Help me understand step response - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 7 Old 10-13-2012, 12:40 PM - Thread Starter
Senior Member
 
splotten's Avatar
 
Join Date: May 2009
Posts: 356
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 38
I have been messing around with unibox for a while, and I think I now have a good grasp of how to work it. I have been modeling different drivers I different enclosures to learn and to prepare myself for building a real vented box.

I have hit a snag though. I can’t seem to understand what the “step response” is supposed to show me and I don’t understand what the estimated values represent or what they are supposed to be. Also it eludes me why the step response changes radically when I apply a high pass filter and at low pass filter.

Please take a look at the following pictures and if someone could tell me what the “response peak”, Qa, Qi, and Qp is supposed to tell me I would be grateful. Are these step responses good or bad?

First: What i think the step response shuld look like.






Second: Same driver in a bigger box and lower tuning. What are those ripples?






Third: Same box as above but with HP and LP applied. Qiute different result...





Dan
splotten is offline  
Sponsored Links
Advertisement
 
post #2 of 7 Old 10-13-2012, 05:31 PM
AVS Special Member
 
A9X-308's Avatar
 
Join Date: Mar 2008
Location: Australia; now run by adults.
Posts: 5,147
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 41
Quote:
Originally Posted by splotten View Post

I have hit a snag though. I can’t seem to understand what the “step response” is supposed to show me and I don’t understand what the estimated values represent or what they are supposed to be.
Step response shows how the system will respond if given a perfect square wave as an input signal. As a square edged step has a bandwidth from DC to light it's not really representative and most people should just ignore it.

The best damped speaker is sealed with a Qts of 0.5 (simulate it yourself and look to see the shape). It means that it will respond accurately to the leading edge of any signal transient, but when the signal is removed, it will return to rest position as quickly as possible. Ported will take longer to settle than any useful sealed box because the energy in the port Helmholtz resonance is less well damped and will take longer to settle.
Quote:
Originally Posted by splotten View Post

Also it eludes me why the step response changes radically when I apply a high pass filter and at low pass filter.
You're changing the signal applied; it's now not a step because the filters have changed it's bandwidth.
Quote:
Originally Posted by splotten View Post

Please take a look at the following pictures and if someone could tell me what the “response peak”, Qa, Qi, and Qp is supposed to tell me I would be grateful.
Are these step responses good or bad?[/quote]If you look at the 2nd graph you posted there is a slight rise between 150-200Hz or so and this is 0.65dB above however Unibox calculates a mean flat response.

From the Unibox page "The full box model takes all losses into account resulting from absorption, leakage and the port. To calculate the vented box frequency response enter your choice of box physical volume (Vb), Q due to box absorption (Qa), Q due to box and driver leakage (Ql) and Q due to the port (Qp)."

Qa is losses due to the type, quantity, density and placement of damping material inside the box. Ql is losses / leakage from the surround and dustcap not being perfectly airtight and the air in the box then not acting like a perfect spring. Qp is mainly frictional losses as the air moves in the port.
Quote:
Originally Posted by splotten View Post

Second: Same driver in a bigger box and lower tuning. What are those ripples?
Port resonance which in the fourth graph is shown at just under 200Hz and the ripples have a period of about 5ms so that correlates.
Quote:
Originally Posted by splotten View Post

Third: Same box as above but with HP and LP applied. Qiute different result...
You're applying a different input signal filtered by the HPF and LPF, so this shows how the system responds to that filtered signal.
A9X-308 is offline  
post #3 of 7 Old 10-14-2012, 01:10 AM - Thread Starter
Senior Member
 
splotten's Avatar
 
Join Date: May 2009
Posts: 356
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 38
OK Thanks :-)

I think I got most of it now.

Could you please verify or reject this: Before (in time) the graph begins the input signal is zero (0) volts. At t=0 the input signal is –instantly- changed from zero to some positive DC voltage different from zero (1) and remains at that voltage, at least, for the duration of the graph. The graph shows us the calculated response to that input and the y-axis is depicting the wave form from the speaker in the timespan from t=0 to t=0.1 Seconds.

-I have simulated the response of the driver in a closed box and it is much quicker to settle back to zero than the vented box:



If I change the Le value of the driver to at a higher value (within reason) the response becomes less “pretty”. To my understanding the inductance will impair/resist the –change- in current flowing through the coil of the driver and with that the acceleration and the ability to accurately track the signal.



Does this mean that high inductance drivers (LAB 15 comes to mind) is best avoided or is it a nonissue since we are applying a low pass filter anyway. The low pass filter seems to have huge influence on the step response. (Because the “step” is no longer a step). The high pass filer seem to have much less negative effect.

Low pass only. 4,th order L-R



4.th order High pass only



I am using “woofer box and circuit designer 3.00” to make the filters and load them in to Unibox. For the low pass filter I can choose between Butterworth, linkwitz-riley and Bessel. Which one is the “correct” one? - I assume one or the other is standard in plate-amps and surround receivers.


Can you please elborate on this: "If you look at the 2nd graph you posted there is a slight rise between 150-200Hz or so and this is 0.65dB above however Unibox calculates a mean flat response."

In dont understand what you mean.

Oh! BTW. Is all of this just a mental exerzise or is it actually important. You seem to think that i shuld not bother too much. Why is that?


Thank you for your time.

Dan
splotten is offline  
post #4 of 7 Old 10-14-2012, 09:18 PM
AVS Special Member
 
tuxedocivic's Avatar
 
Join Date: Feb 2010
Location: Ladysmith, BC
Posts: 4,262
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 106 Post(s)
Liked: 400
Cool topic. One I don't know much about.

Step response is not something to get to wrapped up about. Look at the differences you've pointed out, yet most people don't notice a "bad" step repo as when listening to a high pass, low passed, and vented subwoofer. It's one of those things like, how flat is the 15khz to 25khz of a speaker thing. It doesn't really matter unless you have more time and money than you should.

I always look at it, but it's just a quick glance. I think of it in terms of how less does this track the input signal. Some driver are better than others, but it doesn't matter much.

Thanks for the nice explanation A9X. I hope to hear more on the subject. I've read just a tiny bit about it in a few sources.
tuxedocivic is offline  
post #5 of 7 Old 10-15-2012, 02:58 AM - Thread Starter
Senior Member
 
splotten's Avatar
 
Join Date: May 2009
Posts: 356
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 24 Post(s)
Liked: 38
As already stated I am all new to DIY subwoofers.

By now I realize that people don’t seem to put too much emphasis on the calculated step response. However I have already simulated a few boxes and almost pulled the trigger on one of them. As I go a long and learn more I realize that that there is much more to it than I initially thought and it makes me tighten the “spec sheet” little by little.

I am trying to compile a set of “landmarks” or set of specifications that the eventual subwoofer must be able to do before going ahead building an actual box. Maybe the step response is best utilized when evaluating a complete system (not just the subwoofer) since the step contains all frequencies. At least it would give some indication on the coherency in time between different drivers. Also there might not be –one- right step response but if I had a better understanding I might be able to, at least, weed out less optimal designs.

Cheers

Dan
splotten is offline  
post #6 of 7 Old 10-15-2012, 07:55 PM
AVS Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 20,285
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 35 Post(s)
Liked: 125
Quote:
Originally Posted by A9X-308 View Post

The best damped speaker is sealed with a Qts of 0.5 (simulate it yourself and look to see the shape). It means that it will respond accurately to the leading edge of any signal transient, but when the signal is removed, it will return to rest position as quickly as possible.

I used to think that too, until I did a simple FEA of a spring-mass-damper system, which is what a speaker is.

Turns out that while higher damping does lessen ringing, it adds "stickyness", slowing the response of the cone back to rest position when the signal is removed.

Noah
noah katz is offline  
post #7 of 7 Old 10-15-2012, 10:23 PM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 15,821
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 80 Post(s)
Liked: 808
"I am trying to compile a set of “landmarks” or set of specifications that the eventual subwoofer must be able to do before going ahead building an actual box."

the subwoofer(s) and the environment that they are in form a 'system'. ideally that system will play (at whatever level it is called upon) all bass frequencies, have a flat response, no significant time decay, no significant power compression, and no significant distortion of any kind.

Listen. It's All Good.
LTD02 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