High voltage amplifier using npn transistors

Hello,

Im working on making a high voltage amplifier atleast 600Vp-p. Its really hard, if at all possible to find pnp transistors that suits this voltage, and even more difficult find complimentery transistors.
Im trying to avoid using mosfets.

So Im wondering - is it possible to make an amplifier using pure npn transistors. If it is, could you please show me some curcuit designs?

what output current, what frequency?
here is one solution using mosfets:
**broken link removed**

Look here

The opamp MC 33071 was designed with an all NPN output stage. Maybe that could inspire you.

are you sure about that? i looked at it's datasheet and it seems to have a lot more PNP than NPN transistors...

**broken link removed**

how about high voltage operational amplifier?
i know how it sounds but some opamps pack a punch:
https://www.cirrus.com/en/products/pa94.html
https://www.cirrus.com/en/products/pa95.html

Transistors
FJP5027OTU, KSC5027OTU, FJPF5027OTU
KSC5603DTU
BUL7216
MD2001FX
BUL216
BU508AF
BUL416T
HD1750FX
2SC3743 ****
2SC3979 *****
2SC4960
KSC5042FTU
2SC4004
2SC3405
2SC5465

This should get you started.

Thanks for all the replies.

Its a 5 [ohm] load, so it has to deliver 120 [A]. At 330kHz. The Opamp sure will deliver the voltage, but I don't think they would be a viable candidate as I'll need like 75 000 of them to get the current supply that I need

Thanks for the transistor suggestions. I've allready had a look at a lot of them. But they don't have a complimentary pair. And they're all npn transistors. The problem isnt finding npn transistors that work at 600V, however PNP's working at 600v are difficult to find...

Thanks for the MOSFET amp, I'm going to have a closer look at that

But is it actually possible to make a pure NPN amp? I've googled around, and I havnt had much luck, so if anyone has any curcuit designs that would be great!

don't know for sure, never bothered to think that way. I've seen some amps with only one PNP and it wasn't the output transistor (it was one stage before).
i did see some 600V PNPs for 10A or so but didn't write them down (sorry). btw, what you are after is domain of VFD and servo drives and those use IGBTs,
You can get entire the whole output section as single IGBT module. btw, what kind of amplifier is this? what is it for? what frequency range? does it start from 0 or something low?

panic mode said:

are you sure about that? i looked at it's datasheet and it seems to have a lot more PNP than NPN transistors...

**broken link removed**

Click to expand...

Reread where I said "an all NPN output stage". That is the main point of my suggestion. Not the rest of the circuit.

ooops gotcha

that's 2.88kW.... what are you making? an induction heater?

edit:correction 72kW....... you must have your own power plant?

PSYTROX,

5 ohm load, 120 amp, 600Vp-p, 330khz.
This is not a small project. I have 1000 volt 15A transistors.
HD1760JL 800V 36A I have not used these but 5 in parallel might work.
Is this a home project or do you have money. This is going to cost money!

When I think over it, the current is probably too much. It should be half. Its 600Vp-p. So at the most, its only 300V going through the 5 ohm resistor, giving a current at 60 Amps, not 120.

Its an amplifier to power a magnet, on a sensor used on oilrigs (im sorry i cant be more specific, but its most likely confidential), (Im guessing oilrigs can output the ammount of current that i need - in anycase that part really isnt my problem - I've just been given specifications). The signal is a pulse (a sync looking pulse), so its not going to be on for very long... So I dont really have to worry about it overheating. 95% of the time it works at 330kHz - but sometimes they do use it at lower frequencey's.

Its a bachelors project - theres a company sponsoring it, so I don't really worry so much about what its going to cost

The problem using IGBT transistors is that they don't do very well at high frequenceys.

The company we're doing the project for said they would really prefer using bipolar, as they dont require as much overhead voltage. I think this has something to do with the fact that they want the most juice they can get from 600Vp-p and also that there are laws/rules/protocols for voltages above 600v. That maybe they need more or better (which ofcourse means more expencive) equiptment to make it explosion proof...? Im not really 100% sure - but Im guessing thats why.

Sounds like you are making a magnetic flux based drill stem checking system perhaps?

I have seen something like this in use in the oil fields around here, I forget the correct name at the moment, already used by companies that check the drill pipe for weak or thin spots after its been used for while.

The ones I have seen run off of common 120 Volt portable generators and are hooked up to computers that give an actual graphic image of the pipe wall condition as its being checked.

Neat tech but boring to watch when in use!

PSYTROX,

5 ohm load, 60 amp, 600Vp-p, 330khz.
I have done many projects with 1000V+, 10 to 20A transistors at 100 to 500khz. This type of amplifier will take a long time to make work. I don't know anyone that can design this amplifier with out much thinking. Good luck.

your biggest challenge will be the physical layout and wiring, as the voltages and currents aren't trivial. as far as the amplifier itself goes, paralleling transistors makes for slower amplifiers. the junction capacitances add up quickly

unclejed613 said:

your biggest challenge will be the physical layout and wiring, as the voltages and currents aren't trivial. as far as the amplifier itself goes, paralleling transistors makes for slower amplifiers. the junction capacitances add up quickly

Click to expand...

I disagree. The capacitance does add up but the drive also has to go up by the same amount.
If one transistor has 1A base current and 100pF of B-C cap then two transistors will need 2A and have 200pF. The capacitance/current remains the same.
OR
If you are talking about collector current......same story.

I've been looking at this circuit:

https://1.bp.blogspot.com/-2LRhieMP...s1600/400W+MOSFET++Amplifier+high+quality.png

If you look at the transistors Q4 and Q5. What purpose do they actually serve? What do they really do?

Q1 and Q2 are a amplifier. They amplify the voltage difference Q1-B & Q2-B.
Q4 and Q5 are another amplifier. (or you could say it increases the gain of Q1,2)
This is a typical audio power amp.

Are you trying to get sign wave in your output?
Will a square wave work?

What about D5 and D6?

I wonder what D5 and D6 do? Not zenners, so...