# 5v to 15v level shifter problem

Discussion in 'General Electronics Chat' started by BramLabs, Jun 22, 2016.

1. ### BramLabs Thread Starter Member

Nov 21, 2013
98
5
Hello,
Sir, i want to ask. I've made 5v to 15v level shifter with using a transistor 2n2222.

I'm using 5v PWM (40 kHz) from microcontroller as an input of the base transistor.
The problem is :

1). The transistor itself, become hot.
I've calculated the transistor dissipation :
When transistor in saturation :
Ib = (Vin - Vbe) / R1 = (5v - 0.65v)/1k = 4.35 mA
And then i get :
Ic = hfe.ib = 10 * 4.35 mA = 43.5 mA
And the power dissipation of the transistor would be :
Vce(sat) * Ic(sat) = 43.5 mA * 0.2v = around 8.7 mW (the maximum dissipation of transistor on datasheet is about 1.8 Watt)
I get Ic(sat) = 0.2v and Vbe = 0.65v from the datasheet of transistor 2n2222.

2). The output only give me 15v DC (i think the output voltage is latching to the voltage supply), not inverting of the PWM. I thought that the output would be such as like this :

And also can you give me a recommendation for making 5v to 15v level shifter ?
I've been thinking to use IR2110 as the level shifter, is that good sir ?

For the problem number 2, can i use this kind of circuit, to avoid the 'latching' problem :

And for the problem number 1, i don't have any solution yet.

2. ### Picbuster Member

Dec 2, 2013
376
50
Take the first drawing and replace Q1 with a Nmos fet. reduce the 1K to 100R. max current 15/(2K+Rfet) in mA =15/(2+0,02) 7.5 mA approx.
heat in 0,02Ohm 57 x 10^-6 Watt
Next step; invert PWM signal in software.
That's all.

BramLabs likes this.
3. ### MrChips Moderator

Oct 2, 2009
12,451
3,370
What is the exact part number of your 2N2222 transistor?
There are two very different lead identification. Make sure you get it right.

BramLabs likes this.

Apr 5, 2008
15,649
2,348
Hello,

MrChips is right, the 2N2222 (the metal case) and the PN2222 (the plastic case) have different connections:

Bertus

BramLabs likes this.
5. ### BramLabs Thread Starter Member

Nov 21, 2013
98
5
Thank you sir, can you give me a recommendation for the nmos-fet type ? Because there aren't many type of nmos that works as a logic level on the market in my country, i prefer to choose BJT transistor, rather than nmos-fet. But there're many types of 10v gate mosfet.

Oh i'm sorry sir, i forgot to attach the transistor part number.
The transistor itself : KN 2222A 341
And then it looks like a case TO-92A but with BENT LEAD

And based on your configuration and the configuration i found in internet, i think there's different between the collector and the emitter.
Which one is wrong and which one is right sir ?

I really appreciate it.

Last edited: Jun 22, 2016

Apr 5, 2008
15,649
2,348
Hello,

Looking at the datasheet of the KN2222A, the pinout is as shown below:

Bertus

• ###### KN2222A_KEC.pdf
File size:
220.2 KB
Views:
4
BramLabs likes this.
7. ### BramLabs Thread Starter Member

Nov 21, 2013
98
5
I think i made it right.

Do you know sir, why the transistor gets hot ?
and
Why do the output give only 15v DC, not the inverting of PWM ?

Thank you so much.

Apr 5, 2008
15,649
2,348
Hello,

Have a look at my comments:

Bertus

BramLabs likes this.
9. ### BramLabs Thread Starter Member

Nov 21, 2013
98
5
Hello again sir,
I've tried to exchange the emitter and collector sir, and still, the transistor becomes hot and the output looks like it's latch to the supply, and only give 15V DC, not 15Vpp square wave like i expected T____T.

Thank you before sir for your help ^^

Apr 5, 2008
15,649
2,348
Hello,

It could be that the transistor already has blown.
Do you have an other to try?

Bertus

BramLabs likes this.
11. ### EM Fields Member

Jun 8, 2016
183
32
Here are the results of an LTspice simulation showing that there's nothing wrong with your circuit, so you must have made a wiring error.

BramLabs likes this.
12. ### Alec_t AAC Fanatic!

Sep 17, 2013
5,813
1,105
If the transistor is wired correctly and has a 2k collector load there is no way it should get hot. Max collector current would be ~7.5mA, so if the collector-emitter saturation voltage were say ~0.2V the transistor dissipation would be only 7.5mA x 0.2V + ~5mA x 0.7V = 5mW. I'd guess that either the 2k resistor is shorted out by something or has gone low resistance (unlikely).

Last edited: Jun 23, 2016
BramLabs likes this.
13. ### BramLabs Thread Starter Member

Nov 21, 2013
98
5
Yeah, i think so sir. Because it smells like a burn thing. ahahaha... I'll try later with new transistor ^^
Thank you so much sir for your support !

Yeah sir, i've simulated too with multisim and it works. I thought that maybe because the wiring error as Sir Bertus have already said before, i made an error of the transistor configuration. Maybe i'll try another transistor later.