dc-35 MHz 50Ohm-4000 Ohm Transformer 1

[krytar]

Hi,

I need help to design a tranformer to match my 50 Ohm output from a I-Q mixer to a load impedance of 4,000 Ohm.

The conversion loss from my mixer is about 10.0 dB max (50 Ohm). Rf power is 0 dBm. LO is about +10 dBm.

From my calculation the voltage output at I & Q output is about 70 mV (or 14mA, which translate into 5.6V into 4KOhm). Use quaterwave transformer (N=1) I need a 31" long on alumina (or 98" in air) with the impedance of 447 Ohm.

This is not working in my design since the space is very limited. I have only about .250 X .250 X .075 to work with.

Any idea how to redhce the size of the transformer?

Thanks, Krytar

 

[biff44]

Do you have DC power for an op amp?

 

[krytar]

No,

I don't have DC power out or in.
The I-Q mixer is a combination of a power divider & a quadrature 90 degree hybrid & two mixers (all are lumped elements)

Thanks, krytar

 

[biff44]

Well, there is no way that I know of to impedance match down to DC without an active device, like an op amp or a transistor RF amp. Would you be contented with a good match from 4 to 35 MHz, and a poor match below 2 MHz? There are a number of articles by Ruthroth that talk about broadband transformers for RF. I have a book somewhere that I might be able to find.

 

[biff44]

One idea: use a rectifier diode to generate some DC voltage from the RF LO power. Then use that DC voltage to power a micropower op amp to do the matching. You can use 2 diodes and some caps in a voltage-doubling scheme to get more DC voltage.

 

[krytar]

Hi

I would be more than happy to take 4-35MHz transformer.

Thanks, Krytar

 

[biff44]

Let me look around. It is an age-old problem. Be advised that if either the source or load have big reactive components, Mr. Fano predicts limited bandwidth.

 

[GOTWW]

Maybe I am missing something here, but, generally moderate power levels of DC can easily be provided by means of an RF/IF/LO transmission line center conductor (coax/micro/cpw). But is not with conventional waveguide or the minimized planewave active optical systems. 3-octive matching networks are possible, I would investigate the minicircuits line, to, at least see if simular transformer devices are available. And if they are, then you will have some confidence that a simular unit can be fabricated in a resonable fashion.