Dumb newbie has a question...

Originally posted to sci.electronics.misc, but maybe this is a better place for it...

I am wanting to build a 10 MHz signal source to be used in troubleshooting a preselector/RF amplifier I built to assist my shortwave radio in picking up weak stations, and am wondering how feasable my current approach is. I currently have a 10 MHz crystal oscilator on a chip feeding into the primary of a step-down transformer, the sescondary of which feeds into the primary of a second step-down transformer. This gives me a .02 volt signal on the second secondary, but it is trashy (tons of harmonics, from what I understand about these things.) If I feed this into a 10 MHz tank circuit, should that be sufficient to clean it up and give me a nice sine wave, or am I just full of crap for taking this approach? I am (obviously) a mere experimenter, with more time than money for accomplishing my goals. If this approach is snake-bit from the start, what approach should I take? I am trying to simulate a 200 microvolt signal arriving at the input to the preselector/RF amplifier so I can track it through the circuit and find out where it is failing. I am planning on adding a third step-down transformer to the layout in an attempt to achieve this signal level, and if necessary use a second tank circuit to clean it up one last time. Now that those of you who understand these things have likely wet yourself from laughing so hard, please do me the favor of sharing some of your knowledge and wisdom. :)

Many thanks for any replies,

Dave

Reply to
Dave
Loading thread data ...

Whoa! If you mean you have a 10MHz canned oscillator, that's part of the problem. These generally put out a square wave, the exact opposite of what you need.

I'd get a 10MHz crystal and put it in a one-transistor oscillator circuit. Very simple circuit will do-- google the net for "colpitts crystal oscillator", here's a good reference:

formatting link

Either the colpitts or the pierce oscillator will do just fine.

Don't try any of the logic gate oscillators, they'll have too many harmonics.

Then assuming you have a 50Mhz or better scope, use a 10x probe and find the spot in the circuit with the cleanest sine wave. This will usually be on one end of the crystal.

Then since you need such a small signal, make a capacitor voltage divider, something like 1000pF from ground to your output jack, then maybe 5pF to the sine wave point.

That will give you a clean, low impedance, 5/1000'ths of the sine wave. Put all that in a shielded metal box with just a BNC connector coming out. Then build another little metal box with a 100 ohm carbon composition potentiometer-- hot side of the pot is the input, gound side goes to input and output grounds, the wiper goes to the output BNC center pin.

That should give you a nice weak signal, adjustable over a considerable range.

If the signal is still too strong, build another little 30dB attenuator box, say a 50 ohm resistor from each BNC center pin to ground, and a

820 ohm resistor between the center pins.

If you're going to be doing a lot of this, I'd go to eBay and buy a good attenuatorm like an HP 0-120db one. They're not too pricey. Make sure the seller guarantees it hasnt been burnt out.

feeding into the primary

Reply to
Ancient_Hacker

Wow. Thank you. I had no idea that my canned oscillator was part of the problem. I'd thought about building a Colpitts, but thought I would take the easy way out. First problem.

Thank you very much for the input. I've already got the metal box etc, so all I'll have to do is pull the stuff I have on the perfboard off, and start over. Should not be a problem.

I do appreciate your assistance. (And understanding.)

Dave

Reply to
Dave

You may need to use separate boxes for the oscillator and attenuators. You're asking for such a large stepdown one box may not provide enough isolation.

Reply to
Ancient_Hacker

Hmmm. Hadn't thought about that. No problem. Can do.

One question- the Colpitts oscillator in figure 7 at the website you gave me the link to has a tuning capacitor (2-20 pf). Is this to "trim" the output to exactly what you want? Just curious. I slept through oscillators in school. :\\

One more question: figure 8 at that same website shows an inductor in parallel with the shunt capacitance of a crystal. How can I tell what the shunt capacitance of a crystal is? Hook it up to my DMM? (which measures capacitance).

Thanks much...

Dave

Reply to
Dave

Make a 10 MHz bandpass filter to remove the harmonics from the canned oscillator. You can make a nice filter with a seperate 10 MHz crystal, as well.

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
 Click to see the full signature
Reply to
Michael A. Terrell

It will trim the frequency very very slightly, like a KHz or less.

I'd suggest not using figure 8 circuit, Figure 7 is simpler and doesnt require those funny tuned coils.

Well it's going to be a very very few pF, which is probably below the capability of your DMM.

Reply to
Ancient_Hacker

Gotcha. Was looking in the Allied catalog at their XT49S series of crystals, and notice the specify a "load" and an "ESR Max (ohm)" value for each of these crystals. Any info on what these refer to?

Thanks again,

Dave

Reply to
Dave

PS: the "load" is in pf (20 to be exact).

Dave

Reply to
Dave

Don't worry about it. I've never seen a crystal that wouldnt oscillate nicely.

Reply to
Ancient_Hacker

Okay, thanks. Yoiur help is greatly appreciated.

Dave

Reply to
Dave

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.