Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Rebuilding the Enigma...
Rebuilding the Enigma...
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 14th May 2019, 01:10 AM   #51
ICG is offline ICG  Germany
diyAudio Member
 
ICG's Avatar
 
Join Date: Jul 2007
Location: I had a Déjà Moo - I've seen that BS before!
Quote:
Originally Posted by Brian Steele View Post
Looks like I built this box well. The impedance peaks are 22.8 ohms and 24.3 ohms.
I don't doubt your box works fine but the almost equally high impedance peaks do not imply it works well. In the old days it was a rule of thumb the two peaks had to be almost of the same height, nowadays there are tons of speakers which divert from that rule and work very well too. Back then it was an easy way to approximate how a speaker (sub) works since the fast Fourier transformation could not be applied because of lack of processing power and a correct low frequency measurement was impossible for most folks because of the requirements of the measurement (nearest reflection surface, measurement distance and room modes/influences). A lot of non-equal-height-impedance-peak speakers sound great and there are quite some speakers which fulfill the equal height impedance rule but sound awful. In reality this rule applies only to a narrow range of Qts/Qt (~Qb3 ideal) and can divert a LOT from that to get great results since the room gain and frequency response is different for every setup and room and personal taste.

Quote:
Originally Posted by Brian Steele View Post
There's likely a way to estimate the box losses from the difference between the sim'd and actual impedances, but I don't know how to do this. In any case, this means that the driver is only going to see 35W and 33W at those frequencies, if driven with at 100W amp.
AFAIK WinISD pro is capable of simulating it as long as you stay within the low loss port air velocity range.

Quote:
Originally Posted by Brian Steele View Post
Minimum impedance in the passband is 7.62 ohms however, which translates to 105W at that frequency (around 64 Hz), where cone movement is going to be at a minimum. Hmm...
As I already said, it's not the total power nor the current which is critical, it's the heat dissipation capabilities which count - which are dependent on the ambient temperature and the air exchange and the frequencies. There is no way of limiting it to a 'safe range' if you are trying to limit just ONE parameter (current) since it DOES NOT represent at all what temperatures are actually reached at the VC. Ambient temperature and airflow are much more important for that limit to divert vastly. I've driven 4 15" subs in winter outside with ~500W each and they performed perfectly while a friend used my subs at a summer party and with just ~150W each, two died. I cannot stress it any more, the enviroment is crucial! Thermal design does not depend on just one single parameter.

Quote:
Originally Posted by Brian Steele View Post
Assuming that I get the rest of the work on the box done over the next few days, I'll be able to do some sine sweeps at different voltages on the system this weekend. Basically checking to see how the output changes with input voltage. The sine sweeps are just over 5 seconds long. Let's see if I can get that breaker to trip...
That breaker won't do sh*t (or trigger way too early), depending on the enviroment. If you want to be safe, either use a DSP with peak and rms limiter or measure the pole plate temperatures to stay within the parameters the driver can cope with.
__________________
I had a Déjà Moo - I've seen that BS before!
  Reply With Quote
Old 14th May 2019, 01:43 AM   #52
Brian Steele is online now Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
Quote:
Originally Posted by ICG View Post
I don't doubt your box works fine but the almost equally high impedance peaks do not imply it works well.
That's not what I was using as my point of reference. My point of reference is the magnitude of the impedance peaks predicted by the sim, which assumes no losses. The closer the measured impedance peaks are to the sim, the less "lossy" the box is.
__________________
www.diysubwoofers.org
  Reply With Quote
Old 14th May 2019, 01:47 AM   #53
Brian Steele is online now Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
Quote:
Originally Posted by ICG View Post
As I already said, it's not the total power nor the current which is critical, it's the heat dissipation capabilities which count - which are dependent on the ambient temperature and the air exchange and the frequencies.
Yup, and the more power the driver has to dissipate, the more critical those heat dissipation capabilities. That's why I HAVE to take the power the driver may see into consideration. And yes, it IS critical. If I was running the equivalent of 5W into this driver, we wouldn't be having this conversation ;-).
__________________
www.diysubwoofers.org
  Reply With Quote
Old 14th May 2019, 04:12 AM   #54
Brian Steele is online now Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
Another quick image - the passband response, at various levels, right up to where I measured just over 10% distortion in the passband (around 40 Hz). There is almost no dynamic compression due to vents, etc..
Attached Images
File Type: png 20190513-enigma2 (DA270 power compression).png (35.0 KB, 180 views)
__________________
www.diysubwoofers.org
  Reply With Quote
Old 14th May 2019, 08:11 AM   #55
freddi is offline freddi  United States
diyAudio Member
 
freddi's Avatar
 
Join Date: Aug 2005
Rebuilding the Enigma...
that's pretty impressive. What from this experiment might be applied to other builds?

Brian, which (if any) free spreadsheet programs will run your horn fold sheets in Windows?

Last edited by freddi; 14th May 2019 at 08:15 AM.
  Reply With Quote
Old 14th May 2019, 12:07 PM   #56
Brian Steele is online now Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
Quote:
Originally Posted by freddi View Post
that's pretty impressive. What from this experiment might be applied to other builds?
For a 4th order BP design, use the largest vents that you can, and design the layout such that the position of the driver nulls out the lowest (and most problematic) peak in the out of band noise?

Seriously, there's a reason why I call this particular build the "Enigma". It was built long before I started using Hornresp, and it constantly defies any of my attempts to successfully model it in Hornresp (well, apart from the offset driver thing, which did predict the nulling out of the first out of band response peak).

Case in point: The "Enigma" currently has two flared 3" vents, that have an effective lengths around 21 cm each. The Hornresp modeling that I did for the replacement driver suggested that I'd have to cut the vents down to 14 cm each, which would boost Fb to 71 Hz and flatten the passband. Well, it's a good thing that I decided to do some measurements before proceeding to cut the vents, because the results suggest that, if anything, the vents may already be slightly too ***short***, even though they're a good 50% longer than what the model suggests should be used! Fb is around 64 Hz instead of 71 Hz, but the passband shows a slight upwards instead of downwards tilt. And no, I don't know why this is happening. At least, not yet. I plan to take a much closer look this weekend, to see if there's something that I'm missing.


Quote:
Originally Posted by freddi View Post
Brian, which (if any) free spreadsheet programs will run your horn fold sheets in Windows?
Unfortunately, I don't know of any offhand.
__________________
www.diysubwoofers.org
  Reply With Quote
Old 28th May 2019, 12:56 AM   #57
Brian Steele is online now Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
I had an opportunity to do a few more power tests this weekend. Some notes:
  • I was able to driver the subwoofer with up to 23.3V until measured distortion got over 10% in the passband. The THD significantly increased below this (as predicted by the sim'd driver excursion). Going by the distortion graph, a 30~35 Hz 12dB/oct HP filter will increase the subwoofer's power handling as far as THD is concerned.
  • 23.3V works out to about 68W into 8 ohms, or 70W into the subwoofer's measured minimum impedance in its passband. Current draw at that impedance works out to 3A.
  • The subwoofer's output remained very linear right up to that distortion limit. I suspect that this is one of those bandpass builds that sounds very good at higher and higher SPL levels - until you don't hear any sound at all, LOL. Who says that small signal parameters produce sims that are only accurate at small signal levels?
  • Leaving the amp's gain at the point where it was generating a 23.3V sine sweep, the subwoofer played back a wide selection of music for about two hours without showing any audible signs of distress. It was also loud enough to provide background level music at the bar IMO.
  • The inline PTC did not trigger during the testing. I suspect that was due to the length of the sine sweep test signal, which was not long enough (the PTC device needs to see double the rated current for at least 5 seconds before triggering). It also did not trigger during the music testing.


I used my car's 1.2kW (into 4 ohms) amplifier to do the testing, as my usual test amplifier was not available. I had to add 8dB of gain at the deck to achieve 23.3V at the deck's volume setting of 40, my usual volume setting (unless it's the weekend). This is interesting, as that means that at the amps' normal gain setting, my car's two 12" subwoofers that present a combined 2 ohm load to the amplifier would only be seeing about 43W total, or about 21.5W each, if my calculations are correct. No wonder my car's system seems to have so much headroom at bass frequencies...
__________________
www.diysubwoofers.org
  Reply With Quote
Old 28th May 2019, 01:43 AM   #58
Brian Steele is online now Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
...and here's what the THD looks like (normalized to the FR) at 23.3V. Even at that level it remains around 1.2% from 50 Hz up. Below that the driver's excursion starts to have an impact, and at 39 Hz it's at 10%. That's why I think at 30~35 Hz HP filter should help reduce reduce distortion and increase mechanical power handling, if necessary.

Those little bumps just about 160 Hz? There's something buzzing inside the enclosure at high input levels. I haven't figured out what it is yet. It's quite a bit outside of the subwoofer's useful passband in any case.
Attached Images
File Type: png 20190527-enigma2 (distortion).png (49.6 KB, 140 views)
__________________
www.diysubwoofers.org
  Reply With Quote
Old 28th May 2019, 09:09 AM   #59
BP1Fanatic is offline BP1Fanatic  United States
diyAudio Member
 
BP1Fanatic's Avatar
 
Join Date: Jan 2009
Location: Columbus, Ohio
Great information Brian!
  Reply With Quote
Old 2nd July 2019, 11:52 AM   #60
Brian Steele is online now Brian Steele  Grenada
diyAudio Member
 
Join Date: Nov 2007
Just a quick update to this ongoing saga:

I installed "Enigma v2" in my brother's bar just over a week ago, where it sees somewhere between 8~12 hours of continuous play every day. I also replaced the Technics receiver with a Monoprice Unity 200 amplifier to drive the main speakers (100Wx2) and a Monoprice Unity 100 to drive the subwoofer (100W bridged into 8 ohms). Both amps feature only a power button on the front panel, so they are quite fiddle-proof, something that was needed for this bar.

Feeding the amps is a Dayton DSP408, which provides the filtering and EQ needed. For the sub, a HP filter at 30 Hz (to reduce distortion at low frequencies at higher SPL levels) and a LP filter around 150 Hz (to eliminate out of band noise). I didn't have my measuring equipment with me, so I basically adjusted the gains by ear, to ensure that the DSP408 overloaded just before the amps did, to ensure that the speakers were not over-driven. I might go back in a week or two to fine-tune the gains of the amp and DSP.

Feedback from my brother: "Brian last night I went to the bar and the bar system ( the laptop was not working ) but Kingsley hooked up his phone and damn! Bro that system in the bar is off the chain that is the best I ever heard it sound. Not loud ( even though he could go loud) but real clean with a nice low bass."

So, mission successful, so far.

Will the subwoofer hold up under long-term use? Well, it's only been a week, but it has been used every day of that week for fairly long periods. If it doesn't blow first, I'll do an impedance check on it in a few months to see if there's been any change.
__________________
www.diysubwoofers.org
  Reply With Quote

Reply


Rebuilding the Enigma...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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Another enigma with my "Enigma" Brian Steele Subwoofers 9 4th November 2018 07:27 AM
Audio Analogue Enigma service manual hamradioman Solid State 0 21st August 2015 02:24 AM
Help With DEQ2496 For Subwoofer (Enigma) Gwho Subwoofers 7 15th October 2013 01:11 AM
Someone with Enigma 3 and a scanner: Can you help me? Hybrid fourdoor Music 0 21st September 2003 10:22 AM
I'm rebuilding my amp Lisandro_P Everything Else 2 4th July 2002 08:53 AM


New To Site? Need Help?

All times are GMT. The time now is 08:38 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki