I would really appreciate your help, since I'm not a specialist in engineering, but a biologist.

I'm performing experiments where I apply RF broadband signal. I have a signal generator (Rohde & Schwarze) and an Amplifier (emv - Amplifier Research). The only problem for me is a transmitting antenna.

I need a broadband signal as homogeneous and as clean of extra noise (harmonics etc.) as possible in the 1m*1m*0.5m space.

So far I've been using a loop antenna made of a copper tube. But this antenna is not transmitting good in my range (10MHz broadband signal from 80 to 90 and from 300 to 310 MHz). The diameter of the antenna is 2m, and I thought it should be the major problem.

Does anyone have a suggestion how should I optimize the setup: different size of antenna, dipole antenna instead of a loop antenna?

  • 2
    \$\begingroup\$ Yes, antennas need to be designed specifically for your frequency range. Large loop antennas are typically used for very LOW frequencies, and you're dealing with significantly higher frequencies, so it's absolutely no surprise this doesn't work well. You can actually simply go to electronics suppliers (farnell/element14, digikey, mouser, …) and just buy an antenna that works at your frequency, and it will be better than what you use. What's optimal, however, is totally unclear from what you describe. \$\endgroup\$ – Marcus Müller Jul 7 '17 at 13:04
  • 2
    \$\begingroup\$ Way too much hand waving, inconsistent statements, and a confusing incomplete sentence. \$\endgroup\$ – Olin Lathrop Jul 7 '17 at 13:06
  • 1
    \$\begingroup\$ You would want to ask "What do I need to look for in an antenna for this application:*LENGTHY DESCRIPTION*", generally. The manufacturer of your amplifier tells us nothing about its output powers, etc. 10 MHz at 300 MHz is, by what antenna people would define, not "broadband" (it's only 1/30 of the center frequency), etc, so generally, it's pretty much impossible to recommend something unless you give us a far more detailed description of what you're doing. \$\endgroup\$ – Marcus Müller Jul 7 '17 at 13:06
  • 1
    \$\begingroup\$ The cleanness of the signal has everything to do with the transmitter and amplifier, and nothing to do with the antenna. \$\endgroup\$ – Lundin Jul 7 '17 at 14:04
  • \$\begingroup\$ Give him a break, guys. This isn't his field and he came here for help. \$\endgroup\$ – GroundRat Jul 7 '17 at 14:10

Ok, ignoring the antenna for the moment, there is a bigger issue here.

Your test cell is just way to small to get your target into the far field, you typically need to be several wavelengths away from the antenna (and the cell walls) before measurements become sane, never mind having a homogeneous field.

@100Mhz, you have a free space wavelength of about 3m, give or take, so you would want your test cell to be very much bigger then that, at 300MHz you have a wavelength of ~1m which is going to cause all sorts of fun as it approximately matches the dimensions of your test cell.

I would go for two different aerials for these two frequencies, much less annoying and a dipole or 1/4 wave against ground would be obvious and simple (Lots of web sites showing examples of how to build these, they are trivial).

Go and find your local ham radio club, they usually have someone who will know how to play RF in the VHF region, and some of this stuff is not simple.


I'll give this a try, but I'm making the following assumptions based on your question:

Assumptions: It sounds like you need an antenna that works with a signal with the following characteristics: 10 MHz bandwidth with a center frequency of 85 MHz and another signal with 10 MHz bandwidth and a center frequency of 305 MHz. With those assumptions made, you have two options:

OPTION 1: You build a 1/4 wave dipole or whip for each frequency range (wavelength = speed of light/frequency). So...(wavelength of 85 MHz ~ 3.5 M, so 3.5*.25=.875 M radiating element for the 80-90 MHz range) AND (wavelength of 305 MHz ~ 1M, so 1M*.25 ~ .25 Meter radiating element for 300-310 MHz range).

The easiest way to build the antennas will be a whip consisting of a coax cable stripped at one end to expose the necessary length of center conductor (.875M and .25M). Plug the other unstripped end to your transmitter or pre-amp.

OPTION 2: Alternatively, if you have money to spend, look at AH Systems for an EMC antenna. A Biconical or Log Periodic built for the necessary frequency ranges and power output would be best.

Unfortunately, the space you are working in (1M * 1M * 1/2M) will put you in the near-field (less than 2 wavelengths away) and make accurate measurements difficult. You may need to build a larger screened room/faraday cage.


I need a broadband signal as homogeneous and as clean of extra noise (harmonics etc.) as possible in the 1m*1m*0.5m space

It is important to understand that the antenna plays no role in a "homogeneous" signal nor the generation of noise or harmonics. These are generally functions of your signal generator, amplifier, and transmission lines. There is the notion of a low noise antenna, but this is generally applied in discussions of receiving antennas.

There are a few classes of broadband antennas that achieve that distinction without the use of variable tuning elements over their intended range. Two such examples that you can investigate are the log-periodic and the discone. The log periodic will be directional with a primary lobe. The discone is omnidirectional in the azimuth direction when orientated vertically.

...in the 1m*1m*0.5m space

You say this with no context to understand your intent or requirement. If you mean the antenna dimensions are restricted to this volume, your 2 meter diameter loop violates that condition. If you mean that the RF signal must be concentrated within that volume then you must supply more application details before a solution can be proposed. A drawing would also then be helpful.

  • \$\begingroup\$ I just tried out the idea of @GroundRat with a coax cable. It seems to work at least on a small scale. The problem is, that I don't have a problem of the near-field, since I need strong B-field, not E-field. As a matter of fact, reducing the E-field for me would be an advantage (as far as I understand loop antennas are actually better for this purpose). I'll try to attach the coax cable around my experimental table (ca 1m*1m) and to measure the field. \$\endgroup\$ – user3719737 Jul 13 '17 at 16:58
  • \$\begingroup\$ As far as I've understood the impedance matching on the feed point is important to get high B-field. Have no idea whether I can solve this problem for both frequency ranges with the same antenna. \$\endgroup\$ – user3719737 Jul 13 '17 at 16:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.