I am sorry.

 

I am sorry.

Hi Bordodynov,
i have also changed the source from the library and it works, too. Thank you for your model.
Now i have one more question, do you know which opamp as summing amplifier and performs well? My two inputs are triangular wave with 2MHz, +-0.6mV and normal voltage signal such as a variable voltage between +-4V.
Before i use OPA684 but i think it performs not so perfect.
Thanks
Jackie

 

http://www.linear.com/parametric/Op...os_!enoise_!vosdrift_!lfnoise_!ibias_!inoise_
and
http://www.linear.com/product/LT1806

 

http://www.linear.com/parametric/Op...os_!enoise_!vosdrift_!lfnoise_!ibias_!inoise_
and
http://www.linear.com/product/LT1806

Hi Bordodynov,
As your suggestion i have used the LT1806 as inverting summing amplifier but failed. I don't know why.
Could you help me with this?

My purpose is triangular waveform with +-1.2mV, 2MHz and adde it to a normal variable voltage such as +-2V. So i need a accurate triangular wave generator and a accurate summing amplifier.

Thanks
Jackie

 

I am sorry.

Hi Bordodynov,
i have tried your circuit and it performs well.
But i don't know how to change your amplitude or frequency of your circuit. Could you tell me about it?
And what do the component "bcv61c" and "bcv62c" mean?

Thanks a lot!
Jackie

 

Throw away the opamps . Here's an alternative, using the simple 2MHz oscillator suggested by Dick (post #11).

 

I'm thinking the ADC would probably benefit from seeing a source impedance ~10k or less, so my circuit above could be modified by making C1=91p, R3=10k, R4=10Ω.

 

Hi Bordodynov,
i have tried your circuit and it performs well.
But i don't know how to change your amplitude or frequency of your circuit. Could you tell me about it?
And what do the component "bcv61c" and "bcv62c" mean?

Thanks a lot!
Jackie

The frequency depends on the value of the capacitor C1 and the resistor R4 (F~1/(C1*R4)). Also depends on the amplitude. More amplitude, more period (less frequency). The amplitude depends on the rating of the divider (R1 and R2) and the feedback resistor R3. For linearity, I used current generators. This current generator is built on mirrors BCV61C and BCV62C. The generators are switched on in turn, depending on the output voltage of the comparator. When using a simple resistor, pieces of exponents are obtained. With small signals it's not scary.
To observe pictures on an oscilloscope, I recommend using a probe with a small input capacitance. You can use the repeater on this high-speed operational amplifier in this role. Connect the output via a 22 ohm resistor (for stability).

 

With a transmission factor of -1, you apply 3.5 volts to the input, and a 2.5-volt supply voltage. Either increase the supply voltage, or make a transmission factor of -0.5. Replace the comparator with an operational amplifier and there will be no ejection. The truth is that the frequency decreases, but the frequency can be corrected by a capacitor or a resistor. You also need to add an inverting amplifier.

 

With a transmission factor of -1, you apply 3.5 volts to the input, and a 2.5-volt supply voltage. Either increase the supply voltage, or make a transmission factor of -0.5. Replace the comparator with an operational amplifier and there will be no ejection. The truth is that the frequency decreases, but the frequency can be corrected by a capacitor or a resistor. You also need to add an inverting amplifier.

Hi Bordodynov,

thank you for your answer.
Do you have any suggestion about the choice of opamp. Because i have tried to change many opamps, but my purpose is the accurate triangular wave with 2MHz, +-1.22mV and a accurate opamp as summing amplifier.
So i think you are expert and maybe you can give me advice about the choice (comparator, integrator, summing amplifier and perhaps an inverting amplifier).
Thanks in advance!
Jackie

 

I'm thinking the ADC would probably benefit from seeing a source impedance ~10k or less, so my circuit above could be modified by making C1=91p, R3=10k, R4=10Ω.

Hi Alec_t,
than you for your advice.
I want to select your circuit as an alternative. Could you tell me which resistor and capacitor i can change to achieve my goal of 2MHz, +-1.22mV triangular wave.
Pl can you send me your circuit, too, because it's hard to find 74HC14 spice model.

Thanks
Jackie

 

See

 

Asc file (modified) attached. You can download the 74HC model library and potentiometer model from the Yahoo LTspice User Group.
R1/C1 set the '2MHz' frequency and a trimpot sets the amplitude. The IC 5V supply will need to be very stable if you really need 1.22mV accurate to the second decimal place.

 

Unfortunately, the digital Schmitt trigger has a hysteresis variation and temperature dependence. Individual adjustment of frequency and amplitude is required. The comparator produces a more stable frequency.

 

Unfortunately, the digital Schmitt trigger has a hysteresis variation and temperature dependence. Individual adjustment of frequency and amplitude is required. The comparator produces a more stable frequency.

Hi Bordodynov,
1) Is the principle of your circuit normal comparator and integrator to generate triangular wave? Why you don't use capacitor when you integrating the result of comparator?
2) As you said OUT= (IN + TRI)/2. I need the OUT = IN +TRI, can i directly change the resistor of R8 and R10 to 1k. Is it also accurate? If i use LT1225 as summing amplifier, then i think the power voltage can be +-5V. Then the others are similar as your circuit.
3) I have built the circuit as your photo, but why i can get the triangular wave is sinus-wave? Could you tell me why or send me your circuit?

 

A diagram that is drawn by you will not work correctly. It is important that I use a divisor. The equivalent circuit should contain a voltage source of 2.5 volts and a resistor of 10 kΩ (20 kΩII20 kΩ). Then the Schmitt trigger is obtained. And at you absolutely not so.
Why did you replace 10 kΩ for 1 kΩ.
You want to receive a signal 10 times more. In my circuit, the output of the repeater is 50 mV. After the output divider (1 kΩ and 22 Ω), the voltage is approximately 1.2 mV. Adjust this value by changing the rating of the divider.
In my scheme everything turned out as you wanted. Only the signal is half the time. And us. If you are not confused by the sign of the output voltage, then an inverting amplifier is not needed.

 

I demonstrated the operation of the circuit with a 4-volt input signal. The output was 2 volts. I thought so. The circuit can operate with a signal greater than the supply voltage if the transmission factor is less than unity. I have a transmission ratio of 0.5. The input signal should be less than 4.9 volts.

 

I demonstrated the operation of the circuit with a 4-volt input signal. The output was 2 volts. I thought so. The circuit can operate with a signal greater than the supply voltage if the transmission factor is less than unity. I have a transmission ratio of 0.5. The input signal should be less than 4.9 volts.

Yes, i have seen that. Your ratio is 0.5. But now the output is as you said = (IN +TRI)/2, not = IN+TRI. Then it can't achieve my goal.
So i have an idee to change the OPA354 to such as LT1225, but the result is much worse. Do you have any other suggestion?

Thanks
Jackie

 

I do not understand why you are not satisfied with the half signal. But then use +/- 5 volts power, and that operational amplifier, which I recommended (it will work with a total power of 10 volts). Otherwise, you will not get it.

 

I do not understand why you are not satisfied with the half signal. But then use +/- 5 volts power, and that operational amplifier, which I recommended (it will work with a total power of 10 volts). Otherwise, you will not get it.

Hi Bordodynov,

Thank you for your help.
I think +-2.5V is perhaps enough for the input. I have tried to change to +-5V as source for OPA354, but there are more error. So i will keep it at +-2.5V. Before because i think maybe my input will exceed 2.5V, and the half signal as output i can't directly use it. If i use the inverting amplifier with gain-2, there are also more error. So i quit this idea.
I have tried so many opamp, the one you recommended is the best!

Now i have one last question, your triangular generator principle is also Comparator+Integrator or others? Which capacitor or resistor i can change to change the frequency?

I knew not so much about circuits. Thanks for your help and your understanding.
Best regards,
Jackie