for your information: a collection of articles we find interesting, useful, or amusing
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Nagra VI: Investigating RF "Spray" |
|
|
|
|
(updated with reader feedback) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Since radio mikes are a standard component on most film and television projects, any recorder must be compatible with radio receivers. This is especially important for a recorder that may be used in a shoulder bag and operate in close proximity to the receivers. With some recorders, users have reported diminished radio range and performance when used in a bag . Whether the spurious RF energy comes from the hard drive, or the A-D converters, or some other source, the recent digital recorders seem more prone to excesss RF energy dispersion than the 1/4" machines of years past.
A reader asked how the new Nagra performed with radios and I thought it important to test for this vulnerability.
I reasoned that spurious RF at a level that might interfere with reception ought to be strong enough to show up on the reception meters of the radio receivers. If a receiver were tuned to an open frequency, so that its meter showed no signal, it ought to continue to show no activity when placed atop the recorder. Or, if it did show activity, then there was a potential for radio interference.
With many frequency blocks in common use, this investigation required more equipment than I have in my own kit. I contacted Location Sound Corporation and they offered me the use of their conference room to run the tests and supplied radios from their rental department so we might run a range of tests. They supplied Lectrosonics 411 receivers in Block 19 through Block 28. I am indebted to Dave Panfili, Steve Joachim, and Robert Noone for their cooperation in this venture.
Each radio was tuned to a clear frequency with the Nagra powered off. The receiver was placed on a mouse pad on top of the Nagra so there would not be any chassis contact between the two. Then the Nagra was powered up and any change in the meter observed and photographed. I also put the machine into "Record" to check if that made any difference, With the Nagra, there was no distinction; any interference was observable when the machine was switched on.
|
|
A cluster of radios assembled for the test |
The arrow points to a weak signal on the meter of the receiver
|
|
|
|
|
|
|
|
|
|
|
|
|
The Venue meter, shown here for illustration, is essentially identical to the meter on the Lectro 411 receivers used in the test. Here the meter shows very slight RF activity. The minimal signal that can be displayed is 1uV*; here we see about 2~3 uV.
(1uV = 1 microvolt or 1 millionth of a volt)
|
RF activity seen here is much stronger, perhaps 75 ~ 80 uV. This would be an adequate signal from a transmitter for good audio. If a spurious signal, this would be sufficiently strong to significantly reduce the range of a transmitter. |
This is a full-level RF signal. It's what you want to see from your transmitter. (And, conversely, what you don't want to see BEFORE you switch on the transmitter.) These examples provide context for the readings observed in the pictures, below. |
The front panel of a Lectro 411 receiver with no signal present. |
A typical weak signal observed when the Nagra was switched on. This example on Block 19. |
A detail shot of a typical weak signal observed when the Nagra was switched on. This example on Block 22. |
Blocks 20 & 19, mouse over to show RF activity |
Blocks 21 & 22, mouse over to show RF activity |
Blocks 23 & 24, mouse over to show RF activity |
Blocks 25 & 26, mouse over to show RF activity |
Blocks 27 & 28, mouse over to show RF activity |
There is a consistent pattern in these photos (although it may be a bit diffucult to observe in these small images).
Turning on the Nagra produced measurable but relatively low level RF interference throughout the range of frequencies tested. In each case the interfering signal was quite weak, typically well below 10uV. With this low level of interference, I would expect radio sets to perform normally at distances usually encountered in hand-held recording. However, the interference might be sufficient to slightly reduce range so that a radio set that would otherwise perform to a distance of 250 feet might begin to take "hits" at, say, 150 feet. |
| |
|
|
| All of our tests so far show the receivers atop the Nagra and located near the front edge of the recorder. It seemed reasonable, particularly with a larger recorder like the Nagra, to test for susceptibility in other locations. Moving the receivers to the rear of the recorder, I found that the strength of the interference dropped considerably. |
Blocks 21 and 22 in back position. Only very slight RF activity shows on the meter. |
Blocks 23 and 24 in back position. Only very slight RF activity shows on the meter. |
For most blocks, the signal strength dropped off measureably when the receiver was moved back. For Blocks 27 and 28 the signal disappeared entirely in the back position.
This is useful information because the lower position probably conforms more closely to actual use. No one is likely to want the radios so far forward that the antennas interfere with access to knobs and switches. Moreover, the pouch in the Nagra case (and most cases likely to be used with the Nagra) naturally carry the radios closer the foot than the front panel. |
Blocks 25 and 26 in back position. Only very slight RF activity shows on the meter. |
Blocks 27 and 28 in back position. The meters measure no RF activity at all in this position. |
| How do other recorders fare under the same scrutiny? A fair question but not necessarily one with conclusive answers. On the Cantar, the receivers are mounted in about the only position where operating controls remain accessible. This places the receivers immediately atop the hard drive where one would expect the receivers might be most vulnerable. There was no observable Rf interference with the machine turned on but, in one test, I did observe some RF with the machine in "Record." (It made no difference in the Nagra tests whether the machine was recording or not; any RF interference was observable when the recorder was turned on.) |
Blocks 19 and 28 on an Aaton Cantar show no RF activity with the machine switched on and continue to show no activity when switched from "On" to "PreRecord." (mouse over to observe) |
This is Block 27 on a Cantar. This time I put the machine into "Record" and observed a slight RF signal - one light - on the Lectro 205 meter. (mouse over to observe) |
| Now we come to the puzzling part of the test. The Sound Devices 744T is exactly the machine that users complained would desensitize their radio receivers when they were used in the same bag together. Yet when I made a test, I found only very minimal levels of interference. The Sound Devices unit is so small that there really cannot be much of an issue of where the receiver is relative to the recorder. |
Sound Devices 744T with Block 27 radio. (mouse over to observe interference level with recorder switched on)
Although no mouse pad is used here, rubber feet affixed to the base of the Lectro 205 provide insulation. |
Given all the complaints about the Sound Devices performance with wireless and the knowledge that the company had put forth great effort to ensure that the 788 model would play nice with radios, I thought I shoud repeat the test. I powered both the receiver and the 744 from the cart's PSC Power Station to avoid any contamination from power supplies, too-close wall warts, tangled power cables, etc.
I also tested performance with the recorder in "Record." |
Clear air with the recorder switched off. |
Switching on the recorder produces very minimal RF interference. In this case, the receiver is positioned at the very front of the 744. Others have identified the area around the battery mount, at the rear, as being the location of the heaviest RF dispersion. |
Actually putting the machine into "Record" raises the level of interference noticably. |
| |
|
|
| Additional dispersion of RF energy when the recorder is in “RECORD” would seem to reconcile experiences of users with our initial test showing only minimal interference. In fact, it’s a particular nuisance to have apparent clear air that gets noisy only when actually recording. Still, one wouldn’t expect the receiver to be swamped with RF energy with reception measured at only two lights strong. More extensive testing is beyond the scope of the immediate task of evaluating how the Nagra might fare in a bag situation with radio receivers so this issue must remain, to some extent, a mystery. |
|
After performing all these tests, I learned, from John Owens of Nagra in Switzerland, that the particular Nagra VI used here was a very early example and not part of the regular production run. The front plate of the Nagra is a plastic composite. In the regular production run, that front plate is treated with a metallic coating on the inside. There is some expectation that this metallic coating might reduce the passage of radio signals. So, back to Location Sound we went, Peter Weibel and I, with a production example of the Nagra to repeat the tests. Robert Noone was very gracious and once again assembled examples of receivers in all the radio blocks in common use. Although results with the new machine did not exactly replicate the earlier testing, they were not much different. Where we had previously seen some sign of RF with the recorder turned on, we again saw small but measurable evidence of radio energy. I took no additional pictures because the differences were too small to be illustrated in photos.
From the testing I’ve done, the Nagra does not appear to throw off an alarming amount of RF energy. I would expect that it should be possible to use it in conjunction with radio receivers without suffering any appreciable loss of range or performance. However, there is some measurable activity, as there is with other recorders, so I cannot say with certainty that a user, with a particular radio set in particular circumstances, might not perceive some interference.
Whenever radios are involved, the potential for mischief always exceeds the ability to test. A component that works perfectly today, or under present circumstances, may develop a baffling tendency to make noise tomorrow under slightly different circumstances. No test that we run in our facilities can guarantee that you will experience perfect performance in your work situation. Conversely, you may find that your gear seems to be immune to problems we may identify. If you are seriously considering acquiring the Nagra VI, you should make your own tests, with your own gear, to check performance in circumstances closely matching your expected work conditions. |
John Owens (right) demonstrates the Nagra VI to Giovanni diSimone at the open house exhibit at the Sportsman's Lodge in Studio City in October 2008
 |
| Update Spring 2009 |
|
Since I originally published this test for RF interference I've had some feedback from users.
Larry Fisher, the head of Lectrosonics, pointed out that even just one light or one pixel on a reception display indicates a measureable amount of spurious RF that would have a deleterious effect on range.
Charles Thomaras asserted that the way the test was conducted limited the validity of the results. Rather than doing a sweep test to look at all of the frequencies within a block, we just checked random frequencies. Some of the tests were repeated at alternate settings to gather a wider selection and all of the tests were redone later when we brought in the second Nagra. So, we did check a whole bunch of frequencies. Still, we checked only a fraction all the possibilities. Mr. Thomaras has observed that the Sound Devices 744 recorder emits RF in apparently random spikes so that one frequency might be relatively clean and an adjacent frequency might be severly impacted.
This is a valid point and a possibility we had not really cosidered. In our partial defense, it is very difficult to conduct a test that would examine all the possible points of interference. For one thing, Los Angeles is not a radio friendly area and many frequencies already have interference from radio and TV stations, aero control towers and other sources. Also, without a scanning receiver, one would need to check each and every one of 256 possible frequencies on 10 different blocks to explore all the possibilities. A formidable task.
At this point, one would have to conclude that our test is something less than a definitive examination of the situation.
However, I think the general conclusions are still valid:
1. While the Nagra appears to emit some RF spray, this is relatively weak and not a major source of concern.
2. Susceptibility to the RF spray drops to nothing (or nearly nothing) when the receiver is moved down near the foot of the recorder.
3. Any user expecting to work in a bag with a Nagra and radios should conduct their own test, with their own radios, to confirm that everything will work satisfactorily in expected use circumstances |
|