Need help interpreting an old circuit diagram: Transistor and RF Transformer

[kdausman]

Oh, and sorry... the answer referencing the LM555 just registered in my brain... I clearly need to check the whole thread more carefully before hitting Post Reply.

I am fascinated by this idea, and look forward to the results of your test! But yes, definitely enjoy the day!

 

[Diver300]

Wonderful feedback, everyone! I am thrilled that I have found this forum!

The comments about frequency counters vs. oscilloscopes are well-taken. I am completely inexperienced in purchasing frequency counters, however. I see, for example, **broken link removed** that covers the frequency range I need while reporting what seems to be down to the 100 Hz position. Is that the kind of thing I am looking for? I see other instruments that go up to several hundred dollars. I have the money to make a purchase of a good instrument if it is necessary, but the general applicability of this approach will increase the cheaper the components I can get away with are.

That sort of frequency counter would probably be fine. I think it shows to 10 Hz if the frequency is low enough. Frequency counting isn't difficult so a specialised unit that only reads frequency isn't expensive. However, frequency counting needs the sort of dedicated circuitry which isn't in all oscilloscopes.

At that price I won't feel too bad if that counter's not good enough.

 

[Les Jones]

There is a design for a capacitance and inductance meter that may suit your requirements **broken link removed**. A variation on this design was published in "Everyday Practical Electronics magazine" ( March 2010 ) I have built one and it works well. The way it does the calculations could null out the capacitance of the empty cell. You may even be able to add code to display the results that you want rather than just capacitance.

Les.

 

[JimB]

This afternoon I had a bit of a play around with this circuit, and built a version of it as shown here:

In my version of the circuit, the transformer is taken from the 2nd IF amplifier of PYE Westminster VHF transmitter receiver, and operates at 455kHz.
The capacitor in the tuned circuit is 600pF.
The transistor is a PBC108.
The resistors are 10k Ohm.
I did not fit the 300pF capacitor across the emitter resistor.

Did it work?
Looking at the emitter with the scope, I saw this:

The frequency is 611 kHz according to my frequency counter.

So, the circuit does work, once I got the phasing of the feedback winding correct.

JimB

 

[Nigel Goodwin]

How about adding a supply decoupling capacitor across it, and see if that makes any difference - it's absence is upsetting my sensibilities!

 

[JimB]

How about adding a supply decoupling capacitor across it, and see if that makes any difference - it's absence is upsetting my sensibilities!

An excellent point, it upsets my sensibilities also.

I did try adding a 0.1uF cap to the supply end of the emitter resistor and it made no obvious difference.
Looking at that point with the 'scope, there was no signal to be seen.
I assume that the half metre long leads to the PSU presented a sufficiently low impedance to kill any RF escaping via the resistor.

If I was building this thing for real, I would pit a 0.1uF cap at that point.
I would also build it in a more sturdy fashion than the lash-up shown in the picture.

JimB

 

[Nigel Goodwin]

I'm presuming the distorted waveform is down to the transformer having the wrong turns ratio? - BTW, nice to hear mention of the Pye Westminster

 

[kdausman]

Thank you for running that test!

So am I interpreting your mock-up correctly as follows:

?

(I apologize for the volt-meter symbol for the frequency counter... the online tool I used didn't have an appropriate icon.)

Did I put the supply decoupling capacitor in the right place? And is this a reasonable circuit to use for the desired application, as opposed to a different kind of oscillator? And finally, what transformer should I be looking for to make this work, if I am looking to purchase rather than cannibalize an existing component?

Edited to add: Oh, and you used a different transistor than the one that I started the thread asking about. How much does the specific transistor matter?

Thank you!

 

[kdausman]

As another addition to the discussion, apparently the original circuit diagram is reported as being a modification of the one reported here:

 

[Les Jones]

The circuit in post #29 is the same as in post #1 but re drawn to work with a positive supply. (As I mentioned in post #5)

Les.

 

[JimB]

So am I interpreting your mock-up correctly as follows:

More or less.
I used 10k resistors rather than 12k.
Why?
Because on my bench I have a box of "odd resistors" that have been used in various experiments, and rather than return to their proper storage, just drop 'em in the box.

C2, 300pF not fitted.
The oscillator will work with or without it.
It is a close call which is better, either can be justified.

C4 100pF, should be 0.1uF.
And yes, it is in the right place.

you used a different transistor than the one that I started the thread asking about. How much does the specific transistor matter?

In this circuit, the transistor is not critical.
Any small signal NPN transistor will do, 2N2222, 2N3904, BC108 etc, etc, there are many other suitable types.

what transformer should I be looking for to make this work, if I am looking to purchase

Look here:
https://www.mouser.co.uk/Passive-Co...ignal-Transformers/_/N-5gbg?P=1z0wkwmZ1yjij5b
for part number 42IF106-RC

This is the UK website of Mouser, which is a USA company, they should have this part in the USA.

When using this, you may need to change the 600pF capacitor, otherwise the frequency will be very low.
I used 600pF because that was on the circuit board from where I removed the transformer.

JimB

 

[kdausman]

Wonderful! thank you so much!

 

[Nigel Goodwin]

One further simple point, if you don't like negative supply rails (the -15V used in the original), you could simply replace the transistor with a PNP version, and use a +15V supply instead.

The reason it's been done as it has is in order to get a ground/chassis connection to the sensor.

 

[unclejed613]

it seems the original 1967 article is the source of the "1N3904" typographical error. funny how that propagates to all the other articles describing this circuit. as Nigel said, if you prefer to have a positive supply for this circuit, substitute a 2N3906 in place of the 2N3904.

if you want an analog output for a meter, etc... you can do the conversion with this: https://www.analog.com/media/en/technical-documentation/data-sheets/AD650.pdf which is a lot less expensive than a frequency counter.

 

[dr pepper]

I think the 1n3904 is a typo too, 2n3904's have been around for decades.

The yellow transformer in an old Am radio would suffice for your circuit.
Maybe you could wind your own using a T50-34 torroid, but you'd need a lot of turns.

If you were to do this with a chip a 7555 is the one to use, it goes to 1 mc, the 555 only goes to 100kc officially.

I have a pye westminster too, and mine still works!

 

[kdausman]

After a lot of futzing, only some of which do I understand, I have a working oscillator! Not all of my ordered parts have come in, so I made a lot of strange substitutions for resistances and capacitances to see what made a difference and what didn’t. I may try to draw up what I did and post it for feedback.

Without the external capacitance cell attached, I have a frequency range of about 550 to 585 kHz, based on the adjustable capacitor used to tune the system.

So it seems to get higher frequencies, I may need a lower-inductance transformer. Is that right? And if so, are there tricks you can play with multiple transformers to adjust the inductance the same way you can play with resistors and capacitors in parallel or series?

 

[JimB]

Futzing is good, especially when I leads to working oscillators.

So it seems to get higher frequencies, I may need a lower-inductance transformer. Is that right?

Yes.

And if so, are there tricks you can play with multiple transformers to adjust the inductance the same way you can play with resistors and capacitors in parallel or series?

Putting two inductors in parallel will lower the inductance.
I am not sure that I would recommend that as a practical technique, and I cannot remember ever having seen the technique used in any circuit.

You want a higher frequency, if so how high?
This could get messy.
Remotely trying to select a transformer which can be bought can be a bit difficult from several thousand miles away.
One big difficulty is that it is a transformer (ie two coupled windings).
If I were starting from scratch, I would select an oscillator circuit which used a simple inductor (ie just one winding).
Then using a reel of wire and suitable former, it would be easy (five minute job) to make whatever inductor was required.

JimB

 

[unclejed613]

Futzing is good, especially when I leads to working oscillators.

it's only good if you know why the oscillator started working, or if the futzing helps you understand the circuit better.

one way to make transformers easily (assuming your turns ratio is always 1:1) is to make "bifilar" wire. you use two different colors of magnet wire the same diameter, and twist them together at 1 to 4 turns per inch. if you use wire with red enamel and green enamel, use the red as the primary, and green as the secondary. this wire can be wrapped around cores such as toroids or ferrite bobbins, or it can be used for air core coils. it's not perfect for everything, but you may find it useful in this particular case. if i need magnet wire of various colors, and don't need a lot of it, i go to a hobby/crafts store where they stock magnet wire in different colors in their jewelry department.

 

[dr pepper]

I've wondered about those hobby shop various coloured copper wires, are they actually copper?, I wondered if they'd be aluminium or something.

 

[kdausman]

So the writeup I am working from says to adjust the circuit such that the air/open setting on the capacitor is about 1.3 MHz.

Here's my waveform (my frequency counter has not arrived yet).

I ended up going with the more traditional +15 V arrangement that was discussed earlier in the thread. I played around with the RF transformer a bit, tried several different transistors, and futzed with the resistors and capacitors in the circuit just to see what all of that would do. My circuit is closest to the one in post #29, but with a series capacitor between the output and the oscilloscope. I found that by modifying the two resistors and the fixed capacitor in the diagram for #29 I could modify the frequency, amplitude, and waveform. As I went to higher frequencies and amplitudes, I found that the waveform shifted from an almost sawtooth form (rounded on the rise) to what you see here. I was able to get the frequency up to 1 MHz and the amplitude up to about 500 mV, but as I did so, the bottom of the waveform seemed to get cut off (like you see in the picture, but more extreme) and the smaller extra peak before the main peak grew in prominence. And that extra peak seemed to depend as well on the external cell capacitance, leading me to worry that the frequency counter might start picking them up as separate pulses, screwing up the measurement.

Looking at the **broken link removed**, I don't see specs on the voltage amplitude necessary to count as a pulse. Another site with the same product suggests that it has sensitivity "better than 60mVPP," which I presume means 60 mV peak-to-peak. Given that, should I be aiming for a cleaner waveform at the expense of frequency and total amplitude? Or am I likely to be ok with this waveform?

Thank you all for the amazing help!