wideband HF transformers

EDAboard.com Forum Index -> Professional Hardware and Electronics Design

Author

Message

2000

Joined: 14 May 2001
Posts: 89

10 Jun 2002 0:35 wideband HF transformers

hello friends,

I'm actually wroking on a 2.5W HF power amplifier using the 2sc2166 transistor in a A class topology. The output power is about 2.5W. The problem is... to match the output impedance of this stage to a 50 Ohm load, I'm supposed to use a 1:4 (Z) transformer and that should mean that the voltage in the secondary is supposed to be the double the input voltage in the primary of the tranformer... but its simply the same!!! what is going on? I simply cant work in the A class but I'm being forced in the AB class... the amplitude of the output signal is too large and the transformer is unable to low it in the primary...
Anybody could help me? any interesting site about the topic?

regards,

2000

RF_Router

Joined: 01 Feb 2002
Posts: 47

10 Jun 2002 8:10

You can also use instead of a trasformer just a L match. But now I was thinking about it and maybe it gets to the P1db of your trasformer and it can not do more than that.

unkarc

Joined: 28 Apr 2001
Posts: 280
Helped: 36

15 Jun 2002 14:50 food for thoughts

Hi 2000

Here are some food for thinking over your power amplifier (PA) design.

Let's approach the more or less optimum load resistance of your PA with the 2.5W output power (here I assume you use 12VDC supply for your CB transistor and let's neglect the collector saturation voltage of the 2SC transistor):

RL=(Vc^2)/2Pout=(12*12)/(2*2.5)=144/5=28.8 ohms

Question arises: why do you need the 1:4 (Z) transformer to match?
Instead, you may use approximately 1:1.66 (Z) transformer to match. How can you make such a transformer? The best approximation seems to be using 3 turns at the collector side and 4 turns at the output side. [3:4 turn-ratio, if you make 4 turns bifilarly and unwind 1 turn from one of wires, then you approached it very well;; you should use a toroid with a sufficient permeability so that the 3 turns' inductive reactance (2*pi*f*L) should be at least 4 times of the 30 ohms at your working frequency!]

I cannot comment your finding of the voltage values to be the same at the secondary and the primary windings, it is against common practice.

Can you give some more details on your PA? Collector DC current and voltage, working frequency, the transformer turns, core material or manufactured impedance transformer etc?

rgds, unkarc

eht

Joined: 14 Jun 2002
Posts: 2

15 Jun 2002 19:04

Is the voltage ratio the same across the whole band? Maybe it's core loss - depends on what kind of core you're using. Secondly if your driver output impedance is higher than you think and you're multplying 1:4 your real output impedance may be much higher than the input impedance of whatever you're using to measure the voltage (I assume 50 ohm) - so the loading of the measuring instrument will drop the voltage.

2000

Joined: 14 May 2001
Posts: 89

17 Jun 2002 15:27

hello friends,

thanks for the repplys... I´m uploading the pdf of the circuit... please take a look and see if you can help me. Thanks unkarc, I´ve already found this numbers, but the winds realtions is not working... I see the Vrms at the transistor collector, and for the load that is being used the power should be there but the load is just COOOOOLLLLL. I guess all the power is being dissipated in the 2sc2166 that is REAAALLLLY hot... I´ve already tried taking the secondary of transformer out and connected a load resistor to ground but does´nt work either...
please any help is welcome...

regards,

2000

RF_Router

Joined: 01 Feb 2002
Posts: 47

17 Jun 2002 18:28

Where have you loaded your pdf?

Rayengine

Joined: 18 May 2001
Posts: 314
Helped: 3

19 Jan 2003 7:56 more information

Please find three application notes from Philips and Motorla which talking about HF transformer, hope these can be useful.

TomAss

Joined: 18 Jul 2001
Posts: 20

19 Jan 2003 12:48

Hi there,

i frequently used transmission line transformers (1 to 9 and 1 to 4) and i never had problems. Basic guidelines for design those devices can be found, as RayEngine said, on Motorola AN749 and Philips ECO6907, ECO7703, ECO7213. I have all of these, but not in electronic format, so, if someone is interested, i could scan and "PDFize" them (mail me).
In my experience, transmission line transformers, provide large bandwidths (up to 3-4 octaves but my designs cover the 88-108 MHz band) and with appropriate core selection, low loss.
Remeber always that power transferring in rf transformers is not through the core, that is the core is only useful to rise the inductance of windings at lowest working frequency. For this reason, you might use a coax wound on the core, then arrange the shield and center conductor connections to make the 4:1 transformation you need (look at the sketch).
For a 4:1 transformer you can use a 25 ohm coax or two 50 ohm coax
in parallel. Higher impedances will result in poor bandwidth.
Last consideration is that you 'll have to cancel the primary and secondary winding inductances, compensating them with capacitors in parallel to the primary and secondary side. Often, if your active device has a capacitive reactance at its ouput (in the frequency band of your interest), the primary winding inductance is designed to cancel exactly the capacitive reactance of mosfet or transistor output.
For a deeper sight in the world of transformers, i 'll, if someone is interested, scan those very interesting application notes.

Bye
Tomass

Sorry, but you need login in to view this attachment

unkarc

Joined: 28 Apr 2001
Posts: 280
Helped: 36

20 Jan 2003 13:04

Hi All,

The application notes TomAss mentioned can be freely downloaded from the philips website. See:
http://www.semiconductors.philips.com/acrobat/applicationnotes/ECO6907.pdf
http://www.semiconductors.philips.com/acrobat/applicationnotes/ECO7213.pdf
www.semiconductors.philips.com/acrobat/applicationnotes/ECO7703.pdf
See also
http://www.semiconductors.philips.com/acrobat/applicationnotes/NCO8602.pdf which is also useful.

There is another useful file on designing power amps at
http://homepage.tinet.ie/~yellowbeard/power%20amplifier%20design.pdf

The Motorola AN749 file is available at
http://e-www.motorola.com/brdata/PDFDB/docs/AN749.pdf
See also other Motorola notes on this topic:
http://e-www.motorola.com/brdata/PDFDB/docs/AN721.pdf
http://e-www.motorola.com/brdata/PDFDB/docs/AN1034.pdf

rgds, unkarc

PS: I corrected the last link problem with the last two url's address that were linked together by typing mistake.

Last edited by unkarc on 21 Jan 2003 12:15; edited 2 times in total

g579

Joined: 01 Jan 1970
Posts: 455
Helped: 14

20 Jan 2003 22:06

unkarc, you are great!
Thanks for the links collection, all collected together.
Ps. Caution! The last link contains two links together!

designtech

Joined: 09 Feb 2006
Posts: 52
Helped: 2

03 Nov 2007 6:56 Re: wideband HF transformers

thanks nice notes

byteptr

Joined: 29 Apr 2008
Posts: 61
Helped: 9
Location: France

06 May 2008 18:39 Re: wideband HF transformers

May be this help you making ""wideband-tuned"" transformers.

Moderation:

Deleted, upload available here : http://www.arrl.org/qex/2005/qx3Trask.pdf

Good luck

byteptr

Joined: 29 Apr 2008
Posts: 61
Helped: 9
Location: France

07 May 2008 14:37 Re: wideband HF transformers

Hi,

Here is another application notes, but "linked", not uploaded this time, wich may be usefull for all.

http://www.noding.com/la8ak/12345/n12.htm

khajamoiz

Joined: 17 May 2008
Posts: 11

18 May 2008 10:37 wideband HF transformers

--------------------------------------------------------------------------------

Hi,

Here is another application notes, but "linked", not uploaded this time, wich may be usefull for all.

Added after 2 minutes:

hi
can u pls send the desgine of wideband hf transformers

Added after 2 minutes:

hi what is the meaning of wideband hf transformers

EDAboard.com Forum Index -> Professional Hardware and Electronics Design

Page 1 of 1

All times are GMT + 2 Hours

Abuse
Administrator
Moderators

sitemap