Home |
Search |
Today's Posts |
#1
|
|||
|
|||
Receiver antenna/amplifier matching question
I'm trying to figure something out here...
Assume that I have a 2m antenna that (through whatever means) happens to have an impedance in the ballpark of 50ohms. Also assume that this antenna will generate 1uV open-circuit. Assume that I have a MMIC amplifier that has FET inputs and therefore has a high input impedance... let's say it's 500ohms. If I directly connect the antenna to the amplifier, the amplifier input effectively sees ~0.909uV and then amplifies it. If I build a matching network to make the amplifier "appear" as 50ohms, at the input of the matcing network I'll have 0.5uV, but since energy has to be conserved (my matching network is assumed to be lossless), (0.5uV)^2/50=Vamp^2/500 -- Vamp=1.58uV. This is a gain of 20log(1.56/0.909)=4.8dB over a direct connection. Is that all correct? Now let's go down to HF. My antenna is still 50ohms, but at such low frequencies parasitics aren't that bad and I just use a JFET or MOSFET and have a 5kohm input impedance. In this case, without the matching network, the amplifier sees 0.99uV. With the matching network, it sees 5uV (!), a gain of 14.1dB over a direction connection. Yet many designs for HF/shortwave antenna amplifiers don't bother with a matching network, just feeding the antenna directly into a FET. So why is this? Is the extra gain just not needed on HF? Or too unwiedly to build when you're trying to cover everything from 1MHz-30MHz? Thanks for the help, ---Joel Koltner |
#2
|
|||
|
|||
Receiver antenna/amplifier matching question
On Mar 17, 10:16*am, "Joel Koltner"
wrote: I'm trying to figure something out here... Assume that I have a 2m antenna that (through whatever means) happens to have an impedance in the ballpark of 50ohms. *Also assume that this antenna will generate 1uV open-circuit. Assume that I have a MMIC amplifier that has FET inputs and therefore has a high input impedance... let's say it's 500ohms. If I directly connect the antenna to the amplifier, the amplifier input effectively sees ~0.909uV and then amplifies it. If I build a matching network to make the amplifier "appear" as 50ohms, at the input of the matcing network I'll have 0.5uV, but since energy has to be conserved (my matching network is assumed to be lossless), (0.5uV)^2/50=Vamp^2/500 -- Vamp=1.58uV. *This is a gain of 20log(1.56/0.909)=4.8dB over a direct connection. Is that all correct? Now let's go down to HF. *My antenna is still 50ohms, but at such low frequencies parasitics aren't that bad and I just use a JFET or MOSFET and have a 5kohm input impedance. *In this case, without the matching network, the amplifier sees 0.99uV. *With the matching network, it sees 5uV (!), a gain of 14.1dB over a direction connection. Yet many designs for HF/shortwave antenna amplifiers don't bother with a matching network, just feeding the antenna directly into a FET. *So why is this? *Is the extra gain just not needed on HF? *Or too unwiedly to build when you're trying to cover everything from 1MHz-30MHz? Thanks for the help, ---Joel Koltner You didn't mention noise figure, and that's important. At UHF and above, atmospheric noise is typically very low, and you benefit from a really low noise figure. In that case, you want to match the source to the receiver input in such a way as to achieve the best noise figure, which in general is NOT the same as getting maximum power transfer into the receiver. Data sheets for parts intended as input receiver amplifiers generally have a graph and/or table of optimal source impedance for lowest noise figure versus frequency. Some will have noise contours on a graph for some critical frequency or frequencies. At HF, atmospheric noise is generally very high, and you also have to put up typically with some very large signals. So the design of the receiver input is driven by different needs. You want to keep distortion low, to avoid problems from those large signals, and you want to achieve a "reasonable" noise figure -- but reasonable might well be 15dB, an absolutely terrible noise figure for a 440MHz receiver front end. And you might well put in a pad you can switch in to lower distortion, and out for better noise figure. Also, in some cases such as a car radio, the antenna impedance isn't anything close to 50 ohms (for the AM broadcast band). The design there would of course be different than if you were dealing with a 50 ohm source. Cheers, Tom |
#3
|
|||
|
|||
Receiver antenna/amplifier matching question
Hi Tom,
That's a most informative post; thanks. At work we have a design that uses these: http://datasheets.maxim-ic.com/en/ds...71-MAX2373.pdf -- right after an antenna and a matching network. Too bad that Maxim didn't choose to provide any clues on what antenna impedance it would like to see for the lowest nose -- I just matched it to 50ohms. (And truth be told, we have a slightly wretched antenna to work with anyway... the main reason it's workable at all is due to the antenna designer -- and outside firm -- sticking a 160ohm resistor in parallel across the input leads.) ---Joel |
#4
|
|||
|
|||
Receiver antenna/amplifier matching question
"Joel Koltner" wrote in
: Yet many designs for HF/shortwave antenna amplifiers don't bother with a matching network, just feeding the antenna directly into a FET. So why is this? Is the extra gain just not needed on HF? Or too unwiedly to build when you're trying to cover everything from 1MHz-30MHz? Joel, you might hope that a good receiver provides some degree of band pass filtering prior to the RF amplifier for IMD protection. For best receiver weak signal performance at SHF, you would usually match for optimum noise figure rather than best front end gain. At HF, receiver internal noise is usually lower than ambient noise, so receiver NF is less an issue, but... intermodulation performance is very important. "Noise" can be created in the RF amp due to intermodulation distortion, so designs often focus on optimisation of IMD rather than NF directly, or front end gain. For example, I note that the TS2000 does not use a high performance signal FET for the HF RF amp, but uses a couple of power FETs. The 144MHz case you mention is somewhere between the two, ambient noise in the city is usually much higher than a state of the art LNA, and often higher than a modern transceiver with a NF of 5dB. In my experience, IMD noise is a bigger issue on 144MHz than NF measurements made in a shielded room. A high gain amplifier with poor NF or high IMD noise delivers a high volume low S/N output from the receiver. (Try an IC706IIG on a discone antenna in the city.) Owen |
#5
|
|||
|
|||
Receiver antenna/amplifier matching question
Joel Koltner wrote:
Assume that I have a MMIC amplifier that has FET inputs and therefore has a high input impedance... let's say it's 500ohms. I don't consider 500 Ohms to be "high impedance". I was taught that MMICs were 50 Ohms in and out. |
#6
|
|||
|
|||
Receiver antenna/amplifier matching question
"dave" wrote in message
... I don't consider 500 Ohms to be "high impedance". Compared to 50 ohms it's high. :-) Granted, this is a bit subjective, but in my mind a "high impedance RF probe" is often 500 ohms (a 450 ohm resistor + the 50 ohm terminator in a scope"), and the implication is that it doesn't significantly load down whatever it is that you're probing. (Note that a "1Mohm in parallel with 10pF" "high impedance" probe ends up being ~110ohm by the time you hit 144MHz...) I was taught that MMICs were 50 Ohms in and out. True, most are reasonably close in the bands they were designed for. I probably shouldn't have said "MMIC amplifier" there -- just plain "amplifier." |
#7
|
|||
|
|||
Receiver antenna/amplifier matching question
"Joel Koltner" wrote in message ... I'm trying to figure something out here... Assume that I have a 2m antenna that (through whatever means) happens to have an impedance in the ballpark of 50ohms. Also assume that this antenna will generate 1uV open-circuit. Assume that I have a MMIC amplifier that has FET inputs and therefore has a high input impedance... let's say it's 500ohms. If I directly connect the antenna to the amplifier, the amplifier input effectively sees ~0.909uV and then amplifies it. If I build a matching network to make the amplifier "appear" as 50ohms, at the input of the matcing network I'll have 0.5uV, but since energy has to be conserved (my matching network is assumed to be lossless), (0.5uV)^2/50=Vamp^2/500 -- Vamp=1.58uV. This is a gain of 20log(1.56/0.909)=4.8dB over a direct connection. Is that all correct? Now let's go down to HF. My antenna is still 50ohms, but at such low frequencies parasitics aren't that bad and I just use a JFET or MOSFET and have a 5kohm input impedance. In this case, without the matching network, the amplifier sees 0.99uV. With the matching network, it sees 5uV (!), a gain of 14.1dB over a direction connection. Yet many designs for HF/shortwave antenna amplifiers don't bother with a matching network, just feeding the antenna directly into a FET. So why is this? Is the extra gain just not needed on HF? Or too unwiedly to build when you're trying to cover everything from 1MHz-30MHz? Thanks for the help, ---Joel Koltner Receiving antenna matching is the same as a transmitter loss, except in reverse. As you said if you start with a 50 ohm antenna and transmission line and feed it into a 500 ohm receiver you will have a 10:1 swr and the additional loss of the transmission line will come into effect. If it is very short , it will not mater too much . If the line is long it will. Above around 140 mhz the noise is low enough you need to use a low loss line of a very short length. The noise figure is most important. Usually there is a mismatch when tuned for lowest noise figure. Short line , no problem, long or lossey line will cause much signal loss. On the lower frequencies it usually does not mater too much what you do as the noise comming in from the antenna will usually be the major factor. |
Reply |
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Matching Coax Impedance: To Receiver or To Antenna ? | Shortwave | |||
Matching Coax Impedance: To Receiver or To Antenna ? | Antenna | |||
Antenna impedance matching - old Grundig tube receiver | Shortwave | |||
Mobile Antenna Matching Question | Antenna | |||
Antenna amplifier question | Scanner |