SA602 Gilbert Cell Mixer

[JoeAtl30319]

I'm having an issue with the SA602's on-board oscillator (see photo one). I've setup an LC tank tuned to 10MHZ, but oscillation is intermittent (see photo two). Does anyone have any ideas as to what is causing the issue? It looks like oscillation continually starts, but them stops almost immediately. The on-off cycle is roughly 2.5khz. When oscillation is occuring, it does run at 10MHZ as expected.

Additional notes:
Oscillator: Hartley design
Caps on pins 6 & 7 = .1uf
Input voltage: 5V

[w2aew]

How are you establishing you 5V supply?  I'd that TO-92 a 5V regulator?  Do you have any bypass caps on your supply?  You're going to need a decent well filtered supply and low impedance ground to get clean performance at 10MHz.

Posting a schematic and the other side of your board would help.

[JoeAtl30319]

Here is the schematic with some extra photos. I've also included the output of P7 which has a slightly different waveform. In all cases, it looks like there is a gradual voltage increase preceding oscillation, followed by a slight voltage drop.

[joelby]

Where did you get the oscillator circuit topology from? It doesn't look like the standard Hartley oscillator given in the SA602A data sheet (see figure 7c in the Philips data sheet).

I would expect the '150uF' capacitor to be in parallel with the inductor.

[JoeAtl30319]

I based the topology on the manufacturer's datasheet schematic. It looks funny on my schematic, but the cap is parallel with the inductor. I've tried varying the tap point, but it does not seem to have any effect.

What really confuses me is that the oscillations start, then stop, and then start again. I've built standalone oscillators in the past and they've always maintained one of two states (oscillating or non-oscillating).

Also, I typed 1.68uf on the inductor, but meant to type 1.68uh.

[joelby]

Silly me, you're absolutely right!

I'm not sure, then. I can have a go at building the same circuit and see if I get similar results. I've got a bunch of SA602s kicking around

[G7PSK]

Bad connection on the battery?

[JoeAtl30319]

The battery connection is good. I've verified a relatively stable 5v input into Pin 8 of the chip.

[RCMR]

Where did you get the design parameters for the oscillator components?

Also, looking at the 602 datasheet, I see nothing that looks like your design - they show a Colpits and a Hartly configuration but what you've got is neither.

What's with that 150uF cap -- I see no place for that in any 10MHz oscillator.

[JoeAtl30319]

You're absolutely correct! I'm actually using a 150pf cap, not a 150uf cap.

Also, it looks funny on my "homemade" schematic, but this is a Hartley design. The inductor is in parallel with the cap. Also, the inductor is tapped at 4 turns from ground. I've tried varying the tap point, but it does not seem to help.

[AntiProtonBoy]

Long shot, could it be that some of the coils are magnetically coupling with each other?

[w2aew]

The oscillator is going to be fairly sensitive to the load on pins 6 and 7, particularly pin 6.  How were you connecting to the scope?  1x probes will be too much capacitive loading.  You'll need to use a 10x probe on pin 7, and even that might upset the operation.  You could try employing a buffer and probing the output of the buffer.

[JoeAtl30319]

Playing around with this in Ltsim, lt looks like my issue may be the coupling caps on pins 6 and 7. I'll have to play around with the values tomorrow and see what happens. The caps I'm currently using are .1uf. Perhaps I should be using values like 100pf or so. The results are much better in the simulator anyway...

[vk6zgo]

Here are a couple of links which may help:

http://www.stanford.edu/class/ee133/appnotes/AN1982.pdf
http://techdoc.kvindesland.no/radio/b1/20051213190607573.pdf

[JoeAtl30319]

Thanks! I'll try out some modifications and see if it works.

[JoeAtl30319]

Ok. Think I have this figured out. It turns out that in caps on pins 6 & 7 were too large. Decreasing the value to the pF range seemed to fix the issue. Thanks for everyone's help!