Sennheiser g3 transmitter display reads higher than receiver display at DVinfo.net

Tom Morrow · September 20th, 2012, 01:10 PM

Has anyone else noticed that the AF (signal strength) display on the g3 transmitter read significantly higher than the AF display on the receiver?

I just did some tests to quantify how much... feeding a 1k mic level test tone into the SK100 transmitter from a CT100 through a CL2 cable, I had to set the transmitter to Sensitivity -33db to get the receiver graph to just touch the bar at the top. But then the display on the transmitter is way over the top.

To get the transmitter display to just barely hit the bar at the top, I have to dial it down from -33 to -45db sensitivity, which is 12db difference.

I thought this might be related to the AF setting on the receiver, but adjusting that makes no difference in the display.

So can anyone think of a good reason why these two displays are different? I suppose the best thing is to split the difference and allow the transmitter display to peak a bit pas the mark, but it's discouraging to then see the receiver display only in the middle.

Jay Massengill · September 20th, 2012, 02:10 PM

I have a CT100 and I don't think the tone sounds "normal". Plus there is a note in the manual that the battery strength in the CT100 will affect the tone output level.
I would use something better for tone, like a used Shure field mixer (M267 or FP42) if you can get your hands on one.

Steve House · September 20th, 2012, 03:04 PM

Adjust the transmitter to proper modulation level on the transmitter meter using its sensitivity adjustment and then leave it alone. You don't need or necessarily want full deflection on the receiver AF meter and adjusting the transmitter sensitivity until you get it is liable to result in overloading the transmitter's audio circuits.

Tom Morrow · September 20th, 2012, 04:39 PM

I wouldn't have thought the particular tone device would make a difference, but I just tried using my mixpre rather than CT100 for tone and the displays were closer. The NTI MR2 tone also had closer values.

The table below shows the transmitter settings that were the lowest transmit setting where the display would touch the bar near the top of the display on the Transmitter or Receiver

Device Transmitter Receiver Difference
CT100 -45 -36 9db
MixPre -33 -30 3db
NTIMR2 -33 -30 3db

The CT100 tone is certainly less like a real sine wave than the MixPre and MR2 tone generators. The chart above shows that the CT100 produces a 9db difference between the two displays, while the more "pure" sine waves only have a 3db difference between transmitter and receiver display.

At any rate, we can debate about how much of a difference there is but there's clearly a difference whether it's 3db or 9db or more. I'm thinking that a good rule of thumb might be to think of the horizontal hash mark at the top of the transmitter display as being equivalent to the very top of the the receiver display, as there seems to be about 4-5db between that top hash and the actual top of the display.

John Willett · September 21st, 2012, 03:26 AM

Quote:

Originally Posted by Tom Morrow

Has anyone else noticed that the AF (signal strength) display on the g3 transmitter read significantly higher than the AF display on the receiver?

I just did some tests to quantify how much... feeding a 1k mic level test tone into the SK100 transmitter from a CT100 through a CL2 cable, I had to set the transmitter to Sensitivity -33db to get the receiver graph to just touch the bar at the top. But then the display on the transmitter is way over the top.

To get the transmitter display to just barely hit the bar at the top, I have to dial it down from -33 to -45db sensitivity, which is 12db difference.

I thought this might be related to the AF setting on the receiver, but adjusting that makes no difference in the display.

So can anyone think of a good reason why these two displays are different? I suppose the best thing is to split the difference and allow the transmitter display to peak a bit pas the mark, but it's discouraging to then see the receiver display only in the middle.

The two meters are reading different things.

The transmitter is showing you the AF level before transmission so you can set the sensitivity properly.

The receiver is reading the receive level - they won't be identical.

Tom Morrow · September 23rd, 2012, 07:58 AM

A perfect wireless system would receive the exact same signal that it sends, so what's received should be the same as what's sent (or at least the same level). But I suppose the g3 system isn't perfect in that sense; at a minimum there is companding going on, and it seems like some frequency response shaping too; when I swept the test tone through different frequencies the difference between the receive and transmit AF displays was significantly greater for some frequencies than others. I wonder if the harmonics in the non-perfect CT100 1k sine wave caused the frequency response anamolies to show up in the difference.

The companding circuitry also probably involves compression or limiting; perhaps a good way to think about it is that the transmitter display shows the signal before that compression, and the receiver is lower because of the compression.

Colin McDonald · September 23rd, 2012, 08:39 AM

You do know what "companding" means I hope?

Comp(ressing-Exp)anding

John gave you a definitive explanation for what you are seeing.

Steve House · September 23rd, 2012, 08:54 AM

Quote:

Originally Posted by Tom Morrow

A perfect wireless system would receive the exact same signal that it sends, so what's received should be the same as what's sent (or at least the same level). But I suppose the g3 system isn't perfect in that sense; at a minimum there is companding going on, and it seems like some frequency response shaping too; when I swept the test tone through different frequencies the difference between the receive and transmit AF displays was significantly greater for some frequencies than others. I wonder if the harmonics in the non-perfect CT100 1k sine wave caused the frequency response anamolies to show up in the difference.

The companding circuitry also probably involves compression or limiting; perhaps a good way to think about it is that the transmitter display shows the signal before that compression, and the receiver is lower because of the compression.

Your premise that they should be the same in a perfect system is simply not true - there's no physical reason the two meters should read the same values at all since they are actually measuring totally different things. The meter on the transmitter is measuring the strength of the audio signal passing through the transmitter's audio circuits while the meter on the receiver is measuring the audio in its own completely independent circuits. Harmonics in the signal and companding are red herrings, not relevant to the issue at all. The audio isn't flowing from one location to another.

Greg Miller · September 23rd, 2012, 01:22 PM

I can relate this to the commercial broadcast environment.

There are a number of "ifs" involved.

If the transmitter frequency response is flat, and
If the receiver frequency response is flat, and
If all compression and peak limiting are turned off, and
If all levels and modulation are calibrated correctly,
then within the specified frequency passband of the system, the two meters should read the same thing... at least at 1 kHz.

In other words, if you put in a 1kHz tone and set the transmitter to 50% modulation, then if your receiver is a calibrated modulation monitor, it should read 50%. If you raise the level of the tone by +3dB, or drop it by -3dB, the modulation monitor should show an exact +3dB or -3dB change.

And if, with this calibrated system, you repeat the test with other frequencies (so long as they are within the rated passband of the system), both meters should show the same thing. But... commercial FM uses HF pre-emphasis in the transmitter, so a 0VU input signal at, say, 10kHz will badly overmodulate the transmitter! So high frequency tests are very tricky.

This would be the same situation with an analog device such as an analog tape machine. The levels you see when playing back a given tape should be exactly the same as the levels you saw when recording... If all those above conditions are met. And analog tape, too, uses high frequency pre/de-emphasis. (Besides which, the magnetic media saturates easily.) So again, high frequency tests are very tricky.

Having said all that, at first glance one might expect the same situation with a wireless mic system... and you might encounter similar problems with HF test signals.

One important caveat has to do with frequency response. FM broadcast systems use high frequency pre-emphasis and de-emphasis to improve the S/N ratio. Some wireless mics might do the same thing (but offhand I don't know which ones, or what amount of pre/de-emphasis they use). Certainly the pre-emphasis at the transmitter should exactly match the de-emphasis for a given receiver.

One result of HF pre-emphasis is that the system cannot handle the same input level at high frequencies that it can at low frequencies.

So when you're testing any such system, you need to be sure that you're using pure sine wave signal that doesn't contain any HF harmonics. Harmonics that get up into the pre-emphasis range will certainly skew the test results. You can't use a "test tone" from a mixer... you need a clean sinewave oscillator designed for such tests.

And depending on the amount of pre/de-emphasis, you may need to conduct freq. response tests at a level well below 0 VU. That depends on the amount of pre/de-emphasis used in a given system.

To summarize, it's not as simple as it seems. You need to know a lot of details about your specific system, to know what the results should be. (Unfortunately, I can't give you the needed information (pre/de-emphasis, limiting, etc., about the G3 system.) But there are a lot of reasons that the meters might not match: some might be expected, and some might indicate a problem. Only someone with all the specs and proper test equipment can tell for sure.

Tom Morrow · September 24th, 2012, 02:33 AM

Lots of good info here. Thanks Greg Miller for taking the time to type all that up; very clear now.

The takeaway is not to worry about it; it's more of a measurement anamoly than a problem. The better you measure it, the closer they are.

I think I'll just continue my practice of using the transmitter display for setting the AF out on the transmitter to match the speaker, letting the meter touch the top a bit more than I would on a digital system. The receiver display isn't really used much since the receiver is typically plugged into a recorder or mixer that have a more readable display.

September 20th, 2012, 01:10 PM	#1
Tom Morrow Major Player Join Date: Feb 2011 Location: San Francisco, CA Posts: 416	Sennheiser g3 transmitter display reads higher than receiver display Has anyone else noticed that the AF (signal strength) display on the g3 transmitter read significantly higher than the AF display on the receiver? I just did some tests to quantify how much... feeding a 1k mic level test tone into the SK100 transmitter from a CT100 through a CL2 cable, I had to set the transmitter to Sensitivity -33db to get the receiver graph to just touch the bar at the top. But then the display on the transmitter is way over the top. To get the transmitter display to just barely hit the bar at the top, I have to dial it down from -33 to -45db sensitivity, which is 12db difference. I thought this might be related to the AF setting on the receiver, but adjusting that makes no difference in the display. So can anyone think of a good reason why these two displays are different? I suppose the best thing is to split the difference and allow the transmitter display to peak a bit pas the mark, but it's discouraging to then see the receiver display only in the middle.

September 20th, 2012, 02:10 PM	#2
Jay Massengill Trustee Join Date: Jul 2003 Location: Burlington Posts: 1,976	Re: Sennheiser g3 transmitter display reads higher than receiver display I have a CT100 and I don't think the tone sounds "normal". Plus there is a note in the manual that the battery strength in the CT100 will affect the tone output level. I would use something better for tone, like a used Shure field mixer (M267 or FP42) if you can get your hands on one.

September 20th, 2012, 03:04 PM	#3
Steve House Inner Circle Join Date: Mar 2005 Location: Hamilton, Ontario, Canada Posts: 5,742	Re: Sennheiser g3 transmitter display reads higher than receiver display Adjust the transmitter to proper modulation level on the transmitter meter using its sensitivity adjustment and then leave it alone. You don't need or necessarily want full deflection on the receiver AF meter and adjusting the transmitter sensitivity until you get it is liable to result in overloading the transmitter's audio circuits. __________________ Good news, Cousins! This week's chocolate ration is 15 grams!

September 20th, 2012, 04:39 PM	#4
Tom Morrow Major Player Join Date: Feb 2011 Location: San Francisco, CA Posts: 416	Re: Sennheiser g3 transmitter display reads higher than receiver display I wouldn't have thought the particular tone device would make a difference, but I just tried using my mixpre rather than CT100 for tone and the displays were closer. The NTI MR2 tone also had closer values. The table below shows the transmitter settings that were the lowest transmit setting where the display would touch the bar near the top of the display on the Transmitter or Receiver Device Transmitter Receiver Difference CT100 -45 -36 9db MixPre -33 -30 3db NTIMR2 -33 -30 3db The CT100 tone is certainly less like a real sine wave than the MixPre and MR2 tone generators. The chart above shows that the CT100 produces a 9db difference between the two displays, while the more "pure" sine waves only have a 3db difference between transmitter and receiver display. At any rate, we can debate about how much of a difference there is but there's clearly a difference whether it's 3db or 9db or more. I'm thinking that a good rule of thumb might be to think of the horizontal hash mark at the top of the transmitter display as being equivalent to the very top of the the receiver display, as there seems to be about 4-5db between that top hash and the actual top of the display.

September 23rd, 2012, 07:58 AM	#6
Tom Morrow Major Player Join Date: Feb 2011 Location: San Francisco, CA Posts: 416	Re: Sennheiser g3 transmitter display reads higher than receiver display A perfect wireless system would receive the exact same signal that it sends, so what's received should be the same as what's sent (or at least the same level). But I suppose the g3 system isn't perfect in that sense; at a minimum there is companding going on, and it seems like some frequency response shaping too; when I swept the test tone through different frequencies the difference between the receive and transmit AF displays was significantly greater for some frequencies than others. I wonder if the harmonics in the non-perfect CT100 1k sine wave caused the frequency response anamolies to show up in the difference. The companding circuitry also probably involves compression or limiting; perhaps a good way to think about it is that the transmitter display shows the signal before that compression, and the receiver is lower because of the compression.

September 23rd, 2012, 08:39 AM	#7
Colin McDonald Trustee Join Date: Dec 2007 Location: Glasgow, Scotland Posts: 1,546	Re: Sennheiser g3 transmitter display reads higher than receiver display You do know what "companding" means I hope? Comp(ressing-Exp)anding John gave you a definitive explanation for what you are seeing.

September 23rd, 2012, 01:22 PM	#9
Greg Miller Trustee Join Date: Jan 2011 Location: Pennsylvania Posts: 1,791	Re: Sennheiser g3 transmitter display reads higher than receiver display I can relate this to the commercial broadcast environment. There are a number of "ifs" involved. If the transmitter frequency response is flat, and If the receiver frequency response is flat, and If all compression and peak limiting are turned off, and If all levels and modulation are calibrated correctly, then within the specified frequency passband of the system, the two meters should read the same thing... at least at 1 kHz. In other words, if you put in a 1kHz tone and set the transmitter to 50% modulation, then if your receiver is a calibrated modulation monitor, it should read 50%. If you raise the level of the tone by +3dB, or drop it by -3dB, the modulation monitor should show an exact +3dB or -3dB change. And if, with this calibrated system, you repeat the test with other frequencies (so long as they are within the rated passband of the system), both meters should show the same thing. But... commercial FM uses HF pre-emphasis in the transmitter, so a 0VU input signal at, say, 10kHz will badly overmodulate the transmitter! So high frequency tests are very tricky. This would be the same situation with an analog device such as an analog tape machine. The levels you see when playing back a given tape should be exactly the same as the levels you saw when recording... If all those above conditions are met. And analog tape, too, uses high frequency pre/de-emphasis. (Besides which, the magnetic media saturates easily.) So again, high frequency tests are very tricky. Having said all that, at first glance one might expect the same situation with a wireless mic system... and you might encounter similar problems with HF test signals. One important caveat has to do with frequency response. FM broadcast systems use high frequency pre-emphasis and de-emphasis to improve the S/N ratio. Some wireless mics might do the same thing (but offhand I don't know which ones, or what amount of pre/de-emphasis they use). Certainly the pre-emphasis at the transmitter should exactly match the de-emphasis for a given receiver. One result of HF pre-emphasis is that the system cannot handle the same input level at high frequencies that it can at low frequencies. So when you're testing any such system, you need to be sure that you're using pure sine wave signal that doesn't contain any HF harmonics. Harmonics that get up into the pre-emphasis range will certainly skew the test results. You can't use a "test tone" from a mixer... you need a clean sinewave oscillator designed for such tests. And depending on the amount of pre/de-emphasis, you may need to conduct freq. response tests at a level well below 0 VU. That depends on the amount of pre/de-emphasis used in a given system. To summarize, it's not as simple as it seems. You need to know a lot of details about your specific system, to know what the results should be. (Unfortunately, I can't give you the needed information (pre/de-emphasis, limiting, etc., about the G3 system.) But there are a lot of reasons that the meters might not match: some might be expected, and some might indicate a problem. Only someone with all the specs and proper test equipment can tell for sure.

September 24th, 2012, 02:33 AM	#10
Tom Morrow Major Player Join Date: Feb 2011 Location: San Francisco, CA Posts: 416	Re: Sennheiser g3 transmitter display reads higher than receiver display Lots of good info here. Thanks Greg Miller for taking the time to type all that up; very clear now. The takeaway is not to worry about it; it's more of a measurement anamoly than a problem. The better you measure it, the closer they are. I think I'll just continue my practice of using the transmitter display for setting the AF out on the transmitter to match the speaker, letting the meter touch the top a bit more than I would on a digital system. The receiver display isn't really used much since the receiver is typically plugged into a recorder or mixer that have a more readable display.