Dr Ronald Stiffler SEC technology

[gyulasun]

Hi Itsu,

I think you could connect a 22 to 33 pF ceramic capacitor across the base and emitter of the BF869 transistor, this may help oscillate the circuit. You surely know that the base and emitter is mixed up in the TO-202 casing if I recall correctly http://pdf.datasheetcatalog.com/datasheet/philips/BF871.pdf  (the first pin on the left is the emitter).

Also, the collector current can only be maximum 100 mA peak but still this would be worth trying.

I believe the use of a capacitor across the base-emitter would help reduce tendency for harmonic oscillations too.

When there is a certain resistor value set by the 5 kOhm trimmer, the DC operational point is changing because the 100 kOhm base resistor receives a reduced supply voltage due to the voltage drop across the series set resistor, this would involve the use of a 100-220 kOhm trimmer potmeter instead and control base current in such cases.  Alternatively, you could connect the top pin of the 100 kOhm resistor (or potmeter) directly to the positive input rail, to avoid the voltage drop effect for the base.
The BF869 has a much lower hFE than the earlier BC337-40 so adjusting the operation point is needed.

Gyula

[gyulasun]

Hi Nick,

The 0.5 V voltage level (I assume you meant DC voltage) across the base and emitter can be correct for a bipolar npn transistor.

Would like to ask whether you connected your scope probe across the collector and negative supply ground? to see any waveform?  IF you did check that and saw only the 12V DC level on the collector (flat line at around the 12 V height from the zero line and the time base set to some ten nanosecond), then try to follow what I suggested to itsu with the use of the base emitter capacitor and 100 kOhm trimmer potmeter.
In such oscillator circuit the transistor is normally needed to be DC biased to 10-20 mA collector current by the 100 kOhm but due the various h_FE values of the different transistors, it should be adjusted individually.
What is the DC current draw for you oscillator from the 12 V battery input?  You can watch the input current on a DMM while you close the base and emitter pins of the transistor by a slightly wet finger: if the crystal oscillates then it probably stops when you touch the two pins so input current should change.  But the scope probe across the collector-emitter is better to use for checking.
I can see something sleeved onto the wire which connects the collector to the board: are they ferrite beads? What are they?  IF they are ferrit beads, then try to remove them, they are not needed in this oscillator.

Your choke coil may have come from an AC mains filter part of a monitor or tv set etc input power supply, it may have two windings and I think it is enough to use only one of them in the collector. Did you mean the 2.5 Ohm DC resistance for one winding or both in series? In an unfortunate case when you connected the two coils in series, they may cancel each other so that a very small inductance can remain for the collector in that case, preventing oscillations.   
There are numbers and letters on the label of the choke, try to search for either the top full line or the bottom by google, you may get a hit and the inductance of the choke may turn out. As Itsu suggested, do not connect coil L3 yet, first focus on the oscillator. 

EDIT:  perhaps you have the 1 kOhm resistor in the collector what I suspected to be ferrite bead I guess. And the choke coil is used for supply rail filtering? 


Gyula

[itsu]

I did not do much today, the weather is very nice still here.

I tried to use the BF869 (yes indeed different layout, as with all these transistors, so i always measure where the base is first).

But i could not get it running, not with the extra 20pF across base/emitter either.

Then i tried to attach the 100K to the 12V plus line instead of behind the 5K trimmer pot, but then even with the BC337 it did not oscillate anymore.
The base voltage in either case is around 650mV, so it seems to need the feedback line from the collector to oscillate.

Changing the 100K for a 200K trimmer pot (2x 100K in series) and varying the voltage there did not made it oscillate.

So i decided for now to stay with the old setup (bc337 transistor, fixed 100K to the collector voltage) and lower with the 5K trimmer pot the voltage on the collector so it stays around 30°C  (86°F) and still deliver about 100Vpp at the output.

The spectrum analyzer shows the noise this thing produces (even into the 500MHz range!), see screenshot 1.
The screenshot 2 shows the output signal (yellow), the collector signal (blue) and the base signal (purple).

Still need to redo the L3 coil so it resonates at 13.56MHz with the leds.

Itsu

[NickZ]

    Well guys, I can see that I'm not the only one having a heck of a hard time trying to get this simple little device working.
     I changed the 2SC5200 transistor to a 1815, and it fired up, but only by touching the base, although that's not what its supposed to do. It also fires up by holding the L3 end, or touching it.  I don't have a 100k trim pot.
  I still get no light at the single led, with the diode loop, even though there's about 200v, showing on the scope, but no light on the bulb.
 
   I have no hairs on my head left to pull. I hope that you guys are having an easier time of it. No wonder it took the Doc 6 months just to get as far as he's shown. Makes me wonder what other tricks he has up his sleeve.

   Gyula:  The black (ferrite bead), is actually the choke (1000uH?) with 2.5 ohms of resistance mounted on the red wire going to the collector. There's NO numbers or letters on it
   The tiny ceramic cap (22pf) has only a 22 written on it. So, I'm guessing that it's the right one.
 
   I may remove all the AC filters from the board, and try again.  This small PCB was meant for another Kacher circuit project, but I'm using it, as it's the only board I have left to mount these parts on, until I can get some more of these blank boards to build on.
   
   My idea was to see how high of an output could be obtained by using a higher voltage transistor, but now, if I could only light the single led/loop, I'd be a happy camper.
   
   I ordered two different boost converters about a month ago. They have not arrived from China, by turtle mail, as yet.   
   Itsu, the parts that you shipped have not arrived, either. That's why I'm working on this Dr. Stiffler project, for now.   
I hope that it pans out...

[gyulasun]

Hi Itsu,

Thanks for all your efforts and of course enjoy good weather and take your time as the most convenient.

I think the hFE values of transistors are that crucial, first of all, and cause the problem. Transistor types with lower hFE like say 40-50 (BF869) need at least 15-20 mA quiescent collector current where they already can amplify better.  A 100 kOhm resistor lets roughly 12V / 100k= 0.12 mA base current and with a hFE of 50 the collector current can be only around 50 x 0.12 = 6 mA.
A BC337-40 has a hFE of not less than 400. The same 100 kOhm resistor drives the same 0.12 mA current into the base and the collector current will be at least 400 x 0.12 = 48 mA, a huge diference. The MPSA types the Doc uses have a hFE of at least 120-150 or higher if I recall correctly.

I do not get why the oscillator is not working when the 100 kOhm is directly connected to the positive rail. It is ok that there is feedback created from the collector (or rather from the top wire of the tap) when you insert (set) a resistor value by the 5 k trimmer but it is negative feedback which reduces gain, yet it creates an operational point where the crystal is able to work.
But with the extra collector resistor set by the 5 k trimmer the transistor sees higher collector impedance hence able to amplify maybe better,  I would think. You may say then that the higher collector impedance is present also when the 100 kOhm is directly on the positive rail, yet the oscillator does not work. I know it is easy to say that conditions for oscillations are not insured in the latter case... and I cannot know the actual circumstances, details.Regarding the capacitor between the base-emitter, I may have said too little value (20 pF), it can go up to 40-50 pF or higher too. The 22 pF across the collector-emitter can also be increased but then the 18 turns on the toroidal core should be reduced by 1-2 turns to bring back the resonant frequency of the tank (unless C_var can compensate it).

Thanks, Gyula


Hi Nick,   

Sorry for your finding this hard to build, it is difficult to help, so many issues may develop and even one of such issues is enough for the oscillator not to work.  If the choke in the collector is not a choke with at least some hundred uH value than the transistor with relatively low hFE could not amplify enough.  The 2.5 Ohm choke resistance may indicate it more than likely has 1000 uH, only an L meter can verify it (or there are means to measure it otherwise). The label with letters and number are on the ferrite mains filter and that was I thought you use it in the collector as the choke.   Gyula