VFO: Series-E oscillator

2104/8/3

In blog: VCO: Negative resistance VCO  i got 2 new configuration of negative resistance oscillator, by checking its capability, i found this oscillator had a very good flat amplitude vs frequency response, and very wide band gain, which very suitable for sweep signal source, even without any AGC deploy, in each band it covered,  the amplitude keep incredibly flat.

•  easy start oscillation from several hundreds khz to around 100 MHz depending on what kind of BJT you use.

•  amplitude keep flat in each band it can cover. 

•  band to band amplitude is change, but can be fixed but AGC, which easy to deploy.

the long tail configuration ,VCO: Ultra Negative 100kHz-100Mhz, which easy to start oscillation on khz to 100Mhz, make it amazing. and JFET negative resistance give a even good 2 port negistor. 

all bjt easy going oscillator is very similar, 2 active device, connected as long tail, it actually a push pull oscillator.  and many of them connect one transistor's base to another's collector.  by change the long tail configuration to one pnp and one npn, we get a new 2 port  negative resistor.   

Second new configuration all BJT negistor:

this is actually a   Common Collector Amplifier cascade with a 'semi' common base PNP buffer(PNP base not totally been bypassed).  this is a real TWO PORT device, then easy to deploy as a negative resistor.

before construct the real HW, this sch DC bias could be simplify :

you can not directly connect this to a tank there need a resistor setup the proper bias:

Check the live HW

@ 2V and few mA, oscillator work well, when increase voltage, waveform will severely distortion, you can increase 330R resistor to get perfect sine. (by eye)

Selenium from theradioboard is a quick builder,  here is his regen configuration utilization this 2 terminal oscillator: like the drawing style...

amplitude response

this oscillator had a very stable amplitude vs frequency response, i like call it zoom oscillator. typical amplitude change in 3:1 tune rang <10% ,  if given strong bias(max current), the amplitude will keep no change, but waveform will slightly distortion. 

start from 40 Mhz, the negative resistor lack of enough gain to keep amplitude stable.

( Mar/6, 2016:  actually, start from 40 Mhz the oscilloscope's bandwidth limited it's accuracy, use power meter showing 11Mhz to 50Mhz the amplitude change less 2dB )

MW LF band 

on this band, one of the feedback capacitor must be increasing to make it oscillation, choose a suitable feedback capacitor make it just go at low band.

Improve the Gain control and AGC

the better way to gain control is limit the current flow through the negistor, by inserting a 10k to 51k resistor series with the negistor, supply voltage increase to around 12V, and easy to deploy a AGC control. 

selected resistor should ensure stop the oscillation, then use BJT as bypass, control the current.  even better control might achieve by a current mirror.

if use a tap inductor tank, connect 51 ohm  load directly to the tap, what will happen? oscillation stop, but if you increase the bias current, it oscillation again!  could deliver  4 Vpp, 16dBm output directly to 50ohm Load!  consume 30mA current.

What is it exactly

@8/31 2015

carefully checking the topology of this oscillator, you will find it's based on cascode amplifier, but configed as a franklin style, see following re-draw schematic:

Checking it's ALC performance via SA

May 10/2016

previously i use my oscilloscope to check the amplitude, that's wrong, a 100Mhz Oscilloscope begin mis-report the amplitude start from above 40Mhz. I check that use my Power meter, given a good amplitude stable report <2dB change.

my friend given me several picture of his oscillator and SA output, thanks, very much.