微波EDA网,见证研发工程师的成长!
首页 > 研发问答 > 微波和射频技术 > 天线设计和射频技术 > LC filter changing characteristics on different impedance

LC filter changing characteristics on different impedance

时间:04-06 整理:3721RD 点击:
Hello I have used rfsim99 to make a simple series bandpass filter consisting of one capacitor and one inductor in series.
On 50 ohms input and output the filter behaves as desited. On other impedances the behaviour totally changes.

The problem is that this filter may be used with antennas or transceivers (homemade) of unknown impedances.
Sould I try to match these devices to the filter and how?
If my antenna is 50 ohms and the transceiver s say 300 ohms would the filter have the expected behaviour (with some loss) on 50 ohms?

It was long time ago I played with RF filters. Usually an LC circuit can be designed to work as a filter and a matching circuit as well (I meant at the same time).

Yes some tapped transformers on the I/O can match the filter, but I do not know the impedance of the circuits it could be used. If the transitter for example has not 50 ohms impedance the filter has different responce!

The resonance frequency won't change with real load and source impedances, but the bandwidth (quality factor) does. That's by nature of passive filters.

The filter effect may be reduced or even lost, depending on the impedance. But it will still pass signals of the center frequency unattenuated, if tuned correctly. In so far it should do no harm.

Yes, of course the source and load impedances will affect the response. When changing the load impedance without retuning the filter, you're going to get changes in behavior. Those impedances are, in fact, part of the filter. If you want to keep performance consistent, then you'll need some kind of buffer (but even that will probably not be very consistent).

Right, so you basically use an input and an output buffer, to try to keep the impedances "seen" by the filter stable. Ok, of course you then move the problem to the I/O impedances of the buffers, but this is then translated as a power loss, not a Q degradation.
Am I getting this right?

You are right.
First we consider knowing the optimum impedance of the transmitter load.
Then the antenna of interest usually has nominal impedance (ideally, just a resistance) over a certain bandwidth.
Knowing the load and its impedance that should be seen by the transmitter at the input of the filter, the best practical configuration of the filter could be selected. Then calculating the LC elements could be done rather straightforward (if you find the right reference, I hope others can help you find it).

Talking about impedance, is that easy to calculate the impedance at the collector of Q2 in this circuit? TINY TORNADO QRP TRANSCEIVER
I would like to replace the output filter and C6 with my series bandpass filter designed for 50 ohm I/O

Hi,

As you may have already noticed the original filter is symetrical. We can say that the optimum load impedance is almost equal to the antenna one.
But both impedances (resistances if you like) are unknown (I mean for sure).

On this circuit, the optimum load impedance is proportinal to the RF Vmax/Imax (it depends on the amplification class of Q2).
Vmax is approximately equal to the DC voltage at Q2 collector.
Imax is related to Q2 characteristics (maximum ratings;current and power) and its bias (RF current gain).

Sorry for not being able telling you more because my actual work and since very long is no more related to RF, though my master degree was in communications :)

Kerim

try adding a shunt capacitor at the input, and another one at the output of the filter. That will stabilize the load impedance effect somewhat. But the bandwidth will shrink up a little.

I think you mean my series filter right?
I have tried it, here it is.
The response was a little bit sharper especially on the higher bands, which is desirable. I have also tried to change the I/O loads with this configuration it seems better than before, but when the transistor impedance went to something like 500R and the antenna was at 50R the attenuation was much, but the response was relatively stable.

By adding more capacitance (eg 1nF) the bandwidth was much narrower, that's good! But the attenuation was much more. maybe if this is combined with a narrowband amplifier for the band of interest it will be of some use.

By the way, can I add a varicap in place of the series capacitor, to vary the filter response? for example a bb212. It may not be able to tolerate more than 1W if this filter is used for transmitting?
how should the varicap be connected then?

Copyright © 2017-2020 微波EDA网 版权所有

网站地图

Top