RF Explorer recorded levels

[Jase]

Hi, just a question re the results i am getting during some initial testing of a leaky feeder system.
I have fabricated a testing "cart" to facilitate a power unit, laptop, Marconi comms set and a fixed bracket for the rf explorer unit. The Marconi aerial is connected directly to the set (simple bnc 1/4 whip/helical) and is 0.7m from floor. The r.f. Explorers setup is a little higher (on purpose) at around 1.2m.
I have a uhf and vhf transmitter on constant tx, we walk test the coverage route, record the signal level at known intervals. I was expecting a difference in the levels but not the consistant 30 to 35dBm difference I am recording. Can anyone advise if this is expected and normal? Thanks in advance.

[herbert walker]

Jase,
As I recall from radio theory, receive power falls off as a square of the distance.  This relationship is only valid in the far field, which if I remember correctly starts at two wave lengths.  If your initial measurement is performed in the far field, then your subsequent measurements should only be constant if you are walking in a circle or the transmitter is over loading the RF Explorer.  To test for the latter, first measure the transmitter with your antenna connected directly to the RF Explorer and, next measure the transmitter with a 20 dB attenuator inserted between your antenna and the RF Explorer.  If your second reading is not 20 dB less than the first reading, then the transmitter is overloading the RF Explorer.  This is a test that anyone that uses a spectrum analyzer should have in their arsenal as even laboratory grade spectrum analyzers are prone to having their front end over loaded.  The really expensive ones have a overload warning indication to alert you to this condition.

Herb

[Jase]

Hi Herb

Upon re-reading my post, I may have been unclear (apologies). 

“I was expecting a difference in the levels but not the consistent 30 to 35dBm difference I am recording.”

{Jase} The difference I am referring to is between the two receiver sources (RF explorer and Test Set Spec Analyser) – in truth I didn’t expect to see the same results but I thought they would be similar (within a 9dbm).

"As I recall from radio theory, receive power falls off as a square of the distance. 

This relationship is only valid in the far field, which if I remember correctly starts at two wave lengths."

{Jase} This did not even occur to me, we are rarely `so close` to the aerial when using the spectrum analysers.

VHF (160MHz @ 1.88m, 2 wave lengths = 3.76m ) – so my testing is technically in the `near field` ?

UHF (450MHz @ 70cm, 2 wave lengths = 1.4m) – as above – very border line but its possible it is in `near field` although testing was more than 1.4m away for UHF (although only just).

"If your initial measurement is performed in the far field, then your subsequent measurements should only be constant if you are walking in a circle or the transmitter is over loading the RF Explorer."

{Jase}

As above it would appear I am measuring in the near field for VHF and borderline for UHF, I`m never/rarely more than 3.76 away from the radiating feeder on this project.

"To test for the latter, first measure the transmitter with your antenna connected directly to the RF Explorer and, next measure the transmitter with a 20 dB attenuator inserted between your antenna and the RF Explorer."

{Jase}

This is something I have not done yet with the RF Explorer Unit – this is pending upon some spare engineering time, I`ll check and see if I record the 20db loss with attenuator.

"If your second reading is not 20 dB less than the first reading, then the transmitter is overloading the RF Explorer."  

{Jase} Noted and agreed. 

"This is a test that anyone that uses a spectrum analyzer should have in their arsenal as even laboratory grade spectrum analyzers are prone to having their front end over loaded.  The really expensive ones have a overload warning indication to alert you to this condition."

{Jase} Agreed, a lot of care is needed.  The TX sources are transmitting on low outputs via an attenuator so no overloading is likely."

To go back to your first point (nearfield/far field), if I am within two wave lengths then my method of testing is/maybe invalid or at least borderline? aerials should be replaced with near field items?

Also I`ve not calibrated or checked the RF explorer calibration process - the unit is new.

Any advice would be great.

Jase

[herbert walker]

Jase,

“I was expecting a difference in the levels but not the consistent 30 to 35dBm difference I am recording.”

{Jase} The difference I am referring to is between the two receiver sources (RF explorer and Test Set Spec Analyzer) – in truth I didn’t expect to see the same results but I thought they would be similar (within a 9dbm).

{Herb} Try connecting a 50 ohm signal source to the RF explorer and Test Set Spec Analyzer and see what kind of difference you get.  If I connect a -30 dBm signal at 100 MHz to my RF explorer and then to a calibrated specan, I am seeing about a 3 dB difference. So assuming you are using the same antenna and the measurements are taken at the same distance for both receivers, then I would expect the results would be within 9 dB.  Similar measurements I have performed in the past have either been at an open area test site or in an anechoic chamber.  Both test environments simulate a free space environment that is not influenced by nearby reflecting surfaces.  Also, keep in mind that both measurement receivers must use similar resolution bandwidths or you will need to apply a bandwidth correction factor (20* log10(bandwith_a / bandwith_b)) for coherent broadband signals.

"As I recall from radio theory, receive power falls off as a square of the distance.
This relationship is only valid in the far field, which if I remember correctly starts at two wave lengths."

{Jase} This did not even occur to me, we are rarely `so close` to the aerial when using the spectrum analysers.
VHF (160MHz @ 1.88m, 2 wave lengths = 3.76m ) – so my testing is technically in the `near field` ?
UHF (450MHz @ 70cm, 2 wave lengths = 1.4m) – as above – very border line but its possible it is in `near field` although testing was more than 1.4m away for UHF (although only just).

{Herb} Standard compliance testing is either done at 3 meters or 10 meters (30 - 1000 MHz) to avoid near field effects.  At 1.4m you might consider using a loop probe or e-field and see what kind of results you get.

"If your initial measurement is performed in the far field, then your subsequent measurements should only be constant if you are walking in a circle or the transmitter is over loading the RF Explorer."

{Jase}
As above it would appear I am measuring in the near field for VHF and borderline for UHF, I`m never/rarely more than 3.76 away from the radiating feeder on this project.

{Herb} As suggested above, I would try using a less sensitive transducer (e.g. loop probe) where my initial measurement was in the -30 dBm range.

"To test for the latter, first measure the transmitter with your antenna connected directly to the RF Explorer and, next measure the transmitter with a 20 dB attenuator inserted between your antenna and the RF Explorer."

{Jase}
This is something I have not done yet with the RF Explorer Unit – this is pending upon some spare engineering time, I`ll check and see if I record the 20db loss with attenuator.

{Herb} You're right. Sanity checks do take time.

"If your second reading is not 20 dB less than the first reading, then the transmitter is overloading the RF Explorer." 
{Jase} Noted and agreed.

 

"This is a test that anyone that uses a spectrum analyzer should have in their arsenal as even laboratory grade spectrum analyzers are prone to having their front end over loaded.  The really expensive ones have a overload warning indication to alert you to this condition."
{Jase} Agreed, a lot of care is needed.  The TX sources are transmitting on low outputs via an attenuator so no overloading is likely."

To go back to your first point (nearfield/far field), if I am within two wave lengths then my method of testing is/maybe invalid or at least borderline? aerials should be replaced with near field items?

Also I`ve not calibrated or checked the RF explorer calibration process - the unit is new.

{Herb} I've found the RF explorer to be perfectly suitable for making relative A/B comparisons.  It comes up short at making absolute measurements because the rbw is calculated rather than being manually settable.  For EMC testing, if I know I need to drop a signal by 10 dB and I see a 15 dB change on the RF Explorer after performing an emi fix. I am pretty certain the signal will be down at least 10 dB during compliance testing.  If I need absolute measurements at a specified resolution bandwidth I have to turn to a high cost Rohde Schwarz specan.

Herb