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

Multi-Way Conventional loudspeakers with crossovers

Need help interpreting measurements
Need help interpreting measurements
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 9th February 2018, 02:16 PM   #71
montana1 is offline montana1  United States
diyAudio Member
 
Join Date: Jan 2016
Location: Kellogg, Idaho
Thanks Tony,
You are correct. I believe port tuning is a little higher then expected. My target based on simulations in Unibox and Winisd Pro was 38 Hz. In looking at actual z and nearfield measurements, it looks to be 41 Hz. tuning. If required to obtain the right bass level I can always extend the tube some to lower the tuning point. The boxes are rather deep so there is some room to extend before being to close to the front wall. Even then, I suppose its not against the law to elbow up if need be.

Best Regards,
Rich
  Reply With Quote
Old 9th February 2018, 07:22 PM   #72
montana1 is offline montana1  United States
diyAudio Member
 
Join Date: Jan 2016
Location: Kellogg, Idaho
Hello Guys,
Back once again with another Xover design for consideration. This design is an attempt to address resonant peaks and possible offer better bass extension. The cost as mentioned prior in this thread is reduced overall system sensitivity. I'm learning (in my still newbie state) that compromises are inevitable. With that said, I'm leaning slightly toward the version posted here for a few reasons.

1. Even though the parts count is higher; I already have most all the components so cost will be less then the simple version. Got to like that.

2. Would seem to offer better load presented to the amplifier. Not sure if it is a big consideration but could not hurt either.

3. Rather not have to add a subwoofer if system is bass shy. So achieving the lowest F3 is a design goal.

What are your guys thoughts? BTW, the last photo is system z and phase of the simple version for comparison.

Best,
Rich
Attached Images
File Type: jpg Omega_V1_Xover.jpg (36.4 KB, 71 views)
File Type: jpg Omega_V1_FR.jpg (97.4 KB, 74 views)
File Type: jpg Omega_V1_Inverted.jpg (99.7 KB, 72 views)
File Type: jpg Omega_V1_SYS-Z.jpg (86.8 KB, 73 views)
File Type: jpg omega1_v2_SYS-Z.jpg (89.3 KB, 70 views)
  Reply With Quote
Old 9th February 2018, 09:18 PM   #73
montana1 is offline montana1  United States
diyAudio Member
 
Join Date: Jan 2016
Location: Kellogg, Idaho
Hi jReave,
My sincere apologies. I somehow missed your last post. Unless you feel differently; I'm going to stay with the 6 db of baffle step that was modeled in Response Blender. Unless I'm overlooking something, that program can only display a full baffle step loss. My work around has been to use Response Modeler and apply a given baffle step percentage after splicing in nearfield/farfield data in Response Blender. There probably is a smoother method that I'm not aware of at this point.

These speakers are going in my offfice/exercise room which is rather smallish in size. Speakers will be 4.5 feet from side walls and 28 inches from rear wall to front of speaker. The back of speakers will be 12 inches from rear wall. Given this added information, do you think I should re-visit my 30 or so Xover versions? Like I said, to much time on my hands. The fun soon ends though. I return back to work in less then two weeks.

Thanks in Advance,
Rich
  Reply With Quote
Old 10th February 2018, 12:21 AM   #74
wintermute is offline wintermute  Australia
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Need help interpreting measurements
Have only had a quick look. I've not used impedance flattening before so cannot comment from experience, so won't venture an opinion, other than 500uf is a big cap

When we talk about the Baffle step, it's correct to put the full baffle step into the measurements when you do the blending, it's just when you do the crossover modeling you don't necessarily want to set your target so that it is flat down to the low bass.

I find that a bit of a challenge because your natural tendency is to try and get a nice flat response. If you set your target response curves say for instance 3db higher and then don't worry about the low end, you should be ok.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 10th February 2018, 02:30 AM   #75
montana1 is offline montana1  United States
diyAudio Member
 
Join Date: Jan 2016
Location: Kellogg, Idaho
Hi Tony,
Thanks for once again for sharing your thoughts. Just to make sure I'm following you correctly, the Xover version from post #65 would that meet the target response curve that you are after? Or should I pad down the HF region a db or 2? I have noticed from studying the work of Troels Gravesen, that he likes to have about 2-3 db downward tilt to the response curve. My only uncertainty is does the response curve change much as the listening distance becomes further away then the modeled design. For instance, I took these measurements at 27 inches and based the the Xover schematic in Xsim off that same mic distance. Does the response curve begin to droop some as distances expand (say to 6-7 feet)? My tendency has been to target as flat as possible thinking that by the time sound reached the listening position that the highs would be falling off some.

Thanks in Advance,
Rich
  Reply With Quote
Old 10th February 2018, 04:01 AM   #76
wintermute is offline wintermute  Australia
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Need help interpreting measurements
OK I would say that your target response curve in post 65 was probably around 84db, what I'm saying is make it instead maybe 87db (that's half way between full baffle step compensation and no baffle step compensation).

You will see that your lower frequencies don't seem to be very flat (in the sim), but that will hopefully be evened out by the re-inforcement you get being close to the wall.

On the question of down tilting response, that is a tricky one, because it will be very room dependent as well as personal preference dependent.

If you have a very lively room (lots of hard surfaces) then you probably want the downward tilt, if you have a pretty dead room, then flatter will probably be better.

It also depends on your subjective preferences for treble. What you will find is than in a typical room you will naturally get a falling response as frequency increases. My own speakers I went with pretty flat and I've been happy with that. I have carpeted floor, and curtains on the back wall, and a fair bit of furniture in the room, so it is on the more dead side.

There are different schools of thought on this. Some say tune the crossover to the room, others say make the crossover flat for anechoic and eq for the room.

I made it flat for anechoic, and haven't found that I needed to eq (though I do have a hump that is due to having too much baffle step compensation factored in), but some people may not like my setup (generally it has received praise from anyone who has listened, but I haven't had any hard core audiophiles over )

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 10th February 2018, 04:43 AM   #77
montana1 is offline montana1  United States
diyAudio Member
 
Join Date: Jan 2016
Location: Kellogg, Idaho
Thanks Tony,
I understand what you mean now. Tomorrow, I will try to raise response curve 3 db as you suggested. Before proceeding with purchase of Xover components, I will post another version for critique. Everything you mentioned makes perfect sense. I really appreciate you taking the time to share your thoughts.

Best Regards,
Rich
  Reply With Quote
Old 10th February 2018, 04:57 AM   #78
wintermute is offline wintermute  Australia
just another
diyAudio Moderator
 
wintermute's Avatar
 
Join Date: Aug 2003
Location: Sydney
Need help interpreting measurements
The other suggestion would be when you are purchasing, consider that you may decide you need to adjust, so don't go overboard I'm actually still using my axon caps that I bought for experimenting because they were very well priced. Some people say they don't like them, I've not felt the need to upgrade.

Tony.
__________________
Any intelligence I may appear to have is purely artificial!
Some of my photos
  Reply With Quote
Old 10th February 2018, 06:26 PM   #79
jReave is offline jReave  Canada
diyAudio Member
 
Join Date: Oct 2012
Hey Rich,

Ok, I've had a closer look at a few things. Here are my questions and comments:

Looking at your measurements in post #62, if you look at f3 in the NF and then f3 in your sim, both are at about 67Hz, which means that your sims haven't included the port output which is peaking at around 40Hz. In terms of the simulation accuracy, it's not a big deal because the port isn't affecting the 100 - 200Hz range and that's what you should be looking at to set your target xo levels. So, don't concern yourself with trying to milk more out of the LF because you don't see much in the sim - it doesn't look to me like an accurate representation of reality.

Now when I add in 6dB of baffle step to your NF and blend to your FF, I get a target level of about 84dB, which is exactly where you are in your sims on page 7. So, I'm saying discard your effort from post #72. Although using a resonance compensation filter on the woofer achieves what you were attempting (lowering f3), I think it's totally unnecessary (as explained above) and in fact I don't think I've ever seen it done on a 2-way before - the result is a serious loss in efficiency.

I must admit I am a bit confused by the level of your NF measurement in post #62. Typically when you go from measuring 2 or 3 ft away to a matter of 1/8" or 1/4" from the cone, you are going to see a considerable increase in SPL in the nearer measurement. In Jeff's white paper on the measurement process, there's a difference of 14.7dB, but when I tried blending your files, I needed to add .5dB to the NF to get the best blend with the FF (again, that's with the full 6dB of bsc). ?? I am thinking that you may have changed the amplifier level between measurements whereas it should be kept the same throughout.

Concerning baffle step, your distances to walls don't look too bad to me but a small room may add a little extra to the bass. You might try finding a corner in your present measurement room and set the speaker up with your intended setup distances and this time measure a few ft away with the gate wide open. This will give you a better idea of what you'll really be hearing in a room, although the effects of a small room won't be there.

The difficulty in this instance is understanding how to interpret it, because now you are no longer looking at an anechoic FR and so the target is no longer flat. What you'll be after on the high end is a diminishing curve more similar to the power response in PCD and on the low end, an increasing response due to the additions of the room and its boundaries. Look up House Curves to give you more insight. Or just to open up another can of worms.

Having said that, I would still probably start with the full baffle step and then when you get it in your real room, adjust either the xo or eq to taste. This will probably also apply to the HF level too - whether in room you want to make an adjustment there too and if you want to do it with the xo or eq.

However, this is really a sort of flip the coin scenario - it's just as valid to start with a little less bsc and then listen and adjust from there. It's just a case of adjusting either up or down. The only possible advantage to starting with more bsc is that to make the adjustments, you need inductors of lower value which you can do by unwinding the bigger ones you start with (measuring instrument required) whereas if you start with lower bsc and need to go bigger, you'll need buy new, bigger inductors.

Re bsc, both ways to sim it seem valid to me. You can add the full 6dB of loss and then set a target level above that as Tony suggests or you can adjust baffle loss in Response Modeler and then import that into Response Blender. Although you have to realize in the latter case that you will be blending a FF response with the full 6dB loss with a NF with only say, 3 or 4dB of loss. You may need to adjust that a little bit too.

So that leaves me with the xo sim you did in post #65. Simple and effective. I don't know what that woofer peak at about 3kHz is going to be like, but maybe, as Tony suggested before, you may want to tank that down with a simple capacitor in parallel with L1.

Let's see where those suggestions get you.
  Reply With Quote
Old 10th February 2018, 07:55 PM   #80
montana1 is offline montana1  United States
diyAudio Member
 
Join Date: Jan 2016
Location: Kellogg, Idaho
Thanks for the input guys,
I realized a mistake made in taking nearfield measurements. As jReave suspected, I had different Spl levels for near and farfield. So, the sims in this post will reflect the changes that were made to correct my rookie move.. I will also attach all my raw measurements so you guys can take a look if you like. I'm still finding it challenging to reach 87 db reference level even with factoring in 50 percent baffle step loss. Which by the way creates a wild sag in the lower mid range. The version posted here is with full baffle loss applied in Frequency Response Blender. I really appreciate the in depth and well thought out responses from you guys!

Best Regards,
Rich
Attached Images
File Type: jpg Omega_schematic.jpg (35.9 KB, 23 views)
File Type: jpg Omega_FR.jpg (97.0 KB, 22 views)
File Type: jpg Omega_Sys-Z.jpg (88.2 KB, 18 views)
  Reply With Quote

Reply


Need help interpreting measurementsHide 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
Help interpreting room acoustics measurements kaputt Room Acoustics & Mods 13 20th August 2017 10:20 AM
Interpreting specs akis Chip Amps 5 23rd March 2014 07:55 PM
Need help interpreting measurements jimjimmyjimjim Multi-Way 5 26th September 2013 07:52 PM
Interpreting T/S Parameters fringy75 Full Range 6 19th April 2013 06:46 PM
help interpreting graph getafix Multi-Way 6 12th October 2003 12:52 AM


New To Site? Need Help?

All times are GMT. The time now is 11:31 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki