LC filter can by as 10KOhm resistance?
is it possible to replace 10KOhm resistor w LC parallel or serial filter or any filter
that will be the same as if this 10 KOhm resistor?
i'm using 10KOhmic res and i would like to achive the same effect w/o using resistance that brings me loses!
but using filter at 15MHz freq
Thanks a lot,
Arye
At resonance in parallel LC tank the impedance rises to infinity (ideal) and in serial arrangement the impedance drops to zero (ideal), so if you need 10kohm value and still maintain the filter integrity the only way is to introduce a true resistance of 10kohm ..
Otherwise remove the LC tank and replace it with an inductor that represents an equivalent resistance of 10kohm at 15MHz (low-pass) R[L]=wL,
or
a capacitor that represents an equivalent resistance of 10kohm at 15MHz (high-pass) R[C]=1/wC
IanP
Thanks a lot
if the filter will give more then 10KOhm its great!
i tried one inductor it doesnt stop the signal enoght cause i have a limitation on the inductor to be around 0.7uH, this why i wanted to filter it w parallel tank.
and i even tried LC parallel tank w 200 pF,
yet it didn't do the same effect as if i put 10KOhm resistor instead of the tank.
could you please advice for that?
Thanks a lot,
Arye
How did you connect the tank? In series to your circuit while L and C of the tank in parallell? You may also need to tune the tank frequency and get the resonant ferquency at 15 MHz, which is a dip in s21 if you measure it.
exactly thats how i connected it,
& it didn't block the 15Mhz as 10KOhm did.
As post #4 states you need to tune the LC to your desired frequency. Try a tunable coil and / or tunable capacitor (trimmer).
It's typical to make the coil value many times the capacitor value. That way you get greater voltage swing within the tank circuit. 100 times is a common ratio. Consider making the capacitor a few pF, then calculate a coil value using the formula for frequency of an LC network.
There are also bandpass filters (or notch filters, depending how you configure it) which use only capacitors. These need tuning as well. However you may have an easier time changing capacitors rather than coils, depending on what you have in your parts bin.
May be I missed something - however, did one of you ever hear about phase shift caused by L or C?
It would be interesting to hear how somebody realizes a resistive part without a resistor.
Bradtherad, may I ask you to tell me how you are able to build a bandpass or notch using only capacitors? You should write a paper about it.
Thanks a lot,
so what you say is that it is possible, it strange to me that it didn't work i put 0.7uH and 200pF and it didn't stop the signal the same as 10Kohm did... it is strange for me
this is the practical circuit simulation:
http://images.elektroda.net/61_1318885345.jpg
Tnx
Arye
Some parts of my former answer sound a bit ironic - sorry for that.
But you should consider that in many cases it is the purpose of a resistor to create losses (RC or RL combinations).
And any LC combination without a resistive part can never behave like a resistor (due to phase shifts)!
Therefore, it would be interesting to know: What is the purpose of the resistor under discussion?
Only then one can decide if it can be removed or replaced by any other part.
It combines a high-pass and a low-pass. Each is capacitive.
I found one configuration at this webpage (electronics-tutorials.ws):
Band Pass Filter
Another configuration here (this tutorial is duplicated many times on the internet):
Band-pass filters : FILTERS
There's the twin-T notch filter using capacitors. (Scroll 50% of the way down):
Lessons In Electric Circuits -- Volume II (AC) - Chapter 8
I don't know for certain whether they'll suit the OP's needs.
Hi BradtheRad,
sorry, but I can detect some resistors in all of the circuits you have provided. So - what about filters "using only capacitors" as announced by you?
Unfortunately, the problem is rather vaguely specified. So you can't expect exact answers. An ideal parallel resonant circuit will achieve infinite impedance at only at the resonance frequency. Furthermore, a real circuit has limited Q and thus below inifinite resonance impedance.
In a previous post, you're adressing a 15 - 20 MHz frequency range. https://www.edaboard.com/thread227711.html
This would rather suggest a "bias T", an inductor or "choke" of sufficient high impedance. For th epresent problem, one would think of a 50 - 100 uH inductor, that achieves a self-resonant-frequency around 15 MHz without adding a capacitor.
My apologies for being ambiguous. I meant that the filters use 'only capacitors' in the sense that they don't need inductors.
The OP is probably on the right track by using a coil and capacitor.
---------- Post added at 03:50 ---------- Previous post was at 03:15 ----------
---------- Post added at 03:57 ---------- Previous post was at 03:50 ----------
There's a chance that the coil and cap each must be selected separately, based on resistance in surrounding circuitry. Because the rolloff curve for the coil has to start its drop at 15 Mhz. And similarly the rolloff curve for the cap has to start at 15 Mhz.
Both curves are based on resistance in surrounding circuitry, in addition to the component values.
The reason is that the coil and cap are not functioning as a tank circuit (despite the fact that they form a tank circuit). The coil is there to pass all frequencies below 15 Mhz. The cap passes frequencies above 15 Mhz. You must choose each for the value that will do this. You don't want a great deal of overlap of the rolloff curves. Nor do you want an immense abyss between them.
Will it require a cap and coil that resonate at 15 Mhz? I suppose so. However not every LC combination that resonates at 15 Mhz will do the job. (This means I was premature in recommending a 100:1 ratio above.)
As an alternative to a parallel LC filter, it might also pay to experiment with combinations of series LC, to drain away the undesired signal. Connecting it across the load, or connecting one end to ground and the other to your input, etc. As long as you can tune it to pass 15 MHz.
Another possibility. I believe there's such a thing as a tunable wavetrap. It's the sort of filter used by radio hams to filter out interference. Either in their own equipment, or in a neighbor's equipment picking up unwanted signals from the ham's antenna. And 15 Mhz is in the short wave spectrum, the range for such a device to be used.
Ohh, that`s an old story - back to the early fifties (transistors and capacitors in Sallen&Key topologies).
However, the original question was if there is any other part/circuitry that behaves like a resistor. And - I am afraid - an active filter cannot fulfill such a task.
Thanks a alot for all your answers
the picture that i put above is showing the problem the left coil is like antenna
and the right part is an impedance that shouldn't receive the 20volts of 15Mhz signal
so i try to stop it, to block the 20 lt signal -> 10Khohm resister did it
other citcuits like LC didn't help me.
so for now i'm stuck w it...
---------- Post added at 18:51 ---------- Previous post was at 18:41 ----------
LvW - i asked about resistor cause this what helped me to stop the signal even though i dont want to use resistor.
BradtheRad - i used 0.7 uH and 200pF it is more then 1:100 ratio isn't it, and still it didn't blovk the signal
so you think that it possible to return this 20 volt signal back to the antenna without losses i mean without using resistor.
maybe some more complicated filtering than LC tank,
Tnx very much for the discution, i read it few time and try to extract the knolwage,
Arye
To illustrate the unsuitable LC dimenioning of your circuit, see the below simulation output.
The upper curve shows the achieved signal attenuation with a single 0.7 uH/200pF/Q=50 bandstop, the lower with
50 uH/2.2pF/Q=50
Input voltage is 20V and load 50 ohm in both cases.
The problem is, that the LC filters characteristic impedance is √L/C, so your seemingly large L/C ratio refers to an impedance of only 17 ohm. This value determines the attenuation at slightly off-resonance frequencies. So a much larger ratio is in fact necessary.
As I previously mentioned (yet unnoticed by you), a 50 to 100 uH would be just fine, even without a capacitor, just utilizing the self resonace frequency. The ideal 50 uH inductor has an impedance of about 5k at 15 MHz.
Thanks,
the problem is that in this university final project
i'm not able to increase the L more than 0.7uH
i'm having this coil antenna recieving 15Mhz signal & in parallel there is some load
that i want to prevent the 15Mhz signal to achive the load,
i want the signal "see" the load & go back to the antenna
but puting more than 0.7uH, hurming the load,
thats there i'm stuck
Thanks
If the frequency is actuall 15 MHz, your LC circuit is considerably off-resonancy. You can tune C correctly as a first step. But I also assume, that the inductor has limited Q, my assumption of Q=50 may be already too optimistic. So tuning won't bring you far. Or the frequency is 13.56 and you only didn't tell the exact value for some reason?
Referring to your requirement to achieve a certain isolation (e.g. equivalent to 10k series impedance), there's no way with a 0.7 uH inductor. But I must confess that I didn't understand your considerations about maximum inductance. It may be the case, that you didn't explain your intentions completely. At least I didn't get it. I guess, it would be much clearer if you describe the function of the circuit referring to commonly known RFID problems.
P.S.: Getting a full description of the problem would also allow to understand, if other filter topologies than a series connected two-terminal component come into consideration.
Thanks,
Yes t is sort of RFID coil though i thought it doesn't metter.
the thing is that if i connect just inductor (more than 0.7uH) between the antenna and the paralel load,
the inductor killing the load proper operation!
thats why i didn't increase the inductor and only change the capacitor,
but what you say i may think that if i now increase the inductor and add a suitable capacitor in parallel (for being around 15Meg or 13.56Meg whatever) maybe now i won't kill the parallel load operation,? do you think? and if yes you think i need to use 50uH inductor?
different filter topologies could you please adice for some?
Thanks so much