my call sign at aol.com if you want to talk or do something: yospos radio thread

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > my call sign at aol.com if you want to talk or do something: yospos radio thread

Zap!: May 15, 2002; Nuts.

Crankit posted:

incidentally that's manhattan construction, not dead bug.

Yeah there's a 6 lead DIP in the upper right corner that is laying the wrong way for a dead bug.

Jonny, this thread is awesome.

# ¿ Feb 26, 2014 15:02

Adbot: ADBOT LOVES YOU

# ¿ May 4, 2024 11:47

Zap!: May 15, 2002; Nuts.

Endless Mike posted:

what is a good, relatively cheap (let's say up to $200 used) 2m/70cm radio for a beginner? i got my license almost a year ago and haven't used it at all. i live in dc, so i imagine the longer waves are prob not best for me

Seems like some people like these BaoFeng cheap HTs as beginner radios/throwaways. For $40, seems like a real cheap way to get hooked until you get a better rig.

# ¿ Feb 26, 2014 17:37

Zap!: May 15, 2002; Nuts.

PuTTY riot posted:

are these commercial only or could i get the private jets that fly in here to the uni's airport?

Easy answer is yes.

Complex answer is if they are flying IFR and what class airspace they are currently flying in.

Neat to see that the software is able to do Mode S.

# ¿ Feb 27, 2014 00:08

Zap!: May 15, 2002; Nuts.

Werthog 95 posted:

can i listen to local atc on one of the cheapo radios

Even better is using LiveATC on your computer or phone to listen to ATC chatter.

# ¿ Feb 28, 2014 00:04

Zap!: May 15, 2002; Nuts.

Got my baofeng and could only listen to the local weather channel.

The local VE listed doesn't have a test until May. Guess I'll just have to drive an hour just to take the darn test sooner than May.

# ¿ Mar 4, 2014 15:24

Zap!: May 15, 2002; Nuts.

i never understood why hams in this area would cut you off in traffic while having their call sign on the back of their car.

# ¿ Mar 12, 2014 23:40

Zap!: May 15, 2002; Nuts.

Jonny 290 posted:

Really good to know, hell yes

I think that I want to goof off with building a bandpass or bandstop filter to see if I can figure out how to knock down some of the FM broadcast and uberpagers here. if I can figure out how to switch a filter in and out of circuit I think I can build a box that would hook up to Raspi GPIO pins to turn them on and off. but first i'd just build a single filter, get it working, then add diode switching, scale up from there

When I get home tonight, let me see what I can whip up in ADS. I feel like need to do something useful. Do you have any preference in how quickly the rejection band drops?

# ¿ Mar 18, 2014 13:19

Zap!: May 15, 2002; Nuts.

I recently finished a class on RF and microwave circuits and thought that it was very enjoyable. Real interesting topic. When Jonny 290 mentioned that he was thinking of making a bandstop filter to get rid of 103.9 MHz from messing with his old man signals, I thought it would be the prefect chance to try my hand at some design and make some cool graphs. Grab your soldering iron and your third arm (not that third arm) cause we're about to make ourselves a bandstop filter. Those pesky FM bands won't mess with your VHF enjoyment any more.

We are going to use the low pass filter prototype method of designing this filter. This method uses transformations for each of the four types of filters to replace the logical L or C in the prototype circuit. I've attached my work and some pics of the book below so you can follow along.

Link to the notes with plots

First, we are going to find the center frequency that we want our filter at. For FM broadcast, this roughly means 88 - 108 MHz, so our center frequency is going to be at the geometric mean of this range, 97.3 MHz. Our bandwidth will cover this range, 20% just to work with some round numbers. In reality, this means that 88 and 108 are going to be at 3dB lower than our band. We can estimate what 103.9 MHz is going to be at and compare it to some charts. For this design, we will use a Butterworth or maximally flat filter. Is says that we should see a dB loss of about 11dB with a three element filter. Not bad but not great. It's good enough.

Next, we use the values for n=3 to start calculating the values of our components. Now, this is where we use our filter transformations. So, instead of having an series L first, it's an L and C in parallel for that component. For the shunt C, it is a series L and C. From the work, you can find the values.

* L1' = 16.35 nH
* C1' = 163.5 pF
* L2' = 204.5 nH
* C2' = 13.09 pF

Why not calculate L3' and C3'? Cause in a Butterworth filter, these values are repeated! So we can be lazy and cheap by using the same parts. In class, we used Agilent ADS to simulate our lab designs so let's do the same. Here's our schematic and plot from ADS.

Nice! 97.0 MHz is at -89dB and 104.0 is at -10.98dB! But that's not reality. Reality doesn't have 16.35 nH coils or 16.35 pF caps. But we can get close and simulate. Again

* L1' = 16.35 nH = 16 nH
* C1' = 163.5 pF = 150 pF
* L2' = 204.5 nH = 200 nH
* C2' = 13.09 pF = 12 pF

Whoa! It shifted on us, but in a good way. 97.0 MHz is at -14.6dB and 104.0 MHz is now at -54.99dB.

We can see that we're close. So, I calculated out the values to set 103.9 MHz as the center frequency and tweaked our values. The only thing that changed was L1' and L3'

* L1' = L3' = 15.31 nH = 15 nH
* C1' = C3' = 153.1 pF = 150 pF
* L2' = 191.5 nH = 200 nH
* C2' = 12.25 pF = 12 pF

Eh, 16 nH is probably better for L1' and L3'. Our drop at 104.0 MHz is now -46.4dB.

# ¿ Mar 20, 2014 01:38

Zap!: May 15, 2002; Nuts.

longview posted:

so i did a thing today

fm filter, i used 2/3 of the design posted before except all different values based on what i had around
first version used commercial inductors, but for fun i decided to wind my own, ended up getting about 10 dB extra rejection out of those

frequency response shows decent performance, the loss in the VHF band is acceptable to me since i have a DV hotspot running which often causes problems too
what's also clearly visible is the fact that the two resonant circuits are not perfectly aligned, this is on purpose to get a slightly wider response, the one using discretes did not have this artifact
it was pretty easy to bend and twist the inductors to tune them

bias tee turned out to be way easier than expected

first version used discrete inductors but that turned out terribly, so i wound my own slightly conical one, that worked pretty well

so i decided to make it even more conical

this worked even better, the ceramic component on the left is a tusonix commercial line filter, the black thing is a low frequency inductor
the coupling cap is a special 1nF RF NP0 type i bought for an experiment a few months ago

the reason the inductor should be shaped like that is that the highest frequencies are most sensitive to short connections to the line, so at 4 GHz it only "sees" the first tiny turn or so, at lower frequencies those first turns aren't important but the ones farther behind matter, the distance to the high impedance proportional to wavelength is supposed to be roughly the same at all frequencies, giving a flat frequency response
at the other end of that inductor i can put whatever, it's only seen by lower frequency components, and the regulator IC should only be seen by DC

it performed well, rolling off around 10 MHz and with one notable resonance at around 650 MHz, there's nothing i know of interest there so it's a good solution. performance is very similar to the mini-circuits commercial bias-tees i have, the ripple from 1-6 GHz is me not calibrating the test setup, it's actually very flat

i also integrated a 7805 regulator on board so it can run off 12V from the radio

here's the combined performance of the bias-tee and the FM rejection filter

this RF stuff is fun, now i'm tempted to make a 23cm transceiver for my next project, or a 2.4 gig broad-band voice transmitter

Holy crap it works. Seriously, thanks for trying it and tweaking it. That's really awesome.

# ¿ Mar 22, 2014 01:16

Zap!: May 15, 2002; Nuts.

longview posted:

only downside is the diode is like 1x1mm, how the hell are you supposed to mount them...

what's the package type?

# ¿ Mar 28, 2014 12:22

Zap!: May 15, 2002; Nuts.

Oh, no sweat. Just SMT place it.

You might have to use a bit of solder paste on the pads then use a hot air reflow gun to heat the component pads and melt the solder.

Failing that, you might be able to dead bug the part and just jumper it to some through holes. Should be fine for just dissipating static charges unless you expect some huge current to hit it.

# ¿ Mar 28, 2014 16:09

Zap!: May 15, 2002; Nuts.

I should take pictures of the process our test repair operators use to replace ceramic filters. Those things are a nightmare.

# ¿ Mar 28, 2014 17:19

Zap!: May 15, 2002; Nuts.

just want to say thanks for mentioning Silent Key in this thread cause it is an amazing podcast.

# ¿ Apr 24, 2014 17:51

Zap!: May 15, 2002; Nuts.

This guy has beat the ham radio game. That 20m phased antenna array is simply amazing.

# ¿ Aug 4, 2014 15:25

Adbot: ADBOT LOVES YOU

# ¿ May 4, 2024 11:47

Zap!: May 15, 2002; Nuts.

atomicthumbs posted:

my favorite is the 160m yagi

On second look, this is pretty awesome with the four directors/reflector wires.

# ¿ Aug 5, 2014 13:27

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > my call sign at aol.com if you want to talk or do something: yospos radio thread