Drive-In - recommended FM transmitter power?

[Sean McKinnon]

That would be great if the TX-30 was readily available and you had a license for a signal with that range.

The nice thing with Q-Sys is that for the price of a DCP-300 you get a fully configurable DSP (mix down, limiting, highpass, PEQ, whatever you want, Automation Controller, 16 Channels of AES in on 8P8C connectors, 8 Channel HDMI in and Decoding, 16 Channels of Analog in (phoenix, 8 configurable as outputs if needed), 1 "mic" in on XLR, 1 Stereo 3.5mm in, and 8 analog outputs plus monitor and HI/VI outputs on phoenix, Also a physical main fader knob and a mute button. 4 Relays and GPIO are also included. For a budget conscious customer I used some old SPDT and SPST toggle switches in an old rack plate for selecting Feature, Non-Sync, Bluray, and Mic On/Off. Throw in a $200 Part 15 certified FM transmitter and you have everything you need for a drive in and it is upgrade-able for the future. Hell, you can add a UCI license for less than $150 and give them the ability to operate the entire thing from their cell phone!

[Mark Gulbrandsen]

If it were me I would probably look to use a DSP so that I can create my own "air chain" of limiting, compression, and combining.

Sean, If you combine a USL JSD-60 and the TX30 that is essentially what you have, except in commercial products that directly interface.

[Marcel Birgelen]

Yes, that's the problem, L/R/Ls/Rs/LFE will smother Center in a summed mixdown compressing scheme. I think in the X-Air, one could setup a combination of compressors and limiters on a channel basis that would take care of that.
E.g. have a ducker/sidechained compressor on L/R/Ls/Rs/LFE fed by the center channel. That would always keep the center channel on top IF there is signal on it. Sounds like a lot of fun trying...

This is something I can't even do with the Optimod, as it's just a single program/2ch-stereo device.

I toyed around a bit more last night. I kept the surrounds at -3dB, put the LFE at +4dB and the center at +1.5dB... not based on any real science, just intuition. I ran the stuff through the Optimod and listened to the output on the analog port in monitor mode. Apparently, the monitor port has a delay of 5ms, while the signal to the transmitter has a delay of 15ms. The monitor port doesn't apply stuff like comb filtering and other "FM-processing", but should give a pretty adequate rendition of what you would hear on the radio.

After reading a bit through the manual, I wasn't as intrigued as I was before, the amount of settings are pretty comprehensible. The unit is more of a set-and-forget type of thing, doing a lot of stuff pretty decently right out of the box. It's obviously optimized for radio broadcast channels though. I wasn't displeased with the preliminary result after more or less randomly toying around with the settings. It's clear that the thing will put some emphasis on speech and essentially puts "the midnight mode" (you know, that evil compression setting found on your average home cinema thingy) on the entire mix, which is, what I guess, is what it's supposed to do.

I don't have the transmitter here yet, so, I'm still excited to hear how it will sound through an actual car stereo like a little kid... you know, first time ever we're (hopefully) doing an actual "Auto Kino" around here.

[Marin Zorica]

I did some drive in shows in past few years and i hope there will be some more in next month....

Anyway, here i can get temporarly licence for drive-in transmitter at very cheap cost and is legal. You need to apply some three weeks in advance and check for clear frequency, than you pass location data, frequency etc data in request to regulatory angency and if everything is ok they send you licence with parametres and some fee to pay

From tech point of view, i always used L/C/R summed to either mono if transmitter is mono or stereo like L+C, R+C. In not rate ocasion i notticed that mono is best way to ho, because car systems (speakers placement, some audio processing etc inside of car) are "doing" something to audio you don't want to have while listening dialogue. Some classic compresson was added on main mix to keep max modulation of transmitter up to 75khz, because some radio hates to get clipped. And plus extra compresson on C channel before mixing so you can raise dialogue level, so in case some music etc on L/R, Center will not be masked by other.

As for other processing like optimod, omnia, bw etc i think is not needed at all, while i was working on radio station, it quite different story than to pass clear signal to listener cars, and most of todays fm broadcast processor will sound soooo boosted, which on other hand is go for fm radio.

Mainly i have used half lambda dipole and 50 or 100mW by licence which is enought to cover field like footbal field with good signal without interference. Tried also some cheap to $100 china ebay transmitters but they are no go, since most, like 90% are based upon small bh1441 ic intended for wireless headphones. So either i used professional fm tx, or my own tx i developed back in days with pll and 1w output.

BW is great unit or either some used pro rvr, elenos fm exciter you can find, it cost something like $300 but it will not give you any problems or interference during show.

[Carsten Kurz]

Yes, that's the problem, L/R/Ls/Rs/LFE will smother Center in a summed mixdown compressing scheme. I think in the X-Air, one could setup a combination of compressors and limiters on a channel basis that would take care of that.
E.g. have a ducker/sidechained compressor on L/R/Ls/Rs/LFE fed by the center channel. That would always keep the center channel on top IF there is signal on it. Sounds like a lot of fun trying...

[Marcel Birgelen]

I was just listening with a pair of Sennheiser studio headphones and I was somehow missing the LFE withouth the +10dB boost. Obviously, the headphones will not be able to render the lower band of the LFE, but there's indeed still quite some useful information in the upper bands as well. Putting some kind of HPF on it might be a good idea, if the Optimod doesn't take care of that already. Otherwise, the EQ on the mixing panel can get the job done. Most sophisticated car audio systems do bass management anyway and many of them boost the bass, so it's probably best not to overdo it and put it somewhere in between +3~6dB range. As for the surrounds, I think I'll keep them at around -3dB, and give the center channel a little kick, but it might depend a bit on the audio mix. The problem is that the surround channels seem to mud out the dialog from the center channel otherwise.

I don't know much about the Optimod yet, I'm a bit intrigued by the whole thing, but I'll start to play around with the next few days. Luckily, we've got someone who actually configures those things for a living. I guess it will probably filter the lower frequencies from the signal, but I'm sure there will be some LFE information left in there. The thing also features some special filters that Orban is known for and which is the secret sauce to the "presence" of many known radio voices. Also, stuff like de-essing will probably not be needed. And yeah, I guess we don't really need that secret sauce. What I'm more interested in is the apparent "world class" automatic audio-normalization/compression features the thingy has and how it will affect the audio dynamics of a modern movie. As for the latency, I was assured it was "just miliseconds", because it's a dedicated hardware solution and all processing takes place in dedicated DSPs, so for now I'm not so worried about my ~300ms budget I have on the playout side of things.

I'm going to take a look at the Behringer X-Air too, which might be a good but affordable solution when we don't have some people who do radio broadcasting for a living as a support team.

[Carsten Kurz]

That's an interesting discussion. As a lead in: Occasionally I service a small cinema, more a 'film club', quite a distance from here (NC900). I did supply some surrounds at some point, but they disconnected them later because of complaints. The room is just to small to give an ambient surround experience. It was simply too directional and distracting for 'their' audience. Well, some time later, they played 'Roma, and somehow someone noticed that there was dialog missing. Turned out, the Roma sound mix was sophisticated enough to have a lot of dialog in the surrounds only. That made them rethink their 'position' with regard to surround speakers...

Whatever:
We do all the mix down on a per channel basis in a conventional PA style mixer. So far, for our OpenAir screenings we used a small analog PA mixer with 4 outputs (two main, two groups), so we could have L/R/C and mono surround mixdown from an analog 5/5.1 input.

Now we use a digital mixer. It's a pity it doesn't offer AES inputs right away, but at the price point, it is simply not possible. So we use an otherwise unused DMA8+ in the mobile equipment rack. Doremi/Dolby AUD2DA or USL DACs would be other options.

We feed L/R/C/Ls/Rs into the mixer and leave away LFE. We use three screen speakers with 15" woofers, so I think it's okay to skip the LFE. We usually have between 300-500 people at our openair showings, and the listening situation is near optimal (a 'raked' seating area facing the screen, almost like a greek amphitheatre ). For a larger crowd on a broader estate, we would probably need to add separate subwoofers.

For FM transmission, LFE should be completely unnecessary. I understand, though, there is usually some upper spectrum in the LFE that could be reproduced by car audio systems, so, if you are keen to play and tune, one could probably use a HPF on the LFE and add it to the mix, so those who spent shitloads of money on their car audio systems and have no space left in their subwoofer-cabinet trunks will have fun.

I don't know what an Optimod would do with an LFE signal if it sees one. BTW - what minimal overall latency is possible with an Optimod? Because DCI servers will only allow 280-320ms of timeshift between audio and video, and, for classic radio broadcast, they usually don't care about such small latencies. I know some smaller stations use PC/software based signal optimizers/MPX encoders with in/out latencies of several seconds.

If you use a bluray player, even less delay may be possible in it's AV-Sync setup. To the point where you may need to master your own DCP with a longer pre-delay.

As far as surrounds go - I guess here is why a conventional digital mixer could be a very nice solution because of it's flexibility - you can use separate per channel compression/limiting/EQ, so, instead of just reducing surround levels, you could use compression and limiting, so you can make sure, surrounds are there, but not take over.

For the short film program, we only used a classic compressor. I will take the time now to tune a 5ch multiband compressor and the limiter in order to keep signals in compliance with local MPX constraints. This is, of course, not a commercial broadcast business, so, I am not trying to be as ambitious as they are with maximizing spectrum usage. Keep in mind - FM transmission is targeted at household and in-car receiption, for non-attention-paying audience, and with a lot of environmental noise, especially in a driving car. We can utilize quite a bit more dynamic range in a drive-in cinema, especially since currently, windows have to stay closed.

That small Behringer X-Air offers a heck of signal processing and routing options, and as I said, it is extremely practical you can sit in a car and tune your signal using your iPad while listening to your own standard car audio system (I consider mine as fairly outdated and average, so, I guess not a bad starting point).
The X-Air also offers nice addons - like an integrated USB-Stick based audio player - you can have it playing looped announcements/hints during 'admission', so people have something to tune their radios to when they arrive. Multiple Mic inputs for live announcements, instructions and warnings.

And, of course, everything you do can be saved and recalled easily. It is saved to the unit, and can be called up even if you connect with a different iPad or notebook. Heck, it's 'almost' a Q-SYS ;-)

[Marcel Birgelen]

Did you have a possibility to drive around a bit to see how far your little "pirate radio" is getting?

I've been toying around with our little setup, we're still waiting for the green light from the local authorities, as right now it's not clear if a drive-in cinema violates current rules against mass-gatherings... apparently, the authorities also don't really know how to interpret the ad-hoc rules in place.

At first I was thinking about using the CP750 I recently repaired, but instead we're using a digital mixing panel (with EQ options) with AES inputs to do the down mixing from 5.1/7.1 to 2.0 as this should give us maximum flexibility. I was wondering how everybody is summing up the channels. Right now, I've done a 1:1:1 on all channels to the left and a 1:1:1 on all channels to the right. The LFE also gets dumped 1:1 to both right and left. The sound didn't yet go through the compressor, as someone else with some real Orban knowledge will be setting that thing up.

I used this as a starting point and I was only listening through a bunch of headphones, but it's clear that simply dumping the individual channels "one on one" isn't really the right thing to do, especially in high-octane action scenes with a lot of surround action going on, it's clear the surround channels are way to present. I started to manually tune the surrounds down and the LFE a bit up to a level that it was more comfortable. I also toyed a bit with the EQ, but all in all it sounded pretty decent, almost flat.

So, I was wondering if anybody else is "manually" downmixing their channels to 2.0 and if so, to what proportions do you mix in the surrounds and the LFE?

The biggest uncertainty for me is the LFE. Should I boost it with 10dB or is that just too much for FM transmission?

[Carsten Kurz]

We played a short-film program. For operational reasons, I converted the original Bluray to DCP. It was a small scale operation, but it allowed us to do some real life tests for larger scale operations.

[Marcel Birgelen]

The compression is partly needed to keep within legal requirements regarding your allocated FM band. We're lucky to have a "cheap" Orban processor at our disposal and a broadcast-tech is going to tune it for us. I told him that I don't want dialog to sound like the average outraged radio presenter, so we'll see where we end up.

What movie did you play? And did you play it from a DCP source?

[Carsten Kurz]

We did our first (small) drive-in show yesterday. We used a small and low cost digital mixer, which offers a compressor on the main busses. We simply optimized it by ear. There are expensive devices optimized for commercial radio broadcast, but I think we don't need that for cinema, as in a drive-in, we don't take part in the common loudness war.

I was very satisfied with the results of the simple X-Air mixer, also because we can operate it remotely through Wifi. Which is quite practical, because you can have your gear 'somewhere' while you sit in your car with an iPad and tune things.

Many car audio systems nowadays are quite capable, and I think it is not necessary to apply too much compression.

I was trying to find typical suggestions for optimizing EQ and compression settings on the internet, but I didn't find much, most hits dealt with commercial software and processors. I guess if you look a bit longer, you will find some, but I decided to simply hook up the gear and try myself.

[Juan Jose Garcia Calvo]

We were having the same issue. We ended up installing a Compressor / Limiter between the USL JSD-60 and the TX-5 transmitter. Solved all of those clipping issues.

Can you share with us the compressor / limiter settings (ratio, release time, ....) and if it's a multiband compressor, what settings could we use for each band, and what split frequencies...
This culd be very helpfull

Thanks

[Barry Floyd]

I need to take some time to research compression. Initially our sound was great after installing digital, I found later last year some of the louder scenes started to sound "muddy" to me, though still fine during normal scenes. I'm always very nervous tinkering with the mixer during the show, even though I can see the "clip" light flickering and the TX5 reading " ! ". Not sure whether the DCP's are outputting higher levels lately, but wasn't an issue before.

We were having the same issue. We ended up installing a Compressor / Limiter between the USL JSD-60 and the TX-5 transmitter. Solved all of those clipping issues.

[David Buckley]

Unfortunately, the BWs don't have EQ in them, so such EQ would have to be placed externally, previously to the transmitter. The processor in the transmitter, useful as it is, is a preset-only processor,.

[Harold Hallikainen]

It seems like there should also be some HF rolloff added (about 7 dB at 10 kHz) since you are not shooting sound through a screen.