Integrating Sub - Lots of Questions (gain, noise, crossover, etc.) - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 23 Old 03-10-2013, 07:28 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Hey guys,

I posted this over at HTS, but I thought I'd check in here for feedback as well.

I've just purchased my first sub (well not exactly first, but I haven't had one in more than 10 years), and I'm in the process of integrating it into my 2 channel system (which also serves as HT system). I'm hoping that you can help me get things properly set up.

My system consists of the following:
Denon 3312ci AVR
Oppo BDP-93
Salk SongTower (L,R,C)
Salk Custom Sub (LMS Ultra 5400, sealed, 3.3 cu ft)
SpeakerPower SP1-2400 Amp (for the sub, running on its own 20amp circuit)

Gain Structure and Clipping:

First, I want to be sure that I set the gain correctly on the amp. I read through Wayne's comprehensive discussion of gain structure as a guide (from over at HTS). I measured the sub pre-out on my AVR with a volt meter and found that it produces 6.8 volts (with volume at max and speaker-level for the sub at max). I didn't have a way to identify where this signal begins to clip, so I took Wayne's advice and subtracted 30% from the max voltage to obtain an estimate of clean signal output--this came to 4.76 volts. I then adjusted the volume on the AVR down until the sub pre-out produced 4.76 volts. Then I connected the sub amp (but disconnected the subwoofer) and played the 60Hz signal provided by Wayne. While playing this signal, I then proceeded adjust the gain on the amp until the clip light started to blink. Wayne said to adjust the gain until the clip light becomes solid, but I only had enough gain to adjust it until the light was blinking fairly rapidly. So my question is, have I done this properly? As set, it appears that the amp can take all 4.76 volts from the AVR as it starts to clip.

After setting the gain as described above, I adjusted the speaker-level for the sub pre-out back down to 0db (since it was maxed out at 12db for the gain structure process). Then I further adjusted the speaker-level for the sub down after measuring with REW since it seemed too high (now it is set at -3db). Set here, the sub pre-out now produces 4.76 volts when the AVR volume is set to +10db. So, supposedly this is where the pre-amp signal clips and also where the amp clips.

Should I assume that the signal driving the internal amps in the AVR (for the fronts and center) also clip here? I seem to get distortion in the fronts before reaching this point (+10db on the AVR) and I wonder if this is due to the AVR amp clipping or the speakers not being able to handle the power? Or maybe the amp itself does't produce clean power at this level?

Am I on the right track with all of this?

Noise (Buzz and Hum):

When I initially connected the sub, it exhibited a very disturbing hum. I assumed it was a ground loop problem and quickly discovered that it was largely from the cable box connected to the AVR. I checked the cable outside and found that the cable company literally installed a faux ground cable. It literally ran from the grounding block attached to the coax down a plastic conduit (for the incoming phone line) nearby and attached to....NOTHING. So I ran a new ground to the main electrical ground stake. That eliminated the loud hum, but it did not eliminate the noise entirely. As long as a signal cable is connected to the amp and the AVR is on, there is a hum that is fairly audible (definitely not acceptable). Should I be looking at installing an isolator or some other device? Jensen seems to make a few different products for dealing with this issue, but I'm fairly lost as to what the best approach is. Also, whatever I do will need to be compatible with the addition of future EQ (maybe miniDSP) for the sub.

Crossover:

I set my crossover to 80Hz thinking and it seemed to work OK, but I think I get better results via REW measurements when using 60Hz. I'm new to REW, but this is what I measured (via REW, Tascam 144, Cross Spectrum Labs calibrated mic):

Green Line is Crossed at 80Hz and the Brown/tan line is crossed at 60Hz (1/12 smoothing)


There is no EQ applied yet (that's a bridge I'll cross later), so the sub (being sealed) should roll off at about 30Hz. Given that, do my readings seem high between 10 and 30Hz? Could that be room gain, or am I making a mistake with REW? Which crossover point would you recommend considering that EQ will be added later?

If you've made it this far, sorry for the very long post! But I am very curious to hear some good feedback.
chargedmr2 is offline  
Sponsored Links
Advertisement
 
post #2 of 23 Old 03-10-2013, 11:12 PM
Member
 
Neilsy's Avatar
 
Join Date: Jan 2011
Location: Brisbane, Australia
Posts: 73
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Sorry I cannot give you advice on the gain structure and hum sections, perhaps for the crossover:


Have you run this graph with the sub and speakers set in phase with eachother?

As I can on your graph from the 80hz (green line) see there is a massive dip at 62hz approx with 80hz crossover... Who knows how many dB per octave the Denon rolls off at 80hz... perhaps your mains are still putting out a bit and you're experiencing some frequency cancellation. If you haven't set it already, try putting your mains at 80hz the sub as well, play an 80hz test tone and set the phase so you achieve max combined output of the mains & sub.
In my opinion the preference should be for the sub to do 80hz down - mains generally speaking don't achieve the impact of 60hz, 70hz that your sub can easily. The 80hz crossover will also give them a bit more headroom compared to 60hz.

- - CA 751BD - - CA 650R - - Emotiva XPA-5 - - Velodyne DD18 - - Dali Concept 8's, Center, 1's - - Klark Teknik DN360 GEQs - - Van Den Hul, Kordz & Emotiva interconnects - - LG 60" FHD Plasma - -
Neilsy is offline  
post #3 of 23 Old 03-10-2013, 11:59 PM
AVS Special Member
 
braveheart123's Avatar
 
Join Date: Jul 2011
Location: Lone Wolf McQuade from Islamabad Pakistan
Posts: 1,818
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 116
On the gain structure......that document from wayne mainly deals with full range signal and not specifically with sub and it does not take into account (I THINK) auto EQ systems like MCACC, Audyssey, YPAO, etc. The problem is when you are below reference and dynamic EQ is engaged, audyssey calculates how far you are from reference and accordingly boosts low frequencies to maintain the percieved loudness of bass. But when you go over reference volume (00 on master dial on avr), audyssey does not apply any compensation in my opinion. Coz you are already above reference and +10dB on master volume is way too loud and hence clipping/distortion on mains might be the result.

I am using Klipsch Reference speakers and 2 subs. One is a DIY running off crown xls1000. My avr is denon 3313, which produces 2V at reference volume (00 on the master dial). The problem is usually people including myself dont watch movies at refernce levels and are mostly at -10dBFS i.e. 75dB. So, you need to check your preout voltage at that level. If it is less than the required input voltage of your diy sub amp, you may require a signal booster.

I am using ART CleanBox Pro in my setup and haven't had any complaints since. But I would reckon to investigate this, should you decide to buy signal booster. I discussed this with Wayne and he concurred.
Just my opinion.

History is written by those who have hanged heroes ...

The best EQ is no EQ ...


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


Main System: Klipsch RF-82 II, Klipsch RC-62 II, RS-52 II, Onkyo 5010, Rythmik FV15HP, PSB S300
braveheart123 is offline  
post #4 of 23 Old 03-11-2013, 06:02 AM
AVS Special Member
 
Martycool007's Avatar
 
Join Date: Jun 2011
Posts: 3,594
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 220 Post(s)
Liked: 160
I would take your meter and test the subwoofer pre-out of your Denon AVR with the volume at reference, then also with the volume at what ever level you normally listen. If the sub pre-out is producing enough voltage at both volume levels on your Denon, then you should be good. If the voltage is less, then like braveheart123 said, you might require a signal booster. I have the same Denon AVR-3312 that you have, and am still experimenting with my Behringer EP4000 that I use for my subs, so what works for me might not work for you, as you are using a different amp than I am. Good luck!
Martycool007 is offline  
post #5 of 23 Old 03-11-2013, 07:10 AM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Neilsy View Post

Sorry I cannot give you advice on the gain structure and hum sections, perhaps for the crossover:


Have you run this graph with the sub and speakers set in phase with eachother?

As I can on your graph from the 80hz (green line) see there is a massive dip at 62hz approx with 80hz crossover... Who knows how many dB per octave the Denon rolls off at 80hz... perhaps your mains are still putting out a bit and you're experiencing some frequency cancellation. If you haven't set it already, try putting your mains at 80hz the sub as well, play an 80hz test tone and set the phase so you achieve max combined output of the mains & sub.
In my opinion the preference should be for the sub to do 80hz down - mains generally speaking don't achieve the impact of 60hz, 70hz that your sub can easily. The 80hz crossover will also give them a bit more headroom compared to 60hz.

Neilsy,

Thanks for the tip. I'm still in a major process of learning and I have not considered phase cancellation. I'll give this a shot later today or tomorrow and get back.
chargedmr2 is offline  
post #6 of 23 Old 03-11-2013, 07:23 AM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by braveheart123 View Post

On the gain structure......that document from wayne mainly deals with full range signal and not specifically with sub and it does not take into account (I THINK) auto EQ systems like MCACC, Audyssey, YPAO, etc. The problem is when you are below reference and dynamic EQ is engaged, audyssey calculates how far you are from reference and accordingly boosts low frequencies to maintain the percieved loudness of bass. But when you go over reference volume (00 on master dial on avr), audyssey does not apply any compensation in my opinion. Coz you are already above reference and +10dB on master volume is way too loud and hence clipping/distortion on mains might be the result.

I am using Klipsch Reference speakers and 2 subs. One is a DIY running off crown xls1000. My avr is denon 3313, which produces 2V at reference volume (00 on the master dial). The problem is usually people including myself dont watch movies at refernce levels and are mostly at -10dBFS i.e. 75dB. So, you need to check your preout voltage at that level. If it is less than the required input voltage of your diy sub amp, you may require a signal booster.

I am using ART CleanBox Pro in my setup and haven't had any complaints since. But I would reckon to investigate this, should you decide to buy signal booster. I discussed this with Wayne and he concurred.
Just my opinion.

I should have mentioned that I'm not using any EQ at this point. Everything Audyssey related is disabled. I plan to add EQ for the bass only via miniDSP or other alternative in the future and I will readjust gain structure when I do.

Also, I believe that my master volume puts out 75db when set to 0 (tested with built-in pink noise signal). What would be the correct way to verify this? Should it be set to 85db instead? Would I just raise the speaker levels for the mains until I reach this point?

Thanks for the tip on the CleanBox. If I add a miniDSP into the chain before the amp, where would this enter? After the miniDSP?
chargedmr2 is offline  
post #7 of 23 Old 03-11-2013, 07:31 AM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Martycool007 View Post

I would take your meter and test the subwoofer pre-out of your Denon AVR with the volume at reference, then also with the volume at what ever level you normally listen. If the sub pre-out is producing enough voltage at both volume levels on your Denon, then you should be good. If the voltage is less, then like braveheart123 said, you might require a signal booster. I have the same Denon AVR-3312 that you have, and am still experimenting with my Behringer EP4000 that I use for my subs, so what works for me might not work for you, as you are using a different amp than I am. Good luck!

Thanks for the advice. I'll measure and see what I get. Wouldn't this mean that in almost any setup, if you listen too quietly, there would be a problem? Let's say I'm watching TV and decide to listen at -20db on the master volume. Wouldn't that almost certainly produce too little voltage? If I did need a signal booster (which I hope to avoid), there would still be a point at which the volume could be too low, and hence not enough voltage for the amp. I guess what I am wondering is, what exactly does it mean to have too little voltage? Do amps have a minimum requirement that must be met to work? Obviously I'm a little confused on this issue.
chargedmr2 is offline  
post #8 of 23 Old 03-11-2013, 08:06 AM
AVS Special Member
 
braveheart123's Avatar
 
Join Date: Jul 2011
Location: Lone Wolf McQuade from Islamabad Pakistan
Posts: 1,818
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 116
Quote:
Do amps have a minimum requirement that must be met to work?

Yes. For example iNuke NU6000 DSP requires bare minimum 1.25V RMS (I guess) and Crown XLS1000 1.4V RMS (Um Sure) to be able to perform at their optimum and I have both these amps. I don't know the input sensitivty of your amp so I dont know if your signal is weak from AVR or not. In my case, I had to use signal booster.
Quote:
Wouldn't this mean that in almost any setup, if you listen too quietly, there would be a problem?

Yes and No. Some avrs (NAD for example) put out enough voltage to drive almost any pro amp. My denon 3313 doesn't on LOW volumes. I simply dont care if it puts out 2V RMS at 00 volume coz I am normally watching movies at -15dB blow ref during day and -35dB below ref max during late night. Very rare do I dial the volume to -10dB below ref. It is plenty loud and 0dB on master volume is absurdly loud. Now bare in mind this all is with Audyssey and DEQ engaged. If you are not using audyssey, u can pretty much set your master volume anywhere you want and adjust the channel trims with an SPL meter to whatever level you want (60/70/75/85dB). Now that is your ref max volume. The idea is to balance SPL from all speakers at your MLP. That could be any. Since movies are recorded at 85dB (-20dBFS) with 20dB peak,any signal hotter than that simply gets clipped.

History is written by those who have hanged heroes ...

The best EQ is no EQ ...


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


Main System: Klipsch RF-82 II, Klipsch RC-62 II, RS-52 II, Onkyo 5010, Rythmik FV15HP, PSB S300
braveheart123 is offline  
post #9 of 23 Old 03-11-2013, 08:26 AM
AVS Special Member
 
braveheart123's Avatar
 
Join Date: Jul 2011
Location: Lone Wolf McQuade from Islamabad Pakistan
Posts: 1,818
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 116
Quote:
Also, I believe that my master volume puts out 75db when set to 0 (tested with built-in pink noise signal). What would be the correct way to verify this? Should it be set to 85db instead?

AVR test tones are recorded at -30dBFS (75dB) and movies at -20dBFS (85dB). So SPL measured with avr test tones is same as the SPL in movies. You dont need to further check it IMO. Movies will yeild 85dB at 00 master volume. Also, SPL at 00 master volume is governed by channel trims, speaker sensitivity, and the distance of speakers from your MLP. For example, my RF82 II are rediculously efficient to reach max SPL with -6.5 on channel trim (00 master volume) at my MLP that is 11 feet away.
Quote:
If I add a miniDSP into the chain before the amp, where would this enter? After the miniDSP?

AVR > ART CleanBox Pro > MiniDSP > Power amp for sub ........ This is if the input sensitivity of MiniDSP is higher than the preout voltage of your avr.
Otherwise follow this path
AVR > MiniDSP > ART CleanBox Pro > Power Amp for sub

History is written by those who have hanged heroes ...

The best EQ is no EQ ...


To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.



To view links or images in signatures your post count must be 0 or greater. You currently have 0 posts.


Main System: Klipsch RF-82 II, Klipsch RC-62 II, RS-52 II, Onkyo 5010, Rythmik FV15HP, PSB S300
braveheart123 is offline  
post #10 of 23 Old 03-11-2013, 02:42 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by braveheart123 View Post

Yes. For example iNuke NU6000 DSP requires bare minimum 1.25V RMS (I guess) and Crown XLS1000 1.4V RMS (Um Sure) to be able to perform at their optimum and I have both these amps. I don't know the input sensitivty of your amp so I dont know if your signal is weak from AVR or not. In my case, I had to use signal booster.
Yes and No. Some avrs (NAD for example) put out enough voltage to drive almost any pro amp. My denon 3313 doesn't on LOW volumes. I simply dont care if it puts out 2V RMS at 00 volume coz I am normally watching movies at -15dB blow ref during day and -35dB below ref max during late night. Very rare do I dial the volume to -10dB below ref. It is plenty loud and 0dB on master volume is absurdly loud. Now bare in mind this all is with Audyssey and DEQ engaged. If you are not using audyssey, u can pretty much set your master volume anywhere you want and adjust the channel trims with an SPL meter to whatever level you want (60/70/75/85dB). Now that is your ref max volume. The idea is to balance SPL from all speakers at your MLP. That could be any. Since movies are recorded at 85dB (-20dBFS) with 20dB peak,any signal hotter than that simply gets clipped.

Ok, I went back and took some more measurements. First let me state that my amp lists a spec for sensitivity at +4dBV (which is confusing since I thought that is should read in dBu not dBV). If this is really supposed to be +4dBu (not dBV) then Wayne's article indicates this is equivalent to 1.228 Vrms.

So, on to the measurements. Once again, I played the reference level 60Hz tone. I found that at -10db volume (where we causally watch most movies), the sub pre-out was producing 0.49 volts. At 0db volume, 1.58 volts, and at +5db volume, 2.84 volts.

Keep in mind, however, that the sub speaker level is set to -3db and can therefore be increased quite a bit (to +12db). Of course, if I increase the speaker level to get more voltage (an earlier measure indicated that with max speaker level I can get 4.76 volts with the volume set to -5db), this will cause the sub to run too hot. And this is what is confusing me. Why would I want a minimum of say 1.228 volts (assuming this is actually the sensitivity of my amp) if it causes too much bass (which initially showed up in one of my REW measurements)? What am I missing here? Should I increase the voltage via the speaker levels (until I get 1.228Vrms at say -10db on the volume) and then attenuate that signal via the amp's gain (until I have the proper SPL from the sub)? If I do this, it throws a wrench in the whole grain structure process that I followed from Wayne's article.

Who knew this could be so confusing:eek:
chargedmr2 is offline  
post #11 of 23 Old 03-11-2013, 03:36 PM
Advanced Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 612
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 192
In my opinion, the easiest way to think about this is to start on the acoustical side, then work backward to the preamp sub output with a kind of flow chart like the following:
  • Does my subwoofer have enough acoustical output capability to meet my requirements? If so, I'm fine.
  • Does my power amplifier have a high enough maximum output power to drive the subwoofer to the levels it's capable of? If so, I'm fine.
  • Does my AVR's subwoofer out have a maximum output voltage that's greater than the power amplifier's sensitivity? If yes, this means the power amp will clip before the AVR output does, meaning that the AVR is not the limiting factor in the maximum sub output and I'm fine.
  • Do I have enough adjustment range on my AVR's sub out to allow me to adjust the sub so the REW frequency response measurement can show both a bit too much and a bit too little sub out relative to the mains? If yes, I'm fine.

If the amp's sensitivity is truly 4 dBV and not 4 dBu, then that represents 1.585 Volts RMS. It looks like you're getting higher voltage than this, so all should be well. One way of thinking about this is that the acoustic pressure in Pascal of the sub output at a fixed distance from the sub is proportional to the AVR's sub out voltage in Volts, assuming the sub is not going into compression and the power amp and AVR sub out are not clipping. For example, a 6 dB increase of the master volume should cause the voltage at the sub out to double, assuming it's not clipping. Also, the power amp output voltage will double in this scenario, assuming it's not clipping in either setting.
andyc56 is offline  
post #12 of 23 Old 03-11-2013, 10:09 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Neilsy View Post

Sorry I cannot give you advice on the gain structure and hum sections, perhaps for the crossover:


Have you run this graph with the sub and speakers set in phase with eachother?

As I can on your graph from the 80hz (green line) see there is a massive dip at 62hz approx with 80hz crossover... Who knows how many dB per octave the Denon rolls off at 80hz... perhaps your mains are still putting out a bit and you're experiencing some frequency cancellation. If you haven't set it already, try putting your mains at 80hz the sub as well, play an 80hz test tone and set the phase so you achieve max combined output of the mains & sub.
In my opinion the preference should be for the sub to do 80hz down - mains generally speaking don't achieve the impact of 60hz, 70hz that your sub can easily. The 80hz crossover will also give them a bit more headroom compared to 60hz.



Ok, I got around to checking the phase more closely.

First, I believe that the AVR has a 12/24db crossover. If I understand this correctly, that means the mains roll off 12db between 80Hz (the crossover point) and 40Hz?

I have a 60Hz tone that I've been using so I used this to check the phase with the crossover set to 60Hz and it made no difference in measured SPL. Then I set the crossover to 80Hz and played the 60Hz tone again (I know, I should have used an 80Hz tone, but I didn't have one handy). Anyhow, when I switched the phase this time (opposite of what it was set at for the graph in the first post), it reduced the measured SPL by 10db. So, I assume that the original phase setting was correct?? Might I find different results if I use an 80Hz tone? I'll try that later when I have time to get out the measuring gear and set up REW again.

If the phase was already set correctly, then I must have a null around 62Hz?
chargedmr2 is offline  
post #13 of 23 Old 03-12-2013, 01:48 PM
Advanced Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 612
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 192
If you need a sine wave generator, REW has that. Just click the "Generator" button on the toolbar.

It might also be a good idea to check if the 60 Hz hum is interfering with your measurement at 60 Hz. If you do a waterfall plot and the 60 Hz component is excessive, you'll find that the 60 Hz part of the waterfall, instead of decaying to no signal, will approach a constant-level signal as time increases. Make sure your cable TV stuff is disconnected for the measurements as well, just in case you haven't already done so.
andyc56 is offline  
post #14 of 23 Old 03-12-2013, 06:27 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by andyc56 View Post

If you need a sine wave generator, REW has that. Just click the "Generator" button on the toolbar.

It might also be a good idea to check if the 60 Hz hum is interfering with your measurement at 60 Hz. If you do a waterfall plot and the 60 Hz component is excessive, you'll find that the 60 Hz part of the waterfall, instead of decaying to no signal, will approach a constant-level signal as time increases. Make sure your cable TV stuff is disconnected for the measurements as well, just in case you haven't already done so.

Actually, I'm not sure the remaining noise that I'm getting (after correctly grounding my cable) is 60Hz hum or something else. When I play a 60Hz tone, it sounds distinctly different from the noise that I'm getting (the noise seems to be a distinctly higher frequency than 60Hz). Also, when I disconnect the cable, it doesn't take away the remaining noise. I suspect it has something to do with the fact that I'm running the amp on its own circuit (ground loop issues), but I'm not really sure. I'm going to look into getting a CleanBox or some other device to clear this up (hopefully).
chargedmr2 is offline  
post #15 of 23 Old 03-12-2013, 06:43 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by andyc56 View Post

In my opinion, the easiest way to think about this is to start on the acoustical side, then work backward to the preamp sub output with a kind of flow chart like the following:
  • Does my subwoofer have enough acoustical output capability to meet my requirements? If so, I'm fine.
  • Does my power amplifier have a high enough maximum output power to drive the subwoofer to the levels it's capable of? If so, I'm fine.
  • Does my AVR's subwoofer out have a maximum output voltage that's greater than the power amplifier's sensitivity? If yes, this means the power amp will clip before the AVR output does, meaning that the AVR is not the limiting factor in the maximum sub output and I'm fine.
  • Do I have enough adjustment range on my AVR's sub out to allow me to adjust the sub so the REW frequency response measurement can show both a bit too much and a bit too little sub out relative to the mains? If yes, I'm fine.

If the amp's sensitivity is truly 4 dBV and not 4 dBu, then that represents 1.585 Volts RMS. It looks like you're getting higher voltage than this, so all should be well. One way of thinking about this is that the acoustic pressure in Pascal of the sub output at a fixed distance from the sub is proportional to the AVR's sub out voltage in Volts, assuming the sub is not going into compression and the power amp and AVR sub out are not clipping. For example, a 6 dB increase of the master volume should cause the voltage at the sub out to double, assuming it's not clipping. Also, the power amp output voltage will double in this scenario, assuming it's not clipping in either setting.

Thanks for the info. This is a useful way of thinking about the gain structure issue and what is needed all around.

Let me ask some more questions about amp sensitivity. Is this spec (in my case, 1.585 Vrms or 4dBV) to be considered the minimum voltage to send the amp under any circumstance? From our discussions further up in the thread, this seems to be the case. As mentioned above, my AVR is sending a signal that is weaker than this whenever the volume is set to anything below 0db. Not surprisingly, the volume is not set above 0db very often, so this is a concern for me. Or is it? I'm not certain whether I need to boost this signal or not?

And I also have a question about how amps work. I don't know the correct terminology, but I'll do my best. Conceptually, I'm wondering if the following is correct: The signal from the pre-amp (AVR) is sent to the amp's input stage. The input stage requires a minimum voltage (e.g., 4dBV) to function properly. So long as the input stage receives the needed voltage, the amp's gain control can then attenuate the signal before sending it to the amplifier stage (output stage?).

Thanks for the feedback.
chargedmr2 is offline  
post #16 of 23 Old 03-12-2013, 07:58 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,327
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 113 Post(s)
Liked: 84
Quote:
Originally Posted by chargedmr2 View Post

Let me ask some more questions about amp sensitivity. Is this spec (in my case, 1.585 Vrms or 4dBV) to be considered the minimum voltage to send the amp under any circumstance?
No. An amplifier's sensitivity is the voltage at the input that will drive it to FULL output. So if your amp specifies 1.585Vrms, that's what it needs to reach full power.

Quote:
Originally Posted by chargedmr2 View Post

And I also have a question about how amps work. I don't know the correct terminology, but I'll do my best. Conceptually, I'm wondering if the following is correct: The signal from the pre-amp (AVR) is sent to the amp's input stage. The input stage requires a minimum voltage (e.g., 4dBV) to function properly. So long as the input stage receives the needed voltage, the amp's gain control can then attenuate the signal before sending it to the amplifier stage (output stage?).

Thanks for the feedback.
Nope. Let's start with a power amp with no volume control. It will have a fixed gain of somewhere between 25 and 40dB, but most fall in the range of 28-33dB which means that they will have an input sensitivity of 1-2V or so. The amp takes whatever input you give it, from less than a microvolt (1/millionth of a volt) up to beyond it's rated input sensitivity and amplify that by it's gain and apply that voltage at the speaker terminals. If this was your amp and you fed it more than 1.585V, it would clip.

The volume controls on the front of pro amps usually only attenuate; they cannot increase the signal you give it. This is because in a lot of live use they are in electrically noisy environments and signals sent between devices will often be at the highest level possible to reduce interference and increase the signal to noise. I regularly ran my PA at near 20V from the mixer, and reduced the level at the amps using the front panel attenuator by 20x to increase SNR, and get the level correct for the amplifier so it wouldn't clip.

A little more info. If your amp had 30dB (31.62x) of gain, inputting 1.585V would get you 50.12V at the speaker terminals. With an 8ohm speaker attached would see 314W delivered to the speaker.
P=V^2/R = 50.12^2/8 = 2512/8 = 314W.
In the same circumstance and a perfect amp, a 4ohm speaker would see 628W delivered to the speaker.

If you know the gain of your amp (should be in the datasheet or manual), then Gain = 10^(amp gain in dB/20).

HTH
A9X-308 is offline  
post #17 of 23 Old 03-12-2013, 07:58 PM
Member
 
SpeakerPower's Avatar
 
Join Date: Jan 2013
Posts: 33
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 29
AndyC56 seems to have a good handle on this. As long as the preamp can deliver enough voltage to clip the amp, then you are OK. Perfect gain staging is not that big a deal. I recommend you run the amp gain all the way up. If you have some residual noise you reduce it some, but don't go less than -10dB.

Your noise issue with a cable connection is very common. Any time you are mixing grounded and ungrounded equipment problems may arise. I have a link on my web site to information on this http://www.speakerpower.net/documents.html. The bottom line is: systems work well when either everything is floating (ungrounded equipment like your Denon) or everything is grounded (SpeakerPower amps and your cable connection). Once you introduce a grounded element, then you need to make sure everything is well grounded. You improved your ground by connecting the cable ground, but you probably need to go one more step and connect your Denon to the AC mains ground like you did the cable. This should reduce the currents still running in the signal cable grounds and creating the last bit of noise. The antennae ground may be handy for this purpose.

Brian Oppegaard
SpeakerPower
SpeakerPower is offline  
post #18 of 23 Old 03-12-2013, 08:20 PM
Advanced Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 612
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 192
Quote:
Originally Posted by chargedmr2 View Post

Let me ask some more questions about amp sensitivity. Is this spec (in my case, 1.585 Vrms or 4dBV) to be considered the minimum voltage to send the amp under any circumstance? From our discussions further up in the thread, this seems to be the case. As mentioned above, my AVR is sending a signal that is weaker than this whenever the volume is set to anything below 0db. Not surprisingly, the volume is not set above 0db very often, so this is a concern for me. Or is it? I'm not certain whether I need to boost this signal or not?

The amplifier sensitivity is defined as the RMS voltage (of a sine wave signal) applied to the amp's input required to drive the amp to its full rated output (sine wave) power. The amp will work just fine with input voltages lower than this! The only lower limit is noise, there is no sort of "threshold effect" or anything like that.
Quote:
Originally Posted by chargedmr2 View Post

And I also have a question about how amps work. I don't know the correct terminology, but I'll do my best. Conceptually, I'm wondering if the following is correct: The signal from the pre-amp (AVR) is sent to the amp's input stage. The input stage requires a minimum voltage (e.g., 4dBV) to function properly. So long as the input stage receives the needed voltage, the amp's gain control can then attenuate the signal before sending it to the amplifier stage (output stage?).

Power amplifiers don't work that way. If they did, they'd produce truly massive amounts of distortion. To understand how power amplifiers work conceptually, let's first talk about preamps.

A preamp's job is to boost and control the level of a signal applied to its input. It's a so-called "voltage amplifier". Its job is to make its output voltage Vout be a scaled version of the input voltage Vin. Let's call that scale factor A. So ideally, Vout = A * Vin. Now, the value of A is variable with the volume control setting. For old-style analog-only preamps, the value of A might be 10 or so, but A can go to almost 0 when the volume control is turned all the way down.

In order to do its thing, a preamp usually uses an IC called an op-amp. An op-amp can be configured for nearly any desired gain ("A") by adjusting the values of two external resistors. In doing its job, an op-amp-based preamp may boost the voltage, but it never puts out much current, maybe 10 milliamps maximum.

Now, a power amp is surprisingly similar. Its job too is to amplify its input voltage by a factor A, which is typically about 30. But there's one crucial difference between a power amp and a preamp. In order to do its job of amplifying its input voltage by 30 or so while being loaded down with just a few Ohms of impedance, it may be called upon to put out boatloads of current, having peaks of maybe tens of amps or so worst case. The topology of many power amps is just like an "op-amp on steroids". It's often a discrete-transistor design that nonetheless has a topology similar to an op-amp, except its DC voltage supplies are much higher, and its output stage is capable of putting out boatloads of current.

Edit: I see others have chimed in with similar replies.
andyc56 is offline  
post #19 of 23 Old 03-13-2013, 08:27 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Guys,

These last three posts have been extremely helpful for educating me on this issue. Thanks for taking the time to help me out with this.

Brian - thanks for the grounding tip. I wasn't able to run a ground from my receiver directly to the main grounding stake, but I did run a ground wire from the receiver to the ground of the 20amp outlet that the amp is plugged into. This has reduced my noise issue from intolerable to tolerable. Actually, the hum is almost gone, but not quite.

I also noticed something else about the buz/hum sound: It fluctuates in intensity sometimes rather than remaining at a constant level...and as I'm writing this, I've just figured out the fluctuation is caused by running the washing machine (which is on another circuit, obviously). Hmmm? Is this strange? Can I solve this by using one of the products from Jensen (or some alternative)? Or here's another idea...what if I were to unground the amp? Would that be dangerous (or just stupid)? **Edit to add: I did a little more research (Thanks Brian...Jensen's website has some great info on all of this stuff) and now see that lifting the ground on the amp would be both dangerous and stupid, as I suspected.

And back to the amp gain issue. I should mention that I am not experiencing any problems at all and the amp is working perfectly. But I'm still curious to understand more about this issue of gain, even though I probably don't need to make any changes to my system at this point. So let me prod just a little more if you guys are willing. If I feed the amp 1.58 Vrms (the sensitivity rating of the amp), it should produce it's full potential (2400 watts RMS into 4 ohms). Does this mean the amp should start to clip at this point or just after (i.e., the clip light should start to go off)? The reason that I ask is because when I first set the gain structure I turned the amps gain all the way down. Then I fed the amp a 4.76 Vrms signal (by playing pink noise recorded at 0db) with the sub disconnected. I slowly turned up the gain on the amp until the clip light started to light. The light did not become lit fairly constantly until the amp gain was all the way up. I took this to mean the amp could accept a 4.76 Vrms signal, but this is well beyond the sensitivity figure of the amp (1.58Vrms). What am I missing this time around??

Edit to add: it just occured to me after re-reading the post above that amp sensitivity is measured via a sine wave, but I played pink noise. Maybe this is part of my confusion?
chargedmr2 is offline  
post #20 of 23 Old 03-14-2013, 10:31 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Just as a follow-up to my above post, I've narrowed the hum down to the cable, just as I gather most people do in these cases. Grounding the cable helped substantially and grounding the AVR to the same outlet (and circuit) as the amp helped further. But, I still have hum remaining and when I disconnect the cable from the set top box it disappears entirely. So, I will be purchasing a Jensen VRD-1FF to isolate the cable. Hopefully that does it.

Anyone willing to entertain my gain related question at the end of the above post?
chargedmr2 is offline  
post #21 of 23 Old 03-15-2013, 10:04 AM
Advanced Member
 
andyc56's Avatar
 
Join Date: Feb 2013
Posts: 612
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 192
Quote:
Originally Posted by chargedmr2 View Post

If I feed the amp 1.58 Vrms (the sensitivity rating of the amp), it should produce it's full potential (2400 watts RMS into 4 ohms). Does this mean the amp should start to clip at this point or just after (i.e., the clip light should start to go off)? The reason that I ask is because when I first set the gain structure I turned the amps gain all the way down. Then I fed the amp a 4.76 Vrms signal (by playing pink noise recorded at 0db) with the sub disconnected. I slowly turned up the gain on the amp until the clip light started to light. The light did not become lit fairly constantly until the amp gain was all the way up. I took this to mean the amp could accept a 4.76 Vrms signal, but this is well beyond the sensitivity figure of the amp (1.58Vrms). What am I missing this time around??

Edit to add: it just occured to me after re-reading the post above that amp sensitivity is measured via a sine wave, but I played pink noise. Maybe this is part of my confusion?

I believe your added sentence has hit on the answer. Clipping indicators may vary somewhat from amp to amp, so I'm not sure how to answer the question about the exact circumstances under which the indicator will light up. But these indicators work on the peak value of the signal. For a sine wave, determining the peak value is straightforward: it's the RMS value (what you read on a voltmeter) times the square root of two. Determining the peak value of noise is a different matter though, as noise is a random signal, so it is best understood by statistics. For zero-mean Gaussian noise, its RMS value is its statistical variance. Its peak value is sometimes stated as three times the RMS value, but that is with a probability of 99.7 percent. There will be peaks higher than this, just not often. So the peak value of noise is not well-defined. White noise is Gaussian with zero mean, but I'm not sure about the distribution of pink noise. The concept is the same though: one must speak of the probability that the instantaneous noise voltage will exceed some value, rather than the deterministic peak value of a sine wave.

To add a bit more confusion, your typical AC voltmeter won't read the RMS value of noise correctly. They are calibrated to read the correct RMS value for a sine wave only. To get correct RMS noise measurements, you'd need a so-called "true RMS meter".

The gain adjustment of amps is almost always just a potentiometer at the input which just reduces the signal level before it hits the amplifier proper. The sensitivity should be specified with the gain cranked all the way up. So if you turn down the gain by, say, 6 dB (that's half voltage), the amp will require twice the input voltage to drive it to full power than it did with the gain turned all the way up.
andyc56 is offline  
post #22 of 23 Old 03-15-2013, 11:53 AM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Got it. Thanks Andy. This is all very helpful and interesting info. I'm really enjoying picking up some basic knowledge about the electronics that I'm using daily. Lots left to go, of course!
chargedmr2 is offline  
post #23 of 23 Old 03-22-2013, 08:48 PM - Thread Starter
Member
 
chargedmr2's Avatar
 
Join Date: Jul 2008
Posts: 93
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Just in case anyone is interested, I received a Jensen VRD-1FF today to address the remaining ground loop hum caused by the cable. It cost about $50 and worked perfectly (easily installs in the cable line just before the cable box). Finally, my noise floor is where it should be.
chargedmr2 is offline  
Reply DIY Speakers and Subs

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off