High dynamic range preamp in direct conversion wideband receiver
Because the receiver is very broadband, it is not practical to use "band" filters. So I am thinking of using a high dynamic range RF preamplifier and a variable rf attenuator (just a potentiometer) in it's input.
I was thinking of the pga-103+. It is specified for >50MHz for 50R but I found it to work fine on MW and less as well http://qrp.gr/thf7aa/index.htm the tiny antenna of the handheld is behaving as capacitive and should not be 50R anyway at these low frequencies.
I will use 1uF for the input and output coupling caps and for the inductor shunt capacitor in the PGA103+. But I do not know what should the value of the inductor be for operation down to 3KHz or so? Shall I use a 3.3mH choke?
Also do you think that significant IMD products will arise with this high dynamic range preamp combined with the small antenna and the variable RF attenuator in place?
The PGA103 is a poor choice, you would do better with a MMIC amplifier like the BGA2818. They are MUCH cheaper(~0.5 euros) run from a single 3.3V supply and have around 30dB gain from DC to >2GHz. The 50R output impedance will directly match your mixer. I wouldn't worry too much about input impedance, the antenna will probably vary from a few Ohms to a few K over the frequency range. A 3.3mH choke will be as good as anything over that frequency range. A potentiometer will not work well as the frequency increases but try it and see.
Yes, unless you live in a very isolated location I would anticipate overload problems!
Brian.
Thanks, I will check out the BGA2818. It has a higher NF and a lower IP3 though, so not a particularly high dynamic range... (?) In that sense it could not handle both the small and the large signals without distorting (the way I see it)
Won't the input attenuator help in reducing IMD from large signals? I am thinking of it like this: a 30db gain RF preamp with an input attenuator. Say a large signal is present at the input and also a small one. The large signal produces unwanted distortion. Then the RF attenuator can be used for attenuating ALL signals to say -20db from their initial level. IMD now is not present from that large signal but both signals are attenuated by 20dB. But there will be still headroom for 10 DB amplification of both signald, because of the rf preamp.
What do you think?
Remember, I am talking about rubber duck style antennas for up to a few KHz reception. purely capacitive operation at most.
Hey, I just noticed this BGA2818 does not require an inductor, true?
The BGA2818 does not require an inductor and that is a great thing!. However I cannot find it cheap in this auction site. Would that be possible to buy from you and post me a few (if you have any) just to try them?
Please PM me if so.
I can get them at a silly low price but I will be hit by minimum order charges. You can see if they will supply locally at: https://export.farnell.com/welcome-greece then put 'mmic' in the search box.
I do have some older devices here which you are welcome to a few of, they are Avantek MSA0135 MMICs, DC to 1.2GHz with 19dB gain. They are in 'micro-X' packages, four wired ceramic dots, two are ground, one is an input and one is the output and supply. They are lower gain but are cascadable and unconditionally stable with no inductors. I can send you some if you like. Be careful, they are tiny and fragile!
Brian.
Ok Brian my friend will do an order at the end of the week so I asked him to order these as well. If it does not turn out ok I will let you know. Thank you very much I appreciate!
I like the simplicity. Few external components no inductor. I will test these along with a potentiometer at the antenna side as a variable RF attenuator. I think the attenuator is mandatory for every RX system without any input filtering. At least an attenuator...
Please remind me, shall I connect the wiper at the antenna or at the rf preamplifier input side?
In your case I think the antenna is probably best connected to the wiper. I doubt it would make much difference to operation but to some degree it will help to hide the antenna impedance from the IC. Remember that the MMIC need a coupling capacitor at the input pin (100nF) or else the internal bias supply will be shorted out. Also be sure to add a resistor in line with the output, even if you add a choke as well. It is the load resistor that sets the operating current, if you use only a choke it will try to draw to much and probably be damaged.
Brian.
You mean in the BGA2818 a series 50R resistor from the output to the input of the mixer (through a decoupling cap)?The application diagram in the datasheet does not show a resistor. Please let me know what you mean.
Sorry,my mistake, I was thinking of the MMICs I have here that use the same pin for supply and output. The BGA2818 has a separate supply pin. All you need to do is capacitively couple the input and the output pin as both will have some DC on them. Remember you are adding even more gain so the layout becomes even more critical!
The MMIC amps I use (at 1.3GHz) are constructed on double sided PCB with lots of ground area and top and bottom grounds linked in many places. I also make an upright 'bridge' of double sided PCB sitting between the input and output sides (across pins 2 and 5 on the BGA2818 with a notch cut out over the device) to act as an extra shield. It extends about 1cm each side of the device and is about 1cm high with solder along both sides of the bottom edge to the ground plane.
Brian.
direct conversion receivers are subject to more interference effects from undesired received signals, because there are no IF filters to help to remove them. So adding a broadband preamplifier...will only make those interference signals worse.
It will be ok as long as there are no big interferors nearby, such as driving by an AM radio station tower.
In a digital system, where there is a high dynamic range and high bit number Analog-to-Digital converter used, you can make a virtual "digital" IF filter using DSP techniques.....