test equipment signal generators

[montepollock]

Last? Month posts were made regarding basic tools for building a Pi,and also the best type of soldering Irons.
It was most helpfull as Most of My equipment is obsolete or has been mothballed for years.
I am interested in what features are most necessary in a signal generator.
Also any books helpfull to an aged dinosaur like Me or a novice builder.
Also cables and probes,and signal injectors
Thanks,Mont

[Max]

Last? Month posts were made regarding basic tools for building a Pi,and also the best type of soldering Irons.
It was most helpfull as Most of My equipment is obsolete or has been mothballed for years.
I am interested in what features are most necessary in a signal generator.
Also any books helpfull to an aged dinosaur like Me or a novice builder.
Also cables and probes,and signal injectors
Thanks,Mont

Hi Montepollock,
I'm using a 2MHz signal generator for PI tuning...with sweep and everything.
Really not too complex or newer generator needed...just to have some frequency capability and to be accurate. If you have with freq. meter incorporated is better...I have separate freq. meter.

I have also a good 20MHz signal gen. with built in freq. meter but don't use it for MD stuff. Not required. At least for me.

I found useful a free oscillator circuit...(can't find the schema) to find auto-osc frequency of new made coils for PI. It's very useful if you have.

Cable and probes are not so critical. Frequencies are really low. But Quality is important ever...so avoid oxidation-addicted material. I have almost everything made in japan for probes and cables. Probes with att. 1x and 10x are fine. You really don't need signal separate signal injectors here...just your signal generator and a probe sometime.

A very useful thing is a good and stable LCR meter. It is a must for every experimenter. There are cheapy you can find e.g. on ebay too.
Also a good scope say at least 20MHz BW , double trace is required...but nothing more than this--> again frequencies are low. I've noticed that old analog scopes are better than new digital ones in PI experiments.

About books there are many...you can find a lot on ebay too.

Best regards,
Max

[Carl-NC]

I rarely use a signal generator in detector design. For detector frequencies, it's really easy to throw together a 555 or 8038 and get whatever you want. For the rare times I use a generator, my Wavetek 20MHz combo function/pulse generator does nicely. It's actually overkill, but I bought it at a local auction for $10.

For VLF work, a 1-2 MHz function gen (sine/triangle/square) is plenty. For PI work, you might want a true pulse gen that has sync with delay and pulse width control.

Cables don't matter. At these frequencies you can use anything.

- Carl

[Leto]

?

Originally Posted by Max
I found useful a free oscillator circuit...(can't find the schema) to find auto-osc frequency of new made coils for PI. It's very useful if you have.

This I find very interesting...if coil is connected to this circuit - is the frequency generated same as the frequency of coil self-resonance?

[bbsailor]

Fun with test equipment

Montepollock and all,

If you ever check ebay and find a piece of test equipment called a lock-in amplifier for cheap (under $100), check out what you might be able to do with it by reading the documents on the following web link. http://www.cpm.uncc.edu/lock_in_1.htm

Even if you don't get a lockin amplifier, the above reading will help the curious experimentor understand the inner workings and limits of a PI machine, noise and signal extraction.

In fact, the middle and back end processing of a PI metal detector functions like a lock-in amplifier.

All you need to do is build the MOSFET coil driver. You don't want to drive a PI coil directly with the pulse generator or you will fry the output transistors with the coil's flyback voltage. Just use the pulse generator to drive the MOSFET gate that drives the coil (like in the Hammerhead design). Then, build the NE5534 preamp but with a gain of 20 to 50 including the input clamping diodes. Feed the output of the NE5534 into the lock in amplifier input. Connect the Pulse gen sync out to the sync in of the lock-in amp. You can adjust the pulse generator pulse delay control to move the sample window start point relative to the main pulse in the lock-in amplifier. Now you can expand the sample window size, adjust filtering and increase the gain and see the effects of this on extracting a very small signal from a variety of targets.

What you are doing is using test equipment to actually make a PI machine so you can play with the various parameters and better understand how a PI works. What you will learn is that the big tradeoff is sample time for sensitivity. Longer samples extract weaker signals from the noise. You will begin to see the practical limits of PI design and see what the limits are for sweeping a coil over the ground at a reasonable speed and see why certain common time constants and filter ranges are used in typical PI designs.

While a lock-in amplifier is not as fundamental to making PI circuits as a dual channel scope or a signal generator for measuring coil self-resonance, it can help you have some fun to simulate what is happening inside a PI circuit and control a lot of variables.

I hope this stimulates some experimentors to look into lock-in amplifiers or at least be aware of what they can do.

bbsailor

[Max]

This I find very interesting...if coil is connected to this circuit - is the frequency generated same as the frequency of coil self-resonance?

Hi Leto,
yes...about the same. Circuit introduces only a small capacitance due to osc. transistor (hi freq. one) so you have about the same self resonance you can see using e.g. a function generator with sweep near resonance freq. and a scope (watching at amplitude of signal increase)...cause also the function generator and scope introduce capacitance too in similar values. I've noticed that is better when you have too capacitance due to probes etc... to ues this method...and also faster.

Advantage is that you can read directly frequency on a freq. meter or with the scope as you want and don't need to mess with sweep and freq. searching manually...that requires some minutes...and sometimes more.

Best regards,
Max