NE555 variable frequency circuit 50kHz-1MHz

Thread Starter

-live wire-

Joined Dec 22, 2017
959
Can anyone suggest a good variable frequency circuit with an NE555 timer? It would ideally be possible to do on a breadboard (capacities not having a huge effect). The duty cycle needs to be close to 50%, but it doesn't have to be too exact (±10%). I know that the BJTs can lead to issues, but all I have is the NE555. I haven't really done a lot with 555s yet. I need 50kHz-1MHz, which I am going to use to drive a mosfet. The mosfet will control the primary coil of a high voltage step up transformer.

My arcs haven't been long enough with my 6V supply, nor have they been powerful enough (very difficult to light a piece of paper on fire). At higher supply voltages, it draws too much current. So I need to increase the frequency (to increase inductive reactance). I have been using an arduino, with a PWM of ~150kHz. But the mosfet already heats up a lot, despite the low switching current. It has melted a hole in my breadboard, and required a lot of heat sinking.

I think it needs more current to drive it, and a higher voltage. It also needs its capacities discharged faster. With an output of 14-15V, and up to 200mA, the NE555 seems like it will preform better than an arduino. But I am concerned about the transition. Will it be close enough to a square wave to not have the mosfet in the ohmic region for too long?
 

MisterBill2

Joined Jan 23, 2018
18,167
First thing, an arduino is seldom a good choice as a variable frequency source. A 555 timer IC can serve as an oscillator but the output is not a 50% duty cycle square wave. If you can access an on-line data book it should have an application circuits section for the 555 that will include oscillators and the calculations to determine the frequency of oscillation. If you go to the "schematicsforfree" website you can see a number of circuits to provide high voltages. No instructions but many circuits.
 

MisterBill2

Joined Jan 23, 2018
18,167
Actually, it seems that this thread is very similar to one a while back where somebody wanted to create a sparking device to use for fire starting when camping. I started to ignore that thread when it wandered away from reality. But really, the site that I suggested has lots of circuits that could work. And a hot mosfet needs a heat sink, and probably a bit more drive voltage, as well as a load device with more primary turns.
 

Thread Starter

-live wire-

Joined Dec 22, 2017
959
Whoa! Too many issues in one thread.
Deal with the MOSFET heating problem first.
Show us your circuit.
upload_2018-10-7_16-36-31.png
Mosfet being driven by arduino at ~50%, 140kHz, diode because of inductive spikes damaging gate, pull-down resistor to help discharge capacitance, and HV transformer. It works but the mosfet needs serious heatsinking (huge sink + fan).

First thing, an arduino is seldom a good choice as a variable frequency source. A 555 timer IC can serve as an oscillator but the output is not a 50% duty cycle square wave. If you can access an on-line data book it should have an application circuits section for the 555 that will include oscillators and the calculations to determine the frequency of oscillation. If you go to the "schematicsforfree" website you can see a number of circuits to provide high voltages. No instructions but many circuits.
Can you recommend a particular one for the NE555? And I know there are many circuits to get high voltage, but all of them are basically a HV transformer and an oscillator.

Actually, it seems that this thread is very similar to one a while back where somebody wanted to create a sparking device to use for fire starting when camping. I started to ignore that thread when it wandered away from reality. But really, the site that I suggested has lots of circuits that could work. And a hot mosfet needs a heat sink, and probably a bit more drive voltage, as well as a load device with more primary turns.
The circuit I have right now works. It does light a piece of paper on fire, after a while. But the required initial distance is too small, and the arc could be more powerful. So I am going to use a higher voltage and higher current, but I also have to increase the frequency, and drive the mosfet better.

An NE555 supplying 200mA at ~15V seems much better. But I need a good circuit. And I will also need to adjust the frequency.
 

MisterBill2

Joined Jan 23, 2018
18,167
Probably you need less than a 50% duty cycle, since once the magnetic flux in the transformer reaches saturation the rest of the time the current is only producing heat.
I can describe the 555 oscillator circuit but can't draw it out.
The supply voltage is connected to pins #4 and#8, the supply return (common) is connected to pin#1
A "stablizing capacitor is connected from pin#5 to common.
Timing resistor "rA" is connected from the supply voltage to pin #7, and timing resistor "rB" is connected from pin#7 to pin #6. The timing capacitor "C", is connected from pin #6 to common.
The output is pin #3
The frequency is calculated as F= 1.44/(rA+2rB)xC
and the duty cycle is given by D=rB/rA+2rB
The resistors much have a high enough value to limit the current into pin #7 to less than 200 milliamps.
This data came from a 1979 Motorola IC manual. Old but still very useful
I suggest that you start with a much shorter duty cycle,(on time) and see if that reduces the heating enough. I think that it should. But the duty cycle only needs to be limited to the time for the transformer iron to become saturated. That will take using an oscilloscope to discover, looking at the drain terminal of the mosfet.
 

MisterBill2

Joined Jan 23, 2018
18,167
Which only goes up to 100kHz.
True, but that may be adequate, especially if the primary can be tuned to resonance. And the adjustable duty cycle may be exactly what is needed.
The CD4046 PLL IC, using just the oscillator, is another option but it will need some amplification to be able to drive the gate of the power mosfet. If L.W. already has the 555 devices it would certainly be easy to try.
 

MisterBill2

Joined Jan 23, 2018
18,167
The 4046 oscillator is another option but I see on this very site related posts with the 555 in the megahertz range. It has improved a lot since 1979. And 9it is still worth checking if it can be done easily.
 

Thread Starter

-live wire-

Joined Dec 22, 2017
959
Which only goes up to 100kHz.
If the NE555 only goes up to 100kHz reliably, then maybe I need to consider alternatives. So I am thinking of either this comparator/shunt circuit or driving the mosfet better. Using standard BJTs, or a 2N7000 mosfet (stuff I have), how might I drive the mosfet better, still with the arduino PWM?
 

Thread Starter

-live wire-

Joined Dec 22, 2017
959
True, but that may be adequate, especially if the primary can be tuned to resonance. And the adjustable duty cycle may be exactly what is needed.
The CD4046 PLL IC, using just the oscillator, is another option but it will need some amplification to be able to drive the gate of the power mosfet. If L.W. already has the 555 devices it would certainly be easy to try.
I don't really want to have to get more components, as the shipping can get pretty expensive.
 

MisterBill2

Joined Jan 23, 2018
18,167
What happened ??? I wrote a post and it did not appear.
My comment was that the 555 today is able to run a lot faster than in 1979. So a few hundred kilohurts should not be that painful. (really bad joke there) And still, the reduced duty cycle should reduce the mosfet heating a lot.
 

Wendy

Joined Mar 24, 2008
23,415
May I draw you a concept circuit? First of all a stock 555 can not go to 1 Mhz, You will need something like a TLC555. This is TI CMOS 555 rated to 2 MHz.I would love to help But I do have to be asked first since I will be drawing.

Think a very small capacitor (say 10pf) and design from there.
 

MisterBill2

Joined Jan 23, 2018
18,167
May I draw you a concept circuit? First of all a stock 555 can not go to 1 Mhz, You will need something like a TLC555. This is TI CMOS 555 rated to 2 MHz.I would love to help But I do have to be asked first since I will be drawing.

Think a very small capacitor (say 10pf) and design from there.
Please do draw the circuit that I described. My whole purpose of the description was so that it could be drawn. Presently I do not have any means to draw circuits on this computer, and lightning damaged the computer with the cad program on it. So I am left with doing word descriptions. Thanks for offering to draw it, Wendy.
I doubt that the project needs to go to that since the transformer is an iron core type. Using a 555 will allow reducing the on time so that the current can switch off just at magnetic saturation. I think that is the source of the transistor heating, since the current rises after saturation.
 

Wendy

Joined Mar 24, 2008
23,415
The problem with a standard MOSFET is they require 10V on the gate to turn on fully At 6 V you will need a logic level MOSFET. Have you ever read my intro, I offer a set of templates for M$ Paint I call Paint CAD?
 

Bordodynov

Joined May 20, 2015
3,177
-live wire-
Give me detailed data on the transformer (geometrical dimensions, core material and number of turns) if of course you know them, and I will calculate the timer circuit using LTspice. LTspice allows you to calculate the output voltage and power that is dissipated in the transistor. I also need to know the size of the gap of the discharge gap. This is necessary to calculate the breakdown voltage.
 
Top