Howto Calibrate DIY EMC pre-compliance antenna??

[RFDUK]

Forgot to mention, we use the SSA3032 since January and can highly recommend it, great value. That's used more for radio dev work here.

Our main EMC work horse up to 1.8 Ghz for 20 years is an HP 8591EM. Has a card with EU and FCC limit lines. Screen dumps to PC using HP software. Would highly recommend that too if you can find a good one at the right price. If you are not into fixing faulty RF test equipement, the Siglent 3 year warranty perhaps weighs against a 20 year old machine that may prove unreliable though.

Sounds like you'll be needing CISPR specified quasi peak and average detectors as a minimum, but as you said in an earlier post, the EMC bells and whistles that can come expensive are not must haves. Normal resolution bandwidths of 10KHz for CISPR 9 and 100KHz for 120 are not too much of an issue for pre-compliance tests either.

Martyn.

[dazz1]

Hi
Decent used test equipment here is rare, and because it is rare, people who have it hold onto it, which makes it rarer.
One of the reasons I stay away from ebay is so I don't have to look at the pages and pages of test equipment I can't buy.  The cost of shipping and the risk of damage in transit is too high.

[dazz1]

Hi
I found the attached design document used for the EDGES project.  It includes the design calculations.  I am wondering about the feasibility of using this type of balun on a biconic.  Roberts balun was made for dipoles so the design would need to be recalculated and verified for a biconic.

I don't have access to suitable simulation software to do similar analysis.

Dazz

[RFDUK]

The Roberts balun design not ideal for a biconical as it has a limited operating bandwidth of just over one octave.  Biconical requires 3 or 4 octaves.

Biconical impedance I forget, it depends on the cone angles.

Found thread Topic on this forum: Correctly calculating impedance of a biconical antenna and impedance matching.

Lots of stuff on the web. If you like maths ... http://archives.njit.edu/vol01/etd/1960s/1968/njit-etd1968-006/njit-etd1968-006.pdf

Martyn.

[philpem]

The UK has gone much the same way. The T&M scalpers are widely using ebay to buy equipment, which combined with their asking prices, has massively inflated the cost of even the most clapped-out old bit of "major repair needed" test gear.

You can still get deals, but you need to be ready to play the long game...

[Gandalf_Sr]

... For higher frequencies very low cost PCB Log Periodic antennas are available from Kent at http://www.wa5vjb.com.
Don't be tempted by other products, Kent's antennas are the real deal.

Hope that's interesting, good luck, Martyn.

Wow, not only are those PCB antennas really inexpensive, the References page is amazing - it has the original article by Hidetsugu Yagi on his antenna design dated 1926!

[dazz1]

Hi
I think so too.  I have brought two of them as part of my EMC equipment fit out.

Dazz

[Gandalf_Sr]

Hi
I think so too.  I have brought two of them as part of my EMC equipment fit out.

Dazz

Do you have them yet?  I'm trying to figure out which SMA connector fits the PCBs that take them.

[EDIT] I emailed Kent and he replied - the PCB SMA footprint has 0.2" hole spacings so the TE SMA connectors from Digikey (Pt# A97594-ND) should work and are under $2 each.

[dazz1]

Hi
I just paid extra to get the connectors included loose.  They also have an option to solder them on for you.

Dazz

[Gandalf_Sr]

Oh, I didn't see that option - too late the PCBs have already shipped to me.  Anyway, I order from Digikey all the time so I'll order them tomorrow.

[dazz1]

Hi
I have been looking at precision dipoles.  More than a few implement the Richards balun in solid metal rather than coax cable.  https://avalontest.com/tdk-rf-solutions-dp-1800-precision-dipole-antenna/

This is a metalwork project.

[charliedelta]

Roberts antenna construction attached.

A good reference about antenna calibration.

Theres some excellent information on the NPL web page that you can download after registering. You will certainly understand why these antennas cost so much after reading how these antennas are calibrated.

http://www.npl.co.uk/publications/calibration-and-use-of-antennas,-focusing-on-emc-applications.

[dazz1]

Hi
I find the NPL doc is fascinating reading, not so much for the highly detailed analysis but for the quaintness.

It refers to a photograph  as a "plate".   That type of printing process disappeared decades ago.

The photo of the NPL OATS showing a large 30m x 60m ground plane next to a residential area.  Fig 11 shows the high level of ambient RF.    The most perfect ground plane located on the most imperfect site.   Like setting up a hearing clinic at a rock concert.

This is not a criticism of the content. It is rare to find such a broad coverage of practical antenna design backed up by experience, data and maths.  Nice. 
This looks to be a document written by an old fashioned expert in a well worn dust coat passing on his wisdom before retiring. 

What is missing is details of how to replicate their calculable antennas.

With modern automation methods it should be possible to do in minutes what might have taken a day to do manually.  Calibrating an antenna should be cheap and easy now.

Dazz

[dazz1]

Hi
I have taken a closer look at the NPL calculable antenna design and it would be relatively easy to build.  The main barrier is the cost of a quality WB hybrid coupler like these ones: http://www.richardsonrfpd.com/Pages/Product-End-Category.aspx?productCategory=10136    You would only need 1 coupler for all dipole frequencies

The Minicircuits versions are a lot cheaper but with significantly less BW.

There is also a DIY discussion here: https://www.eevblog.com/forum/rf-microwave/directional-coupler-150mhz-buy-or-diy/


The thing I can't figure out is that Figure 1 specifies an orthogonal coupler (90degrees).  I think it should be a 180deg coupler.

[charliedelta]

Well enjoy the "quaintness" of the NPL document because it wont be long before it is privatized, outsourced and you have to pay for the documents. Much like the current standards that are funded by taxpayers but  you cant access because you have to pay again. I am sure you are familiar with that there in Australia!

Practical construction methods are included in many of the standards.

 I would download a copy of the MIL461G standard which includes a lot of   dimensions needed to build test antennas. Likewise ANSI 63.4  and many of the CISPR standards which give concise details of building the loops for measuring light fixtures  etc etc. The CISPR 25 automotive EMC standards especially the earlier copies had tons of information on building striplines and other test antennas and equipment. There was even details on how to build the conducted emission clamps. Its out there but you need to dig. Not many people like "sharing" the latest CISPR standards. There was also a lot of information published in the NBS and NIST documents and  these documents   had designs for everything needed to do make EMC test equipment. They also included all the maths  associated with calculating and building them. They were good old days when governments freely gave such good information.  Anyway those days are over, all our middle names are PAY according to the many governments of the world.

[dazz1]

Hi

I am designing the mechanics of a biconic antenna.  The attached drawing shows the cone at the base of the antenna. I plan to have an M5 screw going down the centre (not shown). The plan is to support the cone with a PTFE bush (not shown).   See photo.

What I haven't figured out is the best method of connection the balun to the M5 screw.   The Mil-Spec uses a large bolt.   I am thinking that I could use a brass screw and thin down the end so I can solder the balun wire to the end of the screw as shown in the drawing.  Alternatively I could make something like a dome nut and connect that to the balun.

How is this normally done for a biconic??  There are very few photos if inside the antenna termination box on Google.

[coppercone2]

i saw an antenna build like this out of washers, bolts, etc.. no real milling or casting. the guy complained he had very poor performance (on some old radio forum)

Kinda wary but I will try to c ast those holder bits soon