JSD60 headroom in a drive in

[Jay Glaus]

I know drive in sound isn’t something studied as much these days, but I’m curious if anyone has any insight...

On my third screen I’m running a JSD60 in “drive in” mode. I use the two transmitter outputs to feed the transmitter, I use the snack bar output for building speakers, and the field to my field amplifier.

I have the headroom set as low as it will go, but it still seems like there is far too much headroom in my field. The lower scenes are barely heard through my speakers, yet when there is a loud scene it makes my amplifier clip. There is no winning. You either don’t hear the lower scenes to save the strain on the amp and the cones, or you turn it up to make that audible and the field blares when it’s loud.

I run Dolby CP750’s on my other screens and do not have this issue. The low scenes are low but you are able to hear what is going on, and the loudest possible scene might just barely clip the amp if at all, but it’s still a comfortable listening volume.

I’m debating taking the JSD60 out of drive in mode and remaking a pinout to go to a mixer and seeing if that helps.

Anyone have any suggestions?

Thanks,

Jay

[Mark Gulbrandsen]

I did two systems similar to yours that have three separate feeds.You need an independent way to set the input levels at the field and snack bar systems. And why are you using two transmitter outs to feed the field? Do you have two separate field systems or just one? The JSD-60 is a really great DI processor and I have installed quite a few in both DI and indoor systems. It has much newer D to A converters than the CP-750 does.

[Jay Glaus]

Hi Mark,

I’m using the Left Transmitter and Right Transmitter on the transmitter, the Snack Bar output goes to my building speakers, and the Field output goes to my field. I don’t use the Monitor output.

See I don’t think my problem is in the level, I feel like there’s too much range in the volume. I can increase the level of the on my field output but then it’s way too loud and clips when the movie gets to louder scenes. But if I set the level to where it doesn’t get too loud and clips, you can barely even hear softer scenes. If I could just boost the soft scenes a hair, or bring down the loud scenes some in my field, it would be perfect. But setting for the smallest headroom doesn’t help.

[Steve Guttag]

You are misinterpreting the "headroom" setting. Headroom refers to how much internal processing DSP is held in reserve to avoid clipping the DSP due to excessive EQ or volume level. The trade off is noise (Something that is normally a non-issue at a Drive-In). If you don't EQ any of the outputs, then you should be able to run your volume anywhere in the range and not clip the DSP. However, if you have some heave EQ going on with quite a bit of "boost" to either the bass/treble controls or the 1/3-Octave bands, then those levels factor into the levels in the DSP chain and when you raise your volume level beyond 7.0, the headroom setting can keep you in the safe zone...at the expense of noise. To put it another way...let's say that instead of 16-bit or 32-bit or 64-bit (I don't recall what the JSD-60 has) you had 4-bit DSP...or just 16 levels to represent the entire volume of your audio (yes, extremely low quality but very visually easy to see). For easier math, let's also say that the audio has 100dB of dynamic range. If I had 0dB of headroom, then each step of my 4-bit system would represent a 6.25dB change. However, I could never turn my fader up beyond 7.0 nor could I EQ with ANY boost anywhere. Alternately, I could give myself 12.5dB of headroom by trading in 2 of the steps and now each step represents a little over 7dB. But, I can run my fader up to 10.0 (worth about 10dB or I could add some EQ and depending on how much, run my fader above 7.0 with what is left.

Now, with the JSD60, you likely have 32 or 64 bit internal DSP so the resolution of audio is high enough that one can trade in some DSP to pick up some better noise figures, if needed. However, to ensure that for most installations that the internal DSP doesn't clip, even when the volume is run up to 10 and with some bad EQ, they start you off with some headroom (26dB). Moving it down to 20dB only hurts your cause, if you are clipping internally in the processor.

What you may be thinking is a compressor, which compresses the dynamic range. If your field amplifier(s) are clipping, that could be an indication of many things...how are the speakers/transformers in the field? Have you overloaded the amplifier (have you added up all of the speakers, what wattage they are tapped for and compared that to the total output of the amplifier)? There is no real bass on field speakers so you can pick up a bit of dynamic range by rolling off everything below 100Hz and not amplifying that at all. You could gain 10dB there alone (in the Filter Section). And, since field speakers are not the best with HF either, I'd make it a "Band Pass" filter and just get rid of what you aren't reproducing so the amplifier isn't wasting its power amplifying what nobody is hearing.

Ideally, you'll either find what is consuming your amplifier power (issue with the speaker line, shorted or missmatched field speaker/transformer or increase amplifier power to handle the field. However, failing that, adding a compressor could also alleviate any remaining clipping by narrowing your dynamic range...however this is NOT a headroom control, which trades in noise for greater ability to tune a space. Compressors can be all over the place and it takes a bit of finesse to set them up to the point where they do their job without overtly sounding like there is a compressor on line (you don't want to quash all dynamics with an aggressive compressor with fast attack/release times. You want just enough. Depending on your skills with a compressor, you might consider a "canned" solution by RDL, the ST-CL2. Adjust the input to get the compression to the point you are never/rarely clipping and you are done. It is a variable compressor that makes the job pretty easy.

[Harold Hallikainen]

Steve is almost entirely correct. The DSP uses floating point (32 bit or 64 bit floating point) calculations, so there is not a limit (well, there is a VERY HIGH limit) to the calculations. However, there is a 24 bit D/A at the output. If there is boost in the equalization, it's possible for the D/A to clip with a full scale value. The headroom control adds floating point attenuation before the conversion of floating point to integer to drive the D/A. To compensate, the amplifier gain after the D/A is increased a corresponding amount (same as adjusting the output trim). If the headroom is increased (decrease gain before D/A), the analog gain after the D/A is increased. The signal output of the D/A is decreased, but the noise level remains the same, so the signal to noise ratio decreases as headroom is increased. It would be interesting to add logging of when D/A clipping occurs (this would also be a nice feature in amplifiers!).

Steve's idea of rolling off the low end is good. There is a lot of low end content that field speakers cannot reproduce. Since there is a separate equalizer with HPF for each channel (including the field speakers in drive in mode), it should be easy to do.

The drive in mode does the same thing that an outboard mixer would do, but does it as floating point multiply and add (see the advanced tab on the GUI). So, an external mixer would be adding hardware that does not accomplish anything.

You can possibly improve things by adding a compressor. If the JSD-60 is driving a transmitter that includes an audio compressor, you could possibly just drive the field speakers with a mono FM receiver tuned to the transmitter. That would save you from needing to get another compressor.

Good luck!

Harold

[Mark Gulbrandsen]

I would not roll off the low end to the transmitters, especially if they are a higher end transmitter. People that show up with subwoofers will be very disappointed....

[Harold Hallikainen]

Agreed that the frequency response should not be limited when driving an FM transmitter. Field speaker drive should be limited to what they can reproduce.

[Jack Ondracek]

Having run FM at my drive-in since 1987, I think I can weigh in a bit here. As always, this represents my personal opinion, which doesn't necessarily make yours wrong.

The comments about dsp, its capabilities and wide dynamic range are all related to Jay's problem, but the practicalities for drive-ins pretty much boil down to this:

1. Over-the-air, broadcast analog FM has nowhere near the dynamic range of the signals bouncing around even the simplest digital booth. That's a dramatic switch from the pre-Dolby days when we were running monaural, optical sound tracks. Those tracks were limited and pre-emphasized to control the surface noise of the film and the top limits of the sound track modulation. The secondary benefit of this control was that it automatically fit within the practical limits of the common drive-in speaker system. The dynamic range of non-sync sources, such as records and magnetic tapes were within the same relative range, so no additional control was needed.

2. Most in-car listening environments tend to be a bit on the noisy side. In an indoor cinema, you would never tolerate the kind of noise levels that are common in a car full of kids and companions. You subconsciously expect the sound you hear at the drive-in to cut through this noise, just like any other station you'd otherwise listen to. If not, the difference becomes noticeable to you, to one degree or another, and not in a positive way.

3. The average listener will tend to rate the sound he(she) hears from your system as compared to the FM stations that the patron normally listens to. As is human nature, the patron will be more likely to notice bad sound and pass off good sound as normal and expected. This means, as with everything else at your place, you will be more likely to hear complaints than praise.

In a competitive market, you'll be lucky to hear more than a few dB of dynamic range from any FM broadcast station, and this is generally dependent on format. Classical FMs will usually have far more range between the loudest and quietest sounds they broadcast to you than rock stations, most of which put their level meters at the red line and leave them there. Like it or not, this is the kind of sound the listener gets used to and, eventually, considers normal for their experience in that particular (vehicular) environment. As noted above, the sound you offer will be automatically compared. With a bit of work and subjective artistry, you can actually sound better, because you don't have the range of competitive factors to deal with that the commercial station has.

On a good day, the average FM system, from audio source to your radio, will get you around 45dB of dynamic range, making an audio file or cinema sound track, with the kind of range Harold describes, frustrating to listen to. You would be constantly turning the volume of your radio up and down, just to keep the levels within a comfortable range that could be heard above road noise and conversation.

So... here's where audio compression comes in. Simply put, the job of a compressor is to bring the quiet stuff up to a listenable level, while not allowing loud conversation, explosions and car crashes to blow out your speakers. A well-tuned system will let you set your radio to a comfortable position and hear everything, loud and quiet, while still offering enough dynamic range to make the mix sound "natural". This means that the sound of someone, sneaking around in the house, shouldn't go through your radio at the same level as the gunshot that follows.

If you don't have some kind of automatic gain device in your drive-in's audio chain, you are pretty much shot from the start. Again... the audio chain in your booth can generate more than twice the range that can be received in your customer's cars... meaning that no matter how clean your system is otherwise, the need to constantly fiddle will be a negative distraction.

Audio compression is a very mature science, thanks to years of development in the broadcast, entertainment and sound reinforcement fields, so you have a lot of options here.

There are several ways to address the need for dynamic range control. Most (but not all) of the popular drive-in transmitters have some form of compression. BW Broadcast transmitters have a 4-band processor built-in. Some others have a single-band compressor/limiter. There is also equipment that can be connected between your mixer/processor audio and the transmitter. I use a software-based processor called "Stereo Tool". It's a complex, multi-band system I use in many of the AM and FM radio stations and translators I maintain for contracted clients of mine. Simpler devices are available from music outlets (Musician's Friend, BH Photo, etc). DBX and Symetrix make good, single-band equipment that is easy to connect and not very expensive. And... there are many good choices in-between.

What you choose will likely be a balance between how the result sounds to you personally, and your own technical expertise. In any case, just the act of reducing the dynamic range of what you send out, by any means you end up choosing, will make your system sound far better than having no control at all. The rest is subjective.

Oh... one other point. A compressor system is meant to control your theatre's total sound automatically... meaning it replaces the need for you to have any kind of volume control. Installed correctly, you should never have to manually adjust the volume of anything... ever.

Regarding Mark's comment: I do roll off the low end of my sound. Matter of fact, many drive-in owners I know don't bother to connect the subwoofer channel to their system. Most automotive systems won't pass those frequencies anyway, and the people who have subwoofers don't need to impress the rest of the theatre with their personal listening device. In a multiplex, a moderately-loud system in one field can rattle the fillings of customers in another. So, rather than having to fight the problem, I just leave most of those frequencies out. The rest of the issue is easier to deal with.