    Points: 2
    Helpful Answer Positive Rating

IanP

Advanced Member level 5
Joined
Oct 5, 2004
Messages
7,942
Helped
2,309
Reputation
4,620
Reaction score
531
Trophy points
1,393
Location
West Coast
Activity points
66,459
Should work well ..
You can easily increase the value of the pullup from 560Ω to, say, 1kΩ ..

Regards,
IanP
 

    banh

    Points: 2
    Helpful Answer Positive Rating

Sceadwian

Full Member level 3
Joined
Sep 25, 2005
Messages
174
Helped
15
Reputation
30
Reaction score
0
Trophy points
1,296
Activity points
3,588
The 560ohm resistor will give a higher gate bias value. At 560ohms the gate bias is about 2.5 volts (asumeing 5volt VCC) at 1k ohms the gate bias is about 1.8 volts. I'm not sure how this effects the responce of the mosfet, but the closer to 'on' you can bias a mosfet without it actually turning on the better. Considering that lower resistance draws 8ma's which is pretty close to the 10ma limit of the pin it's probably the best value to use, unless the rise/fall time of the I/O pin is too high with the large current draw.
 

banh

Advanced Member level 1
Joined
Dec 16, 2004
Messages
458
Helped
17
Reputation
34
Reaction score
5
Trophy points
1,298
Activity points
3,939
thanks all,

another concern of mine is that instead of the pin being configured as open-drain, it is accidentally configured as normal push-pull output.

in this case Ioh of the pin is only 2mA.

is this a problem? as the pin may source currents to the NPN's, and may exceed 2mA..
 

IanP

Advanced Member level 5
Joined
Oct 5, 2004
Messages
7,942
Helped
2,309
Reputation
4,620
Reaction score
531
Trophy points
1,393
Location
West Coast
Activity points
66,459
The "push-pull" output still will be comfortably drive these 4 transistors ..
Keep in mind, that common NPN transistors have β>>100, and that to drive, say, 10mA collector current of NPN transistor you need I(base) =Ic(ollector)/β = 10mA/100 = 100µA (or less, for β>100) ..
For 4 transistors, therefore, the driving current will be <400µA, so with 2mA Ioh of the micro's pin the driving capability is ≈20 of NPN transistors ..

Regards,
IanP
 

banh

Advanced Member level 1
Joined
Dec 16, 2004
Messages
458
Helped
17
Reputation
34
Reaction score
5
Trophy points
1,298
Activity points
3,939
the transistor i use has hFE (min) of 20.

hence i need to pump into the base around 1mA.
 

IanP

Advanced Member level 5
Joined
Oct 5, 2004
Messages
7,942
Helped
2,309
Reputation
4,620
Reaction score
531
Trophy points
1,393
Location
West Coast
Activity points
66,459
In this case, if you don't want to invert the phase of the signal, add another NPN transistor, and connect its base directly to the microcontroller pin, collector to +5 and emitter (through resistos) to bases of these 4 transistors ..
Obviously, you can do it in one-hundred-and-one other ways ..

Regards,
IanP
 

    banh

    Points: 2
    Helpful Answer Positive Rating

banh

Advanced Member level 1
Joined
Dec 16, 2004
Messages
458
Helped
17
Reputation
34
Reaction score
5
Trophy points
1,298
Activity points
3,939
why can you connect directly the base of the transistor to the MCU pin? shouldn't you have a resistor to limit current?
 

IanP

Advanced Member level 5
Joined
Oct 5, 2004
Messages
7,942
Helped
2,309
Reputation
4,620
Reaction score
531
Trophy points
1,393
Location
West Coast
Activity points
66,459
You can connect additional resistor between the base and the MCU's pin, but it is not needed there as you have resistors wired at the emitter side ..

Rgards,
IanP
 

banh

Advanced Member level 1
Joined
Dec 16, 2004
Messages
458
Helped
17
Reputation
34
Reaction score
5
Trophy points
1,298
Activity points
3,939
thanks IanP,

attached is the modified circuit according to your suggestion.

i'm not so clear about calculating:

- how the current flows in the npn connected to the MCU pin? how much current is pumped out from the mcu pin?
- the current flowing into the base of the other 4 NPN's? are they still (5-0.75)/3K9 for each NPN?
 

IanP

Advanced Member level 5
Joined
Oct 5, 2004
Messages
7,942
Helped
2,309
Reputation
4,620
Reaction score
531
Trophy points
1,393
Location
West Coast
Activity points
66,459
Base current of each of the 4 NPNs:
(5V-0.75-0.75)/3k9 ≈ 0.9mA (second 0.75 stands for Ube of the added NPN)

MCU pin current (assuming H=5V):
≈ (4*0.9mA)/β (of the added NPN) ≈ 3.6mA/20 ≈ 0.18mA

Regards,
IanP
 

    banh

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top