tapping into house PA

[Richard Crowley]

Originally Posted by Paul R Johnson

A DI box is pretty useful, because if you want a mic level signal from the board, then how else will you get it from line level sources?

The traditional/original meaning of "Direct Box" was a device to take the very low level and very high impedance unbalanced signal from an electric guitar and transform it into the low impedance balanced signal we expect from microphone sources. Direct boxes are not really suitable (or designed) for taking line-level signals as an input.

The way to get a mic level signal from a line level source is to use a pad. Some direct boxes may include a pad feature, but that is not their primary job.

It is also very likely that you need isolation from the house system, so a transformer is typically used. Again a DI box may include this feature, but that is not its primary job, either.

However common use of the term has been degraded to the point where "direct box" could mean a number of different things.

I have several of these Rolls DB25 "Matchbox" units and I like them a lot:
Rolls Corporation - Real Sound - Products DB25 Matchbox
They have a good size real isolation transformer along with a 0-20-40 dB switched attenuator and a switched ground-lift.

[Rick Reineke]

Yes I would agree with Richard. And yes the DI was primarilly for musical instrument interface, however the passive one often loaded down acoustic/electric guitars and such. I also have a Rolls as well as a few IL19 iso's and in-line pads. Incidentally, the latest Roll, the DB25-B (I think) has a variable attenuator which is great for more precise gain staging

[Jay Massengill]

Yes, the Rolls DB-25 b model does have a variable attenuator although some sites still show the old photo or even have the old description. I got two of them a couple of years ago right after the change was made so most vendors should have the b model now regardless of the description.

[Robert Bobson]

In what situation would you ever need to use two of the Rolls DB25B at the same time?

[Jay Massengill]

I'm not sure I could name any device that I own where I don't have two of them... Some days I have multiple setups on location in different corporate buildings at the same time and move from one to the other as needed with no time in between for moving equipment.

Even in a single setup there could be a need for different balanced signals to be routed to various recorders or cameras from multiple unbalanced outputs. If the recorder or camera has mic-level-only inputs, a pair of DB25b's is less expensive than a two-channel Ebtech Hum Eliminator plus external pads and adapter cables.
Some of my setups need routing to a PA system, telephone interface, two cameras, computer with screen recording software and a backup audio recorder simultaneously. There's always some need for adapting, padding, balancing and breaking a ground safely.

And there's that fellow Murphy who likes to show up unannounced.

[Paul R Johnson]

Not sure I agree with limiting the use of DIs to guitars - they're an invaluable tool for any on-stage job where it's necessary to convert hi-z unbalanced to low-z balanced. Most of the popular units have a number of available preset gains - with some even coping with loudspeaker level (although they're rarer).

Both pro-level and hi-fi level kit such as MD/CD/DVD/Computer outputs etc etc are perfectly suitable for the popular DIs (I tend to use the BSS ones myself) where the key factor is getting a balanced output. Level isn't a problem. Most designed for broadcast or live sound are extremely tolerant of high input levels, and the transformers (or electronics) are built to cope. It's rare, apart from on bass guitar to DI the feed from the guitars as most players use stomp boxes or other processing, which has a proper line level unbalanced output - and this is what the DI boxes sniff. Now many bass players use processing too - the old purpose of the DI box is changing, with the industry tending to use DIs simply as a convenient method of getting line unbalanced into something with an XLR socket!

Here in the UK transformer isolators are not very common. I've got one in the box somewhere, but I use the BSS DIs on all things really. The 133 has a 40dB pad for speaker connection, and an input impedance of of 1Meg - so isn't a load on sensitive single coil guitar pickups.

The 133 is the latest version of the old 116 which was extremely popular here for years and still give good service. If somebody wants a mic feed from a venue sound system of unknown type, then a DI is a perfect method to provide a mic level balanced signal from the common outs. Some desks will have +4 XLR balanced outs, but frequently, they're too hot for the pads built into popular cameras - people with Sony XD-CAM always say less level - and end up with a fader maybe an inch off the bottom - the gain structure is all wrong - they pads on the Sonys seem too much, or not enough.

A pair of decent DIs are a perfect solution, and I can't understand why US practice finds this a problem. Here, DI boxes are amazingly common in everyone's toolkit - as generic problem solvers.

ALL the current popular DI boxes offer solutions to ground loops, by either lifting the ground, or using a transformer with two windings but no common connection. Very few, apart from budget versions are without pads. All the cheap ones use transformers and only the more expensive types have electronically balanced outputs, and use phantom for powering (often with a battery option).

The current usage of 'DI boxes' is at least 30 years old - the time I bought my first one. It's possible we have a UK/US language problem, but checking the popular manufacturers websites, DI is a common term and I'd even go so far to suggest that what I've explained above is the standard for anyone who needs to connect unbalanced Hi-z to balanced Low-z. People like Alice and others make devices that isolate and in some cases protect, but stick an XLR on the output side and they get re-labelled DIs.

[Benjamin Maas]

They delay you are experiencing has absolutely nothing to do with the console. The largest latency out there in any PA situation that I know about is the old version of cobranet which has a 5-2/3 ms latency to it. Most digital PA consoles are in the realm of 1-2 ms of latency.

What you are most likely experiencing if you are a frame off is the time it takes for sound to travel from the stage to your microphone. Count on about 1 ms per foot which means that if you are far enough away, you could be approaching a frame of delay (especially if you are shooting at 60fps). The console will not be giving you a delayed signal so you will experience a bit of a difference between the two. This is why I tend to sync audio to video in post by eye- not by matching up waveforms. If the camera is close enough, you can make that work, but more often than not, the delay just won't cut it (especially when you get closeups on fingers and such).

As for the mix being suspect- that is absolutely correct. The job of the FOH engineer is to make sure that the house sounds good- not to provide a good sounding feed to video. If they do, it is a bonus, but not the first priority. When I'm mixing, I tend to set stuff up for getting at least a decent recording out of the console, but I'm definitely in the minority of engineers there. I will check a recording mix on headphones on occasion, but most engineers pretty much refuse to use headphones in any way when they mix.

If the PA situation is one where there is very little acoustic energy coming off the stage (ie the sound is all amplified- say large situations inside or outside), you are much more likely to get a usable mix for your video. If you are in an acoustic space (say a concert hall), don't count on getting a good sound as the PA is usually to suplement and balance what is coming acoustically off the stage (between acoustic sounds, amps, monitors, etc...).

--Ben

[William Hohauser]

That's interesting although I have experienced differing amounts of delay in the house feed not the expected distance of the live mike. As I usually shoot 30 or 24 frames (never 60p) in moderate (600 seat) to small houses sometimes the delay seems to be related to the PA feeds I'm receiving not the speed of sound. The worst was a delay of two frames in a small 200 seat theater until I got the tech to send an Aux feed from the mixing board not the main mix via the PA amp processors (or whatever they had rigged up). I did a little research and found that the latest theater shoot I did, the audio turned out to be routed from the lobby A/V feed which had a delay of 1 frame due to being routed thru a digital A/V switcher. That was how this particular house distributed their main mix audio outside of the theater's speaker system. So even though the house feed should be in sync, it isn't. The point I'm making here is that you can't always rely on the house feed and frequently the house staff are interns or students who do not understand how their systems (audio, lighting) work outside of a regular in-house performance.

The four big concerts I worked on where the performances went to DVD for commercial sale, we had a completely separate audio system with a qualified audio crew in place to record the music. The house sound was not reliable for this although both systems shared the inputs. Expensive but the results were CD quality (which were also released).

[Steve House]

Sound travels about 1000 feet per second. An NTSC video frame is about 1/30 of second long, PAL is 1/25 of a second. So for a close approximation, your sound will be delayed by one frame for each 33 feet of separation between the sound source and the microphone in NTSC territory, one frame for each 40 feet in PAL lands..

[Benjamin Maas]

Thank you Steve. Brain wasn't exactly working perfectly and I was a decimal off when I was looking at distance per frame. Yes, sound travels a little more than 1000 ft per second or a bit less than 1ms per foot. A 600 person auditiorium can easily have you 60-80 feet away fom the stage. Depending on the layout, you could be even further away.

Say you're 80 feet away and you're shooting 24P. That means that the acoustic sound hitting your microphone will be about 2 frames off.

If you are pulling feeds off of some sort of distribution amp- especially one that does video, you can introduce even worse latencies. Most of the digital audio switchers out there have latencies of only a couple milliseconds, but I suppose it is possible to have more.

I would suggest that if you are concerned about sync, that you're pulling from analog outs off a console. The more stuff you go through, the more of a chance of problems.

--Ben

[Richard Crowley]

But presumably the microphone(s) (yours or theirs) are located on-stage with the performer(s), not back at the camera. And if you ARE using a microphone 100s of feet away (at a camera location) then it is only capturing ambient room tone and not a primary source.

[Benjamin Maas]

Speaking as "audio guy" here....

I find it exceedingly rare that the video person has the foresight to place mics on stage. depending on the PA situation, a mic placed on stage may not actually capture a good sound. In any case with a band on stage, the stage sound is pretty poor and the sound in the house is much better. The couple times I have had video people want to place mics on stage, it has been a major problem. Classical ensembles where the video person wants to put a big ugly stand in front of a group (in a room where mics are hung). Or, even worse, a video guy coming in where there are stands in front of the group and they just assume that they can clip their mics to mine without asking permission.

Another thing I see a lot of is the video guy that only wants one side of a stereo mix so they can use their own mic in the back of the hall (speaking of delay and lag issues). There are even times when I'm providing a broadcast recording feed and this happens. It is not a mono sum, but rather half a mix and then a camera mic that is out of time with the stage sound.

Don't get me started on all of the unprofessional things I see on a regular basis.

--Ben

[Richard Crowley]

Yes, I was speaking as someone who was doing audio at least a couple of decades before getting into video. In fact my major motivation for getting into video was to try to improve the state of audio for video.

Alas, as a video producer I have seen first-hand how easy it is to think of audio in second-class status behind working out all the video details. But, of course Mr. Maas is spot on with his observation of really deplorable audio recording practices by videographers.

It is one thing for the local news reporter to come in and shoot 5 minutes of B-roll for the 11 PM news, but someone who comes in expecting to capture a whole concert with a single camera and minimal (if any) external audio equipment is just unbelievable. Why do they even bother?

And I find myself more than frequently on one or the other side of the mixer desk. Perhaps I am just unlucky, but most program producers I work with are functionally clueless and useless in coordinating live feeds for video (etc.)