This is topic Setting SPL Levels in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.
To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=002626
Posted by Paul Konen (Member # 41) on 09-04-2001, 10:01 AM:
One benefit that I have received because of the DLP is having THX come out and EQ our auditoriums. While loading the DLP, they setup their R2 units and do EQ.I am raising this subject to see what you all thought about the following.
When setting SPL levels with a CP-500 processor, the tech will set the levels straight across on the processor and use the gain on the amps to reach the proper level.
The idea is, ( I hope I say this right ), to amplify a higher signal with a lower gain rather than amplify a lower signal with a higher gain. With the gain all the way up, you not only amplify the signal but any line noise that may be present on the wire.
Posted by Jon Bartow (Member # 241) on 09-04-2001, 02:47 PM:
While the concept has merit, in the real world it is not that practical for a couple of reasons:
- Without very expensive test equipment (like the R2 for example) one could not detect the S:N difference between "traditional" and the way the DLP guys are doing it.
- If the amps are all set at max then one could switch out a bad amp without resetting the SPL
- If the amps are all different then you are just asking for one of the amp's gain levels to be bumped by whatever (vacuuming out the vents on the amps, someone plugging in a piece of equipment into the back of the rack [How do you know that the DLP guys didn't bump one of the gain knobs when they were packing up their equipment the last time that they were there?], etc.)
If the amps are all set at Max then this eliminates that potential.If you have a private screening room or you operate a single screen theater with projectionists (not operators) that really care and are very careful, then go ahead. But in the modern Megaplex with manager/operators and no projectionists in sight you would be just asking for trouble.
Jonathan Bartow
Posted by Gordon McLeod (Member # 33) on 09-04-2001, 06:46 PM:
Actually it doesn't take a R2 to notice the difference in fact the R2 doesn't really see the differenc e anyway
The idea is that the input to the THX crossover is to match the output buss of the processor and that is 300mv
By useing the amp gains to set the levels the system is dramatically quieter.
Posted by Joe Redifer (Member # 3) on 09-04-2001, 07:38 PM:
Adjusting an analog knob on an amp allows for more precision than the digital "steps" of a processor like the CP500 as well. Usually when I see this done, the first thing that happens is the knobs themselves come off. The adjustments on the amps can still be made this way, but you won't risk "bumping" a setting when you pull the knob off later. Then of course special caps go on so it is difficult to accidentally make an adjustment.
Posted by Mark Gulbrandsen (Member # 72) on 09-04-2001, 11:47 PM:
Gord is absolutely correct on this point....This is not something the R-2 would ever see. You would need an AC millivolt meter to measure the noise levels at the amplifler output. You should set the output of the processor for 275 to 300 mv to the power amps. Doing so should give you a really good signal to noise ratio. This will also improve the apparent(and real) dynamic range of the B chain. No noise at all should be heard from the screen speakers even if you walk right up to the screen. NEVER BUT NEVER crank the amp gains all the way up and then set your processors output levels. You should do your job in a professional manner, not in a manner that is convenient to you. Reasoning like this is why movie theatres have deteriorated to the point that they have. You have to consider that some amps actually go to the plus side of odb when it comes to gains and going over 0db will only lower the signal to noise ratio and you will also loose some dynamic capability as well. Cranking gains up past 0db will get you nothing but noise that can most likely be heard in the center to rear of the autitorium!
Also use only an R-2 analyzer for the absolute best results! The extra effort and actual small expense is well worth it. NOTHING else for general motion pictue work even comes close to the results an R-2 will bring in! As Brad has conformed on this site before, Yes, I am married to my R-2. Its been 6 happy productive years now and counting.
Mark @ GTS
Posted by Joe Redifer (Member # 3) on 09-05-2001, 02:04 AM:
Mark, the R2 is obsolete, you of all people should know that. Especially with the coming of the R3. 
Posted by Mark Gulbrandsen (Member # 72) on 09-05-2001, 09:53 AM:
When the R-3 is available I'll be the first in line too......but I'll have to get divorced from my R-2 first.........but then I do live in Utah......and some have many wives here.
Mark @ GTS
Posted by Pete Naples (Member # 761) on 09-05-2001, 02:40 PM:
OK I'm following this. Would I be right in saying that if one wanted to apply this method to non bi-amped systems, or to non THX cross overs, one would have to modify that 300 mV setting to match the input sensitivity of whatever piece of kit is in use???
Posted by Steve Guttag (Member # 268) on 09-05-2001, 06:08 PM:
Pete, it depends on the nominal output level of the processor and in input of the following device (amp in your case) as to what is the best S/N would be. Normally, but not always, the best S/N is obtained when the processor outputs are set to 0dB (digital) or maximum (analog) though it depends on how the output stage is designed. Some processors actually tell you which is best...Sony, for instance is best at 0dB. Panastereo recommends turning the output trims up all the way and adjusting the amplifier attenuators for proper level.Steve
------------------
"Old projectionists never die, they just changeover!"
Posted by Pete Naples (Member # 761) on 09-06-2001, 03:05 AM:
That makes sense, in a nutshell one is matching the output and input stages as closely as possible.Though I guess one is stuck with adjusting the CP output if the power amps are of a fixed gain design though.
Posted by Antonio Marcheselli (Member # 373) on 09-15-2001, 02:32 PM:
Hi,My tech did the same. He set the SPL from amplifiers.
There is a theater in my town where all amplifiers are set to maximum. I remember that in a quite scene of a movie (or walking near a surround speaker) I heard a lot of noise...
Do you keep your Home Hi-Fi amplifier to maximum and set the CD output knob to have a good result?
Bye
Antonio
Posted by Stefan Scholz (Member # 147) on 09-16-2001, 03:56 AM:
Some amplifiers offer lockable gain controls on the rear, so that you could not change them by coincidance. It is not a great number,I know, but some mfgs. can add the extra attenuator at a small extra cost.
The idea is simple, as power amps have a fixed gain (therefore fixed S/N), you're setting the preamp (processor, crossovers, etc) to run at it's best S/N ratio level, which is normally on the higher side of output voltages.
Then attenuating the signal to match actual SPL in the auditorium the power amp controls come into service. Result can be a dead quiet system, compared to the hissy way with attenuated preamp and power amp attenuator at minimum.
Posted by System Notices (Member # 2357) on 04-04-2005, 11:57 PM:
It has been 1296 days since the last post.
Posted by Tom Wienholt (Member # 1510) on 04-04-2005, 11:57 PM:
So what are the exact procedures for setting the outputs to 300mv? Is this the best thing to do with CP50's and 55's, set the processor outputs to 300mv and adjust the SPL's with the amps?
Posted by Tom Wienholt (Member # 1510) on 04-06-2005, 11:15 PM:
Is it ok to just set the amps at maximum and adjust the processor outputs?
Posted by Pete Naples (Member # 761) on 04-07-2005, 03:49 AM:
Tom, in a word no! Whilst you will end up with the correct SPL levels, the S/N ratio and headroom might be awful. I've cured quite a few noisy systems by 'gain staging' this way, and it sounds audibly better to my ears.
Even if you don't set the CP output to 300mv, you could try putting the CP outputs full up and pulling the amp gains down to get the correct SPL, I often turn the CP full up and then count down, say 3 turns on the pots, not very accurate but better than nothing.
Posted by Tom Wienholt (Member # 1510) on 04-07-2005, 10:01 AM:
The thing that confuses me is that for CP50's, there is an output adjustment on the equalization card for L,C,and R channels plus on the output card itself. How do I properly set the output to 300mV. Do I use the pink noise card or test film to do this. Any help would be great. Thanks.
Posted by Sam D. Chavez (Member # 1841) on 04-07-2005, 11:08 AM:
I did not get in on the beginning of this topic so forgive me if I repeat something already stated.
Regarding the CP50, there were any number of revisions to the manual as to this procedure having to do with the fact that the matrix card continued to evolve.
The best way to set the Cat. 64 pots is inputs all the way down and outputs pots set (to around 200 mVolts)with a generator after b-chain EQ and before final SPL setting.
This method is much simpler, takes wayward a-chain adjustments and the test film out of the loop, pun intended.
You can then set the CP 50 output trim pots to suit your situation.
I taught this method for many years at Dolby Labs
Posted by Tom Wienholt (Member # 1510) on 04-07-2005, 05:34 PM:
So you are saying that the output levels should be adjusted by the output card and the amps can be turned all the way up? What about for a CP55 with Cat64B's that don't have an output adjustment on them? Or a CP50 with Cat64B's? Should the amps be turned all the way up and levels adjusted with the output card? Thanks for the help.
Posted by Gordon McLeod (Member # 33) on 04-07-2005, 05:42 PM:
You can do what ever you want but ideally a soundsystems gain structure is optimal if the SPL is set on the power amps and the processors output trimmed to the processors buss level
Posted by Tom Wienholt (Member # 1510) on 04-07-2005, 11:04 PM:
So basically the optimal processor output level is about 300mV per channel? So I should set the outputs to 300mV using a pink noise generator as the signal?
Posted by Gordon McLeod (Member # 33) on 04-07-2005, 11:46 PM:
You have asked this question in various forms several times so why not just have "your qualified Dolby service engineer" come out and set it up right
Posted by Tom Wienholt (Member # 1510) on 04-08-2005, 12:07 AM:
I know I have. It just seems like different people give different advice. I have set up many sound systems by simply turning the amp gains all the way up and adjusting the levels with the trimpots. This method has worked great but now that I have read these discussions I am considering "re-setting" the levels with the amp gains since I am being told that it will improve S/N ratios and improve sound quality. Just trying to be very clear on the method to use. Thats all.
Posted by Tracy Bellar (Member # 3049) on 04-08-2005, 12:11 AM:
In my theater we keep the amp gains up and set the processor. I know it isn't the best situation. However, in a theater where many employees other than myself have access to the booth and sometimes go in and bump a knob. They should keep thier hands off I know but it happens. We get complaints of the sound being too loud then too quiet. I wish the policy would be noone in the booth but trained people. If this would become a standard then I could see setting the gains down to where they should be. It's a quick fix to stopping someone from changing an amp thinking it was master vol. Amps should have a glass enclosed locked shield to keep hands that shouldn't touch.
Posted by Tom Wienholt (Member # 1510) on 04-08-2005, 12:26 AM:
Im not worried about people messing with the amp gains, they are on the back of my amps. I want to have the best sound quality possible with the equipment I have.
Posted by Tracy Bellar (Member # 3049) on 04-08-2005, 12:30 AM:
ours are on the front. Too many young kids working there that thinks if they have a nice stereo at home that they are qualified enough to change levels.
Posted by Gordon McLeod (Member # 33) on 04-08-2005, 08:52 AM:
The best headroom and lowest noise will be found by useing the 300mv method
Posted by John Walsh (Member # 168) on 04-08-2005, 09:57 AM:
I thought it was a good idea when the QSC USA's had the gain pot on the back ... accessable, but not too easily. Maybe they should include a feature on their next generation of amps where even the gain controls are only accessable with an RS232 port.. !
Posted by Scott Norwood (Member # 30) on 04-08-2005, 10:07 AM:
I don't remember seeing the 300mV information in any of the Dolby manuals. Did I just miss it, or is it not there? If it's not there, then why isn't it? It sounds like it should be in the manual (if it isn't already).
Most installations that I have seen use the "set the pots on the amps to maximum" approach. I can only guess that this is done to make it simple to replace an amp if it ever fails.
Posted by Sam D. Chavez (Member # 1841) on 04-08-2005, 11:19 AM:
The main reason techs run the amps wide open is to keep the staff from turning up the surrounds in particular.
Posted by Steve Guttag (Member # 268) on 04-08-2005, 12:23 PM:
300mV would only be best on certain processors...not as a general statement. It all depends on where a particular piece of equipment's Signal-to-noise is maximized. This is not the same place for all processors nor even all Dolby processors.
In general, the earlier you do amplification and the fewest places you do it, the better. Also, in general, running amps wide-open will not yeild the lowest noise.
Posted by Tom Wienholt (Member # 1510) on 04-08-2005, 12:42 PM:
Is there a way to determine the optimum signal to noise output setting for a particular processor?
Posted by Frank Angel (Member # 248) on 04-11-2005, 10:42 AM:
In one booth the amps are all Crowns. They originally were set up with the pots wide open and even marked with little stick on dots, not to mark levels but to insure that booth personnel could visually see that one pot was to be run at full and the other at zero because they were stereo amps that were used in mono mode -- only one pot controlled the level.
Thing is, if you went up to the stage and listened to the wide opened amp, you did hear some noise component. I pulled the inputs on the amps and still there was noise -- not alot, mind you but it sure wasn't that dead silence that you should get from the amp. In my mind, the amp should never add any noise.
I had someone in the booth turned down each amp's pot as I stayed on the stage listening. It only took a slight reduction on the pots -- maybe at most 1/5 of a turn to get them completely silent (to the ear at least). We remarked the pots and then made adjustments accordingly at the processor.
As Steve says, running amps wide open may not give you the best SN.
Posted by Louis Bornwasser (Member # 3063) on 04-12-2005, 06:42 PM:
Most amps have their level control "first thing" after the input. The noise level should not change when the input is unplugged and the pot is rotated from one end to the other.
When you attach an input (processor) you are then attaching the output stage of the processor to the input of the amplifier.
It is the output stage of the processor that is the noise generator. Some processors are worse than others. Of particular note is the CP-100. With this processor it is correct to run the trimpots up full and reduce the noise with the power amplifier input. In olden days, I made this modification at the Uptown in D.C. Other processors such as the CP-500/650 do not seem to have this problem.
Remember there are only 2 calibrated locations on an amplifier pot: full on and full off.
Posted by Ken Jacquart (Member # 369) on 06-22-2005, 12:18 AM:
Sam is correct. Most installations that i've seen have the amp gains turned maximum. Let's be real here... CP's of today and amplifiers of today are so darn quiet that I challenge anyone to hear the difference between methods. True enough... optimization to 300mV may techncially provide a better S/N, but if you can't hear the difference, should one play it safe and turn the amps to maximum? An amplifier at maximum can't be "tweaked" louder and if it is replace with an identical amplifier, can be set easily...
Posted by Ari Saarinen (Member # 3188) on 06-22-2005, 01:00 AM:
Steve is very right about this issue. Even that todays equimpnets are very high quality, but also same time amplifier powers have gone much higher (at least here in Europe). All most every new multiplex have had tri-amp stage channels allready last 5-6 years. It's allways best way to tune system with correct signal structure.
About todays high quality products... even one three letter brand digital x-overs from California are sometimes more like noise generator than silent ones (SG-01)... but you can avoid it in most cases... if things are done correctly in signal structure to get best possible signal to noise ratio.
Posted by Michael Schaffer (Member # 1204) on 06-22-2005, 02:55 AM:
Dunno which manufacturer you mean, but we just installed 14 systems with digital crossovers made in California, and the systems were very quiet with the amps turned all the way up. I have revised my earlier practice of setting the amps at approx. 2/3 and the processor channel output levels as close to bus level as possible. Man, you wouldn't believe how much booth personnel tinker with equipment here, and how many creative ways they find to break and blow stuff up.
Posted by Ari Saarinen (Member # 3188) on 06-22-2005, 03:07 AM:
I did mean older THX x-over version SG-01...
Posted by Steve Guttag (Member # 268) on 06-22-2005, 06:37 AM:
The QSC amps have a 6-dB hit when you use the "data ports". As such, when you turn your amps wide open, you actually are still 6dB shy of wide open as compared to using the XLR or phoenix connectors. Since I have seen the DCM clip out many times before the amp...I would always turn the amps up full when working with that unit.
As for calibrated pots...nope they are not calibrated at both ends, just the one (zero ohms)...at the other end they will vary a bit but hopefully within the tolerence set by the manufacturer of the pot.
I set the gain structure for the lowest noise, unless I have exposed pots (not much anymore since the DCA amps by QSC have cover plates).
The input attenuator should be thought of as a "sensitivity" control rather than a "gain" or "attenuator"...it should be set to yield the best S/N. I can certainly understand the desire to set to max since it is easy and repeatable and only gives the tinkerer the ability to lower levels.