LT with gain-dependent boost limiter - a new idea? - Page 2 - diyAudio
Go Back   Home > Forums > Loudspeakers > Subwoofers

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 2nd February 2013, 08:10 PM   #11
diyAudio Member
 
Join Date: Mar 2007
Location: California
Quote:
Originally Posted by weltersys View Post
Charlie,

Sounds like what can be done with the dynamic EQ in the Behringer Ultradrive Pro DCX2496.
"The DYNAMIC EQ influences a specific frequency range of the signal, depending on the volume level. It can either boost or cut this frequency range, depending on the gain setting made by the user. If the gain control is set to lower the respective frequency range in level, and if a preset THRESHOLD is exceeded, the EQ modifies the sound by reducing the gain of the frequency range. The amount of gain reduction applied is defined by the RATIO value. When the signal drops below threshold again, the frequency range is “smoothed out” again, i.e. the DYNAMIC EQ stops processing the signal."

Unlike the Bag End device which is fixed for a single application, the DCX2496 can be adjusted for any desired boost/cut frequencies and thresholds.

Art
I have owned a couple of DCX2496 units but never made use of this feature. One problem is that the closest function is the second order shelf, but that is not as flexible in the shapes that it can take on compared to the various shapes of a closed box speaker. You could add on additional EQ bands to mimic the response around resonance of the speaker (e.g. if peaking). You are correct in that the DCX would be adjustable and my circuit can be built in different ways but is not adjustable afterwards.

In any case, while there might be a $300 solution I think that a couple of analog circuit boards can do the same thing for much less. And it makes for something that you can built into the speaker (for a subwoofer or the woofer in a 3-way, for instance) and I personally build speakers that way, with amps and electronics built in so it works nicely for my needs.

-Charlie
  Reply With Quote
Old 2nd February 2013, 08:17 PM   #12
diyAudio Member
 
Join Date: Mar 2007
Location: California
Attached is another plot that shows what is going on with my "boost limiting" circuit. The input was a dynamic burst at 10Hz, with various input levels.

The plot shows the peak output voltage and the corresponding gain (Vout/Vin) for the 10Hz signal, in decibels. At the lowest input voltage, the full gain is available. As input voltage increases so that the output rises to 1.5V the gain is dialed back so that the peak output voltage does not exceed the 1.5V limit/threshold. The 1.5V value is just an arbitrary value I picked and is adjustable via a pot. This kind of response to varying input levels is useful for not driving the amp into clipping, which would result in lots of audible harmonics and other distortion.

Will follow up with a plot of the variable gain versus frequency later...

-Charlie
Attached Images
File Type: jpg 10Hz_boost_limiting_to_1.5Vout_max.jpg (50.9 KB, 70 views)
  Reply With Quote
Old 2nd February 2013, 09:54 PM   #13
diyAudio Member
 
Join Date: Mar 2007
Location: California
Attached is a plot of the output voltage as a function of frequency for a variety of input levels. As before, this data was obtained on a 10Hz burst input signal, the max level of which was varied. This more clearly shows the change in boost.

-Charlie
Attached Images
File Type: jpg 10Hz_voltage_limiting_to_1.5Vout_max.jpg (83.8 KB, 60 views)
  Reply With Quote
Old 8th February 2013, 09:40 PM   #14
diyAudio Member
 
Join Date: Mar 2007
Location: California
A little update on this effort:

I've been doing several iterations on the circuitry that monitors the output level and controls the feedback signal that reduces the boost. I've finally come up with something that has variable controls for:
  • output voltage limit
  • feedback gain
  • attack rate of boost limiter
  • decay rate of boost limiter
This iteration of the circuitry seems to deliver pretty respectable performance in terms of its ability to limit the bass boost while not introducing much distortion.

There are some tradeoffs in this kind of design approach - the attack rate, decay rate, and feedback gain are somewhat interrelated. Attack and decay rates are related to the low frequency limit of operation and so are dependent on the particular application. When operating the circuit with high feedback gain you get very good control of the output voltage and it will not exceed the voltage limit setpoint but for the lowest frequencies there is some pumping that can be observed when the signal period starts approaching the decay time constant. If you reduce the feedback gain the pumping goes away, and if you continue to decrease the gain the limiting action becomes "softer". By making the feedback gain adjustable, the tradeoff that works the best can be left up to the end user.

In the case that "soft" limiting is preferred and/or the user wants to make sure that the output eventually hits a "hard" limit I decided to include a diode shunt limiter on the output. The level at which the diodes start conducting is biased along with the setpoint for the output voltage gain, so that both are controlled in tandem from one setting. For instance if the output voltage gain is set to 2V the diodes will start conducting at around 2.5V and then will clip hard at 2.7V or so. If the output voltage gain is changed to 1V the diodes will start conducting at around 1.5V and then will clip hard at 1.7V or so.

In my modeling I have observed (for low feedback gain) that on some signals the first few cycles of the burst that I am using to test is not sufficiently limited but subsequent output is limited below the setpoint, and the hard limiter seems to be a good way to deal with these brief output surges and prevent the amp from seeing too high an input signal. Again, this type of brief overshoot can also be eliminated by increasing the feedback gain, but for some low-gain settings this seems to be a good secondary measure for limiting the output voltage.

Anyway, that's the latest report of progress on the design. I'll try to post some examples of the limiting behavior but might not get to it for a couple of days.

-Charlie
  Reply With Quote
Old 9th February 2013, 12:25 AM   #15
diyAudio Member
 
Join Date: Mar 2007
Location: California
Attached are a couple of examples of the output of the circuit, given a single frequency (10 Hz) "burst" input.

The first attachment shows the response to increasing input level, with a max output level set to 1.5V. Even up to 1V input, when the gain at 10Hz has been reduced from 24dB to about 4dB, the signal shape is still pretty clean. Not shown is the distortion levels, which are about 1% dominated by 3rd order. There is some evidence of pumping for the Vmax=0.2V input level, but this amounts to only about 2-3dB difference in level.

The second attachment shows a range of feedback gain settings for an input level of 0.5V. At the lowest gain setting, the boost limiter is barely working and the limiting is coming from the diode shunt limiter on the output, causing distortion to be about 14%. As gain in increased (moving down the figure), the distortion levels are 0.5%, 1.25%, 3%, and 5% respectively. Pumping is visually evident in the bottom trace. At the same time, there is a tradeoff in the limiting action - the levels increasingly falling within the 1.5V output voltage setpoint. While the trace second from the top has the lowest distortion, output amplitude is greater than 1.5V but less than the level at which the diode limiter is clipping the output. The gain could be increased or the output voltage setpoint could be reduced slightly.

One note - these results are for 10Hz, which shows the pumping effects the most. Higher frequencies have slightly different behavior but I have not had the time to fully investigate it yet.

-Charlie
Attached Images
File Type: jpg different_input_levels_10Hz.jpg (106.2 KB, 35 views)
File Type: jpg different_feedback_gains_10Hz.jpg (107.1 KB, 36 views)
  Reply With Quote
Old 9th February 2013, 03:02 AM   #16
diyAudio Member
 
Join Date: Mar 2007
Location: California
The attached plots show another set of tests I ran, very much like those shown above except that they were done with a 20Hz burst.

The distortion levels for the feedback gain are (from top to bottom) 10%, 0.65%, 0.44%, 1%, 1.1%. This is a very different profile than above. Even when pumping can be seen, the distortion is not high and only the amplitude is being modulated. It's important to realize that the amplitude modulation is not the same for all frequencies, because it is the low frequency boost that is being modulated by the circuit. The high frequency amplitude remains unchanged.

-Charlie
Attached Images
File Type: jpg different_input_levels_20Hz.jpg (118.5 KB, 31 views)
File Type: jpg different_feedback_gains_20Hz.jpg (115.4 KB, 9 views)
  Reply With Quote
Old 9th February 2013, 05:03 AM   #17
diyAudio Member
 
TundraLTD's Avatar
 
Join Date: Dec 2009
Wouldn't this be similar to what a Bag End ELF processor does, or am I mistaken
  Reply With Quote
Old 9th February 2013, 05:12 AM   #18
sreten is offline sreten  United Kingdom
diyAudio Member
 
Join Date: Nov 2003
Location: Brighton UK
Hi,

FWIW its not new, its a feature of most AV biased subs.

Its in some hifi subs, but they keep quiet about it.

rgds, sreten.
__________________
There is nothing so practical as a really good theory - Ludwig Boltzmann
When your only tool is a hammer, every problem looks like a nail - Abraham Maslow
  Reply With Quote
Old 9th February 2013, 05:19 AM   #19
diyAudio Member
 
Join Date: Mar 2007
Location: California
Quote:
Originally Posted by TundraLTD View Post
Wouldn't this be similar to what a Bag End ELF processor does, or am I mistaken
No, you are not mistaken... but I mentioned this in the opening post, including giving their Patent number on the technology.

They do it differently, and this way is "better" IMO because it uses a biquad filter not a double integrator to create the boost. This means that the response can extend up to whatever frequency you desire to operate the driver, e.g. not "just for subwoofers". It's also more flexible with the response shapes that can be accommodated.

-Charlie
  Reply With Quote
Old 9th February 2013, 05:21 AM   #20
diyAudio Member
 
Join Date: Mar 2007
Location: California
Quote:
Originally Posted by sreten View Post
Hi,

FWIW its not new, its a feature of most AV biased subs.

Its in some hifi subs, but they keep quiet about it.

rgds, sreten.
Why keep quiet about it?
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

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


Similar Threads
Thread Thread Starter Forum Replies Last Post
How to figure gyrator treble boost gain graygem Analog Line Level 2 20th April 2012 06:56 PM
LT/ parametric bass boost of vented enclosure muddasirwaheedmalik Subwoofers 3 6th December 2011 03:42 PM
Noisy gain boost bereanbill Tubes / Valves 17 15th April 2009 04:51 PM
Marshall Valvestate 8080 low gain on Boost Channel mgjimbo Solid State 0 28th May 2008 02:04 AM
Preamp question/ input gain boost 3-LockBox Class D 2 13th June 2006 03:23 AM


New To Site? Need Help?

All times are GMT. The time now is 02:04 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2