A Sub Just Shouldn't Make Certain Noises

[fuzz092888]

So I built this subwoofer using a Pierce Audio driver. My signal path is Onkyo 3009 -> Antimode 2.0 dual core -> Crown XLS 2500. To the problem. Whenever I have things hooked up like this, I get this very unpleasant almost distortion type noise in which the driver stops what it was doing and basically screeches at me. However, when I remove the antimode from the signal path I get no problem. I can run direct from the Onkyo, or my computer and just get teeth rattling bass, but for some reason the antimode causes problems when it's supposed to fix them.

Now I could easily chalk this up to the antimode being defective, but it has done an excellent job with my Outlaw LFM-1 EX. I don't get any of these problems and I've tried cable swaps, input output swaps and every time the Outlaw sounds fine and the new stuff goes bananas.

Now I have this CD that plays sweeps in 10hz increments and each increment takes about a minute and a half. The sub will play whatever tone is being outputted, but only up to a certain notch on the gain knob. One notch up and it starts. One notch back the other way and it stops and starts playing the tone again.

Any ideas? Am I missing something obvious?

 

[Adam]

Adam's simplistic approach.

"It hurts when I do this."

"Don't do that."

Take out the antimode.

-----------------

On a more serious note - is the antimode actively monitoring the sound all of the time (I'm not familiar with it, sorry)? If not, and it's just using an EQ curve that got established when you calibrated the system, then that's curious for sure.

 

[fuzz092888]

Adam's simplistic approach.

"It hurts when I do this."

"Don't do that."

Take out the antimode.

-----------------

On a more serious note - is the antimode actively monitoring the sound all of the time (I'm not familiar with it, sorry)? If not, and it's just using an EQ curve that got established when you calibrated the system, then that's curious for sure.

But but but it makes the FR response so much flatter!!!

No it is not an active monitoring system. It runs a bunch of sweeps, can do multi-point like audyssey , but then sticks with the information it got from that. It also does an infrasonic filter, PEQ filters, and a bunch of other stuff. I only have the EQ run and the infrasonic filter on. I wonder if running XLR to XLR would make anything better instead of running RCA to RCA?

 

[agarwalro]

You may be overdriving the amp on the gain setting that screeches. Have you tried the Anti Mode in the higher sensitivity setting, aka, lower output voltage?

 

[fuzz092888]

You may be overdriving the amp on the gain setting that screeches. Have you tried the Anti Mode in the higher sensitivity setting, aka, lower output voltage?

I've tried the antimode in both settings and it does it on both. I've also driven the driver and amp straight from the receiver and from my computer and cranked the gain setting and the driver and amp don't flinch. Really weird.

 

[slipperybidness]

I've tried the antimode in both settings and it does it on both. I've also driven the driver and amp straight from the receiver and from my computer and cranked the gain setting and the driver and amp don't flinch. Really weird.

I would try to contact the anti-mode vendor and get their input.

 

[TLS Guy]

I would try to contact the anti-mode vendor and get their input.

I suspect that the antimode is passing some suprasonic garbage at the output. It is common for digital devices to do that, and there should by a low pass filter above the audio band to stop it. Class D amps are easily sent into instability by this sort of thing.

Do you have an amp that is not class D to try instead of the Crown? Or else you could do something crude, and just put 0.1 mfd caps across the outputs of the antimode. Since you are using it for a sub, it would not affect performance. So if that cures it leave it.

I do not see much other explanation other than this to account for your problem. It would be very interesting to scope the output of the antimode and look at the squeal it generates in the Crown.

 

[fuzz092888]

I suspect that the antimode is passing some suprasonic garbage at the output. It is common for digital devices to do that, and there should by a low pass filter above the audio band to stop it. Class D amps are easily sent into instability by this sort of thing.

Do you have an amp that is not class D to try instead of the Crown? Or else you could do something crude, and just put 0.1 mfd caps across the outputs of the antimode. Since you are using it for a sub, it would not affect performance. So if that cures it leave it.

I do not see much other explanation other than this to account for your problem. It would be very interesting to scope the output of the antimode and look at the squeal it generates in the Crown.

Yes, I have tried a class A/B pro amp and it makes the same noise. Since I tried a class A/B amp and it still made the noise, should I then not bother with the caps?

I'm not sure I described what is happening all that well either. Basically, when the 30-39hz sweep is playing, the driver is moving in and out as it should. As soon as I pass some unknown volume point, usually pretty low, the driver ceases it's motion and emits a much higher pitched tone. I haven't left it going like that for very long since I have no idea what kind of damage it may do, but I've quickly put my head into the center of the cone to see if it was coming from a specific point, but I didn't gather much from doing that.

I also sent an e-mail to DSPeaker to see if they have any insight as well. Right now I have the subwoofer preout routed directly to the amplifier and everything is fine. It definitely could benefit from a bit of EQ, but it will go just about as loud as I can stand to have it with no problems or odd noises. Meanwhile the Outlaw subwoofer is still connected to the antimode and is doing just fine as it always has.

I'm befuddled.

 

[KEW]

Is the screech a single pitch? If so, what is the frequency?
(I am asking out of simple curiosity, not because I think it will help find a solution)

 

[fuzz092888]

Is the screech a single pitch? If so, what is the frequency?
(I am asking out of simple curiosity, not because I think it will help find a solution)

Not really sure. I haven't broken out the measurement equipment to see. Although that's piqued my curiosity so I may just give it a listen tomorrow.

 

[TLS Guy]

Yes, I have tried a class A/B pro amp and it makes the same noise. Since I tried a class A/B amp and it still made the noise, should I then not bother with the caps?

I'm not sure I described what is happening all that well either. Basically, when the 30-39hz sweep is playing, the driver is moving in and out as it should. As soon as I pass some unknown volume point, usually pretty low, the driver ceases it's motion and emits a much higher pitched tone. I haven't left it going like that for very long since I have no idea what kind of damage it may do, but I've quickly put my head into the center of the cone to see if it was coming from a specific point, but I didn't gather much from doing that.

I also sent an e-mail to DSPeaker to see if they have any insight as well. Right now I have the subwoofer preout routed directly to the amplifier and everything is fine. It definitely could benefit from a bit of EQ, but it will go just about as loud as I can stand to have it with no problems or odd noises. Meanwhile the Outlaw subwoofer is still connected to the antimode and is doing just fine as it always has.

I'm befuddled.

In that case the problem has to be a voltage mismatch. Obviously the problem lies entirely with the antimode, which has insufficient head room to drive the Crown, with your current settings.

This is all explained on pages 13 and 14 of the antimode manual.

You can tell if there is saturation of the antimode input, output or both.

This is how you tell if the input is saturated.

4.3. Input Level Warning
If the input signal level in the Anti-Mode 8033's Analog to Digital Converter (ADC) is near the
maximum input range, the user is warned in two phases. If the input signal level has only 3 dB of
headroom remaining, the BYPASS LED will start flickering. If the level increases from this, the
signal starts to saturate at the ADC input and harmonic distortion increases. If the input level
reaches maximum, the LIFT35 LED starts flashing. When the input level is reduced the LEDs
resume normal operation.
The input level warning is useful in optimizing the signal level that goes into the Anti-Mode. If the
volume of your subwoofer is too high, the AVR's calibration may set unnecessarily low levels for
the subwoofer output. This leaves much of the Anti-Mode's available input range unused and can
cause an audible low level hum or noise from the subwoofer. This may be the case if your AVR
indicates its subwoofer output level is below -3dB.
To take advantage of the full input range of the Anti-Mode, decrease the subwoofer volume and
increase the output level from the pre-amplifier or AVR until you see only occasional 3dB warnings,
then lower the volume a few dB.

This is how you tell if the output is saturated.

4.4. Output Level Warning
The new Anti-Mode correction will raise the overall level of the output depending on how drastic
corrections were applied. This increase and an active lift mode can increase the output level in
some frequencies so much that the signal gets saturated in the output. In this case the LIFT25
LED starts to flicker, and you should decrease the Anti-Mode 8033 input level or turn off lift mode.

If it is the input that is saturated, then you need to turn down the input from your AVR (LFE output), and advance the input controls on your Crown.

If the output is saturating, then you need to advance the volume controls of the Crown and turn down the LFE output.

There really is no other explanation for your problem, and your antimode device, has the instrumentation built in to help you sort this out. I suspect you have too high an input level, as it sounds as if the antimode processor is running out of bits. What you describe is what it sounds like when a processor runs out of bits.

 

[fuzz092888]

In that case the problem has to be a voltage mismatch. Obviously the problem lies entirely with the antimode, which has insufficient head room to drive the Crown, with your current settings.

This is all explained on pages 13 and 14 of the antimode manual.

You can tell if there is saturation of the antimode input, output or both.

This is how you tell if the input is saturated.



This is how you tell if the output is saturated.



If it is the input that is saturated, then you need to turn down the input from your AVR (LFE output), and advance the input controls on your Crown.

If the output is saturating, then you need to advance the volume controls of the Crown and turn down the LFE output.

There really is no other explanation for your problem, and your antimode device, has the instrumentation built in to help you sort this out. I suspect you have too high an input level, as it sounds as if the antimode processor is running out of bits. What you describe is what it sounds like when a processor runs out of bits.

What you're saying is probably true, but I don't have the 8033, I have the dual core 2.0. I've looked through the dual core 2.0 manual and couldn't find an adequate explanation for my problem, maybe you'll have better luck if you have the time. What you're saying makes perfect sense, I'm just not sure how it translates from that unit to my unit. I know you can adjust the digital headroom with the one I have and choose an input sensitivity. Do I need to increase the digital headroom? Do I still need to decrease the sub level on the receiver?

Thanks for all the help.

Oh and here's the manual for the one I have if you get a chance. I think the menus changed slightly because I just installed the latest firmware, but it should be pretty similar.

http://www.dspeaker.com/fileadmin/datasheets/dspeaker/AntiMode20DualCoreEng.pdf

 

[Bryceo]

I've got it take the antimod out and perfect

 

[fuzz092888]

I've got it take the antimod out and perfect

Not exactly.

 

[Bryceo]

Not exactly.

Well don't then have equipment that's angrey at you

 

[TLS Guy]

What you're saying is probably true, but I don't have the 8033, I have the dual core 2.0. I've looked through the dual core 2.0 manual and couldn't find an adequate explanation for my problem, maybe you'll have better luck if you have the time. What you're saying makes perfect sense, I'm just not sure how it translates from that unit to my unit. I know you can adjust the digital headroom with the one I have and choose an input sensitivity. Do I need to increase the digital headroom? Do I still need to decrease the sub level on the receiver?

Thanks for all the help.

Oh and here's the manual for the one I have if you get a chance. I think the menus changed slightly because I just installed the latest firmware, but it should be pretty similar.

http://www.dspeaker.com/fileadmin/datasheets/dspeaker/AntiMode20DualCoreEng.pdf

Thanks for the correct manual.

This unit does not have the, what I would regard as the essential monitoring, the other unit has.

Your problem is common with all equalizers and devastating with digital ones as you are finding out.

These issues are addressed in the manual in several places. This will be a hard climb but not impossible without instruments. However a dual channel scope, signal generator and FET VOM would be very very helpful.

When you are into complex more reference type systems like you are, a substantial investment and education in using test equipment, is in my view part of the basics. I have said this many times before, but it falls on deaf ears.

Here are the relevant parts of your manual.

Use the manual channel balance adjustment to match the levels of the left and right
speakers. Changes in this dialog take effect immediately. Press the OK button on the
remote to accept the new adjustment or press the BACK button to discard changes.
All internal processing is performed with 32- or 40-bit accuracy with the appropriate
number of bits reserved for headroom. However, if emphasizing (amplifying) filters are
used, some input signals could cause saturation / clipping at the output. The unit
automatically calculates the best headroom setting from the gains of the active filters.
However, if you only use the analog input and do not use the full input range, you can
change the headroom setting to a lower value. If the automatically calculated value is
too small (causing clipping at output), you can manually increase the headroom value.
The headroom value is automatically re-calculated any time you calibrate or change
filter settings, so remember to re-do your own adjustment in these cases.
When performing calibration in the advanced mode, you can select different dip
compensation modes: normal, maximum, or no dip compensation. Using the
Compensation adjustment you can manually increase or decrease the lowpass shelve
filter effect that was automatically determined by the calibration. Note that this
adjustment does not affect the before / after measurement display, but you can
perform a user measurement to determine the best value.
Notice: channel balance, headroom, and dip compensation values are associated with
the current sound profile.

The Volume trim menu allows compensation of different volume levels between the
digital and analog inputs. For example, if the digital input is subjectively louder, adjust
the volume trim value towards the left side until the volume levels are equal. Changes
in this dialog take effect immediately. Press the OK button on the remote to accept the
new adjustment or press the BACK button to discard changes.
The Input sensitivity setting adjusts the maximum allowed voltage level of the analog
input. Default value is high sensitivity. Use the low sensitivity setting, if you connect a
device which produces very high voltage levels and you get clipping warnings.
Notice: Input sensitivity and volume trim are global settings, i.e. the values are shared
by all four sound profiles.

6.Advanced Calibration
Advanced calibration gives the user more control over the calibration parameters. The
procedure is similar to the typical calibration, but with some additional steps. Refer to
”Typical Calibration” section of this manual for an overview of the basic process.
To perform the advanced calibration, choose the Advanced Calibration option at the
Calibration menu.
➢ After choosing the relevant speaker configuration, the extra settings dialog will
appear.
Level compensation affects compensation for wide dips in the bass response.
Compensation has three levels, “Off”, “Norm” and “Max”. If the setting is “Off”,
no compensation to boost dips is applied, and the bass level remains as low as
it would be without room effect. This gives maximum headroom for the user to
design custom boosting filters, such as House Curve or Parametric EQs. The
“norm” setting applies a mild compensation, which is the same amount that
“Typical” calibration is using. This setting usually gives the best audible result as
the perceived level of the bass is closer to the uncorrected average bass level
than with the uncompensated “Off” setting. When selecting “Max”, maximum
compensation is used, which gives about +2dB more gain for boosting filters.
This setting is beneficial in spaces where part of the problem is the general lack
of bass, possibly resulted by the absence of proper acoustic barriers. It is good
to remember that destructive reflection nulls can not, will not, and should not be
attempted to be corrected by any DSP.
Note: you can also adjust the compensation level afterwards from the
Compensation setting in the Audio Settings menu.

This is the most important part off all.

”Clipping Detected”: this warning is displayed on the home screen in two cases.
1.”Clipping Detected” is shown when the Anti-Mode 2.0 Dual Core detects that the
analog input signal has reached the maximum input voltage level, causing the
input to be clipped.
Check the output settings of the device you have connected to the Anti-Mode 2.0
Dual Core analog inputs. If you can not reduce its output level, set the input
sensitivity to low in the ”Input Settings” menu. If this does not help, you can try
using the RCA inputs instead of XLR inputs, or use external attenuators.
2.”Clipping Detected” is also shown when there is a possibility of clipping at the
output stage, which could compromise audio quality.
Usually the reason for this warning is that too many boosting equalizers are being
employed. To overcome this, increase the digital headroom value located in the
”Audio Settings” menu. Sometimes also decreasing the input gain may help.

In addition I would make the following observations. Digital clipping is much more serious than analog clipping. From your description you have digital clipping. Whether the monitoring in the antimode detects this I don't know. Most likely your screech is the only warning.

The specs do not say the input level for a specified output on pass through. (No filters engaged). Even worse it does not specify input headroom. Nor does it specify the output headroom.

Equalizers present special problems as you don't have a situation were x input voltage gives y output voltage. The reason as that there are bands of boost. So if an input signal of a volt is boosted between say 40 and 50 Hz then the output in that pass band will be 2 volts even for a zero gain device. If it is boosted 10 db then the output in the pass band will be 10 volts and cause trouble not only in the equalizer, but likely will run the amp input stage out of headroom let alone the power stage.

So here is what I would do. I would advance the controls on your Crown fully, max it out. In that situation 1.4 volts at the input will produce maximum output of you Crown amp. Now lower the LFE output to balance out the input. Note that you can not do your calibration at those settings.

Do the calibration with the Crown controls in the mid position. Then advance them all the way, lower the LFE output to the right range by ear or spl meter and then do the receiver calibration.

If you still have trouble, you will have to interrogate the antimode and look for areas where it has applied a lot of boost and manually reduce the degree of boost. As stated in your manual you may have to build external attenuators. If you get to this point instrumentation is strongly advised.

I could not set up my system without this type of instrumentation. Use of it is a big reason, my rig sounds so clean and effortless.

If results are not satisfactory, then you need the instrumentation I described above or just ditch the antimode. Despite glowing reviews, this unit may well be designed with inadequate head room and my be a less than satisfactory unit.

 

[fuzz092888]

Thanks for the correct manual.

This unit does not have the, what I would regard as the essential monitoring, the other unit has.

Your problem is common with all equalizers and devastating with digital ones as you are finding out.

These issues are addressed in the manual in several places. This will be a hard climb but not impossible without instruments. However a dual channel scope, signal generator and FET VOM would be very very helpful.

When you are into complex more reference type systems like you are, a substantial investment and education in using test equipment, is in my view part of the basics. I have said this many times before, but it falls on deaf ears.

Here are the relevant parts of your manual.









In addition I would make the following observations. Digital clipping is much more serious than analog clipping. From your description you have digital clipping. Whether the monitoring in the antimode detects this I don't know. Most likely your screech is the only warning.

The specs do not say the input level for a specified output on pass through. (No filters engaged). Even worse it does not specify input headroom. Nor does it specify the output headroom.

Equalizers present special problems as you don't have a situation were x input voltage gives y output voltage. The reason as that there are bands of boost. So if an input signal of a volt is boosted between say 40 and 50 Hz then the output in that pass band will be 2 volts even for a zero gain device. If it is boosted 10 db then the output in the pass band will be 10 volts and cause trouble not only in the equalizer, but likely will run the amp input stage out of headroom let alone the power stage.

So here is what I would do. I would advance the controls on your Crown fully, max it out. In that situation 1.4 volts at the input will produce maximum output of you Crown amp. Now lower the LFE output to balance out the input. Note that you can not do your calibration at those settings.

Do the calibration with the Crown controls in the mid position. Then advance them all the way, lower the LFE output to the right range by ear or spl meter and then do the receiver calibration.

If you still have trouble, you will have to interrogate the antimode and look for areas where it has applied a lot of boost and manually reduce the degree of boost. As stated in your manual you may have to build external attenuators. If you get to this point instrumentation is strongly advised.

I could not set up my system without this type of instrumentation. Use of it is a big reason, my rig sounds so clean and effortless.

If results are not satisfactory, then you need the instrumentation I described above or just ditch the antimode. Despite glowing reviews, this unit may well be designed with inadequate head room and my be a less than satisfactory unit.

Thanks again for taking the time to look through those manuals to find an answer.

Ok, so what I tried before your response, was to decrease the volume on the antimode to -20 while also maxing out the digital headroom to 20db. After I did that, I put the gain on the crown on max, then went into the receiver level calibration and adjusted the level with an SPL meter until it read about 75db at the LP. Popped in the sweep disc and got the same issue right around -14 on the receiver.

I'll definitely give your suggestion a try today.

Antimode e-mailed me back and just asked for some basic information.

 

[TLS Guy]

Thanks again for taking the time to look through those manuals to find an answer.

Ok, so what I tried before your response, was to decrease the volume on the antimode to -20 while also maxing out the digital headroom to 20db. After I did that, I put the gain on the crown on max, then went into the receiver level calibration and adjusted the level with an SPL meter until it read about 75db at the LP. Popped in the sweep disc and got the same issue right around -14 on the receiver.

I'll definitely give your suggestion a try today.

Antimode e-mailed me back and just asked for some basic information.

Well we have quite a bit of information now. I think the key points you have made is that the outlaw works OK, and a different amp did not solve the problem when using your new sealed sub. In addition the problem is in the 30 to 40 Hz pass band. Do you have the same problem in the lower pass band?

I'm guessing your Outlaw sub is ported.

I'm starting to feel that your antimode does not have the headroom to compensate for the 12db per octave roll off of your sealed sub.

I would maximise the digital headroom. I suspect this antimode unit is not designed to equalize the roll off a sealed sub. In that case you will need to equalize this second order roll off with an equalizer between the antimode and your Crown. If you were using an O-Amp then this would effectively be the situation, but your Crown has no equalizer.

I bet your antimode is applying 3 db boost from 60 to 40 Hz and the 12 db per octave boost after this and runs out of headroom. This would explain the problem starting below 40 Hz.

From what I can tell your unit is a room correction device and not an equalizer for equalizing sealed subs.

So if your crown amp has a voltage of .1 volt at 60 hz then the voltage will need to be 0.2 volts at 40 Hz at 30 Hz it will need 0.8 Volts. If the input voltage were 1 volt at 60 Hz, then the voltage at 30 Hz would be 8 volts. If you have no subsonic filter engaged, then the voltages at 20 Hz would be 3.2 and 32 volts respectively.

So you can see how much headroom is required. I seriously doubt a digital equalizer would have that much headroom. Now your max input voltage to the Crown, with the volume knobs fully advanced is 1.4 volts. So without subsonic filter, the maximum input voltage to your antimode from your receiver is about 0.05 volts. This would be around - 20 db at your LFE output.

So if your LFE output were set -5 db below the mains, then you would run into trouble at levels above -15 db on your receiver.

So this is why setting gains between units is so crucial.

You one other option is to manually defeat the sealed sub equalization, by setting the antimode manually.

I'm pretty sure these are the rocks you have fetched up on.

 

[fuzz092888]

Thanks again TLS Guy, for all your help. I sent few of your responses along with some additional information to antimode, but I'm still waiting back for a response. It looks like I've got a fair amount of work ahead of me whether the antimode stays or not. Right now it looks as though it may be heading out the door unless antimode has some insight that we're all missing.

I may trying setting just running the room response sweep and setting the antimode manually to see if that yields anything.

Looks like I'm going to need that education and investment in test equipment as well

 

[TLS Guy]

Thanks again TLS Guy, for all your help. I sent few of your responses along with some additional information to antimode, but I'm still waiting back for a response. It looks like I've got a fair amount of work ahead of me whether the antimode stays or not. Right now it looks as though it may be heading out the door unless antimode has some insight that we're all missing.

I may trying setting just running the room response sweep and setting the antimode manually to see if that yields anything.

Looks like I'm going to need that education and investment in test equipment as well

They may have some insights, as they have the circuits. They don't even have a block diagram of how the thing is organized. So that makes it easier for them as we are somewhat in the dark.

One of the things that distresses me these days is that user manuals are much poorer than they used to be. This is a pity as the equipment is more complex. The general assumption is that the user us too dumb to understand a decent manual, including circuits and a parts list. They may well be right, but that is a dreadful take on prevailing engineering and scientific general knowledge. All decent gear used to come with a circuit and description of operation and it still should.

You may well have some work to do, but that should not surprise you. There is no off the shelf solution to a really excellent system. For that varying degrees of customization are required. There is always some to do. The more complex the system the more there is to do.