Author Topic: DC to AC with NPN BJTs?  (Read 1650 times)

0 Members and 1 Guest are viewing this topic.

Offline Windows9ProfessionalTopic starter

  • Contributor
  • Posts: 20
  • Country: at
DC to AC with NPN BJTs?
« on: October 24, 2018, 07:34:13 pm »
First of all: I don't wanna create a 12V to 230V inverter! Since I'm just a 13 years old boy who's interestet in electronics, I don't want to go from LV DC to HV AC. I got a few BJTs layin around on my desk
(BC337-25, Datasheet: https://www.onsemi.com/pub/Collateral/BC337-D.PDF) and I wanted to create an AC source with them. Here in Austria (not Australia), the next big city in my area is Graz. You can't find any electric components shop outside of Graz and my Family isn't often there. So I want to build an Inverter with only NPN Transistors (I got 10 of them, so I can use as many as I want). I got bored of watching some Indian people, building a transformer core out of an NiMh battery (wtf), so I asked the in my opinion best electronics community. The BJTs are capable of 800mA before the magic smoke escapes. So, can someone of you out there recommend me a circuit? Remember, no Transformer!

Thanks!
 

Offline HB9EVI

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: ch
Re: DC to AC with NPN BJTs?
« Reply #1 on: October 24, 2018, 07:52:40 pm »
I'm too lazy to start up the cad, so I stole a schem

something like that is simple and working, but you need a center tapped transformer.

Certainly you cannot use a BC337 for that, since it has a maximal Ploss of 625mA only, and to only get 40mA on the secondary side, you transfer almost 10W of power, so no chance for such a transistor
 

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3925
  • Country: de
Re: DC to AC with NPN BJTs?
« Reply #2 on: October 24, 2018, 08:48:20 pm »
First of all: I don't wanna create a 12V to 230V inverter! Since I'm just a 13 years old boy who's interestet in electronics, I don't want to go from LV DC to HV AC. I got a few BJTs layin around on my desk
(BC337-25, Datasheet: https://www.onsemi.com/pub/Collateral/BC337-D.PDF) and I wanted to create an AC source with them. Here in Austria (not Australia), the next big city in my area is Graz. You can't find any electric components shop outside of Graz and my Family isn't often there. So I want to build an Inverter with only NPN Transistors (I got 10 of them, so I can use as many as I want). I got bored of watching some Indian people, building a transformer core out of an NiMh battery (wtf), so I asked the in my opinion best electronics community. The BJTs are capable of 800mA before the magic smoke escapes. So, can someone of you out there recommend me a circuit? Remember, no Transformer!

Thanks!

Well, first - don't! Capacitor charged to 230V will happily make you fly across the room if you touch it and even a puny inverter powered from a battery will be able to charge that. You obviously don't have neither the knowledge nor the experience how to handle something like that safely yet. I especially don't recommend the various Indian Youtube videos - those guys are often Darwin award candidates and their projects are both extremely poorly designed (will blow up) and dangerous (could kill you).

Second, any inverter will need at least an inductor or you won't have any high voltage. Whether that's a transformer or a simple coil, you will still need something like that there. Without an inductor you could only build charge pump or a Villard cascade multiplier. Those will work but won't be able to deliver much if any current.

Your BC337s are pretty useless for a job like this - that transistor is rated for only 50V and will reliably blow up if you try to use it in any sort of inverter like this. Also that 800mA of current is fairly low if you are planning to draw any current from this - remember, law of conservation of energy applies, so if you want e.g. 50mA of current at 230V, that gives you 230*0.05 = ~12W of power that you need to put into the system as a minimum (that's about enough for a low power light bulb). In reality (much) more because of losses. Then 12W/12V = 1A of current - and your transistor goes POOF. And this is a very optimistic case - in reality the efficiency of such converter will be very poor for any design without a transformer. So a lot more current will be needed to go through the switching transistors.

I do wonder why do you want to build an inverter? High voltage inverters are very complex devices requiring a lot of knowledge to design if they are to be both safe (you don't want to kill someone or set a house on fire!) and reliable. This is definitely not a good beginner project - you will be frustrated when it won't work and in the worst case you can get badly hurt.

BTW, in Austria you should be able to order components from the internet without problems - these days it is very rare to find a store you can walk in carrying electronic components, that has died out some 20 years ago.
 
The following users thanked this post: nugglix

Offline janoc

  • Super Contributor
  • ***
  • Posts: 3925
  • Country: de
Re: DC to AC with NPN BJTs?
« Reply #3 on: October 24, 2018, 08:59:09 pm »
I'm too lazy to start up the cad, so I stole a schem

That circuit is complete (and dangerous!) garbage, don't build that.

Here is a good analysis of these circuits:


* Output frequency varies with load and input voltage
* No feedback/regulation - unloaded voltage shoots up easily to over 700V or more, if you connect load the voltage drops significantly
* No dead time so you have high shoot-through current through the transistors while switching -> losses, heat
* Ton of EMI generated
etc.

 

Offline HB9EVI

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: ch
Re: DC to AC with NPN BJTs?
« Reply #4 on: October 25, 2018, 02:14:22 pm »
Sure, I didn't say it's a good circuit, but you can experiment with it - IF you know what you do

I long time ago stopped emphasizing that it's dangerous to work with high voltages - either the people know it very well themselves and stay cautious - or they don't want to hear about it, because they know everything BETTER.
 

Offline spec

  • Frequent Contributor
  • **
  • Posts: 833
  • Country: england
  • MALE
Re: DC to AC with NPN BJTs?
« Reply #5 on: October 25, 2018, 09:56:24 pm »
+ Windows9Professional

The hand schematic below is the best I could do given the components (just one PNP transistor would have allowed a better and simpler circuit).

I haven't analyzed the circuit in detail, but it should give you something to experiment with.

The multivibrator, Q1 and Q2, does exceed the -VBE limit of the BC337 (-5V max), but the base reverse current would be limited so the multivibrator should work OK.

The output will be approximately 3V RMS, at around 70Hz, and the waveform will be a slightly rounded square wave.  For 50Hz output frequency (mains frequency in Austria) change R1 and R2 to 68K and for 60Hz  output frequency (mains frequency in United States) change R1 and R2 to 56K.

The maximum output current should not exceed about 100mA RMS (minimum load resistance 30 Ohms).

The value of the output capacitor, C4, is not critical: 220uF upwards would be OK. The working voltage of C4 should be 12V or higher.

The value of the decoupling capacitor, C3, is not critical either: 10uF upwards would be OK. The working voltage of C3 should be 15V or higher.

The frequency of the multivibrator is dictated by the values of R1 and R2, and C1 and C2. The best way to change the frequency of the multivibrator is to change C1 and C2. But you can also change the multivibrator frequency by changing R1 and R2, although the minimum value of R1 and R2 is about 10K and the maximum value of R1 and R2 is about 220K.

By the way, the general formula for the multivibrator frequency is 1/1.38RC Hz (R in Ohms and C in Farads) In the case of this multivibrator R is the value of R1 or R2 and C is the value of C1 or C2.

You can also change the mark to space ratio of the multivibrator by having different values for R1/C1 and R2/C2.

You can also try different supply line, but don't go over 12V.



« Last Edit: October 26, 2018, 06:28:48 am by spec »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf