CB and Ham Radio Techs Love Their Bird Wattmeters

[xrunner]

I calibrated the 220 MHz ham band using the Baofeng UV-S9X3 I just got. It's supposed to Xmit on two power levels on each band (2m, 1.25 m, 70 cm), L & H. The funny thing is, it only transmits on one power level on the 220 MHz band. Yet the display toggles the power indicator just like the other two bands. It Xmits 3.2 W - period - on 220. So 3.2 W is the data point I used, but since the response of the AD8307 is linear over the region I'm using, it should be fine. Since I have no other means to go to a higher power on 220, I'll have to settle for that for now. But for what I've seen on the other bands, it should be reasonably accurate for higher powers on 220.

The final two ham bands I needed to calibrate are 2 m & 70 cm. After that, if there are no glaring problems, I can move forward with the next steps. These are the results of the final two band calibrations in units of Watts.

Transceiver Yaesu FTM-7250D

VHF 146.000 MHz

Radio PWR Set	hp 437B	Project	Error
5	5.0	4.8	-0.2
25	24.1	24.4	+0.3
50	45.4	45.8	+0.4

UHF 445.000 MHz

Radio PWR Set	hp 437B	Project	Error
5	4.1	3.5	-0.6
25	22.8	21.7	-1.1
50	45.7	45.9	+0.2

The highest dB error is UHF low power: 4.1 W (hp) vs 3.5 W (Project). That's 0.7 dB, which is reasonable and acceptable for this system. I've calibrated the development system now and I'm satisfied it's worth moving forward. When it finally goes into a nice enclosure (a whole other project) and the breadboarding is gone, I'll recalibrate all the bands.

Next I want to pull the tiny OLED display and use a larger 2.42 inch 128x64 type similar to what I've used on a previous project meter shown in the pic. I think the last unused one I bought is yellow. When I get that going, I have another upgrade for this project in mind ...

[joeqsmith]

Looks like he was using a modulated carrier and did not check it over the entire frequency range but it appears at 7MHz, their's is far more accurate than mine.   

[joeqsmith]

An old review of the MJF 845 shows data for basically the same tests I ran.  Theirs is MUCH better than what I measured with mine.   Did they cut costs?  Forgot how to align them?  I have a defective unit?   Hard to say.  I'm sticking with, save your cash.

http://www.arrl.org/files/file/ProductReviewsForDeb/2016/pr062016.pdf

[joeqsmith]

The enclosed AD8307 finally arrived.   My luck with Chinese products appears to have gone south. 

I wanted to measure S11 for you out of the box and then see if I could improve it.   Yes, I have seen the ham's shove the drill bits into their RF connectors to "repair" them.   That's not going to happen. 

 

[xrunner]

The enclosed AD8307 finally arrived.   My luck with Chinese products appears to have gone south. 

I wanted to measure S11 for you out of the box and then see if I could improve it.   Yes, I have seen the ham's shove the drill bits into their RF connectors to "repair" them.   That's not going to happen.

Oh no. 

You've taken two hits for the team now, stop buying the crap and let me take some hits for a while. 

[joeqsmith]

I could use the output connector but I think I am just going to go ahead with my mods which would have included a better connector anyway. 

There was that recent thread where there were members offered several suggestions on how to repair a section of semi-ridged.  Not 10MHz, but 6+GHz use.  I had some fun damaging my own section and attempting repairs. 

https://www.eevblog.com/forum/rf-microwave/is-is-possible-to-repair-a-hairline-crack-in-semi-rigid-coax/

That's what this video reminds me of.   

[joeqsmith]

It appears similar to the board you show. 

[xrunner]

It appears similar to the board you show.

It is the same (although I can't be sure of the SMD cap and inductor values).


Can you link me the seller of those? I wouldn't mind having those metal cases they use - good RF shielding at least. 

[xrunner]

An old review of the MJF 845 shows data for basically the same tests I ran.  Theirs is MUCH better than what I measured with mine.   Did they cut costs?  Forgot how to align them?  I have a defective unit?   Hard to say.  I'm sticking with, save your cash.

Oh I didn't know MFJ attached their name to that thing. I'm not sure branding it with their name improves the perception of MFJ quality or not at this point. It's hard to say why yours is worse, or if the workers making them know what they're doing. Maybe it's like cars, don't buy a SWR / power meter made on a Friday, but we can't even tell that.

I tested the Baofeng UV-S9X3 again to get a screen shot for you guys. It shows the spurious emissions when the thing is transmitting on the 220 MHz ham band. Also hooked back up the El Cheapo RF amp, and you can see all the spurious emissions ( 0 dBm drive power coming from my Agilent). They decrease in amplitude as you reduce the input drive or board voltage, but of course then it's practically useless to me.

Also tested the new Yellow 2.4" OLED display that I want to use. Had to change jumpers to make it work with I2C. Same thing I did last time with the White one. Jumper R8 must be moved to R9, and R10,11, and 12 must be jumpered.

[joeqsmith]

It appears similar to the board you show.

It is the same (although I can't be sure of the SMD cap and inductor values).


Can you link me the seller of those? I wouldn't mind having those metal cases they use - good RF shielding at least. 

Once I replace the connector, I will measure S11 as it then and see if we are close. 

https://www.amazon.com/Keenso-Logarithmic-Testing-Detector-0-1-600M/dp/B07RDBQSWT/ref=sr_1_2?crid=1CSB8140BKSRJ&keywords=ad8307&qid=1689813432&sprefix=ad8307%2Caps%2C99&sr=8-2

The new display looks good.  Planning to print the case covers again?   

[A.Z.]

Once I replace the connector, I will measure S11 as it then and see if we are close. 

https://www.amazon.com/Keenso-Logarithmic-Testing-Detector-0-1-600M/dp/B07RDBQSWT/ref=sr_1_2?crid=1CSB8140BKSRJ&keywords=ad8307&qid=1689813432&sprefix=ad8307%2Caps%2C99&sr=8-2

The new display looks good.  Planning to print the case covers again?

couldn't resist from cliking on the "visit Keenso store" link at top of page

https://www.amazon.com/stores/Keenso/page/AB1D077B-639C-40D5-AE02-A64DEEBBDEA4

 

[xrunner]

Once I replace the connector, I will measure S11 as it then and see if we are close. 

https://www.amazon.com/Keenso-Logarithmic-Testing-Detector-0-1-600M/dp/B07RDBQSWT/ref=sr_1_2?crid=1CSB8140BKSRJ&keywords=ad8307&qid=1689813432&sprefix=ad8307%2Caps%2C99&sr=8-2

couldn't resist from cliking on the "visit Keenso store" link at top of page

https://www.amazon.com/stores/Keenso/page/AB1D077B-639C-40D5-AE02-A64DEEBBDEA4

 

Car parts and putter covers ... and they want to get into logarithmic amplifiers? They are so proud of the little amp they don't even put their name on it. 

The new display looks good.  Planning to print the case covers again?

Probably, as well as another custom internal mounting bracket for the controller board. But I haven't looked at cases yet. I'll have to see whats available that will work with that Putnam coupler. I can see more custom internal cables also, need to go from N female and N male to bulkhead PL-259 for the hams ... 

[joeqsmith]

Typical exporter.

Was going to just solder in the new connector but the PCB is poorly aligned.  Even with the PCB pushed to the on side, the alignment is off a fair amount.  Can't see machining the case.  Most likely I'll trim the edge of the PCB and call it good. 

[joeqsmith]

Trimmed off the flashing after they tapped the hole for the feedthru to allow for tighter clearance.  I removed as much of the PCB to where the mounting screw holes still should have enough clearance.  It's better but still not centered

[xrunner]

Trimmed off the flashing after they tapped the hole for the feedthru to allow for tighter clearance.  I removed as much of the PCB to where the mounting screw holes still should have enough clearance.  It's better but still not centered

Jeepers ... 

How hard can it be to design a case with proper spacing? I guess they made a prototype and it didn't fit right - f*ck it, go to production!

I can make that case on my 3D printer, it's not too hard to design. Need to look back on your waveguides thread and look at the materials used.

[joeqsmith]

It would be good to move the PCB forward, maybe remove that edge as well.  I'll take a baseline like it is shown.   With you planning on supporting HF,  I'll sweep from 1 to 450MHz, -80 to 6dBm.   If you would like a different range, let me know.   It's all automated so not a big deal to set it up.   After this I'll make the mods I recommended and see how the performance degrades.

[xrunner]

It would be good to move the PCB forward, maybe remove that edge as well.  I'll take a baseline like it is shown.   With you planning on supporting HF,  I'll sweep from 1 to 450MHz, -80 to 6dBm.   If you would like a different range, let me know.   It's all automated so not a big deal to set it up.   After this I'll make the mods I recommended and see how the performance degrades.

I think I got a little to excited about using the ham HF bands with this coupler. The Putnam coupler says the range is from 25 MHz to 400 MHz. When I first got it I proclaimed I could get away with using it up to 450 MHz, and I got away with that bold statement.

However, I am not so lucky with the ham bands below 30 MHz. Well partly lucky anyways. I did a FWD port scan and found out I could get away with using one correction factor down to ~16 MHz. This would cover the ham bands from 17 m to 10 m. but it would not cover the rest of the ham bands below that. So I'm going to stick with the ham bands I originally started with: 6 m to 70 cm.

So you could sweep from 50 MHz to 450 MHz please.

[joeqsmith]

50 to 450 it is.  50MHz & 10dBm steps fine enough?   

Couldn't solder it up with it that far off.   Went ahead and cut the mounting holes then filled them with solder to give the screws something to push against.  Gets rid of much of the air gap.  Still not great but I suspect the results will be fairly close to yours.

I inspected the second Chinese connector and saw no signs of damage.  I then measured it's pin and dielectric depth (0 to -10).  Both were in spec.   The replacement Amp connector also checks good. 

I manually ran a quick sweep with it connected to my meter and RF generator and it appears to function so I went ahead and measured S11 from 1-500MHz.   Yellow is unpowered, Red powered.  It could use some improvements. 

[xrunner]

I manually ran a quick sweep with it connected to my meter and RF generator and it appears to function so I went ahead and measured S11 from 1-500MHz.   Yellow is unpowered, Red powered.  It could use some improvements.

Yea that looks like the plot of my boards before I changed the input components.

New display integrated into project. Primitive look for now. Ugly fonts, no graphics, meaningless numbers. But its working. Plenty of time to pretty it up later on.

Now looking at all options for the case. Many to choose from on Amazon. I must take some time doing this. Also another new item on the way too ... 

[joeqsmith]

Bird 43 with 20 X $300 elements?   

10dB is a bit course.  Dropped it to 1.  Also added a 1MHz sweep.   It's running now. 

[joeqsmith]

1M, 50M to 500M with 50MHz steps, -80 to 6dBm, 2dBm steps.

[xrunner]

Nice set of plots thanks.

[joeqsmith]

Removed all the front end and scraped off some of the mask to attach the foil.   C7 & C8 will be changed to a 1000p NPO.  C6, L1 & R3 are gone.   C6 will now be a short and the termination will be 2 X 105ohm from the old C6 pads to the plane. Basically what I had suggested.   I'll run it later today and see how bad I messed things up.   

[joeqsmith]

Still needs some improvement but a big step in the right direction.   

[xrunner]

Still needs some improvement but a big step in the right direction.

Yea that's a nice improvement. 

On my end I was looking at the way I want to display data on the display screen of my project. The library I'm using for the display (u8g2) has a whole bunch of optional fonts and icons, and it takes a lot of time to check out the ones that I might want to use. Meaning I have to try each one and compile a new image to look at each time. Just looking at the fonts on the web page really isn't good enough.

I also looked at a bunch of enclosures on Amazon and trying to decide which one to order. It has to have room for the Putnam coupler, but at the same time not be too large. I already knew I would have to make custom cables to be able to get this all to fit, a custom mounting plate for the microcontroller board, custom mounting for the AD8307 PCBs, and custom front and back panels (all these 3D printed and done in Fusion 360).

Received several of the items I had already ordered in the mail today, should get the last one delivered tomorrow. Will of course show what I have accumulated in an upcoming post ...