PWM with 555 help

[J_Nichols]

can anyone post a PWM circuit based on the 555?

I have found thi one:
https://www.rmcybernetics.com/projects/DIY_Devices/homemade_signal_generator2.htm

1) Can I use 12 VDC in the circuit posted in the link?
2) In the right end of the circuit I see 3 wires: the output (middle), the ground (low part) and other in the upper part that I don't know where to connect it.

 

[DerStrom8]

1) Yes, you can use a 12 volt supply for this circuit instead of 9 volts
2) The upper wire is still the +9v rail. It is already connected to your V+

Hope this helps
Der Strom

 

[jpanhalt]

I am a little surprised pin4 is not connected to the power supply. Here is another circuit I have had experience with:

It is from: **broken link removed** The configuration with diodes allows the duty cycle to be varied over a wide range.

John

 

[DerStrom8]

I am a little surprised pin4 is not connected to the power supply.

Well if you take a closer look, pin 3 (the output) of the 555 isn't connected to anything, either. The 555 in the RMCybernetics circuit is used much differently than your standard astable timer. I would guess that that's why pin 4 isn't connected to the power supply.

 

[jpanhalt]

I understand that. Not connecting pin 3 should not affect anything adversely; not connecting pin 4 might. From the LM555 datasheet:

When the reset function is not in use, it is recommended that it be connected to VCC to avoid any possibility of false triggering.

John

 

[shortbus=]

I always though that when you used a 555 as shown in John's post for PWM, that when you varied the pulse width the frequency also varied. I f that doesn't matter to the OP all is good. If it does, you have to add a 339 to the circuit to keep frequency stable and just vary pulse width.

 

[DerStrom8]

I always though that when you used a 555 as shown in John's post for PWM, that when you varied the pulse width the frequency also varied. I f that doesn't matter to the OP all is good. If it does, you have to add a 339 to the circuit to keep frequency stable and just vary pulse width.

When you have the 555 set up with the 5818 diodes there, it allows you to adjust the duty cycle fairly independently of the frequency output. However, the 393 and 555 combination circuit is much better at keeping the frequency and duty cycle adjustments independent.

 

[jpanhalt]

However, the 393 and 555 combination circuit is much better at keeping the frequency and duty cycle adjustments independent.

At the sacrifice of using 2 IC's. Other circuits using the 555 to produce a sawtooth wave do connect pin 4 to Vcc for the reason stated. However, if the OP is not wedded to using a 555, a single quad comparator can be used to produce both the sawtooth or triangle wave and the PWM. Here is one way to make the sawtooth: https://www.falstad.com/circuit/e-sawtooth.html. Similarly, a triangle wave can also be used: https://www.falstad.com/circuit/e-triangle.html

John

 

[Reloadron]

I agree with John. The original circuit should work but pin 4 must be tied to Vcc. However, a better way to go about creating a triangular wave is using the latter circuits. The end result is feeding a triangular waveform into a comparator to create a PWM output. The more nice and linear the leading and trailing transitional slopes I would think the better the PWM signal will be.

Just My Take
Ron

 

[DerStrom8]

Yes, you're probably right. It would be a better idea to connect pin 4 to +V anyway, even though it's not shown in the schematic. The triggering system is still connected the same way.

 

[4pyros]

There is another circuit like it on the same site without 4 connected;
https://www.rmcybernetics.com/projects/DIY_Devices/homemade_pulse_controller.htm
Maybe you can use just the comparators without Vcc.

 

[J_Nichols]

thank you very much to all
I want to be able to select frequency and duty cycle independently. Yesterday I started to build the circuit but it was the first time I soldered something, so.... imagine the result.
Now, I've learnt how to solder and I'll post here is the original circuit works or not. Of course, I'll connect the pin 4 to Vcc.

 

[4pyros]

Of course, I'll connect the pin 4 to Vcc.

You should try it without pin 4 connected.

 

[jpanhalt]

You should try it without pin 4 connected.

Why? What would conceivable advantage would that have?

It will probably work without pin4 attached, but will be susceptible to false triggering. Is the user prepared to go through the testing to show that is not likely to happen?

John

 

[J_Nichols]

Ok guys,
I'm finishing the circuit, but I have a question. The pulses that the circuit generates I want to send them to an electromagnet. So the circuit need a transistor and a diode to protect the circuit from the back-emf. Right?
The energy from the collapsing magnetic field goes to the bulb.

The things I don't know are:
1) What kind of transistor I have to use?
2) Do I need a resistante to protect the transistor?
3) What kind of diode I have to use? (between the light bulb and electromagnet)
4) Do I need to use more diodes to protect the circuit from the back emf?

The power supply is 12 VDC 1 Amp.

 

[Reloadron]

The transistor would be determined by the current of the coil it is driving.

The transistor would have a base resistor and like the diode the value of the base resistor value is determined by the coil current the transistor is conducting. If the coil requires considerable current a MOSFET type or Darlington type transistor may be a better choice. This all goes back to coil current and the frequency the coil is being pulsed at.

Any standard diode capable of handling the current. In most cases the lamp isn't necessary. See Back EMF in this link. Since you mention the power supply is 12 volts with a maximum of 1 amp I wouldn't worry about the lamp and would use a diode like a 1N4002.

Ron

 

[J_Nichols]

The power supply is 12VDC 1Amp as you know, and the electromagnet is rated at 12VDC 0.17 Amps.
I'm going to change the pulsating DC frequency going to the electromagnet and also the duty cycle. I'll work with low frequencies up to some Kilohertz, and from 1% to 99% duty cycle. So, the back emf will be different in each situation I think. I've read the document you mention and I've add to my circuit a diode to block the back-emf that goes to the circuit while allowing the back emf going to the light bulb.

About the selection of the diode and transistor I think I need to use one rated at high current since if I'm going to use different frequencies, the inductive kick back could be high in some cases?

I've also a doubt about the VR2. What is the variable and fixed resistance connections? I think the variable resistance is the connection coming from the R4 and ending in - of the IC2. Right?

 

[Reloadron]

You only need a single diode across the coil. There is no need for the additional diode and lamp. Your coil current at .17 amp is really a low current. The frequency also is not that high. So you need a general purpose NPN type switching transistor that will handle the current and frequency. An everyday off the shelf 2N2222 transistor will be more than adequate. This is the data sheet for the transistor. The 2N2222 should do just fine. When power is removed from the coil the current of the back EMF can't exceed the current of the coil. There is a voltage spike. I would use a diode as I mentioned. I would use a general purpose 1N4xxx family diode and the 1N4002 should do fine. The transistor and diode are very common and can be had just about anywhere.

Ron

 

[J_Nichols]

Thank you very much for the aclaration about the diode and the transistor Ron.
I need the bulb since I want to light it using back emf energy only. So, reading your comment I think the first circuit (the first that incorporates the bulb) will work fine. Right?
I used a second diode because I thought that I need to put one to protect the circuit and use other to light the bulb using back emf only.

What about the VR2 potentiometer? In the circuit I cannot see clearly what is the regulable connections and the fixed resistance connection.

 

[audioguru]

A "light bulb" might be incandescent that needs time to heat-up and produce light. The back-EMF might be gone by the time the light bulb begins to glow. Try a neon bulb instead (but then the transistor must be a high voltage one).

R3 provides the base current and turns on the transistor when the output of the low-power comparator goes open-circuit. But an LM393 dual comparator has a minimum output current of only 6mA so the base current of the transistor will be 6mA max then its load current should not be more than only 60mA (not enough to drive the 170mA electromagnet).