CB and Ham Radio Techs Love Their Bird Wattmeters

[joeqsmith]

I should have added that while there was the AD8307 app note about using it to form an in-line tap, our friend down under shows the dual coupler making their life more difficult.  The funny part is that I don't know if he ended up calculating the load power using the two or not.  Seemed like a lot to go through and then not finish it up which is why I wonder if the performance was so poor that really he just gave up on it.   

Just for a Bird prototype, you could chop up some dead slugs.  I doubt the hams are able to recover 100% of them by banging them on their benches.  I would guess Bird kept the span small to get them as good as they are. 

[xrunner]

There are two problems that need addressing to make the accuracy as good as possible. The two areas to work on are -

-73 to -3 dBm

-3 to +13 dBm

If you look at the response plots for each frequency they are linear for the lower power range but the response drops off as you go higher in frequency. This was already understood. As you may be able to tell it's not quite a linear drop as you go higher in freq., but it's not too bad.

So for the lower range (-73 to -3) I picked one power level of -30 dBm (about mid-range). Then I plotted data for frequencies of 1, 30, 50, 100, 200, 300, 400, and 500 MHz "Legally" this AD8310 isn't rated over 440 MHz but I might be able to get away with it.

As suspected you can see a slight curve in the measured power as frequency increases (blue dashed line). I requested a linear "fit" from the calc software and got the green dashed line. I then applied a correction factor based on that equation to the measured data and got the solid green line (ideally we want -30 all the way across). It's obvious anyway that the linear equation will not take out the curve. I could tweak the equation a bit myself by eye and get a little better fit which wouldn't be that bad, but I wanted better.

I used online tools to get a quadratic equation (which is shown on the chart) from the measured data in the chart. When I applied a new correction factor from that equation I got an almost perfect fit (red line). The idea is, as I set a frequency with a rotary encoder, the reading will have a calculated correction applied on the fly. The encoder can be set via the pushbutton to either 5 or 25 MHz step.

So that will take care of the lower power correction. The higher power ranges add a different response. In addition to dropping with frequency it also changes with more power higher than ~ -3 dBm. As if that weren't enough, it's also exaggerated at frequencies over 300 MHz. So a different equation needs to be used.

I will change the correction calculation based on the raw output voltage from the AD8310. I chose 2.15 V as the point where the calculation for corr. factor will switch between the one for the lower power range to the one for the higher power range. You can see a slight improvement in the plots for the higher power ranges. It isn't perfect but it does enough to make the accuracy better than it would have been using the other equation.

All this is now in the software now and it is working. Now I want to pull the two OLEDs and get the new larger display going in the system. So I'll get back to formatting the display so it has all the info I want on it.

[xrunner]

A few operational checks before pulling the dual display setup. Looks good to proceed.

It's going to be a tight fit to get the larger display in there. Not so much the screen size but the PCB it's on. Just enough room for a rotary encoder on the right.

[bd139]

Looking good 

[joeqsmith]

Looks like the display was made for that case.   

Battle scars?     It's been my experience that area gets some abuse over the years....

[xrunner]

Looks like the display was made for that case.   

Battle scars?     It's been my experience that area gets some abuse over the years....

Cat scratches, too much sun, cut through skin with a saw once. 

[joeqsmith]

Being right handed, I normally hold the work piece in the left and knife, dremel, iron.... in the right.     

Replacement AD8310 arrived.  I just connected it to the bench meter and watched it.  It also exhibits the strange switching behavior.  I lowered the level 1dB and it continued.   Guessing that this is inherent to the part and I was just lucky with one of the three.

[xrunner]

You had enough RF power into it to get out 2.25 V, and it dropped from 2.25 to 2.19 V when you observed it for ~ 4 min.?  - OK I might see if mine will do that trick.

I haven't see that except using breadboard connections I have seen the power indication on my test bed drop. But it was due to unreliable breadboard contacts and/or wire jumpers which caused a small error in the Vout of the AD8310 board. It doesn't take much error to cause a bad power reading.

[joeqsmith]

The large step was my changing the output 1dB.  The smaller steps are what I am looking at.     The attached shows more clearly.   Roughly 200uV with that part. 

[joeqsmith]

Attached showing sweeping the AD8310 from -87dBm to 13dBm in 5dBm steps.  The frequency was swept from 1MHz to 600MHz.   

Up to about 80MHz, they overlay very close.  Again, I was only looking for maybe 50MHz but could see using a polynomial fit for each frequency and then interpolating base on the actual frequency.  Maybe tie down the temperature.     

[xrunner]

Attached showing sweeping the AD8310 from -87dBm to 13dBm in 5dBm steps.  The frequency was swept from 1MHz to 600MHz.   

Up to about 80MHz, they overlay very close.  Again, I was only looking for maybe 50MHz but could see using a polynomial fit for each frequency and then interpolating base on the actual frequency.  Maybe tie down the temperature.   

Sure you can take it as far as you want to. Just depends on the amount of work you want to do and the memory available. See right now, with the two curve fits I have and the new display with additional pretty features I'm up to 94% of memory so I can't go to all that trouble now. 

[bd139]

Joe: What are you changing when you switch power? External attenuator or generator output? Trying to work out why there would be a step in voltage. I did stick a 40Hz approx RC low pass on the output of mine originally. I wasn’t interested in measuring changing signals.

I ask because there was a horrible sticky attenuator issue in one of the Marconi 2019’s we had years ago where it’d lag slightly changing the step attenuator output switching between two powers. It dumps all the attenuator bits into one register IC in one write but not all relays are made equal.  So it’d switch out the 20dB then switch in the 10dB rather than the other way round. This would result in a 10dB spike and me scratching head why my AGC was going bananas.

[xrunner]

The large step was my changing the output 1dB.  The smaller steps are what I am looking at. The attached shows more clearly.   Roughly 200uV with that part.

I put in a 0.0 dBm 50 MHz signal from my hp 437B and monitored the Vo of the AD8310 disconnected from microcontroller board. I was monitoring with my digital scope. I don't (or can't) see what you are seeing. All I see is 10 mV of noise and the output voltage is 2.2738 VDC which calculates to 0.26 dBm (with no correction applied). If you had for example two outputs 200 uV say

1.0000 (calculates to -53.333 dBm)
1.0002 (calculates to -53.325dBm

The error caused by 200 uV is 0.008 dB

Besides, I'm taking three measurements and averaging them for each displayed reading, so even if it was happening here and I can't tell it's pretty meaningless.

[joeqsmith]

Joe: What are you changing when you switch power? External attenuator or generator output? Trying to work out why there would be a step in voltage. I did stick a 40Hz approx RC low pass on the output of mine originally. I wasn’t interested in measuring changing signals.

I ask because there was a horrible sticky attenuator issue in one of the Marconi 2019’s we had years ago where it’d lag slightly changing the step attenuator output switching between two powers. It dumps all the attenuator bits into one register IC in one write but not all relays are made equal.  So it’d switch out the 20dB then switch in the 10dB rather than the other way round. This would result in a 10dB spike and me scratching head why my AGC was going bananas.

The generator's output.  The generator uses the Teledyne RF100-5 mechanical relays.  I suspect these are a break before make. 
https://www.mouser.com/datasheet/2/227/TELDS00303_1-2578012.pdf

Normally it doesn't pose a problem for me.  An example is looking at the AD8310.  The software to collect synchronized the Marconi generator and my HP bench meter.  It changes the generator's frequency and amplitude, allows time to settle, collects data from the bench meter, plots it,  then repeats the cycle. 

As I have shown the frequency also has a disturbance at 10.000000 MHz.  I suspect it looses lock and it's tied to the GPS 10MHz ref.  Not normally a problem.

When I showed the glitch, I was running the generator by hand, asynchronous to the HP bench meter. 

[joeqsmith]

The large step was my changing the output 1dB.  The smaller steps are what I am looking at. The attached shows more clearly.   Roughly 200uV with that part.

I put in a 0.0 dBm 50 MHz signal from my hp 437B and monitored the Vo of the AD8310 disconnected from microcontroller board. I was monitoring with my digital scope. I don't (or can't) see what you are seeing. All I see is 10 mV of noise and the output voltage is 2.2738 VDC which calculates to 0.26 dBm (with no correction applied). If you had for example two outputs 200 uV say

1.0000 (calculates to -53.333 dBm)
1.0002 (calculates to -53.325dBm

The error caused by 200 uV is 0.008 dB

Besides, I'm taking three measurements and averaging them for each displayed reading, so even if it was happening here and I can't tell it's pretty meaningless.

 

Thanks for trying to replicate my findings.   

A scope isn't a good choice for it unless you have some crazy scope.   I can easily see it with the Brymen BM869s 500,000 count mode.   Yes, as I stated, very small and was more curious why these two parts behaved so different.   Again, the AD8307 also exhibits this same switching behavior  but to a much lesser degree. 

***
You may be able to use the scoped AC coupling and crank up the vertical gain.   I doubt I could see anything with my old scopes but some of the new high res jobs may show the switching.   

[xrunner]

Currently I'm thinking about using the display presentation shown. The numbers aren't "real" they are just representative strings used to help determine placement.

I was going to say they were "dummy" numbers, but I knew if I said that someone would say "Of course - that's because a dummy put them in!"

I'm also running off a separate 3.3V regulator from the one on the controller board. The OLED draws ~140 mA and the 3.3 supply on board supplies a max of 150 mA. Also I don't want the OLED connected to the 3.3 supply I'm using for the reference.

[joeqsmith]

Same device swept to 1.6GHz.   

[xrunner]

Can't you go higher than 1.6 GHz? 

Yea what's to stop it being useful as long as you know what the error is and correct for it.

And that's exactly what I notice when I'm breadboarding it and I touch near the antenna, the power indication goes up. Why? Because my WiFi router is 3 feet away and it's a strong signal even though it's 2.4 or 5 GHz. It will of course react to any RF power that's strong enough. That's why it needs a shielded case and these guys making projects with this device need to finish up and make a decent case for it. 

[bd139]

You can probably get away with a can over the detector and the associated SMA.

The last two of these I built ended up inside Pomona or Hammond boxes due to similar problems. The real bugger turned out to be my laptop charger at the time which seemed to kick up a few dBm until I unplugged it. This was in the tray under my bench about two inches from the work surface 

[joeqsmith]

Can't you go higher than 1.6 GHz? 

Yea what's to stop it being useful as long as you know what the error is and correct for it.

And that's exactly what I notice when I'm breadboarding it and I touch near the antenna, the power indication goes up. Why? Because my WiFi router is 3 feet away and it's a strong signal even though it's 2.4 or 5 GHz. It will of course react to any RF power that's strong enough. That's why it needs a shielded case and these guys making projects with this device need to finish up and make a decent case for it. 

You're giving up dynamic range.

Yes, 1.6G is a bit overkill for the hams.  As I understand it, hams use 3.843MHz and 7.2MHz.   You can enter these into the YT search engine to find out what ham radio is about.  There are a lot of very good examples.   





I was recently reminded by a viewer that what I show with the VNAs also exceeds what the hams would use it for.   

[xrunner]

You can probably get away with a can over the detector and the associated SMA.

Yea that's true

You're giving up dynamic range.

Yes, 1.6G is a bit overkill for the hams.  As I understand it, hams use 3.843MHz and 7.2MHz.   You can enter these into the YT search engine to find out what ham radio is about.  There are a lot of very good examples.

.

Geez Joe ... your not a very polite person, trying to explain things to people that don't want or need it. You should get off the internet until you get a better attitude. Think about others! 

[joeqsmith]

     When the hams behave so badly toward one another,  I have to ask why anyone would want to be involved with such a hobby.   You have twitter and other social media already if that's your thing.   



Raw
Looking at the raw data from the AD8310 with the HP meter and driving the demo board with the Marconi.   Sweeping 1-500MHz, -83 to 13dBm. 

Ref Least Sq Fit
Least squares for the 10MHz data set.   I then calculate the 4th order polynomial coefficients.

Dif from Ref
Taking the 10MHz data and subtracting it from the raw data.

Poly Fit Dif from Ref
Running the 4th order fitter on the raw data and looking at the difference from the 10MHz ref.

Poly Fit w/ Gain Comp
Compensating for the gain errors due to frequency and running the polynomial fit.

Gain Comp
The difference from the reference after gain and poly fit.   

Give up some of the dynamic range by using -60dBm would certainly do a better job.    Adding a simple gain with a 4th or 5th order wouldn't be too bad.   Fun to play with.
 

[joeqsmith]

Same data sets but using 250MHz as the reference, going to a 6th order polynomial and limiting the range to -67dBm.   

As our friend stated above,  "Hams and other electronic enthusiasts are going to use this thing and you are really going beyond these people and their interests and ability to understand the product and it's use as you explain it.  This means most of them will use it at 1.8 to 30 mhz."  we have really stepped outside the area of interest for the electronics enthusiast....  Oh wait.... I am an electronics enthusiast.  We just don't fit the mold and proud of it!!!   

[mnementh]

...Just for a Bird prototype, you could chop up some dead slugs.  I doubt the hams are able to recover 100% of them by banging them on their benches.  I would guess Bird kept the span small to get them as good as they are.

You bastard... you made me snort iced tea out my nose with that...   

mnem
*toddles off to find something on-topic to post*

[mnementh]

Maybe.  Going in-line like he did adds to the mix.  I think he said he was wanting 100W.   I wonder if you could design a 3MHz to 500MHz, 100W slug for the Bird.  That may pay for a few hobby projects.

The problem with the slug is, I think, more about the casing than the coupling inside. You got to do some precise machining with metal parts. I don't have that capability. I can do 3D printing but that won't cut it for the Bird...

3DP is perfect.... design it using your favorite flavor of not-Frustion360, get the deets worked out, then have the slug body made by one of the 3rd-party metal-printing shops. Boom! Seeya! Done!

mnem
Then, when you're done with that bit of blasphemy, you can make a digital meter replacement for the wiggly needle...