Frequently Asked Questions - Search Results
This is a long posting made to the RAMPS news group. There is a clearer explanation here: see FAQ 105
Some weeks ago we made measurements of a simulated bag system to see what having a 100 mW transmitter 12 inches (30 cm) away from receivers would do to the receivers' sensitivity. If you read the previous post, you will remember I was somewhat surprised at the performance of the UCR210 receiver in this test. I thought the older, helical resonator front end, single frequency UCR195 would carry the day. In fact, the UCR210 performed as well and in some areas was a bit better. Some users on the group were wondering how the UCR201 would perform in a bag even though this was never our intended use for the 201. This time the numbers are more in line with what I would guess, since the 201 is definitely weaker in this test. The first set of numbers are for the UCR210 and the second set are for the UCR201. I'm not going to try to put them on the same line since I think various news readers will mangle the formatting. To see how the test was run, I included all the text of the previous test below the new numbers. Again, I'll put frequency separation in MHz, then dB of desensing, then resulting percentage of range with a percent and then a comma as a line separator.
Again, the explanation of the test setup and procedure are in the text after this posting.
(UCR210 from the previous test)
0.5M 20dB 10%,
1.0M 14dB 20%,
1.5M 12dB 25%,
3.0M 10dB 32%,
4.0M 9.8dB 33%
6.0M 3.6dB 66%,
10.M 2.3dB 77%,
20.M 1.8dB 81%
(UCR201 from a new test)
0.5M 89dB 0%,
1.0M 77dB 0%,
1.5M 18dB 12%,
3.0M 23dB 7%,
4.0M 17dB 14%,
6.0M 11dB 28%,
10.M 6dB 50%,
20.M 3dB 71%,
30.M 2dB 80%
As can be seen from comparing the numbers, the UCR201 needs twice the frequency separation before the ranges are comparable. The 3 MHz number looks funny but that's what we measured. The 50% of normal range is reached by the UCR210 at 5 MHz of frequency difference while the UCR201 needs 10 MHz of separation. I would recommend separation of at least one of our blocks (25 MHz) between the 201 receivers and transmitters in the bag. On the other hand, the UCR210 can operate inside the same block with a little care. The difference is due primarily to the tracking front end in the 210 and secondarily due to the higher power level of the first RF transistor in the front end of the 210. Once the signal is past the front end, both receivers are essentially the same.
We ran the same tests with reduced transmitter power (30 mW equivalent) and found a one to one reduction in desensing. The 5 dB reduction in power (100 mW to 30 mW) reduced the desensing by 5 dB. So you could take the above numbers and reduce the desensing by 5 dB which would be equivalent to a 30 mW system at 12 inches or a 100 mW system at 18 inches (45 cm). Note that the levels drop faster than the square of the distance at very close distances. A 10 mW transmitter would improve the desensing numbers by 10 dB. The reason I mention 10 and 30 mW is those are the power levels of some other popular transmitters on the market. I don't think the matching receivers will be quite as tough as the UCR201 but with the lower transmitter power, the upshot is the 201 numbers probably will describe what other systems would do. The one exception, to the best of my knowledge, is the small Sennheiser receiver that does not have an RF amplifier but takes the antenna input directly to the mixer. It has a very good third order intercept but does not have the sensitivity of other Sennheiser receivers. It would be interesting to measure it and see if the resistance to desensing would counter balance the lack of an RF stage. I'm sure it would. A good universal rule of thumb is that 10 MHz of separation with reasonable quality systems will give you 50% of your range. You will need at least 5 MHz even with the UCR210.
[[[ Below is the text of the original measurements and a complete description of the setup and parameters. The new measurements above were made in the same manner. We tried to make sure we were comparing apples to apples.]]]
We did some interesting RF measurements on a simulated two way bag system to see how much the bag transmitters would affect the bag receivers' sensitivity. A two way bag system will at least consist of multiple receivers to receive audio signals from the talent, a portable mixer to mix the audio and one or more transmitters to retransmit mixed audio to the video cameras. The immediate question is "If the receivers and transmitters are on different frequencies why should the transmitter reduce the sensitivity of the receiver?" One obvious answer is that the RF front end of the receiver is not a perfect filter and can let strong, nearby frequencies pass through and overload the first amplifier. In addition, transmitters do not produce a single sharp frequency but have some noise 5 Mhz or more from the carrier. The levels are very low but bag systems have antennas that are very close together. In the same way, the local oscillator in the receiver produces some noise many MHz away from the desired frequency and acts the same as having noise in the transmitter. Instead of trying to calculate all this stuff it is simpler to just measure a simulated system.
We put a transmitter 12" (30cm) away from an antenna mounted on a power meter and measured an average signal of -5dBm (.5mW) from a transmitter with 20 dBm output (100 mW). This is a very strong signal to bleed into a receiver but will be very typical of a bag system with 12" of separation. We used this level for the interfering transmitter for all the sensitivity tests. We then checked the receiver sensitivity with the transmitter off and then on and measured the reduction in receiver sensitivity for different frequency offsets between the transmitter and receiver. So to simulate a bag system where the talent's transmitter is on 540 MHz and the bag is re-transmitting mixed audio to the camera on 550 MHz, we would inject a 550 MHz signal at -5dBm into a UCR210 receiver set at 540 MHz and see how much that affected the receiver's ability to pick up the desired 540 MHz signal. We attenuated a block 21 UM200C transmitter set at 550 MHz down to -5 dBm and combined it with a weak 540 MHz signal from a signal generator, set the receiver to 540 MHz and checked the sensitivity with the transmitter off and then on. With the transmitter off, the receiver had a normal sensitivity of -107 dBm for 30 dB SINAD.(Same as "signal to noise ratio" at these values) With the transmitter on, the sensitivity fell to -104.7 dBm for a decrease in sensitivity of 2.3 dB. Or the receiver was desensed by 2.3 dB. This means that in the real world, a 10 MHz offset in the two systems' frequencies with the antennas 12" apart, the usable range from the talent to the bag would have been reduced by 23%. This is pretty small and surprised me. I thought it would be much worse. (There is no reduction in the distance from the bag to the camera since the receiver at the camera is not near a transmitter.) At 0.5 MHz separation with the talent transmitter and bag receiver still at 540 MHz and the bag transmitter at 540.500 MHz, the desensing was much worse at 20 dB. This would reduce the talent to bag range to 10% of normal and is a good reason to never operate with only 0.5 MHz frequency separation. Here's some more values for a UM200 UCR210 system and I hope the news readers don't totally mangle the formatting. I'll put frequency, then dB of desensing, then resulting range with a percent and then a comma as a line separator. Something should get through.
0.5M 20dB 10%,
1.0M 14dB 20%,
1.5M 12dB 25%,
3.0M 10dB 32%,
4.0M 9.8dB 33%
6.0M 3.6dB 66%,
10.M 2.3dB 77%,
20.M 1.8dB 81%
The above numbers are for a system with a tracking front end with relatively high power RF amplifiers both of which help. The relatively high power transmitter of course hurts. If you didn't need much range, bag to camera, you could attenuate the transmitter. The better solution would be to move the transmitter away from the receiver. On the other hand, 5 Mhz of frequency separation would leave you with 50% of your range and that would still be a gracious plenty in 99% of the cases. This is all a good reason to have the transmitters and receivers in a bag system separated by a frequency block (25 MHz). But if you have to operate in one TV channel, 6 MHz, off set the frequencies as much as reasonable.
I had postulated before we tried this experiment that the helical resonators in the fixed frequency UCR195 receiver might work better in a 2 way bag system than the higher power RF amps and wider tracking filters in the UCR210. Here's the numbers for a UCR195D at 536.250 MHz and a block 21 UM200C. The measurements were made the same as above,frequency, then dB of desensing, then resulting range and a comma.
0.5M 30dB 3%,
1.0M 28dB 4%,
1.5M 26dB 5%,
3.0M 25dB 8%,
4.0M 7.8dB 41%
6.0M 4.4dB 60%,
10.M 3.1dB 70%,
20.M 2.8dB 72%
The numbers are much worse for separations less than 4 MHz. Basically the dual gate GasFet is wiped out in the UCR195D until the helical resonators can start filtering at 4 MHz off frequency. Note the rapid change in the numbers between 3 and 4 MHz. This is where the helicals really begin to filter. Overall, the higher power RF amps in the UCR210 carry the day, though the helicals are roughly equivalent at 4 MHz and higher. If I can say something in self defense here, that is why the newer Lectros are rightly accused of eating batteries; they burn it in the front ends.
Having been major wrong about the UCR195 and the helicals, I thought we'd test out my hypothesis that the "rock" transmitters (crystal controlled UM195) would have less noise off frequency than the synthesized transmitters such as the UM200. Here we used a UM195 on 536.250 and varied the frequency of a block 21 UCR210 to check sensitivity versus separation. Note the frequency offsets are a little different.
1.5M 28dB 4%,
2.0M 25dB 8%,
4.0M 17dB 14%,
6.0M 7.0dB 45%,
10.M 4.2dB 62%
Notice that the range is much more affected by the UM195 transmitter. The receiver is the same UCR210 used in the first measurements yet the receiver is now much more affected. The only explanation is that the UM195 is noisier off frequency than the UM200C which is not what I would have expected. By the way, this "noise" would only show up under these bag conditions. Twelve inch separations from transmitters to receivers are not usual. We will measure some different UM195's but at the moment the synthesized radios seem to perform easily as well as the fixed frequency units in bag system conditions. Now where's the crow seasoning.
The VRS is the (S)tandard receiver module for the Venue receiver system. The VRT is the more complex (T)racking front end receiver module for the Venue system. There are no other differences between the modules in either the RF sections or audio sections. If you are familiar with our other receivers, the VRS front end is similar to the UCR201 and the VRT front end is similar to the UCR411. For an explanation of what a tracking front end is see the following FAQ.
What is a tracking front end? See FAQ#065-WIRELESS
This is cribbed from the UDR200C manual and it was written years ago but is still very true:
"A number of years ago, the problem posed to the design staff was to retain the RF reliability of the Lectrosonics’ fixed frequency designs but add the frequency flexibility of a frequency agile design. The universal (but not best) way to build frequency agile systems is to design a wide open front end that will pass any frequency within the tuning range of the system. This leads to compromised RF performance in the front end with the possibility of interference, forcing the user to switch frequencies in an attempt to sidestep the interference. This makes frequency agile receivers a self fulfilling system; you have to use the frequency agility to get away from the problems caused by the frequency agile design compromises. The problem of frequency agility is further compounded when you realize that frequency changes “on the fly” cannot be made on any type of wireless system. For example, if there is suddenly an interference problem with a system in use, on stage for instance, a frequency change cannot be made without interrupting the program. Basically, the show must go on. In multichannel applications, changing the frequency of one system will usually produce all kinds of new intermodulation problems with the other systems operating in the same location. Frequency agility is not the universal panacea for interference problems. It is only another tool and a limited tool at that. The first line of defense must be the system’s basic immunity to interference. That required a new look at frequency agile receiver design.
"FREQUENCY TRACKING FRONT-END
Our solution to the wide open front end problem was to design a selective front end that can be tuned to the frequency in use. Since we wanted this front end to be equivalent to our fixed frequency front ends, this was a daunting task. Lectrosonics has always used front ends with more sections and much more selectivity than any other wireless manufacturer. The final design consisted of a total of 12 transmission line resonators with variable capacitance applied to each resonator by a microprocessor. This allows each resonator to be individually tuned by the microprocessor for any user selected frequency in a 25 MHz band. This sophistication produced a front end that was as selective as fixed frequency designs, yet could cover the entire 25 MHz range.
"HIGH CURRENT LOW NOISE AMPLIFIERS
The gain stages in the front end use some rather special transistors in a feedback regulated high current circuit that combine three parameters that are generally at odds with one another. These are: low noise, low gain and relatively high power. It is easy to understand the advantages of low noise and high power capability but why is low gain desirable? The answer is that in a receiver, low gain allows the front end to handle stronger RF signals without output overload, which is “increased headroom,” so to speak. The result of a design that takes all three of these parameters into consideration at once, is a low noise RF amplifier with a sensitivity rating equal or better than the best conventional design with a hundred times less susceptibility to intermodulation interference. Combining the high power gain stages with the tracking front end produces a receiver that is unusually immune to single and multiple interfering signals close to the operating frequency and in addition strongly rejects signals that are much farther away."
To see what kind of interfering levels would exist in a bag, we put a transmitter 12" (30cm) away from an antenna mounted on a power meter and measured an average signal of -5dBm (.5mW) from a transmitter with 20 dBm output (100 mW). This is a very strong signal to bleed into a receiver but will be very typical of a bag system with 12" of antenna separation. We used this level for the interfering transmitter for all the sensitivity tests. We then checked the receiver sensitivity with the transmitter off and then on and measured the reduction in receiver sensitivity. We then repeated the measurements for different frequency offsets between the transmitter and receiver. To simulate a bag system where the talent's transmitter is on 540 MHz and the bag is re-transmitting mixed audio to the camera on 550 MHz, we would inject a 550 MHz signal at -5dBm into a UCR210 receiver set at 540 MHz and see how much that affected the receiver's ability to pick up the desired 540 MHz signal. We attenuated a block 21 UM200C transmitter set at 550 MHz down to -5 dBm and combined it with a weak 540 MHz signal from a signal generator, set the receiver to 540 MHz and checked the sensitivity with the transmitter off and then on. With the transmitter off, the receiver had a normal sensitivity of -107 dBm for 30 dB SINAD. (Same as "signal to noise ratio" at these values) With the transmitter on, the sensitivity fell to -104.7 dBm for a decrease in sensitivity of 2.3 dB. The receiver was desensed by 2.3 dB. This means that with a real bag system having a 10 MHz offset in the two systems' frequencies and with the antennas 12" apart, the usable range from the talent to the bag would have been reduced to 77% of normal range. This is a pretty small reduction and surprised me. I thought it would be much worse. (There is no reduction in the distance from the bag to the camera since the receiver at the camera is not near a transmitter.) To simulate a worst case situation, we reduced the frequency separation to only 0.5 MHz with the talent transmitter and bag receiver still at 540 MHz and the bag transmitter now at 540.500 MHz. The desensing was now much worse at 20 dB. This would reduce the talent to bag range to 10% of normal and is a good reason to never operate with only 0.5 MHz frequency separation. Here's some more measured values for a UM200 UCR210 system. I'll put frequency and then resulting range as a percent and also in actual feet, assuming 300 feet for a normal system.
Here are the results of UM200 and UCR210 at 12 inches apart:
Here are the results of UM200 and UCR201 at 12 inches apart:
A quick measurement with the antennas between the bag transmitter and receiver at 18 inches instead of 12 inches as above, showed a reduction in interference power of 5 dB. This is a huge change, is faster than the usual square of the distance rule and would allow you to more than double the range for some smaller frequency separations.
The results for all of this are: