Coupling transformer phasing in class C rf amplifier

Thread Starter

Elerion

Joined Sep 11, 2017
125
Hi everyone.

I built a simple amplifier to be used as a transmitter.
There's a transistor oscillator circuit, coupled via a transformer to a class C rf amplifier (30 uH choke at the collector, and 68 ohm resistor from base to ground). That's all.
rf_classC_phasing.png
It is soldered neatly. No protoboard. Working at 7 MHz.

For some reason, transformer's secondary phasing affects the circuit performance (in other words, swapping the secondary leads)
With secondary in phase with primary:
- I get more power.
- The waveforms at the base of the rf amplifier are "noisier".
- The collector voltage shows a second "peak" (this is later filtered by a pi filter, and the output to the load is sinusoidal).

Attached, real measruments of both pashings.
With some transistors, the power difference is small. For others, it is quite large.

Could anyone explain why the secondary phasing (which, by common sense, should not have an impact) changes the base and collector waveforms and usually gets higher outputs?

Thank you very much!
 

Attachments

Last edited by a moderator:

k7elp60

Joined Nov 4, 2008
562
Hi everyone.

I built a simple amplifier to be used as a transmitter.
There's a transistor oscillator circuit, coupled via a transformer to a class C rf amplifier (30 uH choke at the collector, and 68 ohm resistor from base to ground). That's all.
View attachment 172760
It is soldered neatly. No protoboard. Working at 7 MHz.

For some reason, transformer's secondary phasing affects the circuit performance (in other words, swapping the secondary leads)
With secondary in phase with primary:
- I get more power.
- The waveforms at the base of the rf amplifier are "noisier".
- The collector voltage shows a second "peak" (this is later filtered by a pi filter, and the output to the load is sinusoidal).

Attached, real measruments of both pashings.
With some transistors, the power difference is small. For others, it is quite large.


Could anyone explain why the secondary phasing (which, by common sense, should not have an impact) changes the base and collector waveforms and usually gets higher outputs?

Thank you very much!
I noticed that the second image has less distortion. Looking at the two images it is hard hard to tell which one has the greatest output as the scales are different.
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
it is hard hard to tell which one has the greatest output as the scales are different.
That really depends on the particular transistor used. In the case of the data I posted, the difference on the output is quite small. I chose those just to illustrate the waveform shapes.
On other cases, I measured two and even threefold difference in output amplitude (which is HUGE) for a circuit where the secondary phase should not have an impact, IMHO. This is why I'm asking, as I suspect that I'm missing some (maybe) non obvious issue here.
 

Bordodynov

Joined May 20, 2015
3,177
There is an error in your scheme. Not enough separation capacitor between the collector of the first transistor and the primary winding of the transformer. Now you have a power supply short to a common point. If your circuit corresponds to reality, then one of the inductances should be burned.
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
There is an error in your scheme.
Yes, I mixed by mistake two different setups.
L5 is actually the primary (now renamed Lp), not directly attached to the supply, but through a 220 ohm resistor. Sorry for the mistake.
This is the correct circuit from the experimental results

classC_corrected.png
 

Bordodynov

Joined May 20, 2015
3,177
Apparently between the inductance of the transformer and the inductance of 30 μH there is a parasitic connection. There is a parasitic capacitive coupling between the secondary winding and the inductance of 30 μH (collector capacity of the transistor). This capacitance has a variation that affects the phase relationships. Try to shield (or rotate 90 degrees) the inductance of 30 µH
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
There is a parasitic capacitive coupling between the secondary winding and the inductance of 30 μH (collector capacity of the transistor)
I tried to rotate, but didn't see any change.
I also tried to place the 30 μH inductance outside of the board, at a few centimeters, wired. This adds some parasitic inductance, but I think that is ok in this situation. Nothing changed either (or was not apparent).
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
Do you mean how it is built?
It is an iron powder toroid. Primary is wounded next to the secondary. 7:1 turns ratio.
The inductances shown in the schematic match real values closely.
 

Thread Starter

Elerion

Joined Sep 11, 2017
125
How are the windings made on the toroid?
Like the narrow band (page 2, Some practical aspects of Toroid) and the one attached.
EXCEPT I didn't overlap the secondary.
I wounded the secondary next to the primary (each of them "sitting" right on the toroid surface).
In the same sense.

EDIT: I attached an image of the actual toroid.

Very nice articles, by the way. Learned a few things from them.
 

Attachments

Last edited:

Thread Starter

Elerion

Joined Sep 11, 2017
125
Rewinding the toroid to look more like the schematic view attached in my last post did make a difference. There's still some output power and waveform shape difference, but much closer now.

Can anyone explain why such a difference between both ways of winding the toroid? Intuitively, parasitic capacitance should be higher now (as pri-sec are overlapped), but it works much better that way.
 
Top