Forums

I need a good text on low noise amplifier design ASAP

Started by Unknown January 20, 2017
I'm struggling with an amplifier design for a radio front end. 

Ham 600 meter band (475kHz)

The antenna is a ferrite loop (can't change that) with a resonating capacitor on the primary, and a few turns on the secondary, so I have a pretty low input impedance.

I have several designs in spice, with common emitter and common base configurations, but I am struggling to get the noise below 300nV/Hz^1/2.
I need to get the noise as low as practical, as my input signal is somewhere less than 300nV.  Basically, how low can I really go in a practical sense?
I need to stay with transistors that are currently available, vintage unobtanium won't work for me.

This is followed by a mixer and crystal filter with 200hZ bandwidth that is performing well, but the front end is just not hacking it.  I keep orbiting the drain with gain and noise.

The amp has to run from 2.8V. 

In the end, I will need three preamps for three antennas, and the ability to switch inputs quickly without hammering the IF, but that's a secondary problem.


So, can someone point me to a text, preferably available online (Kindle ok) so I can get through it this weekend?
On Fri, 20 Jan 2017 13:48:05 -0800 (PST), kc6ete@gmail.com wrote:

>I'm struggling with an amplifier design for a radio front end. > >Ham 600 meter band (475kHz) > >The antenna is a ferrite loop (can't change that) with a resonating capacitor on the primary, and a few turns on the secondary, so I have a pretty low input impedance. > >I have several designs in spice, with common emitter and common base configurations, but I am struggling to get the noise below 300nV/Hz^1/2. >I need to get the noise as low as practical, as my input signal is somewhere less than 300nV. Basically, how low can I really go in a practical sense? >I need to stay with transistors that are currently available, vintage unobtanium won't work for me. > >This is followed by a mixer and crystal filter with 200hZ bandwidth that is performing well, but the front end is just not hacking it. I keep orbiting the drain with gain and noise. > >The amp has to run from 2.8V. > >In the end, I will need three preamps for three antennas, and the ability to switch inputs quickly without hammering the IF, but that's a secondary problem. > > >So, can someone point me to a text, preferably available online (Kindle ok) so I can get through it this weekend?
A BF862 jfet has noise around 1 nv/rthz. A decent bipolar should be similar. Too bad the antenna sec turns are so few. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On Fri, 20 Jan 2017 13:48:05 -0800, kc6ete wrote:

> I'm struggling with an amplifier design for a radio front end. > > Ham 600 meter band (475kHz) > > The antenna is a ferrite loop (can't change that) with a resonating > capacitor on the primary, and a few turns on the secondary, so I have a > pretty low input impedance. > > I have several designs in spice, with common emitter and common base > configurations, but I am struggling to get the noise below 300nV/Hz^1/2. > I need to get the noise as low as practical, as my input signal is > somewhere less than 300nV. Basically, how low can I really go in a > practical sense? > I need to stay with transistors that are currently available, vintage > unobtanium won't work for me. > > This is followed by a mixer and crystal filter with 200hZ bandwidth that > is performing well, but the front end is just not hacking it. I keep > orbiting the drain with gain and noise. > > The amp has to run from 2.8V. > > In the end, I will need three preamps for three antennas, and the > ability to switch inputs quickly without hammering the IF, but that's a > secondary problem. > > > So, can someone point me to a text, preferably available online (Kindle > ok) so I can get through it this weekend?
You get low voltage noise with lots of junction area, to reduce the noise voltage. The cool vintage unobtanium solution is to use an LM194. I don't know if there's a current part available, but the LM194 is, more or less, a bazzilion low-noise transistors all on a die in parallel with emitter loading resistors. What about stepping the voltage up with a transformer so that your antenna impedance is closer to the ideal noise match for a transistor that you do have? -- Tim Wescott Wescott Design Services http://www.wescottdesign.com I'm looking for work -- see my website!
On Fri, 20 Jan 2017 14:10:28 -0800, John Larkin wrote:

> On Fri, 20 Jan 2017 13:48:05 -0800 (PST), kc6ete@gmail.com wrote: > >>I'm struggling with an amplifier design for a radio front end. >> >>Ham 600 meter band (475kHz) >> >>The antenna is a ferrite loop (can't change that) with a resonating >>capacitor on the primary, and a few turns on the secondary, so I have a >>pretty low input impedance. >> >>I have several designs in spice, with common emitter and common base >>configurations, but I am struggling to get the noise below 300nV/Hz^1/2. >>I need to get the noise as low as practical, as my input signal is >>somewhere less than 300nV. Basically, how low can I really go in a >>practical sense? >>I need to stay with transistors that are currently available, vintage >>unobtanium won't work for me. >> >>This is followed by a mixer and crystal filter with 200hZ bandwidth that >>is performing well, but the front end is just not hacking it. I keep >>orbiting the drain with gain and noise. >> >>The amp has to run from 2.8V. >> >>In the end, I will need three preamps for three antennas, and the >>ability to switch inputs quickly without hammering the IF, but that's a >>secondary problem. >> >> >>So, can someone point me to a text, preferably available online (Kindle >>ok) so I can get through it this weekend? > > A BF862 jfet has noise around 1 nv/rthz. A decent bipolar should be > similar. > > Too bad the antenna sec turns are so few.
I wasn't paying attention to his actual figures. Dammit. Yes, that noise figure seems high -- maybe just use a nice low-noise op- amp with a GBW product of 10MHz or better? I've used 'em for IF amplifiers at 455kHz. I don't know how the situation is now, but in the past I've never gotten the impression that SPICE is really good at predicting noise performance. I could be wrong, but I think this is one area where you're better off scrutinizing data sheets and doing some pencil-and-paper work. -- Tim Wescott Wescott Design Services http://www.wescottdesign.com I'm looking for work -- see my website!
On Fri, 20 Jan 2017 16:35:13 -0600, Tim Wescott
<seemywebsite@myfooter.really> wrote:

>On Fri, 20 Jan 2017 14:10:28 -0800, John Larkin wrote: > >> On Fri, 20 Jan 2017 13:48:05 -0800 (PST), kc6ete@gmail.com wrote: >> >>>I'm struggling with an amplifier design for a radio front end. >>> >>>Ham 600 meter band (475kHz) >>> >>>The antenna is a ferrite loop (can't change that) with a resonating >>>capacitor on the primary, and a few turns on the secondary, so I have a >>>pretty low input impedance. >>> >>>I have several designs in spice, with common emitter and common base >>>configurations, but I am struggling to get the noise below 300nV/Hz^1/2. >>>I need to get the noise as low as practical, as my input signal is >>>somewhere less than 300nV. Basically, how low can I really go in a >>>practical sense? >>>I need to stay with transistors that are currently available, vintage >>>unobtanium won't work for me. >>> >>>This is followed by a mixer and crystal filter with 200hZ bandwidth that >>>is performing well, but the front end is just not hacking it. I keep >>>orbiting the drain with gain and noise. >>> >>>The amp has to run from 2.8V. >>> >>>In the end, I will need three preamps for three antennas, and the >>>ability to switch inputs quickly without hammering the IF, but that's a >>>secondary problem. >>> >>> >>>So, can someone point me to a text, preferably available online (Kindle >>>ok) so I can get through it this weekend? >> >> A BF862 jfet has noise around 1 nv/rthz. A decent bipolar should be >> similar. >> >> Too bad the antenna sec turns are so few. > >I wasn't paying attention to his actual figures. Dammit. > >Yes, that noise figure seems high -- maybe just use a nice low-noise op- >amp with a GBW product of 10MHz or better? I've used 'em for IF >amplifiers at 455kHz. > >I don't know how the situation is now, but in the past I've never gotten >the impression that SPICE is really good at predicting noise >performance. I could be wrong, but I think this is one area where you're >better off scrutinizing data sheets and doing some pencil-and-paper work.
I've done noise analysis in LT Spice and the numbers looked reasonable and the hardware worked about right. Phil has noted that Spice might not always be trusted for noise. He'll probably need a tuned LC drain load on a jfet, to run at that low a supply voltage. Or maybe just a biggish inductor. A manual analysis should be pretty simple here. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On Fri, 20 Jan 2017 13:48:05 -0800 (PST), kc6ete@gmail.com wrote:

>I'm struggling with an amplifier design for a radio front end. > >Ham 600 meter band (475kHz) > >The antenna is a ferrite loop (can't change that) with a resonating capacitor on the primary, and a few turns on the secondary, so I have a pretty low input impedance. > >I have several designs in spice, with common emitter and common base configurations, but I am struggling to get the noise below 300nV/Hz^1/2. >I need to get the noise as low as practical, as my input signal is somewhere less than 300nV. Basically, how low can I really go in a practical sense? >I need to stay with transistors that are currently available, vintage unobtanium won't work for me. > >This is followed by a mixer and crystal filter with 200hZ bandwidth that is performing well, but the front end is just not hacking it. I keep orbiting the drain with gain and noise. > >The amp has to run from 2.8V. > >In the end, I will need three preamps for three antennas, and the ability to switch inputs quickly without hammering the IF, but that's a secondary problem. > > >So, can someone point me to a text, preferably available online (Kindle ok) so I can get through it this weekend?
My favorite... <https://www.amazon.com/Low-Noise-Electronic-Design-C-Motchenbacher/dp/0471619507/ref=sr_1_1?ie=UTF8&qid=1484953604&sr=8-1&keywords=0471619507> ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | I'm looking for work... see my website. Thinking outside the box... producing elegant solutions.
On Fri, 20 Jan 2017 16:35:13 -0600, Tim Wescott
<seemywebsite@myfooter.really> wrote:


[snip]
> >I don't know how the situation is now, but in the past I've never gotten >the impression that SPICE is really good at predicting noise >performance. I could be wrong, but I think this is one area where you're >better off scrutinizing data sheets and doing some pencil-and-paper work.
Spice is as good at predicting noise only as accurately as the MODEL portrays it. See "InputReferredNoise.zip" on the Device Models & Subcircuits page of my website for elements that can be added to just about any active device model so that the model portrays noise properly. ...Jim Thompson -- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at http://www.analog-innovations.com | 1962 | I'm looking for work... see my website. Thinking outside the box... producing elegant solutions.
On Fri, 20 Jan 2017 13:48:05 -0800 (PST), the renowned
kc6ete@gmail.com wrote:

>I'm struggling with an amplifier design for a radio front end. > >Ham 600 meter band (475kHz) > >The antenna is a ferrite loop (can't change that) with a resonating capacitor on the primary, and a few turns on the secondary, so I have a pretty low input impedance. > >I have several designs in spice, with common emitter and common base configurations, but I am struggling to get the noise below 300nV/Hz^1/2. >I need to get the noise as low as practical, as my input signal is somewhere less than 300nV. Basically, how low can I really go in a practical sense? >I need to stay with transistors that are currently available, vintage unobtanium won't work for me. > >This is followed by a mixer and crystal filter with 200hZ bandwidth that is performing well, but the front end is just not hacking it. I keep orbiting the drain with gain and noise. > >The amp has to run from 2.8V. > >In the end, I will need three preamps for three antennas, and the ability to switch inputs quickly without hammering the IF, but that's a secondary problem. > > >So, can someone point me to a text, preferably available online (Kindle ok) so I can get through it this weekend?
Hi, Dave. Second the BF862- that's really what it's designed for. You should be able to get it to operate from 2.8V since typically cutoff is 800mV (1.2V max). --sp -- Best regards, Spehro Pefhany Amazon link for AoE 3rd Edition: http://tinyurl.com/ntrpwu8
On Friday, January 20, 2017 at 1:48:09 PM UTC-8, kc6...@gmail.com wrote:
> I'm struggling with an amplifier design for a radio front end. > > Ham 600 meter band (475kHz) > > The antenna is a ferrite loop (can't change that) with a resonating capacitor on the primary, and a few turns on the secondary, so I have a pretty low input impedance.
How low? What value capacitor? You want to match impedance, after all, or the best low-noise design won't give you low signal-to-noise. Maybe I'm being naiive, but the low frequency should mean you might be able to neglect self-resonance of the ferrite...
kc6...@gmail.com wrote:
> > I'm struggling with an amplifier design for a radio front end. > > Ham 600 meter band (475kHz) > > The antenna is a ferrite loop (can't change that) with a resonating capacitor on the primary, and a few turns on the secondary, so I have a pretty low input impedance. > > I have several designs in spice, with common emitter and common base configurations, but I am struggling to get the noise below 300nV/Hz^1/2.
** So you have not built anything yet ? I suspect you are making a bad error somewhere, as common audio pre-amp circuits have noise figures under 10nVrtHz. .... Phil