CB and Ham Radio Techs Love Their Bird Wattmeters

[joeqsmith]

During my continued search of Watt meters, I came across a ham who was making their own.  I decided to try and find the other parts to their video and came across this gem.   Someone should tell these hams that we invented something called a heatgun and soldering iron.   

[joeqsmith]

Looking at 1.8,3,7,14,21,28 & 50MHz from -73 to 13dBm in 2dB steps.    You can see why I am simply doing a point slope calc.  The two parts behave different enough that they each have their own coefficients but really, that's about it.   The heart of it is the coupler.  Getting something to work over this wide of a range was the difficult part.

[joeqsmith]

Of course, you run the part to a couple of GHz,  SI=SO.   

[bd139]

Nothing you can't fix with a poorly conceived quadratic regression

[xrunner]

Of course, you run the part to a couple of GHz,  SI=SO.   

Good, good - keep going ... 

I integrated the new display and code into the main program, substituting actual variables for test strings. Except for a few tweaks it looked just like I planned. Testing with 0.0 dBm 50 MHz from hp power meter.

Also included for any parties interested are two pics, one with the system on the bench (4 feet away from my dual-band WiFi router) and the other next to it. With no antenna on a breadboard it picks up RF energy and raises the noise floor from -73 to -68 dBm. Actually, it goes even higher (like -55) but I couldn't catch the higher bursts from the router on the camera, but I could easily see them on the display. With a small antenna on the bench it easily captures the RF energy (2.4 & 5 GHz).

As can be seen from Joe's graphing, it doesn't stop amplifying RF inputs just because the data sheet says it's rated to 440 or 500 MHz. By the way I'm going to get the AD8307 part and see how she goes. May use it instead of the AD8310.

[joeqsmith]

The problem with the devices next to it are easy enough to solve.    While the ham tried to isolate the two ports,  I doubt you would see any need with the eval boards.   This is the reason I never isolated the two.   I did however make an attempt to separate them from the digital source.   

If we limited the last set of data to 1.4GHz and -27 to 13dBm, you could maybe make use of it.   If we were to attempt to fit each band, it would do much better.   

[xrunner]

Shifting gears to front and rear case panels, some progress at least.

[joeqsmith]

Nothing you can't fix with a poorly conceived quadratic regression

True.  Fitting to that chopped photo of the Bird meter is by far the most complex math I have running.      

Demo of my final setup.   

[xrunner]

How about an AD 8317? 1 MHz to 10 GHz. I think some of the Chinese power meters (no case) use it but they don't specifically say so. Looks like It might be underneath when looking from the side.

[joeqsmith]

Looks like the dynamic range is pretty limited.   Still, if you're interested in demonstrating how to design a design a bi-directional 1M-10GHz coupler for 100W with a directivity of 30dBm and a +/-0.5dB flatness, I would be very much interested in following along.   

There are a few videos showing some OTS meters as well as a few homemade ones.  I did not find anything in-line with a bi-directional coupler.   You could be the first!!!   

Here's an OTS one

[xrunner]

Well, I think I can do a better user interface than what I've seen maybe that's a good place to start. 

I want to get the AD8307 which should be here today. I might change the correction factor method a bit - maybe a function per 100 MHz span. That's only five functions for a 500 MHz range. Still not burned out yet.

Going to print out a front panel today in green PLA and see if it fits the first time out. It's been known to happen from time-to-time. 

I had been using the Arduino IDE 2.0 RC and it updated itself yesterday to a new version (has before). But later in the day I tried uploading to a board and it refused to comply, said "Exit Status 1" which doesn't tell me a damn thing. It looked to me like it was confused about COM ports. Well I found that they had a bug in the release in that if you had the serial monitor port open it wouldn't upload. Lost time trying to figure that out, but they said if you close the serial port it would work for now. I did, and it did. Supposed to be fixed in an overnight build guess I'll install a nightly build and see, could be fixed but have a different problem I suppose.

 

[bd139]

That sort of shit pushed me to use AVR gcc tool chain instead ages ago. That just works and stays working.

[xrunner]

That sort of shit pushed me to use AVR gcc tool chain instead ages ago. That just works and stays working.

OK, well I'll look into that thanks.

[xrunner]

Got the AD8307. I guess I forgot to pay the extra $1.25 to get a hole drilled for the output voltage pin. Check the original picture vs mine. Come to think of it neither have a hole for the power supply positive.

 

[xrunner]

Front panel of case done. Came out right the first time no editing needed - Fusion 360 and 3D printer all cooperated for once - go figure. 

I'll need to modify the slide-in mounting plate for the controller board because there is little room for the front edges of the plate now. Shouldn't be a big deal to do.

Did an initial checkout of the AD8307 device. Log intercept appears to be -85.2 dBm which is within specs. Better behaved at high frequencies with higher power as can be seen in the datasheet. So I'll be using this part in this power meter.

[bd139]

Looking good. It does look like it's "flipping the bird" at us though at first glance 

[xrunner]

Looking good. It does look like it's "flipping the bird" at us though at first glance 

Yea I had thought about that but it kindof grew on me so I left it as-is. 

[joeqsmith]

Flipping the Bird...   

The cost to drill those two holes in your AD8307 eval boards is offset by the added SMA you can salvage for other projects.  Mine do not have the output cap shown but I did add one.

[bd139]

Another one for the list…. LT5534. 50MHz to 3GHz.

https://www.analog.com/media/en/technical-documentation/data-sheets/5534fc.pdf

Playing with DIY bolometers in my head. May go nowhere 

[xrunner]

Flipping the Bird...   

The cost to drill those two holes in your AD8307 eval boards is offset by the added SMA you can salvage for other projects.  Mine do not have the output cap shown but I did add one.

If they would have put on an LED like the AD8310 board all would have been forgiven. 

Another one for the list…. LT5534. 50MHz to 3GHz.

https://www.analog.com/media/en/technical-documentation/data-sheets/5534fc.pdf

Playing with DIY bolometers in my head. May go nowhere 

We'll have DC to daylight power measurements covered by the time this is over, maybe we should start a new company. 

[bd139]

So who’s renting a garage in Palo Alto? 

[free_electron]

During my continued search of Watt meters, I came across a ham who was making their own.  I decided to try and find the other parts to their video and came across this gem.   Someone should tell these hams that we invented something called a heatgun and soldering iron.   

shit like this makes me want to live on an island far away from the rest of the world

[free_electron]

So who’s renting a garage in Palo Alto? 

i can rent you mine

[bd139]

Didn’t actually watch that video before.

Have commented appropriately.

[bd139]

Back to poking hams. This is wonderful:



I assume my comment will be deleted so I've left it here too:

Well that was exciting. A dull, cheaply furnished, uninspiring office without a bit of natural light resembling a poor attempt at cloning an IT operations centre. With a couple of radios chucked underneath the monitors. Perfect for the modern caged walt ham.