different active antennas

OPA659 is pretty noisy. OPA858 is much better, and only 0.6 pF.

Your 27 ohm resistor makes noise too.

--

John Larkin      Highland Technology, Inc 

The best designs are necessarily accidental.

Am 30.12.2020 um 17:11 schrieb snipped-for-privacy@highlandsniptechnology.com:

That is out of interest for frequencies below 30 MHz. The outside noise is much more.

Am 30.12.2020 um 17:11 schrieb snipped-for-privacy@highlandsniptechnology.com:

BTW, You didn't understand the theory of an active antenne. We work with noise matching.

Then the antenna will be a bit longer.

The passive antenne part is so calculated that antenna-noise is amplifier-noise.

Am 30.12.2020 um 18:19 schrieb Leo Baumann:

All other antennas with integrated amplifier are NO active antenna.

What does "noise matching" mean? You wouldn't want to match noise sources the way you match impedances, would you?

Am 30.12.2020 um 20:12 schrieb John Larkin:

The lenght of the passive antenne part is so calculated that antenna-noise is amplifier-noise.

Am 30.12.20 um 20:12 schrieb John Larkin:

I'm just building a low noise preamplifier for 432 MHz ham radio. For the Sky 67150 eval board we measured a 0.3 dB noise factor. I set up my 8670B + HP346C + 10 dB Rosenberger precision PC3.5 attenuator to 4.9 dB excess noise.

That makes me duplicate the 0.3 dB result for the evaluation board, but my own boards are constantly at 0.6 db. :-( The circuit is the same, just 2 stages and I added 2 saw filters and a PIN diode attenuator. Gain is like expected. Selectivity beautiful. Input to the first SKY 67150 is unchanged.

But it makes 0.3 dB too much noise. I even transplanted the input circuit and the phemt chip from the Eval- to my board. The eval board is Rogers 4035 and mine is FR4, but the input series 12nH+20pF is just

4 mm in size. I just can't imagine that this short transmission line makes such a difference on 432 MHz. The eval board must work at > 3 GHz with other RLC.

And John, thanks for the 8670B help! With an empty CMOS RAM battery I can calibrate till the cows come home.

cheers, Gerhard

Am 30.12.2020 um 09:46 schrieb Leo Baumann:

Active antennas are often vehicle antennas of vehicles that operate transmission systems themselves. They are better than passive long wire antennas with baluns because the cores in the baluns are also noisy.

Looks like you used sewage pipe as tubing. Yuck :-)

Anyhow, active antennas are usually best constructed as small loops with the amp having a very low input impedance. Electric field antennas are more susceptible to man-made noise or QRM in ham-speak.

I just started ham radio again and realized that it's not going to be fun without a separate receive antenna, on account of noise from all those solar panel installations on neighbor's roofs.

The amp is going to be easy, probably just a BFR93 in grounded-base and another as a follower to drive the coax. The loop is the trick, to make it survive the sometimes high winds up here and the occasional botched landing by a turkey vulture.

You'll be in 2021 in 20 minutes, we'll have to wait another nine hours.

--
Happy New Year, Joerg 

http://www.analogconsultants.com/

Am 31.12.2020 um 23:39 schrieb Joerg:

By definition, these are not active antennas. Active antennas are electrically short antennas in which the active part is integrated into the antenna. Noise matching is established between the passive part and the electronics. I.e. the passive part is designed in such a way that the antenna noise is equal to the noise of the electronics.

Here the Theory of active Antennas:

Am 31.12.2020 um 23:39 schrieb Joerg:

Yes - Lined with copper foil to achieve a thick radiator with high capacity

Maybe it was new.

Cheap inverters? No doubt plastered with UL and FCC and CE stickers.

No prob. Couple of thoughts:

- What happens if you elevate the input matching components from the surface of your PCB? Does that change the NF?

- Looks like they are using a 10-mil thick laminate between top copper and the ground plane beneath it. Some fab houses will use a single layer of prepreg between the top two layers, so you end up with a lot more capacitance to ground than you thought you had. Have you double-checked the stackup?

Don't forget, you'll need to enter the ENR table as well as run an IF calibration each power-up cycle until the battery is replaced.

-- john, KE5FX

Maybe I should try it.

I do not have their stackup. I did a test shot on a JLBpcb 1.6 mm epoxy board as an add-on to a different payload I had to do anyway. From SMA to gate is just a 0603 Murata Hi-Q inductor (Optional 1206) and a 0603 20 pF Cap. They have a 0402 cap and inductor. The eval board documentation says that the coil is 0603. My L&C work on the evaluation board, and the Murata coil even seems to be slightly better than theirs.

and below that 2 layers FR4. The range of the rogers layer varies more then the nominal thickness! Does not seem to be very precise.

But for the length of 2 0603 parts you cannot yet speak about a transmission line on 432 MHz. I'll redo the layout on 0.5mm FR4. I need to change the pin attenuator anyway or I must live with a

15V control voltage for a 5V amplifier.

<

>

I did that as soon as possible. Panasonic CR2032. Without it, the box is a royal pain.

A happy new year to you! Gerhard, DK4XP