cant get crystal oscillator to oscillate

[Teekaytk]

hello

i am trying to build this circuit

http://www.webx.dk/rc/xtwin/convert17012006.gif

I cannot get the oscillator part to oscillate in real hardware. so i made a multisim model and it also does not oscillate. can some one please help me with this

thanks

[lacy]

I can get this oscillating by going to SIMULATE>INTERACTIVE SIMULATION SETTINGS and selecting "set to zero" under intial conditions. The problem is after doing this I still do not believe it is oscillating correctly. I see a damped waveform and this should not occur in a transistor ocillator that is behaving correctly. I tried installing a crystal into the circuit, but it still doesn't appear correct. This may due the crystal models not working as they should (I don't know this for a fact yet).

You might try using a tank circuit consisting of a coil and capacitor to simulate the crystal, but that will require that math be done to determine the component values at resonance.

Also, you might want to check your component values as a couple of them do not seem to be correct with the picture. R1 in the picture is 56K not 56 ohm and R3 is 100K instead of 100 ohm. I do not know if this was intentional or just an oversight, but I thought I would let you know about it.

I have done all I can on this for the moment. I hope this helps.

[RnDMonkey]

I found it easier to use a parallel crystal configuration than a series one, but oscillators can be tricky in Multisim. Here are a couple things to try:

Just wait longer. Sometimes oscillators take a while to stabilize, especially when very small currents are involved.

Shorten your timestep. It may seem like it goes against the whole idea of waiting, but there are times when you have to slow down the simulation for it to start oscillating in any reasonable amount of time, or sometimes to oscillate at all. I didn't fix your file, but go ahead and have a look at the circuit I have attached. Start off with a timestep of 1e-007 (0.0000001 Sec) and watch as it gets to 10 ms into the simulation. Then try with 1e-005 and watch it take way, WAY longer. If you set it to 1e-003 it won't oscillate at all. (At least it didn't for me) Use the VI probe to watch as the Vp-p keeps climbing.

[lacy]

Thanks morgol for that useful information. I can understand the theory behind what you say. It allows more time for the simulator to do its thing of number crunching. I did try this on the original circuit and I still only get a brief burst of oscillation and then it just fizzled out. I have no explaination as to why (this is even with a crystal in the circuit).

My theory is that NI is going to have to examine all these problems with oscillators and come up with a better solution. The simulator is not treating ocillators correctly in my opinion. I not only have problems with basic transistor oscillators, I have problems getting CMOS oscillators to work. This has been an issue for quite some time and I think NI need to pay speacial attention to it now.

[RnDMonkey]

I think it might be helpful to have another model parameter in crystals that kick-starts them, that way you at least know that the would oscillate once they got going. The problem is that sometimes they just take so long to get going that it appears they aren't working at all. Other times, you might wait for them for a long time, knowing this fact, but not have the conditions right anyway. An initial waveform of maybe 5 periods or so, even if it was a square wave, would allow the caps to get into the right state and the circuit should be able to take it from there. The mistake I made was assuming that if I put voltage across a crystal, I could instantly see a waveform. It seems like it would work, but Multisim just doesn't do that.

[Teekaytk]

thanks people

i will now go back to my hardware and see what IT doesnt  oscilate

also will see if i can remove the dampness

thanks

TK

[Teekaytk]

hi Lacy

I have been able to replicate exactly your circuit and i get the same dampened signal that dies. I did correct some values of compoenents that were worng in my real hardware, but it still does not oscillate in real. can you see what might be wrong

Thanks

[RnDMonkey]

Do you have the datasheet for the crystal you are using? It might help to look at the application notes for it. You might have better luck with a Colpitts-style configuration.

[_user32]

All,

Although this does not solve your specific circuit, here are some example circuits showing a equivalent model for a 25Mhz crystal with a 'kick start' source as the previous user mentioned - it is basically a damped high voltage sinusoid which will get the oscillation started very quickly).

Regards,

Patrick Noonan
Business Development Manager
National Instruments - Electronics Workbench Group
50 Market St. 1-A
S. Portland, ME 04106
Email: patrick.noonan@ni.com
Tel. (207) 892-9130
Fax. (207) 892-9508