14398 views|27 replies

541

Posts

1

Resources
The OP
 

The operational amplifier forms a three-point capacitor oscillation [Copy link]

 

A schematic diagram found on the Internet. Now replace the L in the diagram with a passive crystal oscillator, as shown below

Can this achieve oscillation? Does oscillation require R1 to be much larger than R4?

This post is from Analog electronics

Latest reply

任何震荡器的基础电路都沒有稳幅固形之效, The so-called amplitude stabilization of the LC oscillator is actually a routine similar to PWM or thyristor phase-cutting voltage regulation (Class C status is required).   Details Published on 2023-10-17 23:10

2w

Posts

0

Resources
2
 

“Can this achieve shock?”

Yes, absolutely. There is no problem at all. In fact, starting from the 48 series of single-chip microcomputers, all single-chip microcomputers have such an inverting amplifier inside. Connecting a quartz crystal and two small capacitors to the output and input of the inverting amplifier can generate oscillation. The single-chip microcomputer already has a resistor inside to realize DC feedback to stabilize the DC operating point, so there is no need to add a resistor outside the chip.

This post is from Analog electronics

Comments

So is it a sine wave or a square wave?  Details Published on 2019-9-24 11:31
 
 

2w

Posts

0

Resources
3
 

“Does R1 need to be much larger than R4 to start oscillation?”

Yes.

In fact, R4 can be omitted (short-circuited), and only R1 is used to ensure the correct DC operating point bias.

This post is from Analog electronics

Comments

Remove R4, only R1, adjust the value of R1, the circuit has no waveform, the op amp is powered by a single power supply, how do you understand "the DC operating point bias is correct"  Details Published on 2020-7-20 15:22
 
 
 
 

2w

Posts

0

Resources
4
 

This circuit does not even need to use an op amp (op amps that can work at several MHz to tens of MHz are relatively expensive), just use the inverter in the digital circuit. For example, there are six inverters in the 74HC04 chip, and you can use one of them to form the quartz crystal oscillator circuit you need. When using 74HC04 to form a quartz crystal oscillator, C4C5 is usually 15pF to 30pF, R1 ranges from hundreds of kilohms to one megohm, and R4 is not used.

This post is from Analog electronics

Comments

I won't consider the cost for now. The agents of ADI and TI will give me free samples. Now the main thing is to find a sine wave with as small harmonic components as possible above 5M. This is quite difficult. Do you have any good methods to achieve it?  Details Published on 2019-9-24 11:35
 
 
 
 

541

Posts

1

Resources
5
 
maychang posted on 2019-9-24 09:59 "Can this achieve oscillation?" Absolutely, no problem at all. In fact, starting from the 48 series of MCUs, all single-chip...

So is it a sine wave or a square wave?

This post is from Analog electronics

Comments

"So is it a sine wave or a square wave?" It is difficult to get a sine wave, nor is it a square wave. In fact, it is a very complex waveform.  Details Published on 2019-9-24 12:14
 
 
 
 

541

Posts

1

Resources
6
 
maychang posted on 2019-9-24 10:04 This circuit does not even need to use an op amp (op amps that can work at a few MHz to tens of MHz are more expensive), just use the inverter in the digital circuit, such as 74HC ...

I won't consider the cost for now. The agents of ADI and TI will give me free samples. Now the main thing is to find a sine wave with as small harmonic components as possible above 5M. This is quite difficult. Do you have any good methods to achieve it?

This post is from Analog electronics

Comments

In order to obtain a waveform that is as close to a sine wave as possible with fewer harmonics, the oscillation circuit must have a limiting circuit. There is no limiting circuit in the quartz crystal oscillation circuit, and the amplitude is limited by the power supply voltage. Such an oscillator without a limiting circuit will produce a very poor waveform that is far from a sine wave.  Details Published on 2019-9-24 14:44
In order to obtain a waveform that is as close to a sine wave as possible with fewer harmonics, the oscillation circuit must have a limiting circuit. There is no limiting circuit in the quartz crystal oscillation circuit, and the amplitude is limited by the power supply voltage. Such an oscillator without a limiting circuit will produce a very poor waveform that is far from a sine wave.  Details Published on 2019-9-24 14:41
In order to obtain a waveform that is as close to a sine wave as possible with fewer harmonics, the oscillation circuit must have a limiting circuit. There is no limiting circuit in the quartz crystal oscillation circuit, and the amplitude is limited by the power supply voltage. Such an oscillator without a limiting circuit will produce a very poor waveform that is far from a sine wave.  Details Published on 2019-9-24 12:23
In order to obtain a waveform that is as close to a sine wave as possible with fewer harmonics, the oscillation circuit must have a limiting circuit. There is no limiting circuit in the quartz crystal oscillation circuit, and the amplitude is limited by the power supply voltage. Such an oscillator without a limiting circuit will produce a very poor waveform that is far from a sine wave.  Details Published on 2019-9-24 12:18
 
 
 
 

2w

Posts

0

Resources
7
 
S3S4S5S6 posted on 2019-9-24 11:31 So is it a sine wave or a square wave?

"So is this a sine wave or a square wave?"

It's hard to get a sine wave, and it's not a square wave either. In fact, it's a pretty complicated waveform.

This post is from Analog electronics

Comments

任何震荡器的基础电路都沒有稳幅固形之效, LC震荡器的的所谓稳幅,其实是类似PWM或可控硅裁相调压的套路 (需要丙类状态)。  Details Published on 2023-10-17 23:10
 
 
 
 

2w

Posts

0

Resources
8
 
S3S4S5S6 Published on 2019-9-24 11:35 Cost is not a consideration for now. The agents of ADI and TI will give me free samples. Now the main thing is to find a sine wave with harmonic components as small as possible and above 5M...

To obtain a waveform that is as close to a sine wave as possible with fewer harmonics, the oscillation circuit must have a limiting circuit. There is no limiting circuit in the quartz crystal oscillation circuit, and the power supply voltage is used for limiting. The waveform obtained by such an oscillator without a limiting circuit is quite poor and far from a sine wave.

The main requirement for a quartz crystal oscillator is an accurate and stable frequency, and there are no very high requirements for the waveform.

This post is from Analog electronics
 
 
 
 

2w

Posts

0

Resources
9
 
S3S4S5S6 Published on 2019-9-24 11:35 Cost is not a consideration for now. The agents of ADI and TI will give me free samples. Now the main thing is to find a sine wave with harmonic components as small as possible and above 5M...

If the frequency is required to be accurate and stable, and the waveform is required to be close to the sine wave, then a limiting circuit must be added to the quartz crystal oscillator circuit. The function of the limiting circuit is to reduce the feedback amount after the amplitude reaches a certain value, that is, to reduce the loop gain. When the loop gain is reduced to 1, a sine wave can be obtained.

There are many commonly used limiting circuits, such as a diode detector that controls a field effect transistor to shunt the positive feedback signal, or uses the nonlinearity of a diode (the equivalent resistance decreases when the voltage across the diode increases), or uses a thermistor. When the amplitude reaches a certain value, the thermistor value increases or decreases to control the feedback amount.

This post is from Analog electronics

Comments

How to add a limiting circuit to the oscillator circuit I mentioned last time? I considered using a diode before, but I thought that the frequency was relatively high and the diode junction capacitance would affect the circuit.  Details Published on 2019-9-24 13:29
 
 
 
 

541

Posts

1

Resources
10
 
maychang posted on 2019-9-24 12:23 If the frequency is required to be accurate and stable, and the waveform is required to be close to the sine wave, then a limiting circuit must be added to the quartz crystal oscillator circuit. The function of the limiting circuit...

How to add a limiting circuit to the oscillator circuit I mentioned last time? I considered using a diode before, but I thought that the frequency was relatively high and the diode junction capacitance would affect the circuit.

This post is from Analog electronics

Comments

"How do I add a clipping circuit to this circuit?" It is more complicated to add a clipping circuit to this circuit (not impossible). In addition, don't you want to use an op amp to achieve oscillation? And the agent will give you free samples? Why are we back to the discrete transistor circuit?  Details Published on 2019-9-24 14:36
 
 
 
 

2w

Posts

0

Resources
11
 
S3S4S5S6 Published on 2019-9-24 13:29 How to add a limiter circuit to the oscillator circuit I mentioned last time? I considered using a diode before, but I thought that the frequency was relatively high, and the diode...

"How do I add a limiting circuit to this circuit?"

It is more complicated to add a limiting circuit to this circuit (not that it cannot be added).

Besides, aren't you going to use an op amp to achieve oscillation? And the agent will give you free samples? Why are you going back to discrete transistor circuits?

This post is from Analog electronics

Comments

The discrete transistors were tested last time, and the op amps are still under design. If the limiting circuit of the transistor is not very complicated, you can use the PCB you have to test and see the effect.  Details Published on 2019-9-24 15:43
 
 
 
 

2w

Posts

0

Resources
12
 
S3S4S5S6 Published on 2019-9-24 11:35 Cost is not a consideration for now. The agents of ADI and TI will give me free samples. Now the main thing is to find a sine wave with harmonic components as small as possible and above 5M...

Using a limiting circuit is the first method. The second method to get a sine wave is filtering. The waveform is not sinusoidal because it contains many high-order harmonics. As long as the high-order harmonics are filtered to a very small content, the waveform becomes a sine wave. Either low-pass filtering or band-pass filtering can be used. When the frequency is not too high, active low-pass filtering can be used (second-order filtering usually cannot filter thoroughly and requires a higher order), and LC filtering can be considered when the frequency is higher.

This post is from Analog electronics

Comments

What is currently used is filtering, but after low-pass filtering, the AC component is attenuated and the DC component remains unchanged. The signal we ultimately want to obtain is a sine wave with a DC bias. The amplitude of the sine wave is 2.5V, and the DC bias is 2.5V, so we get a sine wave with a minimum value of 0 and a maximum value of 5V.  Details Published on 2019-9-24 16:05
 
 
 
 

2w

Posts

0

Resources
13
 
S3S4S5S6 Published on 2019-9-24 11:35 Cost is not a consideration for now. The agents of ADI and TI will give me free samples. Now the main thing is to find a sine wave with harmonic components as small as possible and above 5M...

Two methods have been described above. The third method is to use a DDS chip to output a sine wave. This method can produce a fairly good waveform.

This post is from Analog electronics
 
 
 
 

541

Posts

1

Resources
14
 
maychang posted on 2019-9-24 14:36 "How to add a limiter circuit to this circuit?" It is more complicated to add a limiter circuit to this circuit (not that it cannot be added). In addition, you don't...

The discrete transistors were tested last time, and the op amps are still under design. If the limiting circuit of the transistor is not very complicated, you can use the PCB you have to test and see the effect.

This post is from Analog electronics
 
 
 
 

541

Posts

1

Resources
15
 
maychang posted on 2019-9-24 14:41 Using a limiting circuit is the first method. The second method to get a sine wave is filtering. The waveform is not sine because it contains many high-order harmonics. As long as...

What we are using now is filtering, but after low-pass filtering, the AC component is attenuated and the DC component remains unchanged. The signal we ultimately want to get is a sine wave with a DC bias. The amplitude of the sine wave is 2.5V, and then the DC bias is 2.5V, so we get a sine wave with a minimum value of 0 and a maximum value of 5V. After low-pass filtering, the AC is attenuated and can be amplified appropriately, but due to the limitation of the DC in the AC, the amplification factor cannot be too large, otherwise the DC bias will be greater than 2.5V, and the output of the op amp will be distorted. I originally wanted to use band-pass filtering, but there is a DC component in it, which is not easy to do. In short, filtering is possible, but if you want a relatively pure sine wave, the filter needs several levels. After several levels of filtering, the signal is attenuated. If it is amplified, the harmonic components in it are also amplified, so it is not easy to do. I asked about a band-pass filter before. If the ground is biased, as shown in the figure below

Can this make the bandpass filter filter out AC without filtering DC?

This post is from Analog electronics

Comments

"Can the bandpass filter filter out AC without filtering DC?" In this circuit, the DC component of the input signal is not reflected at the output end. It is not possible to "let the bandpass filter filter out AC without filtering DC".  Details Published on 2019-9-24 16:33
"Can the bandpass filter filter out AC without filtering DC?" In this circuit, the DC component of the input signal is not reflected at the output end. It is not possible to "let the bandpass filter filter out AC without filtering DC".  Details Published on 2019-9-24 16:31
 
 
 
 

2w

Posts

0

Resources
16
 
S3S4S5S6 Published on 2019-9-24 16:05 Currently, filtering is used, but after low-pass filtering, the AC component is attenuated and the DC component remains unchanged. The signal we ultimately want to obtain is a signal with a DC bias...

“Can this make the bandpass filter filter out AC without filtering DC?”

In this circuit, the DC component of the input signal cannot be reflected at the output end. It is not possible to "let the bandpass filter play the role of filtering out AC without filtering DC".

This post is from Analog electronics
 
 
 
 

2w

Posts

0

Resources
17
 
S3S4S5S6 Published on 2019-9-24 16:05 Currently, filtering is used, but after low-pass filtering, the AC component is attenuated and the DC component remains unchanged. The signal we ultimately want to obtain is a signal with a DC bias...

"The signal we ultimately want is a sine wave with a DC bias. The amplitude of the sine wave is 2.5V, and the DC bias is 2.5V, so we get a sine wave with a minimum value of 0 and a maximum value of 5V."

If you want to get such a waveform, it is strongly recommended to use the DDS method.

This post is from Analog electronics
 
 
 
 

541

Posts

1

Resources
18
 
maychang posted on 2019-9-24 10:01 "Does the oscillation need to satisfy that R1 is much larger than R4?" Yes. In fact, R4 can be omitted (short-circuited), and only R1 is used to make DC work...

Remove R4, only R1, adjust the value of R1, the circuit has no waveform, the op amp is powered by a single power supply, how do you understand "the DC operating point bias is correct"

This post is from Analog electronics

Comments

"Remove R4, only R1, adjust the value of R1, the circuit has no waveform, the op amp is powered by a single power supply" 1. Do you mean the picture on the first floor? 2. What model is the op amp? What is the power supply voltage? 3. What is the nominal frequency of the quartz crystal? 4. The two capacitors C3 and C4 are  Details Published on 2020-7-20 15:52
"Remove R4, only R1, adjust the value of R1, the circuit has no waveform, the op amp is powered by a single power supply" 1. Do you mean the picture on the first floor? 2. What model is the op amp? What is the power supply voltage? 3. What is the nominal frequency of the quartz crystal? 4. The two capacitors C3 and C4 are  Details Published on 2020-7-20 15:51
 
 
 
 

2w

Posts

0

Resources
19
 
S3S4S5S6 Published on 2020-7-20 15:22 Remove R4, only R1, adjust the value of R1, the circuit has no waveform, the op amp is powered by a single power supply, what you said is "DC operating point bias...

"Remove R4, only R1, adjust the value of R1, the circuit has no waveform, the op amp is powered by a single power supply"

1. Are you referring to the picture on the first floor?

2. What model is the op amp? What is the power supply voltage?

3. What is the nominal frequency of a quartz crystal?

4. What is the capacitance of the two capacitors C3 and C4?

This post is from Analog electronics

Comments

"1. Do you mean the picture on the first floor?" "Yes" "2. What is the model of the op amp? What is the power supply voltage?" The op amp is ADA4891, and the power supply is 5V 3. What is the nominal frequency of the quartz crystal? The nominal frequency is 8M 4. The capacitors C3 and C4 are 22pF  Details Published on 2020-7-20 17:23
 
 
 
 

2w

Posts

0

Resources
20
 
S3S4S5S6 Published on 2020-7-20 15:22 Remove R4, only R1, adjust the value of R1, the circuit has no waveform, the op amp is powered by a single power supply, what you said is "DC operating point bias...

How do you understand "the DC operating point bias is correct"?

The DC operating point should be biased to the area where the op amp can amplify linearly.

This post is from Analog electronics
 
 
 
 

Guess Your Favourite
Just looking around
Find a datasheet?

EEWorld Datasheet Technical Support

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京B2-20211791 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号
快速回复 返回顶部 Return list