Desired value of L and C in colpitts oscillator?

[samu]

Hello everybody and welcome Chemelec,

The circuit is from an old metal detector that worked in a little mine. This metal detector crashed and the owner didn´t find the manufacturer because is too old. Hi asked my to repair it if i could, and i made a new board and coil exactly as the old one. Then i repaired 4 o 5 more metal detector form the same manufacturer because they have finished the business and they have assistance no longer.

The circuits works great but the only problem is the fake detections. I know that the original ones worked without fake detections and in the model with a larger coil there is a 5,6nF cap in parallel with the 10nF one for change the frequency. But the question is: which is the good frequency?

Chemelec: Do you try to simulate or you built the circuit?. When i mounted the circuit it always oscillates and almost with any coil, but i can´t simulate it.



This is the Net List i tried to simulate

.EXTERNAL OUTPUT Vosc
R_R1 N00915 N01096 15k
R_R2 0 N00915 15k
R_R5 N02268 N01096 680
V_V1 N02268 0
+PULSE 0 24 1m 0.01m
R_R4 N01096 N01123 4.7k
R_R6 0 N02268 27K
C_C4 N01321 VOSC 2n
R_R7 0 N07105 1
C_C5 0 N02268 220u
R_R3 0 VOSC 470K
L_L2 N07105 N01321 100uH
Q_Q3 N01321 N00915 N01123 BC640/SIE
C_C2 N01123 N01096 100n
C_C1 N01321 N01123 10n
C_C3 0 N00915 10n

[samu]

I did it!!

The simulation already gives me expected results. I added a 220uF cap between the 680ohm resistance and ground and now it works fine. This cap was in the original scheme but i thought it was unnecesary

I didn´t have time to change coils and capacitors, i´ll come back in a while.

Thanks

[chemelec]

Hi Samu, I Always Prototype Circuits I design using real parts.
And this is easy for me as I usually have All the parts.

Besides, I find some circuits that get simulated don't work, either in simulation or later when made in real life.

Gary

[Qiaozhi]

I did it!!

The simulation already gives me expected results. I added a 220uF cap between the 680ohm resistance and ground and now it works fine. This cap was in the original scheme but i thought it was unnecesary

I didn´t have time to change coils and capacitors, i´ll come back in a while.

Thanks

You should also be able to get the oscillator to simulate in SPICE without the 220uF capacitor, as long as you define an initial condition. I would suggest setting an initial condition of 0V on the base of the transistor. You can either use ".IC N00915=0", or modify the last line to read "C_C3 0 N00915 10n IC=0".
I'm surprised that simply adding the 220uF cap causes the oscillator to start, as the calculation for the DC operating point would result in the capacitor being charged at the start of the transient analysis, thus resulting in the same starting conditions. Unless, of course, there is an initial condition set on the capacitor that you haven't told us about.
Anyway, that's good news. Hopefully you can now proceed to fix the other problems.

[chemelec]

You should also be able to get the oscillator to simulate in SPICE without the 220uF capacitor, as long as you define an initial condition. I would suggest setting an initial condition of 0V on the base of the transistor. You can either use ".IC N00915=0", or modify the last line to read "C_C3 0 N00915 10n IC=0".
I'm surprised that simply adding the 220uF cap causes the oscillator to start, as the calculation for the DC operating point would result in the capacitor being charged at the start of the transient analysis, thus resulting in the same starting conditions. Unless, of course, there is an initial condition set on the capacitor that you haven't told us about.
Anyway, that's good news. Hopefully you can now proceed to fix the other problems.

I also added that capacitor to my breadboarded prototype, and it Fixes the problem. Even a much smaller cap works.

But I find it a bit strange, as to why.

On the circuit I posted previously, the .022 cap isn't needed.
http://www3.telus.net/chemelec/Temp/Det-Osc.png

Unless you want to adjust the frequency downward.
But I didn't mean to show it there.

[Qiaozhi]

I also added that capacitor to my breadboarded prototype, and it Fixes the problem. Even a much smaller cap works.

But I find it a bit strange, as to why.

On the circuit I posted previously, the .022 cap isn't needed.
http://www3.telus.net/chemelec/Temp/Det-Osc.png

Unless you want to adjust the frequency downward.
But I didn't mean to show it there.

The 680 ohm resistor in the power line might the problem. Perhaps this causes the oscillator to become current starved, and the 220uF cap helps to smooth out the power line. Otherwise, with the cap removed, the oscillator just kills itself.