Newb Asks About Feedback

Forums Forums GLD Forums Archived GLD Discussions Newb Asks About Feedback

Viewing 12 posts - 1 through 12 (of 12 total)
  • Author
    Posts
  • #23686
    Profile photo of MattManMattMan
    Participant

    Hey all –

    We have the GLD80. I love it. I have learned so much and I am nearly giddy with all the power and capability I have now compared to the analog board I moved from (RIP Mackie).

    I cant wait until we get a better manual – but until then I have to ask for some expertise here. It is probably a very basic question. I searched through the existing GLD Forums to see if it had been asked, but I dont see it.

    For handheld mics, I have no problem.

    But for any hands-free mics such as over ear and lapel, I REALLY struggle to get volume. Especially when the crowd is loud, and I want the speaker to talk over them. I simply cant push up the gain any higher without getting feedback. When the crowd is quiet, its no problem. But is there any FX, addition or setting I can work through or research to figure out how to get LOUD volume of spoken word without ever present feedback?

    Thank you!

    Matt

    #33529
    Profile photo of bigbeanbigbean
    Participant

    Discovering new types of feedback is the sound guy’s No 1 job. It takes precedence over all other jobs even turd polishing and creating tripping hazards for performers and audience members.
    But seriously….

    Assuming the mics aren’t in front of the speakers……… (If they are you’re boned- re install the system correctly)

    If you have an RTA use it to help you find the problem frequencies.

    If you don’t try this:

    High pass the mics at 75.

    Start by creating two notches of 9 db. One at 350 w/ a Q of 7, one at 530 same Q.
    Sweep them up and down until you located the problems.

    Also compress them at 3.5 to 1 with 3 db makeup gain. Set the threshold so it knocks the tops off the more energetic speech.

    The surest way to windup with nothing is to wait for everything. On the other hand, the 2nd mouse always gets the cheese.

    #33533
    Profile photo of Chris93Chris93
    Participant

    If a sound goes into the mic and arrives back at the mic louder than it was the first time…and then it does it again…and then again…

    All microphones in the same room as the PA they are routed through are feeding back all the time. The reason it isn’t normally a problem is that the signal gets quieter each time it goes from the mic, through the mixing console, amplifiers, speakers, air in the room… and back into the mic again.

    As you increase the gain* you will reach a point where the signal will go from getting quieter on each cycle to getting louder on each cycle.

    *Gain meaning any gain, not necessarily preamp gain.

    If it isn’t loud enough before this happens you have a problem. [:)]

    If you can get the desired source louder into the mic you can use less preamp gain to achieve the same input level. You can do this by having them produce a louder sound and/or moving the mic closer to whatever makes the sound.

    You’ll also want to reduce the amount of sound from the PA picked up by the mic. Because this will be sent back to the PA… back to the PA…back to the PA…

    Get the mic out of the coverage pattern of the PA.

    Get the PA out of the “coverage” pattern of the mic.*

    *Headset and lapel mics are usually omni-directional because using a directional mic very close to the source would produce a lot of unwanted proximity effect. They also generally have a flatter frequency response.

    Directional (normally cardiod) headset and lapel mics do exist but have much more handling noise. They aren’t normally needed.

    Your handheld mics are very likely cardiod and are probably normally held pointed back towards the person.

    After that’s done narrow down a couple of PEQ bands as narrow as they go, I think it’s 1/9th octave. Turn on the “input full range” option in the pull-up tab. Cut them some amount. 9 db if you want but you might not need that much.

    Have someone wear the mic as normal, stand in the normal area and speak as normal. Push the fader until the mic begins to ring.

    Sweep a band around until the feedback stops. If it’s not loud enough push it until it rings again and cut that frequency too. If it’s the same frequency as before just cut it more.

    Only cut as much as is needed to prevent feedback. Any more will harm sound quality for no real benefit.

    Compression is nice on speaking mics but it won’t help with feedback. Used so that it only goes into gain reduction on louder peaks won’t be a problem (and it’s nice to have when people sneeze wearing a headset mic). [:p]

    If the speech is being compressed at normal speaking volumes you will compensate for this either with the make up gain or with the fader. When the person stops speaking and the compressor releases feedback may start as a result.

    Basically you want your system to behave as:

    SoundSoundSound

    And not:

    SoundSoundSound

    Chris

    #33537
    Profile photo of TJCornishTJCornish
    Participant

    Distilling down some of the good information posted: The loudest source at the mic wins. If the loudest sound at the mic is a copy of itself from the PA system, you will have feedback. You will likely have some ability to affect this with EQ, but the most complete solution is to either decrease the distance from the mic to the source, and/or increase the distance between the mic and the speakers.

    General things:
    – make sure you’re not sending these mics to any monitors.
    – Lapels suck, and are borderline useless for live sound reinforcement. Move to headset mics to get the element closer to the mouth.
    – Train your speaker to project. Many people are afraid of hearing themselves in the system, and as you turn them up, they speak even more softly. Reassure them that they need to speak loudly and clearly, and that you will control the volume.

    #33538
    Profile photo of GSLC-TechGSLC-Tech
    Participant

    All the other posters have the real answers, but sometimes that just can’t be done. We have children’s choirs in front of the mains and are forced to use EQ to notch out the offending frequencies. Like someone else said, use an RTA. If I read the release notes correctly, there is an RTA with the GEQ. During a sound check, SLOWLY bring up the fader (or gain) until you just hear feedback. The RTA should peak and you can take that frequency down a little. After that you might find another. We can usually get much better gain before feedback after dropping a couple of the offending frequencies.

    Rick Kohrs
    Good Shepherd Lutheran Church – West Campus
    Verona, WI

    #33553
    Profile photo of Jono-MorganJono-Morgan
    Participant

    A trick i’ve found with the GLD is go to your parametric, do yourself a favour and buy yourself a samsung s-pen(they work! its amazing) and narrow the middle bands up and drag them up and sweep sideways till you find the offending frequencies. then pull them down until you don’t get feedback. Just remember a comfortable speech volume is about 65-70db. don’t try for 90.

    #33693
    Profile photo of MattManMattMan
    Participant

    So there were a lot of excellent and thoughtful responses. Some of the information was new to me, some was old and some was over my head :). But in the end you all have helped confirm what I thought I knew, pointed me to what I needed to know and gave me some starting points.

    A few details to clear things up:

    1) OK, using lapels just sucks. We can forget them.

    2) The over the ear mic – I had the pastor just keep it closer to his mouth and its been fine.

    3) The last offender is a podium mic by Audio technica (U859QL). This one is really driving me crazy. The podium is underneath the main speakers (hung center above stage), but behind them (only a few feet offset).

    I did find a few frequencies that seem to be repeat offenders and cut them out a little in the overall main GEQ, and then some at the source input PEQ – is there a better place to make that adjustment?

    I was really hoping there is some FX that was a feedback killer magic module… but given that no one said “Man, you havent used the anit-feedback FX yet?!” It seems that might not be the case :)

    BigBean: Your response got a good chuckle out of me. I will probably remember you (digitally) for a long time, thinking about polishing turds. It may not be how you want to be remembered, but it stuck.

    But I didnt quite understand what this meant:

    “Start by creating two notches of 9 db. One at 350 w/ a Q of 7, one at 530 same Q.Sweep them up and down until you located the problems.”

    I cant say I know what “Q” is. I am gathering you suggest a frequency generator and run it (sweep) to find frequencies that cause the mic to have issues.

    Rick Khors – You said something about an “RTA” in the GEQ. I dont see that in the GEQ settings for the main outs… and I only see PEQ on the individual input. Beside that, Im not sure what an RTA is – but I can google it :)

    Cornish – the only outputs for the mics in question are the main house speakers. All monitors are in-ear, so there is no other output source.

    Jono – Ill google the S-pen. Thanks!

    Chris – yeah, like you suggested I’ve had limited success using compression to control feedback. It seems to be the wrong tool for the problem generally speaking.

    BigBean, Rick, Chris, Cornish, Jono – thank you all for the help.

    #33697
    Profile photo of Chris93Chris93
    Participant

    If you alter the main GEQ you will affect all sound passing through your PA, rather than just the mic that’s causing the problems. Although it may be the case that it’s feeding back there because the PA is producing too much of that frequency.

    Input EQ would generally be better, assuming the PA sounds correct.

    In the GEQ window you will see red bars lighting up to indicate which frequency is feeding back, though you may have to let it squeal a bit to see this clearly.

    “Q” is filter quality. It is basically referring to the bandwidth of the filter (the EQ band). Higher Q values will give a narrower band.

    Chris

    #33703
    Profile photo of GSLC-TechGSLC-Tech
    Participant

    You might be able to knock out the frequency using one of the 4 channel eq’s. As most have said, use a narrow Q, bring up the volume for frequency level and then slide the frequency up and down until you find the offending frequency. The touch screen view of the EQ will be more helpful as you can see the Q value better.

    This might be a waste of a group, but you could assign the pulpit to a group. Doing this, you have a 1/3 octave GEQ for a single channel. We do this for our 3 choir mics.

    Don’t do this with the LR GEQ, you can loose important room frequencies.

    Rick Kohrs
    Good Shepherd Lutheran Church – West Campus
    Verona, WI

    #33704
    Profile photo of GSLC-TechGSLC-Tech
    Participant

    Forgot to add. Chris explained how to use the GEQ for finding the bad frequencies. An RTA is a Real-Time Analyzer. There is an RTA available for an ipod touch or ipad. Do a search for AudioTools. Studio Six is the creator. It does cost $20. Other toys are also included (SPL meter). The mics on ipads and ipods are quite good and I know of a few very reputable sound engineers that swear by the package.

    Rick Kohrs
    Good Shepherd Lutheran Church – West Campus
    Verona, WI

    #33705
    Profile photo of BobWitteBobWitte
    Participant

    Podium mics under speakers – very tricky. The podium itself is a fantastic audio reflector that can focus energy onto the mic itself (even it is not directly in the path of the speaker, the lower frequencies – even mids, still hit pretty hard). One thing that has helped is some type of “cushy” or heavy cloth covering on the podium to help absorb some of the energy. Change the angle if possible too. Using a group as discussed earlier and apply GEQ (try the different types) along with an RTA (AudioTools is great) to assist in identify the frequencies is also good. The big problem will remain though in not removing too much “content” so that the speakers voice still sounds close to the actual speakers voice.

    Bob

    GLD80, AR2412, 2xAR84, Dante

    #33706
    Profile photo of MattManMattMan
    Participant

    Excellent. This is all very helpful information.
    Thank you.

Viewing 12 posts - 1 through 12 (of 12 total)
  • The forum ‘Archived GLD Discussions’ is closed to new topics and replies.