555 - HV output for the piezo with 400-450V @ 25-30kHz.

ian field

Joined Oct 27, 2012
6,536
Hello, I am so glad that you decided to help me out with this.

Well how much current is not really matter to me. It can mr from 1mA to max of 555 output 200mA. Because the load is piezo material. So only thing I need is suply my piezo with 400-450V @ 25-30kHz.

I see you suggest me the power mosfet. Will that need transformer for specified output or?

And could I use the backwards transformer?
I already tryed backward with simple circuit but the voltage just jumps up and drop around 120 to 500V. If i only could stabilze that.
If you resonate the piezo element - it will draw peak current.

If you don't resonate the piezo - the whole exercise is pointless.
 

Wendy

Joined Mar 24, 2008
23,415
I meant the output from the transformer (if he has one), if that's a sinusoidal 450V 30kHz I always thought you can't switch that on and off directly with semiconductors? ( but I am no expert lol) I know solid state relays can be used on low frequency AC but I've had trouble finding ones for higher frequencies.
You are misinformed. A 30khz square wave will translate just fine through a transformer, minus the DC component. Not sure where the heck you are coming from on transistors. Theit two mkost efficient states are on and off, the basis of all digital electronics.
 

Wendy

Joined Mar 24, 2008
23,415
@Uskok

My feeling is a transformer would work fine for your application. The really are not hard to fabricate. Willing to try? Otherwise you will have to make a much more complex circuit.
 
Last edited:

Thread Starter

Uskok

Joined Aug 19, 2015
30
Well. I it is not problem for me to try. But which one to use. Backwards transformer. Or to wind one my own?

Can someone make circuit for me. Pls
 

MikeML

Joined Oct 2, 2009
5,444
Capacitance is 12.84 nF
I think you way underestimate what it takes to drive that piezo.
Here is a hack at it. I didn't bother with the 555; V1 is a proxy for it.

Note the magnitude of the current in the piezo, I(C1), violet trace.

You could also do this with a full H-bridge. In both cases, you will have to find 500+V PFETs and NFETS. R2 needs to be rated for 20W. M2 and M3 each dissipate about 40W, so will have to be on a huge heatsink, or under a fan.

1.gif

If you drive the piezo with a transformer as suggested above, the turns ratio with have to be about 450/10, or 45:1. If the piezo currents are as shown, the peak primary current will have to be about 100 Amps, which, the last time I looked, is about 500 times what a 555 can do...
 
Last edited:

GopherT

Joined Nov 23, 2012
8,009
Use BJT transistors from an analog TV HV horizontal output transistor. They are high-speed and handle 1500 volts and 50 Watts.

2SC1172B is a good starting point.
 

Wendy

Joined Mar 24, 2008
23,415
Hmmm, that current spike seems a lot high.

I'm trying to get away from a high voltage power supply. You can always add drivers for the 555 for the transformer. It is a matter of simplification.

A question for @Uskok , do you have an oscope and know how to use it?
 

MikeML

Joined Oct 2, 2009
5,444
Hmmm, that current spike seems a lot high...
To charge 13nF through 450V in 16.67us takes a constant current of 0.35A. If you want the drive to look more like a square wave than a sawtooth, then the current needs to be at least 10X more than that...

2.gif
 

ian field

Joined Oct 27, 2012
6,536
Use BJT transistors from an analog TV HV horizontal output transistor. They are high-speed and handle 1500 volts and 50 Watts.

2SC1172B is a good starting point.
Back in the 80s/90s a typical horizontal output transformer had a gain in the range 2 - 8.

They probably improved a bit before the demise of CRTs, but the driver circuitry still won't be trivial. Electronic ignition transistors may have better gain.

I'd go for a full-bridge driving the primary of a transformer. As long as its not a low Vcc application, I'd go for complementary darlingtons, they can't switch all the way to the rails so you lose a little bit of voltage swing - but they fall just shy of saturation which should reduce switching losses at the stated frequency. The second pair of transistors can be driven by a simple MOSFET inverter like a 2N7000 or BS170 and pull up resistor.

A good source of SMPSU transformers that could be used back to front was the old plastic case home computers - these days a more likely source is TV set top boxes of various kinds.
 

Wendy

Joined Mar 24, 2008
23,415
I'm waiting for input from the OP about the oscilloscope.

He might be able to get by with a store bought transformer, but I suspect not. Most power transformers are rated line frequency, 50Hz in his case.

But transformers are really easy to make. You wind a certain number of winding around a chunk of iron, do it again for the secondary, and it is done. Transformers and inductors are among the easiest components for a DIY to fabricate. You could try using a 6.3VAC/220VAC transformer to see how it might work, as an experiment.

The alternative is to create a 450VDC power supply, which could be quite dangerous if not treated with respect, and use transistors to switch it to your piezo circuit. Many more parts, greater hazard, there is a reason I'm testing the waters for a transformer type circuit first.

It may be possible to buy an off the shelf design to do this. That would probably be best, but I am not aware where you might get something like this.

I could design a circuit using the 1st 2 ideas, as could most of the folks helping you. The transformer option really is the simplest in terms of circuity, you just have to do some mechanical fabrication, as in limited drilling, screwing, and winding of magnet wire.

Most people subscribe to KISS, Keep It Simple Stupid.
 

MikeML

Joined Oct 2, 2009
5,444
... The transformer option really is the simplest in terms of circuity, ..
Most people subscribe to KISS, Keep It Simple Stupid.
Hey Wendy, you will have to convince me that a low-voltage, high-current (100A?) bipolar driver plus a ferrite pulse transformer is any simpler than what I already posted...
 

Thread Starter

Uskok

Joined Aug 19, 2015
30
Hello... Yes I have an oscilloscope and I know to use it...

I need to tell you that I don't use the ordinary piezo transcuder.. I use the Macrofiber composite material, is that make any diference?

And yes I decided to use transformer... I'll probably listen your advices and wind my own... So the ratio is 45:1 right?
 

Wendy

Joined Mar 24, 2008
23,415
@MikeML

Which goes back to option #2, a high voltage 450VDC power supply. You show a circle to describe it, but it will need built in the real world. It is a hazard for a beginner too, if they aren't careful. All of that can be done with beefy drive circuit (4 transistors at worst) and a transformer.
 

Wendy

Joined Mar 24, 2008
23,415
Hello... Yes I have an oscilloscope and I know to use it...

I need to tell you that I don't use the ordinary piezo transcuder.. I use the Macrofiber composite material, is that make any diference?

And yes I decided to use transformer... I'll probably listen your advices and wind my own... So the ratio is 45:1 right?
I can't help you much with the piezo end, but the electronics is pretty easy.

As to the turns ratio, if you were to feed a 1V P-P signal you would need a 1:450 wind ratio. But you will be feeding what is more likely to be a 10V P-P signal, which puts your ratio to 1:45.
 

Thread Starter

Uskok

Joined Aug 19, 2015
30
Okay... So my input voltage is 12-15V...
So I need to wind primary 10 turns and secondary 450turns... Am I right?

And connect that in circuit with power transistor?
 

Wendy

Joined Mar 24, 2008
23,415
I'll show you some drivers in a bit, assuming someone doesn't beat me to it. If you used a large torrid (I wouldn't, best design but big hassle) I would use 10 winds primary and 450 secondary. The relatively high frequency works in your favor on this.
 
Top