Hartley oscillator: amplitude not stable

Bordodynov

Joined May 20, 2015
3,177
There were, and I think there are now inductances on the cylindrical rods. Take the rod inductance of 60 mH and rewind a few turns in the center of the inductance. Measure it and correct the number of turns. There are formulas, but they are very cumbersome for such a case. The effective magnetic permeability for the rod is 4 - 10, so there will be several turns. Take into account the resistance of the throttle of the oscillatory circuit in the calculations. It may be necessary to increase the current of the transistor oscillator.
There is an alternative (without feedback winding), see below:2018-07-09_09-07-51.png
 

Attachments

Thread Starter

patpin

Joined Sep 15, 2012
401
I have been thinking about the calculated variations with LTSPICE with the .TEMP. In fact the load L3 is a coil. I guess that the magnetic field is proportional to the current and not to the voltage. So there will be very large variation in the magnetic field to be expected between -40 and +85°C and when I do the pickup of the signal produced by L3 (by means of L4 : not on diagram)(the L3-L4 forms a transformator with variable core e;g. a ferrite moving near the E shaped transformator core), I guess the picked-up signal will be proportional to the magnetic field and the distance of the ferrite... Any ideas how to stabelise the current?
 

Bordodynov

Joined May 20, 2015
3,177
But you're pulling out the voltage, not the current!
In this case, it is necessary to stabilize using feedback on the current. Also it will reduce the maximum power! If you want to stabilize the magnitude of the magnetic field, then it is possible, but this other generator circuit will be different. And I do not understand, your attachment to the transformer.
 

Thread Starter

patpin

Joined Sep 15, 2012
401
But you're pulling out the voltage, not the current!
In this case, it is necessary to stabilize using feedback on the current. Also it will reduce the maximum power! If you want to stabilize the magnitude of the magnetic field, then it is possible, but this other generator circuit will be different. And I do not understand, your attachment to the transformer.
The rotor of the motor distributor has small ferrite rods imbedded. Those have to be detected when they pass the very small (11mm high E ferrite) transformator. One of the windings on it is the L3 from the diagram, generating 600mH 12V sinus. Another winding detects the inducted signal. Detection rate can be up to 10.000 times X 6 (for 6 cylinders)/minute. Due to the winding of the transformator the signal at the pick up coil's ampitude becomes lower (1Volt) when the ferrite passes. This is detected by another circuit. So it is important that the peak amplitudes don't change a lot due to temperature since this will interfere with the passing of the ferrite rod.
How can one stabilize the current? Is there a simple way for this Hartley oscillator. Maybe with NTC/PTC's?
 
Last edited:

Bordodynov

Joined May 20, 2015
3,177
I do not know of a simple way to stabilize the current.Only complex ways come to mind.I present the circuit as follows:
1. a small resistor connected in series with the inductance (this is a current-voltage converter);
2. Amplifier on the operational amplifier;
3. Peak detector;
4. The filter;
5. Scheme of adjustment of the oscillator current;
6. Buffer amplifier.
 

Thread Starter

patpin

Joined Sep 15, 2012
401
I do not know of a simple way to stabilize the current.Only complex ways come to mind.I present the circuit as follows:
1. a small resistor connected in series with the inductance (this is a current-voltage converter);
2. Amplifier on the operational amplifier;
3. Peak detector;
4. The filter;
5. Scheme of adjustment of the oscillator current;
6. Buffer amplifier.
1. You mean L3?; how small?
2. Not present in my diagram; don't understand.. as the rest
5. this attrcat's me . How do U realise that
 

Bordodynov

Joined May 20, 2015
3,177
Is it possible to understand the mechanism as follows: The ferrite rods are periodically approaching the inductor, which causes a change in inductance?
Do you want to get the modulation frequency and control the rotation speed?
Until I understand what you want, I can not help you.But the current stabilization circuit will not be easy!
 

Thread Starter

patpin

Joined Sep 15, 2012
401
Is it possible to understand the mechanism as follows: The ferrite rods are periodically approaching the inductor, which causes a change in inductance?
Do you want to get the modulation frequency and control the rotation speed?
Until I understand what you want, I can not help you.But the current stabilization circuit will not be easy!
Yes the inductance changes when the ferrite rods comes near
I want the modulated freq. and then demodulate it and then fire the plugs at the time the ferrite rods go away. That part of the circuit is no problem.
 

Janis59

Joined Aug 21, 2017
1,834
Generally, the voltage instability may have at least three reasons.
First is instability of PS itself. Remedy - 78XX.
Second is thermal point of set instability. Bordoyunov is solved that brilliantly.
Third is - what the generation regime You are looking for? If mild, it is principially instable yet has good (low harmonics) form factor. Choosing a harder loop factor You get a hard generation regime what gives a stable voltage but rather bad form factor. Thus, its always the compromis between both mentioned, quality and quantity.
 

Thread Starter

patpin

Joined Sep 15, 2012
401
Generally, the voltage instability may have at least three reasons.
First is instability of PS itself. Remedy - 78XX.
Second is thermal point of set instability. Bordoyunov is solved that brilliantly.
Third is - what the generation regime You are looking for? If mild, it is principially instable yet has good (low harmonics) form factor. Choosing a harder loop factor You get a hard generation regime what gives a stable voltage but rather bad form factor. Thus, its always the compromis between both mentioned, quality and quantity.
PS is with LDO regulator at 8V.
Amplitude is more important than frequency in my case I suppose..
 

Bordodynov

Joined May 20, 2015
3,177
I need to know the inductance resistance of 60 μH, or even better the Q-factor at some frequency.This is required to determine an additional, series resistor with this inductance, and the value of the current through the inductance, which must be stabilized.A small resistor is needed to isolate a voltage proportional to the current through the inductance.Further, this voltage will be amplified, straightened and filtered.The resulting voltage is used to change the mode (and transconductance of the transistor) of the oscillator.
 

Thread Starter

patpin

Joined Sep 15, 2012
401
I need to know the inductance resistance of 60 μH, or even better the Q-factor at some frequency.This is required to determine an additional, series resistor with this inductance, and the value of the current through the inductance, which must be stabilized.A small resistor is needed to isolate a voltage proportional to the current through the inductance.Further, this voltage will be amplified, straightened and filtered.The resulting voltage is used to change the mode (and transconductance of the transistor) of the oscillator.
Sounds interesting. Resistance of the coil is 1.2 ohms. I possibly have to adjust the inductance, so it would be interesting to have the formulas used. Many thanks for yr help!
 

Thread Starter

patpin

Joined Sep 15, 2012
401
Post 36 it is a generator with a series resonance, and this is my attempt to get a stable current by a simple method.
My models:
http://bordodynov.ltwiki.org/
Looks very nice, very stable current I(L1) in #36 ... but since I am not an expert as you apparently are.... , can you explain yr voltage source B1 and V(I)?… and the functioning of yr current stabilization. Thanks fr yr time and effort!
 
Last edited:
Top