MC1648 how stable is it?

[neazoi]

Hi, I consider the MC1648 as an oscillator in LC mode for HF CW transmitting.
I would like to know to things:

1. What sort of output power and impedance should I expect out of it? (in order to see what will be the drive to the power amplifier)

2. What sort of frequency stability should I expect out of it? Is it enough for CW transmitting on HF amateur bands? I am thinking of a low span, changing coils for each band.

Or maybe I should use a CD4046 instead?

My goal is to make the oscillator as simple as possible.

- - - Updated - - -

Update:
The newer MC12148 has enough frequency stability for ham use, provided that a stable PSU is used. But the MC1648 (I like DIP packages) does not specify frequency stability, although the two chips internal diagrams are identical as far as I can see.
Any guess?

 

[betwixt]

1. fully optimised with a matching network - about 13mW.
2. it is intended for things like data recovery where a PLL error voltage controls the frequency. In 'free-run' it probably isn't very stable, I would anticipate several KHz/C drift with temperature at say 10MHz and the need for an extremely stable power supply.

You will probably get similar results with a CD4046, being CMOS it isn't intended to drive low impedance loads like an antenna. If you add a frequency divider and reference oscillator, both will be quite stable, if low power but the circuit gets significantly more complicated.

Brian.

 

[neazoi]

2. it is intended for things like data recovery where a PLL error voltage controls the frequency. In 'free-run' it probably isn't very stable, I would anticipate several KHz/C drift with temperature at say 10MHz and the need for an extremely stable power supply.

You will probably get similar results with a CD4046, being CMOS it isn't intended to drive low impedance loads like an antenna. If you add a frequency divider and reference oscillator, both will be quite stable, if low power but the circuit gets significantly more complicated.

Brian.

You get the point, simplicity is greatly appreciated here.
I am thinking of a HAM sitting on the bench, trying to make some CW contacts. Long term frequency stability is not too important, if the oscillator drifts a bit after a few minutes he can correct it.

However short term frequency stability is more important, as he should not continuously correct the frequency during a CW session (<2-3 minutes?).

Do you think that the MC1648 can cope with this, provided that a stable PSU is used, or the operator would need to correct the frequency every few seconds? (yes, once I have build such a bad vco...)

- - - Updated - - -

I also investigate if the oscillator of the SA612 should be better used, because with one chip you get an oscillator and a mixer.

 

[betwixt]

I understand what you are trying to do, a close friend G3RJV edits the 'Sprat' (Small Powered Radio Amateur Transmissions) magazine for the GQRP club (www.gqrp.com).

The point I'm making is the VCO in these IC's is optimized for fast tuning so they can rapidly track and lock to changes in incoming data, that makes them particularly susceptible to voltage variations which is exctly what you want to avoid for good stability. Used normally, they are the VCO in a PLL system where the tuning is automatically corrected, if you manually tune them it will work but be prone to environmental and power supply variations. They will perform no better than a simple single transistor oscillator and have similar power outputs. Also consider that their outputs are essentially digital so you still need filtering to eliminate the harmonics.

If you are willing to experiment with SA612 tyes of devices, it would be worth investigating the possibility of mixing a fixed crystal frequency with a narrow tuning range VCO. The VCO could tune say 1MHz to 1.5MHz and be left running for good stability and the crystal oscillator keyed to upmix to the frequency you desire. You still need filtering but keying a crystal oscillator will give a relatively clean signal and at low frequency you can make a fairly stable VCO. The 0.5MHz should be enough to cover all ham bands.

Brian.

 

[neazoi]

If you are willing to experiment with SA612 tyes of devices, it would be worth investigating the possibility of mixing a fixed crystal frequency with a narrow tuning range VCO. The VCO could tune say 1MHz to 1.5MHz and be left running for good stability and the crystal oscillator keyed to upmix to the frequency you desire. You still need filtering but keying a crystal oscillator will give a relatively clean signal and at low frequency you can make a fairly stable VCO. The 0.5MHz should be enough to cover all ham bands.

Brian.

Thanks Brian,
I have also found a useful document with some nice hints about the sa612 **broken link removed**
The interesting part is that he uses the harmonics to cover different bands with one tuned circuit.
Some tweaking is needed though.
I also like the idea with the mixing you mentioned, although it would require a se cond chip/oscilaltor.