Enjoy the benefits of a premium membership!
Upgrade Account
-
TalkBass has been independent since 1998. Add your voice.
Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
Join free Log in
Want zero display ads or expanded classifieds tools? Compare plans.
Overall System gain, gain staging, and positioning of speakers as best you can given what bar/club/wedding type venues allow are your first steps. If you just turn on feedback suppression you'll never actually learn what's causing your issue.
I definitely recommend checking out Nathan lively's sound design live podcast and his info on his website as a start.
And turn those channel gains down if you are constantly seeing the clip light. I don't know that mixer, but 0 may not be "analog 0," (0dbv), it may be digital 0 (dbfs), in which case youre running at something like +18 and then choking that off later in your system (that's what gain staging refers to- setting each step at the optimal level to get to the next step).
It will be the best money you ever spent on your gigging to hire someone for even one night to show you on your system, and if that's not available in your area, then look for a good online class or good books or good youtube videos (but beware, there's a lot of incorrect info out there). There are different opinions and different approaches, but physics is physics so anything you do to fight physics has limits.
Oh, one thing - again, I don't know your mixer (meaning I have never used it, i know what you have from your post), but on my x32 the output delays are on the output routing page, on yamaha mixers it's on the output setup page, and on the soundcraft impression I think it's on the output setting of each output channel. The impression may be the most similar to your mixer, or with the way soundcraft designs user interfaces and software, it may be nothing like it at all! Speaker positioning and output delay is a much better first step than the graphic eq, and is a better tool overall for tuning your system. Are there times when you need eq to fight feedback? Yes, but that should be the exception, not the norm.
You can use the delays for 2 purposes (which is what the post that mentioned collocated is referring to)- one is when you have speakers that are not together, you delay your farther speakers so the sound from those hits the farther portion of the audience at the same time as your mains (which is why those speakers are often called delay fills or just delays). When your speakers are on the same spot, delay helps you do the same but on a micro level, to align phase of signals, which is a compromise because all frequencies can't align in perfect phase, but you find the best compromise. There is science and measuring that shows this (that I don't yet fully understand but know the very basics and am learning), but ultimately you are using your ears to determine what sounds "best," which does not always agree with "ideal." But it will be close to what science shows as best because , again, physics is physics.
I hope that helped a little bit.
I definitely recommend checking out Nathan lively's sound design live podcast and his info on his website as a start.
And turn those channel gains down if you are constantly seeing the clip light. I don't know that mixer, but 0 may not be "analog 0," (0dbv), it may be digital 0 (dbfs), in which case youre running at something like +18 and then choking that off later in your system (that's what gain staging refers to- setting each step at the optimal level to get to the next step).
It will be the best money you ever spent on your gigging to hire someone for even one night to show you on your system, and if that's not available in your area, then look for a good online class or good books or good youtube videos (but beware, there's a lot of incorrect info out there). There are different opinions and different approaches, but physics is physics so anything you do to fight physics has limits.
Oh, one thing - again, I don't know your mixer (meaning I have never used it, i know what you have from your post), but on my x32 the output delays are on the output routing page, on yamaha mixers it's on the output setup page, and on the soundcraft impression I think it's on the output setting of each output channel. The impression may be the most similar to your mixer, or with the way soundcraft designs user interfaces and software, it may be nothing like it at all! Speaker positioning and output delay is a much better first step than the graphic eq, and is a better tool overall for tuning your system. Are there times when you need eq to fight feedback? Yes, but that should be the exception, not the norm.
You can use the delays for 2 purposes (which is what the post that mentioned collocated is referring to)- one is when you have speakers that are not together, you delay your farther speakers so the sound from those hits the farther portion of the audience at the same time as your mains (which is why those speakers are often called delay fills or just delays). When your speakers are on the same spot, delay helps you do the same but on a micro level, to align phase of signals, which is a compromise because all frequencies can't align in perfect phase, but you find the best compromise. There is science and measuring that shows this (that I don't yet fully understand but know the very basics and am learning), but ultimately you are using your ears to determine what sounds "best," which does not always agree with "ideal." But it will be close to what science shows as best because , again, physics is physics.
I hope that helped a little bit.
Last edited:
Thanks everyone. I read all the posts. My takeaways:
1. Lower the pre-amp gain (better gain staging). No clip light.
2. Position speakers away from walls and overhead roof when possible
3. Drive sub with a dedicated aux channel. Implement crossover in mixer.
4. Remove vocals from sub channel, also guitars? (we mic our amps)
5. To balance main/sub volume use Wasnex's idea of a/b'ing with sound generator at crossover freq
6. Use the feedback suppression built into the soundcraft mixer (it's dbx afs2)
7. Hire a sound expert for a gig or two to help optimize our setup
8. Consider polarity switching
9. HPF anything that doesn't have LF content
10. Output delay adjustment is not an option with my mixer. But I'll keep it in mind if I upgrade.
Hey, this is 9 ideas that I wouldn't have had without your help. TB is indeed an awesome resource. Restores my faith in humanity.
Before I use these ideas at a gig I'll test them out in a venue type setting. I have a buddy with a big metal barn that should be sufficiently hideous for feedback. Then I can a/b the different methods and quantify improvement in gain-before-feedback.
1. Lower the pre-amp gain (better gain staging). No clip light.
2. Position speakers away from walls and overhead roof when possible
3. Drive sub with a dedicated aux channel. Implement crossover in mixer.
4. Remove vocals from sub channel, also guitars? (we mic our amps)
5. To balance main/sub volume use Wasnex's idea of a/b'ing with sound generator at crossover freq
6. Use the feedback suppression built into the soundcraft mixer (it's dbx afs2)
7. Hire a sound expert for a gig or two to help optimize our setup
8. Consider polarity switching
9. HPF anything that doesn't have LF content
10. Output delay adjustment is not an option with my mixer. But I'll keep it in mind if I upgrade.
Hey, this is 9 ideas that I wouldn't have had without your help. TB is indeed an awesome resource. Restores my faith in humanity.
Before I use these ideas at a gig I'll test them out in a venue type setting. I have a buddy with a big metal barn that should be sufficiently hideous for feedback. Then I can a/b the different methods and quantify improvement in gain-before-feedback.
I still have some questions
1. Do kick and toms go in the mains as well as the sub?
2. Any recommendations for crossover frequency and slope?
3. Any suggestions on how to find a really good sound person?
1. Do kick and toms go in the mains as well as the sub?
2. Any recommendations for crossover frequency and slope?
3. Any suggestions on how to find a really good sound person?
I think this is a very good explanation for what was going on with us. So this answers my question about why we were getting feedback outdoors. Thanks.
After reading through these things I realize my mental model for sound reflection was wrong. I was using an analogy of light beams (speaker output) reflecting off of mirrors (surfaces), so very tight beams bouncing around the room. Now I see a better analogy is a bare light bulb in a dark room lighting up the room everywhere. Any surface that is illuminated (that returns light to your eyes) is a surface that will reflect sound to your ears. This captures both the omni-directional aspect of the sound source (speaker radiating sound in all directions) and the dispersion aspect of the reflections (sound reflecting in all directions, like ripples on a pond, when it bounces off a surface). With this model in mind I can now imagine how an overhead roof eave in front of a speaker can be part of a reflecting path to a wall behind the band.
Agree, the point mainly being that oddities like glass walls can play hell on overall response in a room, and it's a great place to start. And, I rarely have much of an issue with vocal mics in this venue, since it's 100% IEM, and our singers have good mic technique (unlike the doofs that stand 2 feet back and wonder why they sound thin and it rings . . .). Any yes, polar pattern is huge, and selecting the 'right tool for the job' is critical in that regard as well, as well as understanding polar patters of your PA cabs as well, and aiming to avoid reflections wherever possible.
But ultimately physics rules, and you hit the point where you are at max gain, and any more rings. I'd just argue that this is rarely the issue with gigs at clubs and such - typically other factors are in play - mainly setup . . .
Last edited:
I would definitely put kick in the sub. I might put the floor tom, but probably not the other toms.
I believe 24dB/octave Linkwitz Riley is pretty much the standard for crossover slope in pro applications. As far as the frequency, it varies a bit with your gear. Big pro rigs may run the subs at 80Hz or lower. If you're running small low-power mid/high packs you might want to run the subs up to 150Hz, so the mid/high packs have more headroom. Read the specs of your gear and try to make an informed decision; then experiment and use your eyes and ears based upon available meters and resultant sound.
I get the impression finding a good audio tech can be a challenge...good luck.
When using a compressor, feedback will depend on how much make up gain you employ. Compression, per se, shouldn't cause feedback if you use it right. You just always have to be aware of the overall gain of your system. Same as always.
Yes, if run correctly a comp won't leave you feedback haunted. And yes, like everyone else I'm diagnosing based over the internet...But the details here don't read like the root cause is the odd acoustic signature of the venue's performance space.
Yes, the feedback at this venue is probably made worse by the glass wall's acoustic reflections and some standing waves in the "stage area (an outdoor space, but tucked in the corner of exterior walls *and* roofed—almost certainly sloped down from the back of the stage to the front to shed the weather rather than raised at the front to project stage sound).
But...the OP says his band *is* haunted by feedback, *not* just in this acoustically odd performace space, and he regularly gets PA feedback in a band that plays at volumes low enough for patrons to talk over.
All those details say to me it's worth taking 15 seconds to look at the compressor settings on the band's board. They might not be the root demon here, but I couldn't justify skipping a peek, if just listening weren't enough evidence.
Yeah, I'd have to agree with that, but we're talking about a live performance here, so speaking from a purely feedback standpoint, the EQ needed to make mic 'A' sound good with person 'A' may make
mic 'B' with person 'B' cause feedback.
I do agree with you, however, because the system shouldn't be so 'on the edge' anyhow.
That's why I figure, if you're compressing a source so much that you need makeup gain, you might be using too much compression in the first place.
Indeed, I agree completely! (FWIW) Absolutely a good observation and excellent diagnostic technique. Systematically check everything on the soundboard (and everywhere else!), and certainly include the compressors! If the thresholds are set up improperly on a hot output, that could sustain feedback that ordinarily might merely be transient. I'm just saying the culprit is ultimately going to be the make up gain (on the compressor). Not that this shouldn't be carefully and systematically checked. Your right on the money there!
Now, I'm not real familiar with how these here new fangeled digital soundboards are set up. So I don't know how they process make up gain. On a black box that says audio compressor on it, when adding compression, you unavoidably loose output level. Any audio compression will cause this to happen. That's why there's both an input gain and an output gain control on the unit (like the venerable DBX160, for example). You employ the output gain to "make up" for the loss of output level from the compressor as you, well, "compress" the dynamic range of the signal. The more you compress an audio signal, the more loss of output level there 's going to be. You can't make up for that with additional input gain without eventually over driving your unit. So compressors provide an output gain usually with some soup to it for this purpose. The process of using this output gain to make up for loss of signal level as a result of audio compression is called "make up gain". I'm sure you already know all this. It's just a matter of how we're communicating the ideas. That's all.
Just "compressing" dynamic range in and of itself shouldn't create feedback. In fact, without compensating for the loss of output level from audio compression, compression would make feedback less likely. Them compressors can be pesky little devils, though. Sometimes, you have to shoot em.
Last edited:
You may be on to something. Make-up gain is 8.9dB. I used the preset compressor setting for male vocal.
Also the eq curve on this channel has a 2.8dB boost at 136Hz, not far from the 185Hz problem freq. This also is a preset for male vocal.
the make up gain is to bring it back close to unity. I think it may be that 136Hz bump on the EQ, flatten that out and see how it goes. As a general rule presets are not a bad starting point but in my experience they don't work, you need to tweak it to get the best out of it.
Holy Cow! That's a lot of compression on that vocal, partner! You may wish to try reducing your compression ratio on down to let's say 2, to at the very max 3 to one. I can't help you on threshold without being there, but wow, just from what I see it sure looks like your digging in there pretty deep. ??? For vocal compression, you may want to set it up so you don't need more than let's say 3 or 4 dB make up gain (more or less) to get back to unity as a starting point. On a live stage, a little vocal compression goes a long way. At best, it will just give you a touch more articulation. Vocalists like to work the mic themselves. If you crush the dynamic range real hard they can't do that, and besides, it takes the life out of the vocal. Of course, every situation is different. I'm just saying a little compression on a single mic usually goes a long way, that's all.
You don't necessarily have to put very much compression on a vocal mic. You can use buss compression or compression just before your line drivers to glue everything together in your mix. There are several ways to do that you can look into including blending the compression into an uncompressed buss and side chaining. Back in ancient times, the compressor often was the line driver, if you had a good one. The old DBX160 worked great for this purpose, and was widely employed like that. The DBX Drive Rack series incorporates that characteristic if you don't already have that kind of capability built into your soundboard. It would be highly desirable to have something like a DBX Drive Rack on your FOH.
Now, hard limiting on a vocal, that's another matter. Depending on the vocalist its not a bad idea if you don't already have brick wall limiting on your line driver sends as mentioned above, but you shouldn't hear the hard limiting, per se. It's primarily a safety feature to roadie proof your system. To set up hard / brick wall limiting you have to know how much power your potentially sending into your mains (particularly your HF drivers), and that's math. One vocalist I knew blew out all the Altec 290e's at the Fillmore West during an audition for his band! Opes! Things were pretty primitive back then. Typically for that time period, there was an LA2A up on that loft at the back of the hall where you had to mix from through a pile of Altec 1567's. Who knows where the LA2A was set at just then. I was up there a few times. A few times too many, actually. During a show, the cigarette smoke (and other varieties of smoke) up there on the loft would get so thick your eyes would tear up and you literally couldn't see! It was brutal. Oh yeah, that guy that blew out the HF horns? He didn't get booked.
I don't know about no 2.8 dB boost at 136 Hz. Why ask for trouble? Sounds like your using Shure microphones, so you got that right. Give yourself half a chance and begin your vocal channel eq flat and hi passed at let's say 80 to 100 Hz. You can always dial in EQ a little at a time if you think you need it. A little channel eq goes a long way too, and will indeed cause feedback in unstable situations. As you are no doubt aware, you just have to sort of feel your way along in a live mix a little at a time. Like, well, you know. Good luck!
Last edited:
That could do it.
OP, while you're under the hood, you might also have a look whether the mains have their own compression assigned. I've seen bands set their PAs with every individual mic channel heavily compressed (drum mics, too!), and then cascade those channels into aux and main channels that clamp down with more compression.
To trouble-shoot the band's habitual feedback, ease off on all of that. Then you can bring back a bit of comp one channel at a time to address any specific problems. As you do, HPF each channel just below where it starts contributing meaningful low end in the band's mix. And as a default for compression in live sound, set any channel comp's make-up gain so that the channel seems to stay at the same volume when you toggle compression on/off. (Unity gain on comps is not an iron law, if the SE knows the system and its limits. But if your immediate goal is a feedback Banshee exorcism, then any successful compression ceremony begins here.)
OP, while you're under the hood, you might also have a look whether the mains have their own compression assigned. I've seen bands set their PAs with every individual mic channel heavily compressed (drum mics, too!), and then cascade those channels into aux and main channels that clamp down with more compression.
To trouble-shoot the band's habitual feedback, ease off on all of that. Then you can bring back a bit of comp one channel at a time to address any specific problems. As you do, HPF each channel just below where it starts contributing meaningful low end in the band's mix. And as a default for compression in live sound, set any channel comp's make-up gain so that the channel seems to stay at the same volume when you toggle compression on/off. (Unity gain on comps is not an iron law, if the SE knows the system and its limits. But if your immediate goal is a feedback Banshee exorcism, then any successful compression ceremony begins here.)
BTW, same rule-o'-thumb with channel eq: if toggling it on significantly boosts the channel's volume, then you're looking at a likely culprit if and when feedback strikes.
Gain structure is the key to this I think. Next time you are at that location, before sound check, set your front of house (mains) and your channel faders to unity and then bring up your channel gain controls during line check.
You may be getting the channel strips to hot trying to get the stage volume where you want it. By running your master and channel faders up near unity you will find that the channel strips don't have to come up as much to get the same volume out front.
When you open the gain controls up to hot you blow your gain structure. Headroom goes out the window and feedback takes it place!
Best of luck!
Zombie follow-up question: @golfnsubie – when you ring out the monitors, are you using a pink noise generator? If so, can you describe the process using the ui24r's built-in soundcheck features?
