Circuit Analysis of IR transmitter/receiver

[enJayneering]

Hi everyone,

I am a complete beginner when it comes to circuit analysis that involves anything more than a few resistors and capacitors. Currently, I am working on a project that requires the transmission and reception of IR signals to control a motor. I know the requirements for the motor I must use and the amount of voltage I am able to work with. My question is, will the voltage have any effect on the frequency output of the 555 timers? In addition, I'm having trouble selecting the correct transistor to provide the right amount of power to the DC motor. Any help is much appreciated!

Here is the picture for the schematic that I am working off of:

I am fairly comfortable with how the 555 time operates in astable and monostable configurations. I'm just having difficulty calculating numbers.

 

[davenn]

My question is, will the voltage have any effect on the frequency output of the 555 timers?

it shouldn't do so unless the PSU cannot supply the voltage and more specifically the required current so that the Vcc rail voltage doesn't sag when the motor is running under load ... I would also be adding some bypass capacitors on that Vcc rail so that the 555 doesn't get spike noise from the motor

I'm having trouble selecting the correct transistor to provide the right amount of power to the DC motor. Any help is much appreciated!

it needs to be able to handle the current that the motor you choose requires ... you have given no info on thatDave

 

[Baluncore]

Try to avoid controlling the Vcc power terminal of an LM555. One improvement to the transmitter circuit would be to gate the second LM555 with it's RST input rather than the Vcc power supply terminal. When an LM555 switches it takes a short but high current spike from the supply and so needs a bypass capacitor on Vcc. That upsets the internal voltage divider and comparators. If the Vcc is hanging off another LM555 then you cannot use a bypass capacitor on Vcc of the second LM555.

You might also use the open collector Discharge output from the first LM555 with a pull-up resistor to Reset on the second LM555. Those changes will give better regulation of output frequency.

 

[enJayneering]

Thanks for the suggestions davenn. Is there a way to calculate/estimate the amount of noise produced by the motor? I'm trying to figure out what value of capacitor to use.

it needs to be able to handle the current that the motor you choose requires ... you have given no info on that

Here are the requirements of my motor:
Torque: 0.1 Nm
Total Power: 11 Watts
Maximum Current: 2.44 A

Also, I'm working with a 4.5 V power source and will use a potentiometer to control the frequency/pwm.

@Baluncore
I'll look into those other methods as well. I'm still getting use to 555 timers and the different modes of operation that they have. Thanks for the suggestions!

 

[enJayneering]

Hi All,

I've done some research into bypass capacitors and I'm finally starting to grasp its purpose and implementation within a circuit. However, I am having trouble testing my circuit and seeing where it can be improved. I don't currently own an oscilloscope so I am wondering if there are others ways of signal analysis. Does anyone know of online simulators or alternative methods that could help with the design?

Thanks

 

[Baluncore]

You could download a free copy of LTspice from; http://www.linear.com/designtools/software/

Here is an LTspice model of an example IR transmitter with the example output shown.
I left out the bypass capacitors but changed the gate method.

All the attached files are text files inside, but need a txt extension to attach to this post.
You will need to remove the .txt extensions from the attached files to use them.
The .asc and .plt files go in your working directory.
The .sub and .sym files for an idealised NE555 need to go in the installed library directories.

 

[enJayneering]

Wow, I can't explain how much this helps out! Being able to see the pwm signals being modulated is extremely informative. If I want to experiment with different bypass capacitors would I get rid of the gate and place in varying capacitors. I ask because currently I do not see any noise.

Also, to clarify a few things in this circuit:
1) Do the two resistors in parallel act as a potentiometer?
2) Am I correct in saying that this type of modulation would be amplitude modulation?

 

[Electron Spin]

Also, to clarify a few things in this circuit:
1) Do the two resistors in parallel act as a potentiometer?
2) Am I correct in saying that this type of modulation would be amplitude modulation?

1) Typically no they do not..two r's in // act as current dividers. Unless you picked the V from the middle node of the 2 resistors giving you a single voltage somewhere between VCC and grd.

2) To obtain a true AM wave output, generally you would excite 2 xtls. your schematic as shown would output two positive going AM'ed alterations, I believe.
Kinda' like a full wave bridge with intelligence 'inside' the DC waveform.

Below, would be one example.

Again as my colleagues suggested, a oscilloscope would be invaluable to see this outputted waveform.

 

[Baluncore]

If I want to experiment with different bypass capacitors would I get rid of the gate and place in varying capacitors. I ask because currently I do not see any noise.

If you avoid turning the power on and off to the chips it will not take much bypass capacitance to keep the circuit happy. I would put an 0.1uF ceramic capacitor as close as possible to the end of each NE555 chip, between pins 1 and 8.
To simulate supply noise I would add some resistance and inductance to the Vcc1 voltage source. There are no big circuit currents with the gating circuit I used in the example. The NE555 model I used is idealised, so it is simple and computes fast. I would test the real circuit rather than load a more realistic model of the NE555.

1) Do the two resistors in parallel act as a potentiometer?

R4A and R4B make the 10k pot. The “.param pot 50” is the 50% position. Edit that “50” to test other duty cycles. The two {equations} are the resistance values that make the two sides of the potentiometer.
They are drawn next to each other since they are used as variable resistors in this circuit. The diodes make sure that current only flows through one side of the pot at the time. If it was a voltage divider I would have drawn them end-on in series,, more like a typical potentiometer.

2) Am I correct in saying that this type of modulation would be amplitude modulation?

It is probably best not to think of it as AM since when it is there it is a fixed amplitude carrier. It is a version of pulse-width or duty-cycle modulated continuous wave. MCW.

Have you got LTspice installed and running yet? If you have problems, ask in this thread.
For the best help, post your filename.asc by appending .txt, to filename.asc.txt then others can run and test your circuit.

 

[Electron Spin]

It is probably best not to think of it as AM since when it is there it is a fixed amplitude carrier. It is a version of pulse-width or duty-cycle modulated continuous wave. MCW.

Excellent point indeed. I was thinking what you verbalised Balun, as in CW in HF communications. I'm a old bag of bones and still pound away with a bug or straight key! Thanks for pulling me back on track!

ES

 

[enJayneering]

Thanks guys. I really appreciate the help!

I did manage to get LT spice running and simulating the circuit you sent. Being able to probe each section of the circuit has allowed me to see the different square waves that are generated. It has helped me greatly to visualize what is going on. Now I just need to look into the MCW that you mentioned.

Also, I'll look into seeing if I can pick up an oscilloscope for cheap. Maybe my University has an old one they don't need.

If I decide to make a more realistic simulation with noise, I'll update this thread to see what you guys think.

There is only one more question before I start ordering parts and putting everything together. What is the best way to estimate noise and thus calculate a bypass capacitor for my DC motor?

 

[Baluncore]

There is only one more question before I start ordering parts and putting everything together. What is the best way to estimate noise and thus calculate a bypass capacitor for my DC motor?

That will depend on the type and rating of the DC motor. Maybe you do not need a capacitor.

But you MUST use a reverse biassed power diode across the motor, to catch the expected positive inductive voltage spike when the transistor turns the motor current off. Without that flyback diode the TIP122 will be destroyed by the excessive positive collector voltage. It would be best to use a fast recovery diode rated at the motor current, similar to that used as a rectifier in a switching power supply.

Any bypass capacitor should go across the transistor and motor, that is between the motor positive = Vcc, and ground.

 

[enJayneering]

A quick update of progress

First, I just wanted to say that I appreciate all the advice I've received. You've given me plenty to research and learn.

1)
So I decided to visually wire up the schematic on a breadboard so that wiring up a prototype is easier.
Here's what I have so far:
https://circuits.io/circuits/3383536-ir-transmitter

One question that I have is in the wiring of the potentiometer. I'm pretty sure the wiper should be in series with the 1k resistor but I don't know if the other two pins are connected correctly. The led does light up when I run the simulation, but I just want to double check my work.

2)
For the lt spice simulation I'm getting interesting values for the frequencies. After probing and measuring, I am getting a value of 3.8 kHz for the output square wave. Is there a correct way to go about measuring the frequency of pwm in lt spice? Thanks

 

[Baluncore]

One question that I have is in the wiring of the potentiometer.

The pot, 1k0 and diodes on the breadboard look OK to me.

Is there a correct way to go about measuring the frequency of pwm in lt spice?

The first cycle of a burst produced by an NE555 often has a stretched first cycle because the capacitor must be charged from zero to 2/3Vcc, rather than 1/3Vcc to 2/3Vcc. So avoid the first cycle, measure the period of a later cycle.
First zoom in on a couple of later cycles of the signal. You could do that with the mouse on the plot window, or by specifying the transient analysis simulation time and the time to start recording data.
Select two cursors for the trace. Place cursor 1 on a rising edge, then cursor two on the next rising edge. Read the Freq. If rising edges are irregular, try falling edges.
Or measure over ten cycles to average the period, remember to multiply the frequency, or divide the period, by ten.

I would use an opto-isolator to simulate the IR LED to phototransistor link. But I am a bit confused by your original circuit for the IR receiver, because the pin numbers are not clearly readable. I suspect it would be best to use a re-triggerable monostable configuration.
Can you please attach a clearer circuit of the proposed IR receiver to your next post.

 

[enJayneering]

Select two cursors for the trace. Place cursor 1 on a rising edge, then cursor two on the next rising edge. Read the Freq. If rising edges are irregular, try falling edges.

Perhaps I am not using the correct method or not placing the cursors at the right location. However, my values always end up being around 3.8 kHz. The readings are even less if I try measuring more cycles.

I suspect it would be best to use a re-triggerable monostable configuration.
Can you please attach a clearer circuit of the proposed IR receiver to your next post.

As you mentioned, the 555 timer is configured in the monostable mode. The purpose is to demodulate the input signal to it's original pwm duty cycle wave form. From what I understand the low pulses generated by the IR sensor are the trigger for the 555 timer. The output is configured to be slower than the input pulses thus changing the 38 kHz frequency back to original pwm.

How would an opto-isolator work in place of this? Thanks!

 

[Baluncore]

Perhaps I am not using the correct method or not placing the cursors at the right location. However, my values always end up being around 3.8 kHz.

I am not sure why you have trouble, but 3.75kHz is about right. Try displaying a shorter period of time by editing the .TRAN script to show only a very short time with two positive edges.

The output is configured to be slower than the input pulses thus changing the 38 kHz frequency back to original pwm.

Yes, changes it back to the PWM envelope of the signal. But your monostable is not retriggerable.
Attached is a “retriggerable monostable” version using the NE555. It fills in the gaps between the gated carrier pulses and so prevents gaps of drive to the motor. I think you will see those gaps if you modify my model back to your circuit by removing the PNP transistor.

How would an opto-isolator work in place of this? Thanks!

The optocoupler is a simple model of the IR path, but has very low losses compared to real paths. The real path loss suffered by the IR signal may result in a need for amplification. The usual reason that an IR signal uses a PWM string of regular pulses is so it can be extracted from the background IR environment by using a tuned amplifier with automatic gain control.

I'm not sure if you have the darlington transistor models. I would use an Nchannel MOSFET to drive the motor rather than a darlington. You can change the darlington to a "logic level" Nchan MOSFET from the LTspice devices list, or let me know if you need me to give you a set of standard darlington models.

Remove D3 from my model to demonstrate the inductive voltage spike that, on the first turn-off edge without a flyback diode, would destroy the U4 motor switch.

 

[enJayneering]

I am not sure why you have trouble, but 3.75kHz is about right.

Okay, that's my fault. My intention is to have an output frequency of about 38 kHz so it can be read by my IR sensor. I'll have to change the capacitor and resistor values to obtain that. I assume that the second 555 timer is being run in an astable mode, correct?

Is there a certain method for choosing a resistor/capacitor combination that's most efficient? Right now I am using an online calculator to determine the values that output 38 kHz (I also have the written equations for this). Should I aim to make the capacitance as high as possible while keeping the resistances low?

It fills in the gaps between the gated carrier pulses and so prevents gaps of drive to the motor.

So this means that the circuit will still work in a non-retriggerable mode, but the pwm pulse will not be continuous and thus the duty cycle will be less than anticipated?

The optocoupler is a simple model of the IR path, but has very low losses compared to real paths.

So this is just a way to simulate the transfer of IR signals digitally and not a physical component?

Sorry for the many questions, but I want to make sure I know how everything works and how to optimize it. You've been a great help!

Also, here's an update of progress. I managed to breadboard the circuit and gain access to an oscilloscope.
Here's some results:

After some calculating the cycles vs time period the frequency was found to be about 3.3 kHz. It does make sense that the frrequency is not 3.75 kHz as shown in the simulation because I had to use different resistances (R1= 330 ohms & R2=2.2 kilo ohms). However, it also deviates from the theoretical frequency of an astable timer. The calculated output frequency should be 3.05 kHz. Is there any explanations for this difference?

 

[Baluncore]

I assume that the second 555 timer is being run in an astable mode, correct?

Yes, it produces a rectangular wave when not reset by the PWM signal.

Should I aim to make the capacitance as high as possible while keeping the resistances low?

No. Make the resistor big, but not bigger than 100k when surface leakage currents become a problem in dirty environments. Smaller capacitors need less charge so use less power per cycle. The currents being switched are also smaller so make less RFI.

Indeed, the position of the decimal point is important. That is why in electronics we write 2k7 rather than 2.7k. The frequency you want is 38k0 Hz not 3k80 Hz. To change from 3.78kHz to 37.8kHz, try simply changing Ct from 100nF to 10nF. Test it, then each time you increase resistors by a factor of 10, reduce Ct by a factor of ten. Avoid Ct less than about 220pF as capacitance of the prototype breadboard then becomes a problem.

So this means that the circuit will still work in a non-retriggerable mode, but the pwm pulse will not be continuous and thus the duty cycle will be less than anticipated?

The NE555 is usually non-retriggerable so there will be small gaps after recovery while waiting for the next trigger. The added PNP prevents all that unnecessary switching of motor currents. It also permits shorter monostable times and so follows the PWM tail closer.

So this is just a way to simulate the transfer of IR signals digitally and not a physical component?

Yes, I needed to get the simulation signal from the TX model to the RX model. The common 38kHz IR receiver modules handle the IR receiver design problems for you.

I have not tried to explain the simulation versus reality difference.

I have changed the motor drive to a logic level MOSFET because it is more efficient than the darlington which wastes 30%, 1.5 volts of the 5V available. I used the IRFH6200 because it is in the standard LTspice library, I would use something cheaper in reality.

 

[enJayneering]

To change from 3.78kHz to 37.8kHz, try simply changing Ct from 100nF to 10nF

This is actually one of the things I tried while I was in the lab. I forgot to take a picture but I measured cycles that ended up giving me around 45k0 Hz. That was using a 10nF capacitor and 330 ohms for R1 and 2.2 kilo ohms for R2. Based on my calculations I should be getting around 30 kHz instead. I know there is some error in reading exact values from the oscilloscope, but is there another reason that could cause this discrepancy?

I used the IRFH6200 because it is in the standard LTspice library, I would use something cheaper in reality.

Would you mine sending me the model for this exact part? I would like to make sure that I'm following along with your data.

PS I have finals next week so my response times may be a lot slower.

 

[Baluncore]

Would you mine sending me the model for this exact part?

It should be in the LTspice library. Specify nmos device, then right click and Pick New Mosfet.
The Nchan MOSFET in my circuit has a 10R series gate resistor to prevent parasitic oscillation.

I trimmed the carrier frequency closer to 38 kHz. Use .TRAN 0 12m3 12m to see ten cycles for frequency measurement.
I removed pullup resistor from DIS of U1 and now take gate signal from OUT of U1, to RST of U2.

Maybe you do not need a monostable with the IR receiver module. What model module ?

 

[enJayneering]

Thanks for the explanations of each component

Maybe you do not need a monostable with the IR receiver module.

Are you suggesting that I might not need a 555 timer?

What model module ?

When I opened the schematics you sent before it said that I didn't have the darlington transistor. However, I have all the components for the updated schematic you sent. You mentioned before that the darlington wasted 30% voltage. Could you explain to me why this is the case? Also, how would you go about determining which mosfet would be a good fit as a cheap alternative?

While I was in the lab, I noticed that I wasn't seeing much noise from the frequency generator. Are there setting that I must change to see these noise spikes?

Thanks!

 

[Baluncore]

Are you suggesting that I might not need a 555 timer?

Only the receiver re-triggerable monostable. Demodulation is usually done by the $3 IR receiver module. As an example, take a look at the top of page 3 of this .pdf;
http://www.farnell.com/datasheets/2049436.pdf

You mentioned before that the darlington wasted 30% voltage. Could you explain to me why this is the case?

A MOSFET can now have a very low resistance when switched on, often 0.1R or less. That will drop very little voltage so there is more for the motor. This is especially true for low supply voltages. A darlington pair must drop at least one Vbe, base emitter voltage, say about 0.8V when on, so only 4.2V of the 5V supply remain for the motor.
See; https://en.wikipedia.org/wiki/Darlington_transistor
And; https://en.wikipedia.org/wiki/Darlington_transistor#Disadvantages
The complementary darlington, or Sziklai pair, overcomes some of the loss by overlapping the Vbe, so only Vsat remains.
https://en.wikipedia.org/wiki/Sziklai_pair

Also, how would you go about determining which mosfet would be a good fit as a cheap alternative?

The specification of the motor switch MOSFET and flyback diode will depend on motor specification and selecting a MOSFET with Vgs(on), if possible, to work directly with the output voltage from the IR RX module.
Once you know your needs, the best way to select a MOSFET is with the selection filters available on some suppliers sites. Then sort by price.
Try this; http://au.element14.com/mosfet-transistors-lt-600v

While I was in the lab, I noticed that I wasn't seeing much noise from the frequency generator. Are there setting that I must change to see these noise spikes?

It is designed to be quiet, why do you want to generate noise? Simulation or real electrical noise? Remove supply bypass capacitors and introduce series resistance into the spice voltage sources to simulate supply noise, (right click the source). The NE555 model is idealistic and so probably not good enough to show the NE555 transient spike. Use an oscilloscope to see supply noise in the real circuit.

 

[enJayneering]

Demodulation is usually done by the $3 IR receiver module.

Okay, awesome. I didn't realize that was the case. I know that when the sensor receives a signal it produces a low output. Does this mean I will have to flip flop these pulses to turn the mosfet on?

A MOSFET can now have a very low resistance when switched on, often 0.1R or less.

After looking at some data sheets I can see exactly what you're talking about. Thanks for the heads up.

It is designed to be quiet, why do you want to generate noise? Simulation or real electrical noise?

I don't want to generate noise, but I do want to observe the effect of the bypass capacitors. When I was in the lab I did not see a difference in noise levels between having a bypass capacitor and not.

 

[Baluncore]

Does this mean I will have to flip flop these pulses to turn the mosfet on?

N-chan MOSFETS are cheaper than P-chan, so it may be an advantage to invert the signal and use a P-chan.
You can invert a signal and get a schmitt-trigger effect with an NE555, just put the signal onto pins 2 and 6, take the output from pin 3, tie RST high.

 

[enJayneering]

Hi, so I finally got a chance to continue to work on this project and I have a few questions moving forward

N-chan MOSFETS are cheaper than P-chan, so it may be an advantage to invert the signal and use a P-chan.
You can invert a signal and get a schmitt-trigger effect with an NE555, just put the signal onto pins 2 and 6, take the output from pin 3, tie RST high.

I want to try driving the motor with both a p-channel and n-channel MOSFET to see how both work. I understand the configuration for the n-channel but I am unsure of the p-channel
1. I noticed that p-channel MOSFETS require a negative voltage. When the IR sensor receives a signal and pulls down the voltage, is this considered providing a negative voltage?
2. Is there a way to determine how low a IR sensor will pull a voltage. On the data sheet the range is between -.3 and 5.5 V.

Thanks

 

[Baluncore]

I understand the configuration for the n-channel but I am unsure of the p-channel

Try turning the schematic upside down and reversing the supply polarity to flip between P and N channel models.

1. I noticed that p-channel MOSFETS require a negative voltage. When the IR sensor receives a signal and pulls down the voltage, is this considered providing a negative voltage?

Yes, but an NE555 used as a schmitt trigger inverts that to drive the P-channel MOSFET.
If you need a polarity inversion, consider reversing the position of the photo-diode and pull-up resistor so that the resistor becomes a pull-down resistor. That may upset the retriggerable monostable.

2. Is there a way to determine how low a IR sensor will pull a voltage. On the data sheet the range is between -.3 and 5.5 V.

When an IR sensor is illuminated it generates a current. That diode current, Id, is proportional to the amount of light captured. Id will drop a voltage across the series resistor, Vr = Id * R. In a pull-up resistor circuit, the diode voltage will then be Vs – Vr. You must select R based on your diode response and illumination. The maximum diode voltage will be Vs so you need to keep Vs below 5.5V to protect the diode.

 

[Baluncore]

If you use one of the Vishay TSOP34x38 series IR receiver modules, the output will go low when it receives illumination pulsed at 38kHz. The IR module active-low output could directly control a “high-side switch” = P-channel MOSFET on the same +5V supply, switching a load connected to Gnd. That MOSFET would need to be selected for gate threshold voltage to suit the logic level.

It is usual to run the IR RX module from the Gnd and +5V supply rail. Your motor will probably share the same ground rail potential, while running on a separate higher voltage supply. For an active-low IR module output;

To control the motor with a high-side, P-channel MOSFET, you need to lift the logic level control voltage to the +ve motor supply rail, but you do not need inversion. I would avoid that P-channel MOSFET solution.

To control the motor with a low-side, N-channel MOSFET, you would need an inverter between the IR module and the MOSFET gate. I would use an inverting NE555 as the gate driver of a low-side, N-channel MOSFET.

It would be good if you could provide a circuit schematic with your next question.

 

[enJayneering]

Hey, so I've finally gotten the opportunity to work on this project again. I have made a prototype of the circuit on a bread board and I am currently testing.

Based on some results from an oscilloscope, it seems that the transmitter is working. from the first 555 timer I am getting a square wave with a pwm that is controlled by the potentiometer. Furthermore, the second 555 timer is outputting a square wave at around 38 kHz which is perfect. However, after using a test led I found that the light is constant. The led does not pulse when the potentiometer is turned.

Should I be able to see this pulsation?

Thanks

 

[Baluncore]

Can you please remind us, which circuit you have built ? A post number or filename would help.
If you have a dual channel CRO you could look at, and synchronise to, the PWM envelope from the first NE555. Then with the second channel, look at the transmit LED. You should see many rectangular pulses filling the PWM envelope.

 

[enJayneering]

Thanks, I didn't think to try that. At this time, I have only probed the output of the 555 timers individually.

I am still working off the same schematics as before. Thus, the IR_TX_RX_8 model you sent is a good representation. I have some video of the circuit and the LED which I'll try to upload.

In the meantime, I'll try synchronizing the channels like you suggested and will post the results tomorrow

 

[enJayneering]

Alright, I finally got a chance to upload some videos of the circuit in action!

I can't tell why I'm not getting the correct output for my second 555 timer. I have double checked my wiring and everything seems fine.
As shown by the videos the first 555 timer works perfectly, but the second one is not receiving the envelop signal. Instead it's outputting just a constant frequency

 

[Baluncore]

Sorry, but I don't have the bandwidth to watch videos over my internet radio link, and I do not know how to download your embedded video files to my system once. Maybe if they were attached as files or I had the link I could do it.

. from the first 555 timer I am getting a square wave with a pwm that is controlled by the potentiometer. Furthermore, the second 555 timer is outputting a square wave at around 38 kHz which is perfect.

So you reported that the output, pin 3 of U2 is a continuous 38kHz rectangular wave.
That means pin 4 of U2 being held high.
1. Check the signal on the variable duty-cycle output, pin 3 of U1. Does in vary with pot position? Or is it always high?
2. Check the signal on pin 4 of U2. Is it the same variable duty-cycle envelope? If not, why not.

 

[enJayneering]

Okay, I'll check that when I get a hold of an oscilloscope again.

Here are the link to the two videos:



You can see the pwm signal with the blinking LED. I also show it on the oscilloscope as well

 

[enJayneering]

Hey, so just a quick update. I realized the problem with my circuit and why the envelope was not working. I was mixing the old schematic with the new one and I had a few wires in the wrong place. I believe I have fixed it and am now seeing the envelope, which is awesome. Now I just have to fine tune the frequency and test the sensor out.

I will post a video if I am able to get it to work

Thanks

 

[enJayneering]

Update
Since last post, I've made progress in sending the IR signal to my sensor. I am getting a visible PWM square wave that corresponds to the position of the potentiometer. However, I have not been able to drive the motor with my current setup. The schematic that I am using is a simplified version of the "IR_TX_RX_8" circuit. Here is a picture of what it looks like:

I removed the 555 inverter along with the extra transistor because I noticed that the output of the IR was already correct
Probing the signal of the IR sensor on an oscilloscope produced this square wave:

I tried to use this signal to drive the motor as such but I have had no luck.

Any ideas of why this circuit would not work?