Maker Pro
Maker Pro

3V to high voltage inverter

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
Hello, I wanted to make 3v to 1000 volts inverter circuit. I already soldered circuit but it doesn't work. I wanted experts to look for errors in my circuit or even in author circuit. Here is link: http://circuitschematicelectronics.blogspot.com/2011/03/3v-to-high-voltage-inverter.html
I may have used inappropriate parts but I couldn't find original so I used such alternatives: instead of 24D506 transistor I used C547B 75d, instead 22M of resistor I used two 10M resistors. Also I don't know if I made properly tranformator. It has same number of turns but I ignored wire and core thickness. I used very thin copper wire(about ~0.1mm) and 5mm thick ferrite core.
High+Voltage+Inverter+schematic.jpg

There is no current in coils L3 and L2 too. There is current flow in transistor from emitter to collector.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I could not get the link to work properly
Have you got the transistors numbered correctly?
The BC547B is 45V 200mA
The 2SD506 is 100V 12A 150W
Neither is the same number as you have and they are very different.

The circuit looks strange to me with a resistor in the collector lead. Also, the output is labeled AC when there are rectifiers.
Could you put component values on the diagram?
If there is current flow emitter to collector there must be current in L2.
This is a blocking oscillator, it needs the windings connected the right way round to work. You could try changing over the connections to either L1 or L2.

Some pictures may help.
 

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
If link doesnt work simple copy/paste "3V to High voltage inverter Schematic Diagram" in search engine and you should see it.
http://circuitschematicelectronics....3v-to-high-voltage-inverter.html#.UQKu-vJYWSo
Transistor that I used is not powerful and small. 100V 12A 150W probably has to be large transistor that can screwed to cooler. I have bunch of them at home. How to identify needed?

Resistor
R1____1K5 Ω
R2____4K3 Ω
R3____22M Ω (I used two 10M)

Capacitor
C1____100n
C2____100n 400V
C3____0.2uF 400V


Diode
D1___1N4007
D2___1N4007

Transistor
Q1___ 24D506 (I used C547B)

Transformer
L1___#100 turns
L2___#100 turns
L3___#1000 turns
 

TedA

Sep 26, 2011
156
Joined
Sep 26, 2011
Messages
156
sauroman1,

You should know that the schematic, both as seen in your original post, and on the linked page, has a typo that will prevent the circuit from working.

The transistor's base and collector connections are reversed.

This will tend to destroy the transistor when power is applied, depending on how much current your DC source can provide, and the DC resistance of the transformer winding. The transistor's base current capability will obviously also matter.

The transformer, and how it is constructed can vary a great deal, but if too out of whack can keep the circuit from working at all.

The linked circuit has several shortcomings.

The transistor's base and collector must be connected to oppositely phased transformer windings, so the base goes positive as the collector goes negative.

Conventionally, when drawn as we see in this schematic, the transformer connections are shown the wrong way for circuit operation.

The L1 and L2 windings are connected so the transistor will just sit there and get hot.

The secondary (L3) also appears to be connected backwards, since to get a large voltage step-up, we want the diodes conducting while the transistor is off, rather than when it is on.

Three dots in the right places would be one way to fix this. However, it's not a bad practice to draw windings so the phasing dots are not necessary, then add the dots for emphasis.

The entire secondary circuit does not make much sense. The 1N4007 is a very slow diode. The large C2 and R2 values will tend to clobber the output.

I'll leave comments about the transistor types to others. Random subs might work if you get lucky.

I hope all this is not too discouraging; do keep tinkering and posting questions. That's how you learn.

Ted
 

TedA

Sep 26, 2011
156
Joined
Sep 26, 2011
Messages
156
sauroman1,

I had a chance to look into your transistor. Is C547B the complete part number? Who is the manufacturer? Can you post a link to a data sheet? I think this number might signify a GE SCR. That won't work in your inverter circuit!

Often the marking on the part is not complete. If the transistor is from Japan, it might be a 2SC547B. If from Holland, a BC547B. Not at all the same. Give us some clues! Sometimes, even a photo of the thing will help.

And to get more useful answers, you might tell us what test equipment and instrumentation you have available for use in getting your circuit working. And if you reveal your location, you may find that one of the forum members is nearby and might help in person. Location also gives us some notion of parts availability.

Ted

PS I see another flaw in the posted circuit. Depending on many factors, the transistor's reverse Vbe rating will likely be exceeded each cycle of oscillation, when the transistor turns off. Energy that should go to the load will instead degrade the transistor. Perhaps gradually, perhaps very quickly. This is assuming that the circuit is changed to make it a flyback inverter.
 

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
Transistor has such markings "C547B W 5d". I also last day tried using KT829B russian transistor. but circuit also didn't work properly. I live in Lithuania
Nuotrauka0199.jpg

Nuotrauka0202.jpg

Nuotrauka0203.jpg
 
Last edited:

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
Transistor C547B W 5d is dead. In other forum I was told also that I improperly connected transistor. Just say what characteristics needed transistor must posses to work properly.
 

Rleo6965

Jan 22, 2012
585
Joined
Jan 22, 2012
Messages
585
If all component were correct and still not working. Reverse the winding of L1.
 

TedA

Sep 26, 2011
156
Joined
Sep 26, 2011
Messages
156
sauroman1,

Thanks for the pictures!

It is still not clear what sort of "C547B" you have. I don't recognize the manufacturer's logo. If that is the only one you have and it is now dead, perhaps it does not matter.

Don't power-up your circuit again with any transistor, until the schematic errors are corrected. Dead transistors are the expected result, with the circuit shown. Sketch a schematic and post it and we will comment.

I think I might try some sort of small power transistor at first. It would be harder to blow up. Something rated at 2 to 10 A collector current and 150 to 200 volts collector voltage.

( If you are getting 1kV output, the transistor's collector voltage will exceed 100V during circuit operation with a 1:10 transformer ratio. )

I don't know what you have, or can easily get in the way of transistors.

You might investigate some dead compact fluorescent bulbs to see if the ballast uses bipolar transistors. These should stand a chance of working.

Ted
 

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
I reversed base and collector connections and current is flowing in all three coils. L3 shows AC 220 volts, but when I touched connections I felt nothing. Maybe my old type multimeter is not precise, but even fuse button pops out meaning there is a lot volts. Although I get nothing from AC out connections.
 

TedA

Sep 26, 2011
156
Joined
Sep 26, 2011
Messages
156
sauroman1,

Sounds as if you are making progress.

Note that the AC frequency may be too high for your multimeter to measure properly. Often the AC ranges on these are only good for power line, and possibly low audio frequencies.

It is possible that touching the connections overloads and stalls the oscillator.

The schematics have connections labeled "AC Out", but these should have mostly DC on them. Were you using the DC Volts range on your meter here?

For troubleshooting such a circuit an oscilloscope is almost a necessity. You should look around for a nearby amateur radio club or technical institute where you might find some help. Hams and teachers of all sorts often are glad to help someone interested in their specialty. ( Or, you should start an electronics "Make" club, if there is not one nearby already.) Using a 'scope will also help you understand how such circuits work.

I'm trying to think how you might even see what frequency you are getting. If you have only a multimeter, perhaps you can use a MW or LW radio to tune the harmonics of the oscillator. The space between the signals on the radio's dial should be the oscillator frequency.

What are the voltage ratings for the transistor you are using now? As you get the circuit working better, the voltage stress will become more important. Both Vce and Vbe may be too much for your transistor.

At some point, you will want to see about reducing the value of C2. But C2 may be the only thing between an under-rated transistor and instant POOF.

Perhaps the cost of components is an important consideration. You might find useful parts in old CRT-type televisions and computer monitors, as well as discarded light bulbs. Computer power supplies, too.

You should figure out how to post a schematic that you have created, so you can let us know how your circuit is actually wired-up. I guess a very dark pen on paper, then taking a picture can work. If you draw big.

Thanks for letting us know where you are located. Perhaps you should change to a MOSFET circuit, so you can use a product of International Rectifier. But for a couple of nasty dictators, that company might well have been right there in your country. http://www.irf.com/whats-new/nrfounder.html ( Actually, it is a bit hard to get a MOSFET to work from such a low supply voltage.)

Keep us posted with your progress.

Ted

PS Can you tell us about the other forum? A link to it? It would be interesting to know about it.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
One source of low voltage to high voltage convertors is the flash circuit of disposable cameras. There was a shop near here which would develop the films while you waited.
I asked for a couple of cameras and was given two, complete with good alkaline batteries. I was told to be very careful and found out why when I got a bite!

The flash circuit may not give as much voltage as you require but it should be possible to add a voltage multiplier.

One I looked at had two small transistors in parallel, presumably to get sufficient current at high frequency.
 

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
One source of low voltage to high voltage convertors is the flash circuit of disposable cameras. There was a shop near here which would develop the films while you waited.
I asked for a couple of cameras and was given two, complete with good alkaline batteries. I was told to be very careful and found out why when I got a bite!

The flash circuit may not give as much voltage as you require but it should be possible to add a voltage multiplier.

One I looked at had two small transistors in parallel, presumably to get sufficient current at high frequency.

Does disposable camera flash unit produce more then 25hz and how much voltage?
 

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
I can't understand multimeter reading, it shows 220V but feel nothing. It seems that voltage differs every time I measure. Then I tried circuit from 6V fluorescent lamp - multimeter shows 220 volts and when I touch connections get strong shock.

I don't have any way to measure with oscilloscope. Here are characteristics of transistor KT829B:
Maximum collector power dissipation (Pc), W: 60
Maximum collector-base voltage (Ucb), V: 80
Maximum collector-emitter voltage (Uce), V: 0
Maximum emitter-base voltage (Ueb), V: 5
Maximum collector current (Ic max), A: 8

Should I try buying 100V 12A 150W transistor? Or should you recommend me making new circuit? If it was possible to modify circuit of lantern: to amplify voltage to ~1000V, use 3V and simplify circuit because there are two power supplies for fluorescent lamp that can't be connected into one.
SAM_6148.JPG


SAM_6149.JPG
 
Last edited:

TedA

Sep 26, 2011
156
Joined
Sep 26, 2011
Messages
156
sauroman1,

Disposable camera inverters usually provide a 200 to 300V DC maximum output. The frequency is often a few tens of kHz at full voltage. The frequency is lower as the flash capacitor begins to charge. You may be able to hear it. These circuits are designed to work well over a range of output voltages, since their job is to charge-up a big capacitor.

If your start with one of these camera inverters, you might use the existing transformer with a 3X or 4X voltage multiplier circuit. ( Don't try this with 1N4007 diodes!) Or you could rewind the transformer with more turns of finer wire.

Are your 220V readings taken using the DC ranges of your meter? An ordinary meter may not provide useful AC readings at the inverter frequency. The output of the posted circuit is mostly DC, not AC.

When my voltage readings do not make sense to me, I often feel not "nothing", but confusion, or frustration! Or are you engaging in some "digital" measurements? Best to keep those digits dry.

The output of inverter type fluorescent lamp ballasts is AC. And likely to be at a frequency too high for your meter.

If this is still your transformer:

L1___#100 turns
L2___#100 turns
L3___#1000 turns

the output voltage will be limited by the turns ratio to about ten times the transistor breakdown voltage.

With the circuit you originally posted, after the connections to the windings are corrected so the circuit will oscillate, this limit is set by the transistors base to emitter breakdown voltage. Presumably the actual part breaks down at some voltage value in excess of the 5V maximum rating. Many silicon transistors will go to 9 or 10V Vbe, before breakdown. Due to leakage inductance in the transformer, you may get an output voltage somewhat more than ten times the Vbe breakdown voltage. But much of the power is going into the transistor, instead of your output. And the transistor is being degraded by operation in Vbe breakdown.

So you need a transistor with a higher collector voltage rating, and circuit changes to avoid excessive reverse voltage on the transistor's base. ( Generally, there are no transistors to be had with much higher Vbe breakdown voltages.)

Or, you might try increasing the transformer turns ratio, so you can get more output voltage at a given transistor Vce. If you do modify the transformer, the base drive winding might be wound with fewer turns than the primary. There is far more voltage across the 100T winding than you need to drive the transistor base.

A thought about the transformer: The one in the photo with a cylindrical rod core has a very long air gap. Far longer than required for operation. You could gain some advantages by using a different core, such as the one on the lantern inverter. Far fewer turns would be required to get the inductance you need. The single-ended inverter circuits still require an air gap in the core, but not one of several cm.

Ted
 

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
Have you got the transistors numbered correctly?
The BC547B is 45V 200mA
The 2SD506 is 100V 12A 150W
Neither is the same number as you have and they are very different.

How you determined 2SD506 transistor parameters? It seems that this transistor doesn't even exist because I can't find any information about it on internet. May be culprit too weak transistor?
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I have a book - remember them. Unfortunately when I tied to look up the transistor again it wasn't there. I will have to get a new pair of glasses.

The circuit you show has so many faults that you should start again. In particular, turns ratios will need to be recalculated and matched to the transistor characteristics. The output uses 1N4007 diodes which are slow and so no use at high frequencies, they are shunted with a capacitor which will severely reduce the output and there is a resistor to waste power.

To start from such a low voltage to go to such a high voltage is difficult which is why I suggested a flash circuit which is known to work.
 

sauroman1

Apr 3, 2012
34
Joined
Apr 3, 2012
Messages
34
Could I just simplify circuit to get it to work? I think 300 volt output should be enough.
I saw on Youtube that people make high voltage inverters also called "Joule Thief" that are simple - use transformer, transistor and maybe resistor.
I even could replicate circuit I showed in photos, only I can't also find transistor 0882 Transun U11 datasheet and neither I know how much windings there are on that transformator.
116160-283x424-Portable_lantern.jpg
 
Last edited:

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Sorry to seagull this thread... I haven't read much of it. Just wanted to say to TedA that the "C547B" is a BC547B, a very common European transistor - NPN, TO-92, 45V, 100 mA, pinout the reverse of the 2N3904 (CBE not EBC).

Also, that design is very strange. I had a look at the link in the first post and I have to wonder if the guy is just trolling and trying to waste people's time. The description text is worse than useless. A project like this needs a detailed specification for the transformer, at least. As well as the base and collector being swapped, the secondary circuit looks very odd.

Possibly it's been reverse-engineered incorrectly? It looks rather like the design of a fly zapper in the shape of a tennis racket that I bought once. The secondary of that circuit uses two diodes and two capacitors as well, but they're connected as a voltage doubler, not as shown in that crazy circuit. The racket does work for zapping flies BTW.
 
Last edited:
Top