CB and Ham Radio Techs Love Their Bird Wattmeters

[joeqsmith]

Did you find out if it is asymmetrical?

[xrunner]

Did you find out if it is asymmetrical?

The datasheet says nothing about this. So I just now checked the resistance looking in one end with no load on the "output".

Looking in one end (female SMA) I measure 52.5 ohms.

Looking in the other end (male SMA) I measure 48.7 ohms.

If I connect a 50 ohm load to the opposite end and repeat these measurements, I get the same results.

What say you on this?

When I checked the attenuation on the spectrum analyzer I had the tracking gen. going into the 52.5 ohm end.

[joeqsmith]

If your goal isn't to provide us with details about this part, I would just sweep the power in the direction you plan to use it.   

.... All this will be done for the entertainment of all who care to read this thread.    ...

... It's just irritating that they put this stuff out with claims that are not being met. ...

consider the cost is under $50, I wouldn't expect it to perform like a name brand part.  Then again, I recently looked at $700 Keysight handheld meter that also came with lots of claims.   

[xrunner]

Oh I'll keep doing a few more tests like the power handling then I'll decide whether to return it. I'd say it's a symmetrical attenuator since it is close to the same value of resistance both ways and adding a 50 ohm load doesn't change it much, which would be expected since it is a large value of attenuation.

[joeqsmith]

I don't follow why you feel the input output resistance would tell you anything about the power symmetry. 

Some time ago we were playing with attenuators and I built up a 50dB. I'm imagining the same values for each stage but the input side being rated for higher power. 
https://www.eevblog.com/forum/projects/20db-rf-attenuator-seeking-feedback-to-improve/msg2906026/#msg2906026

[xrunner]

I don't follow why you feel the input output resistance would tell you anything about the power symmetry. 

Well I might be wrong about it. That's why I gave you the measurements for your opinion. What is your opinion of the device's symmetry given the measurements I gave you, and considering I cannot remove the inner substrate? Do you think it's a symmetrical or asymmetrical attenuator?

[joeqsmith]

I don't follow why you feel the input output resistance would tell you anything about the power symmetry. 

Well I might be wrong about it. That's why I gave you the measurements for your opinion. What is your opinion of the device's symmetry given the measurements I gave you, and considering I cannot remove the inner substrate? Do you think it's a symmetrical or asymmetrical attenuator?

My opinion has very little weight.  Without documentation,  I would have to buy a few of them for destructive testing.   

[xrunner]

My opinion has very little weight. 

I thought most here thought otherwise, or so I had observed. 

Without documentation,  I would have to buy a few of them for destructive testing.

OK.

This afternoon I'll take another look at how to remove the inner attenuation part. I really don't care if it breaks all that much, I've spent way more that $19 on dates and had less fun. If I still don't think it will come out after reasonable tapping force, I'll move on to testing the power handling aspect of it.

[joeqsmith]

I wouldn't have expected these smaller attenuators to be unidirectional but there have been posts on this site of others finding otherwise.  I have not attempted to verify their claims.   If your attenuator isn't even marked for the value, I can believe they also wouldn't mark which connector was the input/output. 

The parts I bought from Barry Industries to rebuild that dead JFW attenuator are bidirectional.  Those were rated to 150W.  That larger attenuator I have is also bidirectional.   

[xrunner]

Here's the test setup to test for temperature at max power (10W). I am monitoring the temp with a K thermocouple. I have a DMM monitoring the voltage drop across the 40 dB att. There is a small 50 ohm resistor (~ 0.25 W size) acting as the load and since we have a 40 dB att. it dissipates very little power.

In the pic is shown settings for 1 W being dissipated by the 40 dB att. The temp is stable @ 42 C (108 F). More tomorrow ...

P.S. Oh and a fan ... if needed.

[A.Z.]

Close-ups of the 40 dB attenuator construction. The resistive elements are on a ceramic substrate. Tried to extract it to no avail.

You might think the end contact point is a little dirty (and the opposite end as well) and may be a good thing to brighten it up. Maybe that's causing the attenuation to be off ... oh wait the attenuation is too low we're going in the wrong direction with that.

 

Can it take 10W? Place your bets.

well... it could take even 1KW, it all depends from the effect you're trying to achieve !

[xrunner]

well... it could take even 1KW, it all depends from the effect you're trying to achieve !

I'm trying to create the possibility of a dramatic event for the audience, or at least the "anticipation" of one. I fear making power meters is too boring for them. I might be more entertaining with this setup. It's like a reality show series. Like yesterday I did 2W, today I increased it to 5W. Tomorrow it will be 10W. Keep the audience coming back to see what happens.

As stated above, I dialed in 5W. I waited for the temp to stabilize, and as you can see, it reached 93 C with no forced air in about 2 minutes.

The max operating temp as stated is 100 C. Since we're at 93 C already @ 5W she ain't gonna handle 10 W (for a reasonable amount of time) without forced air cooling. Should need a little over 22V. So I will record the time it takes to reach a case temp of 100 C with 10W and no forced air. Then I will see if the fan I have now will let it run @ 10W for several minutes keeping the case at 100 C or less. In the end, if it can take 10W and be kept below 100 C with a fan, and it doesn't blow, it may be a useful attenuator for my purposes.

[joeqsmith]

With the BNCs on there, any plans to flip the orientation to see if it has any effect?   

I'm trying to create the possibility of a dramatic event for the audience, or at least the "anticipation" of one. I fear making power meters is too boring for them. I might be more entertaining with this setup. It's like a reality show series. Like yesterday I did 2W, today I increased it to 5W. Tomorrow it will be 10W. Keep the audience coming back to see what happens.

Are you going to be able to loan out the Bird meter for the grand finale? 

[xrunner]

With the BNCs on there, any plans to flip the orientation to see if it has any effect?   

I can do that yes.

Are you going to be able to loan out the Bird meter for the grand finale?

I wish I could get it. A real face-off contest. A Bird model 43 with a selection of elements vs. my home made power meter with just one attenuator of my choice. That would be very entertaining. 

[joeqsmith]

I was curious about the accuracy of the data you collected for your attenuator using the Rigol.   They spec 38.8 - 41.2dB to 3GHz where you are showing it out be more than 5dB at 1.5GHz.

https://www.eevblog.com/forum/rf-microwave/cb-and-ham-radio-techs-love-their-bird-wattmeters/msg4415494/#msg4415494

So I bought a few of the same part from the same supplier along with another 10W 40dB attenuator from BECEN that cost the same amount.  The BECEN part has a lot more surface area and may do a better job.

I checked and inspected the connectors.  The seem alright.  Female pin depth was a bit low on the ZETECH but all well within spec.   I then swept all three from 300kHz to 3GHz using the LiteVNA.  The two parts from ZDTECH are well within their spec and the one part is right down the middle.  The BECEN is a bit out but I am not even sure what they spec this part to.   

We paid about $130 USD for the LiteVNA.  Add on say $100 for my homemade cables, adapters and my sorted load standard.   I'm fairly confident with the results and suspect if I were to run these on my Agilent VNA using a better calibration, we could clean up a lot of that ripple but you would get the same basic results.

I then repeated your temperature test using the BECEN part.  First, I removed that sticker from the heatsink to allow it to breath as much as possible.   Shown with roughly 5W (V=15.8V) after about a half hour.   The ambient temperature is shown on channel 2.  So about a 42 deg C rise.    I tried flipping the orientation and the results were the same.    I tried not to create any air flow in the office when running this test. 

So, I wonder now, what's up with that Rigol?  It isn't by chance a model 43 is it?   

[xrunner]

I was curious about the accuracy of the data you collected for your attenuator using the Rigol.   They spec 38.8 - 41.2dB to 3GHz where you are showing it out be more than 5dB at 1.5GHz.

...

I was just completing my 10W testing and I read this. I re-did the plot and it's shown below with the first one. I did that plot before I opened up the attenuator for inspection. The high frequency response is now better. I suspect the reason was that the attenuator simply works better now because I cleaned some of the contact points inside. As you can see in the pictures I took with the microscope there was a substance on the threads I suspect to prevent water ingress. Also the end contact point on the ceramic substrate was burnished on both ends by me because it didn't look very clean. So now the response is better.

I will post the results of today's testing later tonight

[joeqsmith]

Where you have the marker is 38.1 but even that isn't the worse area.   Is the Rigol really this bad or is something else going on?   You have the Narda part.  How does it compare?

***
Attached, using the best of the three as a reference and subtracting that error off the other two.  This at least removes the majority of the ripple and gives us a better idea of how they behave.

[xrunner]

Where you have the marker is 38.1 but even that isn't the worse area.   Is the Rigol really this bad or is something else going on?   You have the Narda part.  How does it compare?

We both have different physical parts and systems. Unless you come over and get my part to test or I do the same with yours we're doing the apples and oranges thing. Maybe your parts are out of spec but your test instrument is out of cal so it's actually making it look better than it is - you know how the game goes ... 

But I'll test a different attenuator tomorrow.

[joeqsmith]

Look forward to it.   

No doubt I would also expect some slight differences but your setup looks like it was out by 7dB and your not even at half the rated frequency.  This is why I questioned it.   I was surprised when you jumped to the conclusion the part was bad without showing any further data which is why I procured a few to test.   

For the LiteVNA I use the supplied standards with the V2Plus4 and the ideal model.  I did replace the load with one I had sorted from a batch from Mini-Circuits parts using a set of Agilent standards as a reference.   I could characterize these standards and use the data to tighten up the measurements but never felt the need.   

For the Agilent VNA, I have a set of home made standards that were characterized against a set of Agilent & HP standards.  I use the best load I sorted as my SMA reference.  One of the nice things about that system is it supports calibration methods beyond SOLT.  I also own a set of purchased standards for it that were also characterized against a set of Agilent standards. 

While my LiteVNA and poor mans cal standards show all three basically in spec,  I can also fire up my old Agilent and remeasure them a bit more accurately if you like. 

[xrunner]

Next up was 10W. It took 2m 16s to reach 100 C and would have kept rising had I not turned on the fan. With the fan on, the temp decreased to 75 C and stayed there. I left it on for a couple of minutes but no Majik Smoke appeared, so I shut it down.

I reversed the connections to the attenuator and noted no difference in readings.

So, it can take 10W for minutes at a time without a fan, and certainly it could have withstood 10W for longer than that with a fan. If I was testing a transmitter I wouldn't key it up for minutes at a time even adjusting the power per the manual.

So it looks like it can be used in a power meter made from the AD8310 to extend the range to something like -73 to +40 dBm which is a pretty darn good range. I could get fancy by sensing the temp and turning on the fan automatically, maybe at 60C. But The simpler approach would be to just have the system turn on the fan anytime you are selecting to measure power over +13 dBm (switching in the 40 dB att.) which was my max power level in the power meters I made to date.

I only chose the small footprint 10W 40 dB attenuator so the case wouldn't be too big. But there is no reason a person could not use a bigger case and fan and make a power meter to handle -73 to +50 dBm 100W (100W) ... or more.

I'd have a manual switch to select the higher range, which would flip the coaxial relays and turn on the fan at the same time, so the user has to realize to switch in the attenuator. But, even with the expensive commercial power meters, if you the user aren't paying attention and apply too much power with bad settings or setup, you are going to pay for it. Not paying attention is the reason why some people can't have nice things. Can't get past human error in the end.

To paraphrase some lines from 2001: A Space Odyssey

"Well HAL how do you explain this power meter burning up when it had a clearly labeled switch for the user to select a built-in attenuator?

"Well, I don't think there is any question about it. Things like this have cropped up before, and it has always been due, to human error ... "

[joeqsmith]

I'm sure it's not the case but in every photo of the setup, it almost looks like the attenutator's heatsink has Kapton tape over it with hot glue holding the TC in place.   Obviously covering the heatsink would make it very ineffective.   

[xrunner]

I'm sure it's not the case but in every photo of the setup, it almost looks like the attenutator's heatsink has Kapton tape over it with hot glue holding the TC in place.   Obviously covering the heatsink would make it very ineffective.

I has kapton tape holding in some thermal padding so the thermocouple is held tight so the temp is measured accurately. Air can flow through the fins just fine.

[xrunner]

Is the Rigol really this bad or is something else going on? 

Something else ...

The tracking generator has three power levels : 0, -10, and -20 dBm

For higher attenuation levels being measured I needed to set the default power to 0 dBm. After that it gives the plot below. Just had a very low level being measured when using -20 dBm power and 40 dB attenuation.


*****

Hey BD are you going to post anything on your recent adventures into the power measuring area? I thought you got another part ... ?

[bd139]

Still here. Not had the time yet. Also the AD8307's I landed are possibly fake. Need to do a test rig

Also other projects (note the new gear...)

Build out of Jay Diddy B ESR adapter...



Tek 321A...

[joeqsmith]

You had mentioned earlier, 
"Well, I don't think there is any question about it. Things like this have cropped up before, and it has always been due, to human error ... "

The old Agilent VNA I own is a bit of a pain to drive.   Most of the time, I run it from the PC using the same software I wrote for the other VNAs I own.  The advantage of having a common interface is they all basically drive the same.   

The LiteVNA with Dislord's firmware allows me to change the drive strength for both synthesizers.  Its a pretty crude and the noise floor isn't all that great.   Still, for the $130 they sell them for with a few bits added to the mix, it throws up some decent numbers.   Its not as good as the original NanoVNA for lower frequencies.  Then again, my Agilent VNA is limited to 300kHz so I still keep my old HP around for low frequency experiments.   

Is the error at the start of the data sets an artifact how the Rigol works?   Is it effected by changing the start freq or slowing the sweep rate?   Is there an AC/DC coupling that you can set to DC?   Seems like there would be a way to correct it as I doubt it's due to the attenuator.

Can you export the data to a Touchstone file and upload it here?   We could then do a better job comparing results.