Circuit analysis I need help

[Astrid]

I'm trying to understand the circuit in the picture, I need help.
The purpose of the 14MHz TTL square wave make 14MHz sine wave high power (current 3A).

What is clear about.

TTL step
C6 - blocking DC voltage 5V from 7404
R5 - make current over C6
L4 - I do not know, first of all I'm not sure what the bond L4 and C gate of Q1 ?

24V step
C1 - blocking DC coltage from V1 over L1
R2 - make current over C1 and L2, is not clear to me just has so little value?
C2,3- I do not know, you do?
L2 - Again I do not know and I do not see the link to the C gate Q3 1.86 nF?

Current from V1 are approximately 3A, I find it too much to drive Q3

High voltage step
C9-11 - Blockiing DC voltage from V2 over L6
R1 - load (Just an idea)
What is impedace resources in terms of load R1?

 

[chuckey]

You do not say what your problem is. You do not give the output voltage you require or get. In general the output stage should have a high impedance load ( 90µ H - good!), and then an impedance matching network which matches Vsupply/Isupply : 20 ohms. This can be either a PI or a T, or tapped tuned circuit. It will increase the voltage at the output.
Frank

 

[carpenter]

yes impedance is one of the issues. In my opinion, impedace L6 should be a major component impedace resources for load R1. Physically, the L1 and L6 identical component, toroid size T68-2A with 10 threads. Material toroid I could not identify, and I've not measured impedance (molded silicone). 90uH stated in the manufacturer's service documentation, and I quote (10 toroid turn 90uH), but more and more I tend to misprints and real impedance 9uH or rather 900nH.
However, in this case, it is also 9uH Q1 and then this excitation degrees will flow a large current because the C3 = C2 = 1nF 1nF (measured) and R2 = 25Ohm (in real 3x75ohm).

Voltage V2 is regulated 10-45V.
Between output C9-C11 and the load R1 is in reality a big LC low pass filter 9th-order.
The purpose of all this is from TTL 14MHz square wave low power do 14MHz sine wave with high power (0 - 90W).

 

[FvM]

Are you referring to additional information about the circuit or just guessing? In the former case, please share your knowledge.

My first question would be if it's a known working circuit or an unvalidated design idea?

 

[carpenter]

OK so what is certain

L4 - 560nH - Toroid T50 red/clear ==> material 2 10.75 turn =560nH in documentation write 10.75turn.
C6,C1,C9-C11 - 100nF 650V PP Film identified as described in the Part
R5 - 100k 1W identified as described in the Part and measured
Q1 IRF510 identified as described in the Part
R2 - in real 3x 75Ohm 1W parallel , measured and identified by color code
Q3 is
 IXFH12N50F identified as described in the Part
V1 is 24V
V2 is regulated power supply with output voltage 5 - 45V max. 3A
Load R1 on schematic, is in real inductor with measured impedance between Z= 18-40 Ohm
it's actually a heating element in which power is converted into heat
L7,L3 is ferrite core with 4 hole.
L5 is big inductor on toroid 680uH is write in documentation.

What is not 100%
L2 - in real 7 turn air inductor d=5mm l-22mm ,I calculated 70nH
C3,C2 - if disconnect L2 and measured capacitance over C2 is measured as 2nF I'm not sure if c1 do not measure over some parasitic capacitance. In real is SMD film or silvermica size 1812
L6, L1 - in real toroid T68-2A unknown material all white, 10.75 turn. I thing iron powder Carbonyl TH material 7 but this is white/clear and this toroid is full white.

Piece to which I have access works, I try to understand and possibly copied.
I wanted to do in Multisim SPICE model, but not for IXFH12N50F SPICE model

Edit:
I make any simulation and first step (circuit in gate Q1 IRF510) works as expected. If 74HC04 produce square wave with Duty cycle 65% (gren) is on gate Q1 nearly sinus wave (red)

 

[chuckey]

1. R1 MUST resonate with C9/10/11, to provide a real resistance.
2. output power = 90W into 20 ohms , V^2 = 90 X 20 = 1800, V ~45V.
3 So you do not need a matching network, just feed the drain via a RFC with a impedance 10 X 20 = 200 ohms. 200 = 2PI L 14 X 10 ^6, L = 200/2 X PI X 14 micro henrys ~1.3 micro henrys. 90 micro henrys might have a self resonance frequency below 14 MHZ.
4. What purpose does R2 provide? It reduces the gain of the first transistor to virtually zero.

This is a RF amplifier, the use of multiple inductors in the drain circuits is a waste, the use of 100 MF "decoupling" capacitors will not work in practice as they will be highly inductive. Just use 100 or 10 NF caps.

Get hold of the AARL handbook or the RSGB handbook, this is an amplifier operating in the amateur radio band and these books will give you an insight into the techniques used.
Frank

 

[carpenter]

In short
1. As I have said, between C9-C11 and R1 is in real LPF 9th order with relatively big toroid T150,
2. Output power P=U^2/R P=(45*45)/20=101W . R1 is in real any inductor of feromagnetics material, the characteristics of temperature changes see Curie temperature. Z=20 is measuren on one piece on room temperature.
3. Why 10 x impedance load? What material do you recommend for the coil? How about T68-7
5. Yes, use Electrolytic capacitor C4 =10uF and C7 = 100uF also surprised me, but actually there are. My choice is prop up it with 330nF Film and 100nF ceramics.

The others look tonight.
In any case, thanks.

 

[carpenter]

I had no choice than to disassemble the device again and re-measure the coil
Here is the result


I remove and measure L6 =12.6uH size is T68A with 10turn, unfortunately I do not know materiál, color full white
I draw whole output LC stage L9 - C16 , remove and measure L12 =420nH , L9,10,11 is calculated from number of turn and core size/material T130-6
C2,C13,C14,C15,C16 in in real many C parallel.
In simulating (Netvork analyzer) betwen point A and B , this is not LPF (as I originally thought) , but tuned circuit.
Partly, I measured the load and in coud have L=420nH +R=1.3, L decreases with temperature, on the Curie temperature is a step change
I quite said that Q2 operates as an amplifier in class E, but it was missing C between D and S Q2.
Although we moved, but we still did not understand the whole structure

Edit : Repair picture
+ add guestion
C in output stage is SMD Silver Mica. Buy smd silver mica is quite a challenge, price € 1-2 / ks.
How far it is possible to replace the ceramic NP0?

 

[chuckey]

Your attachment is invalid - it will not open.
Do you know what resonance is?, do you know what impedance transformation is?, have you looked at the books I have suggested? Until you get some basic RF facts under your belt, its going to be a long design time.
Frank