Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

antenna testing chamber

Status
Not open for further replies.
mihir08

mihir08

Member level 4
Joined
Dec 28, 2009
Messages
70
Helped
7
Reputation
14
Reaction score
6
Trophy points
1,288
Location
India
Activity points
1,787
hi folks, we are testing radiation patterns for different types of antennas in our lab . let me explain you the setup : we have a antenna test kit from signet , basically a tx source at 800 Mhz using folded dipole and at the rx we have a stepper motor turntable on which we fix Rx antenna under test . the problem is that lot of interference occurs and even if some one walks from the test region the readings are affected. please suggest some ways to isolate our test bench . if i use a metallic mesh i think reflections will occur and cant use expensive obsorbers . want to make something effective enough from scratch - wood , foam , aluminum foils :) etc .. !!!!

thanks and cheers ,, always good to be on eda
 

mihir08 said:
hi folks, we are testing radiation patterns for different types of antennas in our lab . let me explain you the setup : we have a antenna test kit from signet , basically a tx source at 800 Mhz using folded dipole and at the rx we have a stepper motor turntable on which we fix Rx antenna under test . the problem is that lot of interference occurs and even if some one walks from the test region the readings are affected. please suggest some ways to isolate our test bench . if i use a metallic mesh i think reflections will occur and cant use expensive obsorbers . want to make something effective enough from scratch - wood , foam , aluminum foils :) etc .. !!!!

thanks and cheers ,, always good to be on eda
Click to expand...


How about 'Air'?

As in an OATS (Open Air Test System) range - out in the parking lot maybe?

Jim


Testing for EMC compliance: approaches and techniques - Mark I. Montrose, Edward M. Nakauchi - Google Books
 
Last edited:

RF_Jim said:
How about 'Air'?

As in an OATS (Open Air Test System) range - out in the parking lot maybe?

Jim


Testing for EMC compliance: approaches and techniques - Mark I. Montrose, Edward M. Nakauchi - Google Books
Click to expand...

Cellphones work around the 800MHz frequency, don't they? You'd have much interference and you can't take away the multipath effects (it still reflects of the concrete floor or w/e).
Best setup would be something like this: 20120213_155520.jpg
Although I have no idea what material was used.
 

john fiction said:
Cellphones work around the 800MHz frequency, don't they? You'd have much interference and you can't take away the multipath effects (it still reflects of the concrete floor or w/e).
Best setup would be something like this: View attachment 69596
Although I have no idea what material was used.
Click to expand...


OATS testing is an industry accepted test method, although attention must be paid to nearby potential reflectors.

Recall, Johnny, the RADAR equation and the 1/x^4 power as it relates to returned power from 'targets' (or any nearby reflector). Remember, many times these tests are performed on the 'cheap' or very quickly; if something more permanent needs to be set up, that is a different issue.

And suitable selection of an RF test frequency and us of the narrow BW on the test receiver (with generator set for CW) can mitigate potential interfering terrestrial RF sources. It would also help to use a directional antenna on the source, to minimize 'stray' RF that might reflect off walls and such.

For antenna performance, it is usually necessary to find the angular 3 dB points; this can even be done indoors, ignore the 'reflections' that may only be down 10 or 15 dB and find the 3 dB beamwidth points for that measurement, and realize this is a first-order approximation (some data, even rough data, is better than no data).

Before indoor ranges and so-called compact ranges OATS ranges were the only option. I have have had access and use of all three.

Did you make reference to the cited text by any chance, where some of these issues are addressed?


Jim
 

RF_Jim said:
OATS testing is an industry accepted test method, although attention must be paid to nearby potential reflectors.

Recall, Johnny, the RADAR equation and the 1/x^4 power as it relates to returned power from 'targets' (or any nearby reflector). Remember, many times these tests are performed on the 'cheap' or very quickly; if something more permanent needs to be set up, that is a different issue.

And suitable selection of an RF test frequency and us of the narrow BW on the test receiver (with generator set for CW) can mitigate potential interfering terrestrial RF sources. It would also help to use a directional antenna on the source, to minimize 'stray' RF that might reflect off walls and such.

For antenna performance, it is usually necessary to find the angular 3 dB points; this can even be done indoors, ignore the 'reflections' that may only be down 10 or 15 dB and find the 3 dB beamwidth points for that measurement, and realize this is a first-order approximation (some data, even rough data, is better than no data).

Before indoor ranges and so-called compact ranges OATS ranges were the only option. I have have had access and use of all three.

Did you make reference to the cited text by any chance, where some of these issues are addressed?


Jim
Click to expand...

No I didn't refer to the cited text. But yeah, you could say that testing in the environment you are going to use the antenna is the best testing environment.
 

If you want a precise measurement, and can not afford an anechoic chamber, a test outdoors is the way to go! Such tests work very well at 800 MHz.

Just make sure you load up any control/DC lines to the unit, or to your stepper motor, with tons of ferrite clamp-on beads. I have a special wooden table and wooden stools set aside for just such testing.

Rich
 
Last edited by a moderator:

thanks for your replies . i cant take my setup out doors as this is a college lab work ( difficult to convince prof :) . and we have batch of 20 guys performing experiments . cell phones i think is a major problem . i can change my tx-rx frquency . i am not sure about the range but its around 800Mhz - to some XX ghz . will changing the frequency help ?
 

You are 20 guys doing these things without thinking? And your prof., is he just smiling?
Almost everything is wrong set up in your antenna measurement chamber, if you intend to measure antenna radiation pattern.
1. An antenna test chamber without absorbers is not usable to measure antenna radiation pattern at all. You will get reflections from all directions. If you understand ground principles for how radio-waves behaves when radiated against metallic walls, must that be pretty obvious?
2. A dipole antenna as measurement antenna in a chamber, fine if you want to measure everything that comes from ALL directions, but for measuring radiation pattern is it almost forbidden.
3. Antenna under measurement as RX antenna in an anechoic chamber for measuring radiation pattern is a general error. Try google to find basic information how and why to use a anechoic chamber for antenna measurement.

Why not to use a dipole as measurement antenna => you will reduce quite zone to nothing. A common preferred antenna selection is a horn antenna with correct directive radiation pattern and well defined polarization.
Functional absorbers are not easy to do by yourself as absorber geometry is vital for total chamber function including were the quiet zone will be localized, were the turntable should be placed.
See this which shows an example of a simple anechoic chamber. It can as well be a rectangular chamber, but it cost more absorbers. Even if the chamber have absorbers all around are they not total absorbing all RF energy, so do not use a measurement antenna that have a radiation patter suited to receive 1:st reflections (as a dipole antenna would do).
It is possible to simplify the chamber further then what is showed in the picture, by only having absorbers on the wall behind the turntable, at cost of reduced dynamic as even good absorbers not attenuate 100% and a horn antenna also will see some of the reflection coming from surrounding outside of the wall, but it is better then nothing.
If we in this later case assumes opposite signal direction, DUT is used for RX will you measure signal from all directions which would result in reduced dynamic in your measurement results for radiation pattern and phase&polarization calibration for the chamber would be depending on DUT antenna properties. Full chamber calibration for each new DUT measurement, is a vast of energy.
There are a lot of other things, not covered here, that are more complex and that also must be handled if you should be able to have a chamber with repeatable results and low measurement error. Keeping measurement result reliable and repeatable do more require that you know what you are doing, then quality of chamber as that part is static.

In many ways is it much less complex to measure in free field. However do you still need to think a bit about what you are doing. Assume for example that you want to measure a cell phone antenna radiation pattern and you can not use a rural enough measurement location, free from interfering transmitters.
Then can you at least reduce interfering levels compared to signal level from DUT a lot by also here use a directive measurement antenna for RX as you then can select quietest direction and better reduce ground reflections if DUT is placed high enough. Another common measurement error is due to measurement coaxial cables that is connected to DUT and extends DUT ground-plane, adds standing waves and reflections. For a cell phone, if you can control it via a communication tester, can it be set transmitting without external cables, which reduces cable errors and simplifies that measurement part as no external RF generator is needed.
Drawback with using a directive measurement antenna, is that a such antenna needs to be placed at a longer distance, compared to a dipole, to still be in fare field, and if it not is a static measurement range, must antenna orientation be more exact verified then if a omnidirectional antenna is used.
 
Last edited:

I sounds like you do NOT want an "antenna testing chamber".

Instead, it sounds like you want 20 EMI shield boxes, where experiments can be done inside the box by 20 students at the same time.

You can get copper screen and form a box with no bottom. Then have a copper plate on the table top. Put your devices on the plate, put the copper box over it, and put some weights on the top to get contact between the two.

If you have more money, you can buy a professional shield box, like this:

**broken link removed**
 

E Kafeman said:
You are 20 guys doing these things without thinking? And your prof., is he just smiling?
Almost everything is wrong set up in your antenna measurement chamber, if you intend to measure antenna radiation pattern.
1. An antenna test chamber without absorbers is not usable to measure antenna radiation pattern at all. You will get reflections from all directions. If you understand ground principles for how radio-waves behaves when radiated against metallic walls, must that be pretty obvious?
2. A dipole antenna as measurement antenna in a chamber, fine if you want to measure everything that comes from ALL directions, but for measuring radiation pattern is it almost forbidden.
3. Antenna under measurement as RX antenna in an anechoic chamber for measuring radiation pattern is a general error. Try google to find basic information how and why to use a anechoic chamber for antenna measurement.

Why not to use a dipole as measurement antenna => you will reduce quite zone to nothing. A common preferred antenna selection is a horn antenna with correct directive radiation pattern and well defined polarization.
Functional absorbers are not easy to do by yourself as absorber geometry is vital for total chamber function including were the quiet zone will be localized, were the turntable should be placed.
See this which shows an example of a simple anechoic chamber. It can as well be a rectangular chamber, but it cost more absorbers. Even if the chamber have absorbers all around are they not total absorbing all RF energy, so do not use a measurement antenna that have a radiation patter suited to receive 1:st reflections (as a dipole antenna would do).
It is possible to simplify the chamber further then what is showed in the picture, by only having absorbers on the wall behind the turntable, at cost of reduced dynamic as even good absorbers not attenuate 100% and a horn antenna also will see some of the reflection coming from surrounding outside of the wall, but it is better then nothing.
If we in this later case assumes opposite signal direction, DUT is used for RX will you measure signal from all directions which would result in reduced dynamic in your measurement results for radiation pattern and phase&polarization calibration for the chamber would be depending on DUT antenna properties. Full chamber calibration for each new DUT measurement, is a vast of energy.
There are a lot of other things, not covered here, that are more complex and that also must be handled if you should be able to have a chamber with repeatable results and low measurement error. Keeping measurement result reliable and repeatable do more require that you know what you are doing, then quality of chamber as that part is static.

In many ways is it much less complex to measure in free field. However do you still need to think a bit about what you are doing. Assume for example that you want to measure a cell phone antenna radiation pattern and you can not use a rural enough measurement location, free from interfering transmitters.
Then can you at least reduce interfering levels compared to signal level from DUT a lot by also here use a directive measurement antenna for RX as you then can select quietest direction and better reduce ground reflections if DUT is placed high enough. Another common measurement error is due to measurement coaxial cables that is connected to DUT and extends DUT ground-plane, adds standing waves and reflections. For a cell phone, if you can control it via a communication tester, can it be set transmitting without external cables, which reduces cable errors and simplifies that measurement part as no external RF generator is needed.
Drawback with using a directive measurement antenna, is that a such antenna needs to be placed at a longer distance, compared to a dipole, to still be in fare field, and if it not is a static measurement range, must antenna orientation be more exact verified then if a omnidirectional antenna is used.
Click to expand...

This is seriously a good post. Just too add, you CAN ONLY AND MUST DEFINITELY INFINITELY measure radiation pattern in an anechoid chamber. The anechoid chamber is full of pyramidal sponges that is creepy looking, the purpose of that room is to prevent reflection of signals on disturbing your antenna measurements.

Read about anechoid chamber to the threadstarter!
 

I fully agree with E kafeman, use a directive antenna (that points towards a region with low reflection (for example non-coated window class or light weight wall) ).

That directive antenna can be anything, as long as it is directive. I frequently use 4 element patch arrays (as they are not too big and have good directivity/side lobe suppression). For calibration you change the DUT for a dipole or other known antenna. Check whether the results are within expectation to avoid or confirm other (measurement) errors.

Use the shortest distance between directive antenna and DUT, but make sure you are in the 1/r2 field decay (fraunhofer) region. Best is to verify this experimentally. You have to find a compromise between errors because of reflections and errors because of not being in the fraunhofer region.

If you move the complete setup within the college room (horizontally and vertically), you can assess the influence of reflections. If influence is too large, show this to your professor to convince him you need a bigger (empty) room.

With such a setup you can measure the parts of the antenna pattern where there is significant gain, but don't expect that you can measure low side lobes and nulls with good accuracy.
 

thanks a lot for your replies . what i can interpret from this discussion is that instead of dipole we should use a directive antenna for tx . Dut is a t rx side and far feild region is greator than normal dipole . anecoid chamber is unavoidable . please correct if i am wrong some where . thanks once again
 

mihir08 said:
thanks a lot for your replies . what i can interpret from this discussion is that instead of dipole we should use a directive antenna for tx . Dut is a t rx side and far feild region is greator than normal dipole . anecoid chamber is unavoidable . please correct if i am wrong some where . thanks once again
Click to expand...

Yes anechoic chamber is unavoidable. If you have to do the measurement elsewhere, it pretty much useless and inaccurate your data would b.

Directive antenna such as a horn would be good. Remember and understand why you measuring the radiation patterns, whether co-polar or cross-polar.
 

frodonet said:
Yes anechoic chamber is unavoidable. If you have to do the measurement elsewhere, it pretty much useless and inaccurate your data would b.

Directive antenna such as a horn would be good. Remember and understand why you measuring the radiation patterns, whether co-polar or cross-polar.
Click to expand...

Whether or not an anechoic chamber is necessary depends on several things.

If you have a spectrum analyser as receiver, you can definitely find a frequency where you can do your measurements without interference from intended emissions. You will probably use a narrow band "transmitter" so you can reduce RBW of the analyzer to reduce the effect of interference.

Around 800 MHz, distance between DUT and directional antenna will be around 1 m, so influence of reflections can be controlled. A relative large room with one wall not fully reflecting, in combination with a directional antenna may reduce reflections to 20 dB or less, resulting in inaccuracy due to reflections of < 1 dB (for directions of relative high gain of the DUT). You may upgrade your large room by mounting absorbers on one wall only (the wall that is illuminated by the directional antenna).

If you want to measure accurately the nulls and parts of the radiation pattern with low gain (w.r.t. direction of maximum gain), then reflections have to be reduced further. In that case you need an OATS or very good anechoic chamber.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top