4
\$\begingroup\$

I would like to build an oscillator that will satisfy a few requirements. The main requirement is to have exactly one capacitor. Another one is to have high Q and sine wave output.

I am considering Hartley's oscillator design. Since it has only one cap. I don't what to built the biasing network and select a transistor that will suit the oscillator's needs, so I figured that there must be a way to build Hartley's oscillator around an operational amplifier as shown on the figure below.

enter image description here

I have made the simulation as shown above, and I believe it should oscillate, since the gain of the Op Amp negative feedback loop is greater than \$\frac{L1}{L2}\$ ratio. But the oscillations start and die. I can see that the frequency of the decaying oscillations is correct, but they would not sustain. I event tried to crack the gain up to some unreasonable number, but this doesn't help at all.

Can anyone advise me as to what I'm overlooking and how to make this circuit oscillate?

PS. The simulation uses 741. If is is not the best op-amp for the job, then what would be a good replacement?

|improve this question
\$\endgroup\$
  • \$\begingroup\$ Your circuit cannot work. The Hartley oscillator needs a feedback path with 180deg phase shift at one single frequency only. This is accomplished by a 3rd-order highpass. Realize: The inductor L2 acts as a load - nothing else. It has no influence on the feedback patrh. \$\endgroup\$ – LvW Feb 2 '18 at 16:51
  • \$\begingroup\$ Instead of a 3rd-order highpass you can of course, use 3rd-order lowpass (L,R,C). But it must be 3rd-order circuit - otherwise you have no phase shift of 180 deg. Why Hartley and not Colpitt? \$\endgroup\$ – LvW Feb 2 '18 at 17:11
  • \$\begingroup\$ An opamp setup as a negative resistance generator driving a LC tank circuit would be a simpler approach \$\endgroup\$ – sstobbe Feb 2 '18 at 17:13
  • \$\begingroup\$ As another - and simpler - alternative, you could use a RLC bandpass in the positive feedack path (zero phase shift) and you are completely free to select the proper gain using the resistive negative feeedback. \$\endgroup\$ – LvW Feb 2 '18 at 17:24
  • \$\begingroup\$ See my detailed answer. \$\endgroup\$ – LvW Feb 2 '18 at 17:41
2
\$\begingroup\$

I think you have a few issues that you need to address.

The primary issue is that the Q of your tank is very low. The Q of a parallel RLC circuit is:

$$Q=R\sqrt{\frac{C}{L}}$$

Considering that your resistive feedback is effectively in parallel with the tank, we can calculate Q:

$$ 1500\Omega \sqrt{\frac{1nF}{15\mu H}}=Q= 12 $$

The Q of 12 is VERY low, so this is probably your main issue. Increasing the magnitude of R1 and R2 should be sufficient for your circuit to oscillate properly. See my example simulation here. If your Q is too low, you will need more gain to compensate for the losses.

A second potential problem is that the op-amp you selected, the 741, has a bandwidth that is a bit low for the frequency you selected. The datasheet indicates a bandwidth of 1.5 MHz, and your oscillator frequency is 1.3 MHz. This may result in your op-amp not providing enough gain for the oscillator to function properly. There are MANY op-amps that would provide an improvement over a 741.

Another possible issue is that oscillators are a bit tricky to get working in simulators. While it sounds like this is not an issue for you, it is a potential pitfall. Often, the random noise that usually starts oscillators in reality does not occur in a simulator. Often a noise source or impulse is required to kick-start the oscillator.

|improve this answer
\$\endgroup\$
  • \$\begingroup\$ Thanks for the advice. Will try. What's the purpose of the 100Ohm resistor at the Op-Amp output in your simulation? \$\endgroup\$ – udushu Nov 29 '11 at 17:49
  • \$\begingroup\$ I've changed the circuit to oscillate at 100kHz with the Q of 7800. The feedback resistors R1 and R2 are 25k and 100k respectively. The oscillations now sustain for approx. 2ms but then die. Is there any other pitfalls that might contribute? \$\endgroup\$ – udushu Nov 29 '11 at 18:06
  • \$\begingroup\$ @udushu The 100 ohm resistor on the output is to simulate a finite output resistance. I tried building the circuit in PSPICE, and I got similar results as you. I had to significantly decrease the frequency to maintain stable oscillation. \$\endgroup\$ – W5VO♦ Nov 30 '11 at 4:26
  • \$\begingroup\$ @udushu Given the behavior of the oscillations, it looks like there is significant loss in the system. I suspect the op-amp is the culpret, as the phase starts to roll off before 100kHz - well before the gain becomes an issue. This topology may be unsuitable for any frequency << the bandwidth of the op-amp. \$\endgroup\$ – W5VO♦ Nov 30 '11 at 4:29
  • \$\begingroup\$ I've tried to replace the 741 with an ideal op-amp but I am still getting the same results. This is funny, because, I got the Hartly's working on a single BJT and it is working like a charm. But the op-amp just would not give up. \$\endgroup\$ – udushu Nov 30 '11 at 4:44
1
\$\begingroup\$

W5VO has explained things well, oscillators in SPICE can be really tricky, and sometimes just don't work correctly at all.
The opamp model you are using can sometimes stop things working (sometimes they are not accurate), lack of real world noise makes it difficult to start etc.

Follow W5V)'s advice (e.g. about resistors), plus here's a few extra thoughts on making it work with the 741.

Tips for getting oscillation to start:

You can use .startup to start the supply rails at 0V

You can set initial conditions e.g. `.ic I(L1) = 10mA, or V(n001) = 2V and so on.

You can apply a voltage source in series e.g. a sine wave at the expected frequency of oscillation for e.g. 100 cycles to get it started. It will not affect the circuit when it's completed N cycles as long as you don't set the source series resistance > 0 ohms.

The 741 is a very old opamp with a low GBP. You would need to lower you frequency of oscillation. Try setting C1 to 100nF, and using .startup. If that doesn't work include a voltage source in series between C1 and opamp out of 1V for 100 cycles at ~129kHz.

|improve this answer
\$\endgroup\$
  • \$\begingroup\$ I used the 741 because it is an OpAmp I am familiar with. I agree that something more modern should be selected. Thanks for the advice with the frequency. I definitely should lower it. As for the initial conditions, I used to use the starter circuit, but then I realized that oscillations start even without it, but decay very fast. \$\endgroup\$ – udushu Nov 29 '11 at 17:44
  • \$\begingroup\$ @udushu - the initial ringing may not even be enough to start if it's lower than say, 100mV. I would maybe try using the series voltage source to have the best chance of it working. If you have no other opamps in your SPICE to try, look on manufacturers sites (e.g. TI, LT, AD etc) for spice models. Alternatively, LTSpice has a decent selection of opamps included. \$\endgroup\$ – Oli Glaser Nov 29 '11 at 18:00
1
\$\begingroup\$

The circuit cannot work because the opamp has a voltage output in contrast to a transistor which provides a current output at the collector node. Therefore, an additional resistor Ro is neessary between node (6) and the upper node of L2.

In this case, we have 3rd-order highpass in ladder topology (Ro-L2-C1-L1) which can provide a 180 deg phase shift for feeding the inverting path through R1. This is the principle of an opamp based Hartley oscillator.

In addition, I like to recommend somewhat larger resistor values for R1 and R2 (just the ratio is important). This can avoid heavy loading of the RLC highpass.

|improve this answer
\$\endgroup\$
-2
\$\begingroup\$

you may want to try LF347 4mhz bandwidth quad jfet input amp.# focQs200 kHz

  • 10V peak sinusoidal output swing without slew limiting to 200 kHz
  • See LM148 data sheet for design equations
|improve this answer
\$\endgroup\$
  • 4
    \$\begingroup\$ Why are you yelling? \$\endgroup\$ – Bort Feb 2 '18 at 16:20
  • 1
    \$\begingroup\$ @Markku: Welcome to EE.SE. You're on a technical Q&A site so it's important that you format your posts properly for legibility and credibility. This includes correct punctuation and capitalisation. e.g., 4 mhz (millihertz?) should be 4 MHz. \$\endgroup\$ – Transistor Feb 2 '18 at 17:52

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.