555 Motor Speed Controller

[mike_mike]

Hello, I just built the schematic from the following link on a breadboard and I am getting the following screenshots on the oscilloscope.

This is the link: https://www.multisim.com/content/jcWPXBhUCutR4dmEitXfLg/high-frequency-pwm-with-ne555/open/

The screenshots from the oscilloscope does not look like the screenshots from the simulation, and I don't know why.

I want to use this circuit to control the speed of my Proxxon mini drill. The power source is a 12V/5A SMPS.

Can this problem appear because I build the circuit on a breadboard instead of building the circuit on a PCB ?
What can I do to solve this problem ?

[Zero999]

The oscillograms look pretty good to me. Where is the proble? Presumably the output? What are your concerns?

Obviously the 555 will not be able to directly drive the motor directly. A transistor and freewheeling diode will be required.

[mike_mike]

The probe is connected to pin 3 and the alligator clip to the gnd.
I though that the output on the oscilloscope should look like the output in the simulation. I built some time ago another 555 pwm controller, which worked at lower frequency, and the output looked like in the simulation, and I though that this should work the same.

I know that the 555 will not be able to directly drive the motor. I will use a mosfet (n-channel). Should I use a gate resistor, for example 100R ?
I will also add a freewheeling diode.

[Zero999]

Sorry, I can't see much difference. The oscilloscope waveform looks almost identical to the simulation to me, barring a little noise and over/undershoot, which is to be expected.

Yes, a MOSFET and 100R gate resistor will be fine.

[Chriss]

I'm also familiar with Zero999. The scoped signals are pretty good.
Put the 100R to the gate.
When you put the mosfet under load then measure again the driving signal on the 555 pin3 check also the drain signal with scope. Just to be sure does the mosfet is completely closing...

Sent from my GT-I8260 using Tapatalk

[mike_mike]

I built the circuit on a PCB and the results look like the attached screenshots:
1. On pin 3 of 555 (I used 555P from TI) DS0655.jpg
2. On the 100R resistor DS0656.jpg
3. On the G of mosfet (I used IRF630) DS0657.jpg
4. On the output (I used a 27R resistor as load) DS0658.jpg and DS0659.jpg

Are those wave forms normal ?

Later Edit: I uploaded the schematic.

[strawberry]

My design version.

[mike_mike]

I also built on a bread board a PWM controller using Arduino, an IRF630 and PWM.h library.
The problem is that I saw some oscillation (maybe ringing ?) of the mosfet when the probe of the scope was connected to the load resistor (27R resistor, power supply voltage = 12V).
The gate resistor was 100R with a pull down resistor of 10K.
I tried to use higher values for the gate resistor (220R and 510R) but the square wave began to show like a distorted wave.
The last screenshot is the signal in the Gate of the MOSFET, when using 100R gate resistor.

Are the results good or should I try to reduce the oscillation (ringing) ?

[soldar]

I think we are making this more complicated than it needs to be.

- Get the 555 working and check with the scope the output is correct.
- Connect the switching transistor to the output of 555 and connect a light load and check with the scope the PWM load switching is correct
- Connect the motor as the load.

Each step will confirm what is right up to that point.
If the next step creates problems it is relatively easy to deduce what is the problem.
The two first steps should be no problem.
Then the motor could create some problems depending on inductance, transients, power supply filtering, etc. In the worst case you could install a filter as a buck SMPS.

[mike_mike]

I tested again the circuit with 555. I used an IRF630, 100R gate resistor, 1N4007 protection diode.
The results are as follows:
1. With the probe on load (I used as load a 27R resistor, the supply voltage was 12V).
0665, 0668, 0669.jpg

2. On the Gate of IRF630:
0666, 0670.jpg

3. On pin 3 of NE555P:
0667, 0671.jpg

4. Using another 555 IC, with pin 3 disconnected from the rest of the circuit:
0672.jpg

I attached the schematic.

Please have a look at the screenshots and tell me what you think.

[Bulb]

Hello!
If I may, I would advice you to put two transistor for driving MOSFET in totem pole configuration. For example https://electronics.stackexchange.com/questions/299552/what-is-the-purpose-of-this-pull-up-resistor-in-totem-pole
Also, I would recommended of using UF4007 diode instead of 1N4007 for flayback diode. Also with totem pole configuration you can try out to reduce gate resistance and observer output wave forms of voltage across MOSFET. Decreasing reistance you speed up switching and reduce switching losses.

Edit:
If i see good (from scope wave forms), your switching frequency is 160kHz. You should reduce it to a round of 5kHz to 20kHz and all should work just fine. Then, totem pole isnt needed.
Buf if you are intresed in reasherce you can build one totem pole driver just for analayzing. Also, here are switching losses small, but if you use for example higher voltage sorce (for example 100V), switching losses also increase.
Hope it helped.

[mike_mike]

I tried the attached schematic, based on the information from: https://www.quora.com/What-is-the-purpose-of-a-MOSFET-gate-driver
The results are as following:
1. On the Base of the 2 transistors (PNP and NPN): 0673.jpg
2. On the Gate of the Mosfet: 0675.jpg
3. On the RLoad: 0676.jpg

The power supply of the entire circuit was 12Vdc.

1. Please have a look at the attached screen shots and tell me what you think.
2. Can you please explain why the Vpp is 35.6V in the first screenshot, why it is 34.4V in the second and why it is 24V in the third while the supply voltage is only 12V in all 3 cases ?

[Bulb]

I tried the attached schematic, based on the information from: https://www.quora.com/What-is-the-purpose-of-a-MOSFET-gate-driver
The results are as following:
1. On the Base of the 2 transistors (PNP and NPN): 0673.jpg
2. On the Gate of the Mosfet: 0675.jpg
3. On the RLoad: 0676.jpg

The power supply of the entire circuit was 12Vdc.

1. Please have a look at the attached screen shots and tell me what you think.
2. Can you please explain why the Vpp is 35.6V in the first screenshot, why it is 34.4V in the second and why it is 24V in the third while the supply voltage is only 12V ?

This is good example when using high switching frequency, good MOSFET driver is need. This wave forms are now 
I recommend  of reduceing frequency around 20kHz. You can keep it higher than 20kHz to reduce sound noise of PWM.
To sum, this wave forms now look good.

Edit:
High voltage spikes are due parasitic inductens of traces. To reduce it, you can increas a gate resitance. At 10 ohm resistance, mosfet is fast switching and voltage across indactens is raisng very fast, U_inductor = L*di/dt. Adding a capacitor close in DC bus, neer the load can also help to remove parasitic inductance. But tehnici you cant eliminate parasitic inductance.

[soldar]

Yeah, 160 KHZ seems unnecessarily high. Probably 40 KHz would work better in reducing switching and other losses.

[Bulb]

One more thing I forgot... Peak current through BC338 and complement transistor is around peak apsolte curret for that transistor. Beacuse, at 10 Ohm resistance at gate, peak current is around 1A. So, for selected transistor in totem pole configuration I advice of puting gate resistance of around 20Ohms (for example series combination of two 10 ohms resistors).

[mike_mike]

Can you please explain why on the oscilloscope, the Vpp is about 35V, while the supply voltage of the circuit is about 12V (please have a look at the latest screenshots) ?

[Bulb]

Can you please explain why on the oscilloscope, the Vpp is about 35V, while the supply voltage of the circuit is about 12V (please have a look at the latest screenshots) ?

With 10 ohm resistor in gate path, switching speed of MoSFET is very high (under 1 microsecond). What that means? That means that mosfet will shut down under 1uS and current through load should stop. With RL load that isnt possible because L. This why is parallel connected diode with RL load. When mosfet dont conduct, energy stored in the inductor is transfered via diode back to load. Beacuse diode cant start conduct imidiatly there is voltage spike on load. Decreasing switching speed by increasing value of mosfet gate resistor, you will reduce it. To sum, spike, or Vpp is ove 12V (power supply voltage) because high switching speed, inductive load and nonideal diode.
This why (beacuase L) there is undershoot (negative spike). So oscilloscope make difference between max and min peak voltgae value and display it.

[mike_mike]

I used the attached schematic and I tried to reduce the frequency by increasing the 100pF capacitor to 470pF.
I made the tests with R7 and without R7. I used as load a 27R/5W resistor.
The results are:
1. With R7=1k and C=470pF
a. pin 3 of 555: 0688.jpg
b. E of the PNP and NPN transistors: 0689.jpg
c. G of mosfet: 0690.jpg
d. on Rload: 0691.jpg

a,b,c were measured with the crocodile clip on GND.

2. Without R7 and C=470pF
a. pin 3 of 555: 0692.jpg
b. E of the PNP and NPN transistors: 0693.jpg
c. G of mosfet: 0694.jpg
d. on Rload: 0695.jpg

a,b,c were measured with the crocodile clip on GND.

1. Please have a look at the screenshots and tell me what you think.
2. Is R7 really necessary ? Does 555 have pull down resistor ?

[Bulb]

Waveforms looks good. I would recommend to put a snubber in parallel with load and use of UF4007 diode instead of 1N4007. Reason for it is voltage spikes on load and MOSFET. This voltage spikes are high (double of power supply voltage) and can hurm MOSFET if higher currents and voltages are used. With snubber you can reduce voltage spikes.
What is snubber and how to dimension it is discraibed in this pdf https://www.illinoiscapacitor.com/pdf/Papers/RC_snubber.pdf
Snubber is RC circuit connected in paralled with inductive load and purpose of it is to reduce voltage spikes caused by fast switching of mosfet (or any other switch).

No, R7 is not needed. Output of NE555 is also totem pole configuration (there are two transistors, one switch to positive power supply input and other to negative power supply side (gnd)) but it can deliver max +/- 0.25A. With external totem pole you can achive higher currents (like +/-0.5A) that are useful for fast switching of MOSFET.

Nice work 

[mike_mike]

The snubber is needed even if the max load current is 5A and max supply voltage is around 12V ? Or the snubber is needed when the voltages are bigger than 12V ? I'm sorry for asking again but I did not understood clearly from the initial message.
In the last screenshots, the frequency was reduced to about 50KHz, from about 160KHz.

[Bulb]

I recommend of using snubber now, for this situation 5A and 12V.

If is possible, can you measure voltage waveform across MOSFET drain source electrodes?

Do you plan of using this type of load for always? In your first post you mention proxxon drill. Motor of drill has its own inductance, is it same as one used in test (measurments)?

Probably this will work without snubber because IRF630 can work up to 200V, but to be safe, I recommend of using snubber.

For the first test you can use snubber with R around 10 ohms / 2W and capacitor of 4.7uF (working voltage over 16V, around 25V).
RC snubber connect parallel to (with) flyback diode.

[mike_mike]

If is possible, can you measure voltage waveform across MOSFET drain source electrodes?

Do you plan of using this type of load for always? In your first post you mention proxxon drill. Motor of drill has its own inductance, is it same as one used in test (measurments)?

I will measure tomorrow (today it is too late) the voltage waveform across MOSFET D-S.

I will use only PCB drills with this PWM controller. I will start by using the proxxon, but I am planning to use some other drills. I do not know the inductance of the motor, and it is not with the one used in measurements.

The 4.7uF capacitor needs to be upolarised or needs to be polarised ?

[Bulb]

Capacitor need to be nonpolarised, for example ceramic capacitor https://en.m.wikipedia.org/wiki/Ceramic_capacitor

Ok, I think that RC snubber with asumed values can help, because snubber will conduct until diode turns on.

[mike_mike]

I searched for 4.7uF unpolarised capacitor at the local electronics shop and in my boxes, but I did not find any 4.7uF capacitor.
I cannot order only a few 4.7uF capacitors from the internet...
I there any other solution ?
I will also try a simulation in LTSpice.

[Bulb]

You can also try with 1uF, 3.3uF, 2.2uF or 6.8uF. Any of these values should help. You can also make a parallel of 2.2uF and get about 4.4uF.
The use of the RC snubber in this particular case is to reduce the surges (voltage spikes) and prevent the possible destruction of the MOSFET.