How Flat is "Flat"? (XT18WO + 27TBFC-G = MTM)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have been using the FRD tools to model an MTM with two XT18WO Vifa woofers and a 27TBFC-G tweeter. The plot in the attached file takes into account the Baffle Step of the Box (Partsexpress MTM enclosure).

The tweeter will be offset 1"

The active XO network would consist of (shown in the .xls file):
- 4dB 470Hz Baffle Step Compensation (variable gain)
- 5dB 1100Hz Q=1.414 Notch Filter
- 1200Hz 4th Order Linkwitz Riley XO
- Tweeter will be padded +2.5dB

These two drivers look like they would sum nice and flat. However, the baffle step predicted by the BDS simulator really throws a wrench in the works in terms of final response (Yellow = Woofer, Cyan Tweeter)

I have been able to keep the predicted response flat within 3dB without complicating the XO too wickedly.
Is this flat enough?

From what I see when I look at the response the tweeter section is the only part left that's up and down, but it looks more like driver/BS that's causing this. I think I would need at least three notch filters to get it much flatter, and they may not even help out (I have not tried modeling them) and would be extremely high Q.

From here I would like some feedback and suggestions. Does this look good from a frequency response standpoint? Are there any obvious design flaws?

TIA
 
That BDS simulation looks worse than it should be. Did you run it in complex mode, with a 1/2" rounded baffle? Did you also model off axis response?

Have you considered the impact of the notch filters on off axis response and phase? The notch filters may solve on axis problems, but make off axis response worse.

My current project is an OB using the Seas 27TBFC/G and Vifa XG18. I designed the baffle to be easy to fix. The midrange has one large peak well below the crossover frequency, and the tweeter is flat to half an octave below the xover frequency.

Dan
 
After your post I tried modeling 5 and 10 degrees off axis to the left and right and the BDS plots look to have little difference.
I will still make net speaker response plots from these BDS simulations to see what the final result looks like.

I for the previously attached .xls plot I ran:
Quadruple
Complex
Point
Quadrangle
Separate
Sum Down
Edge = 0.5"
Axis Dist = 36
Angle = 0
Baffle Tilt = 0

My thinking has been that since the Notch and BS compensation will be applied to the signal BEFORE the XO that any effect they have will occur on both drivers.
ie: the phase may be off, but it should still be continuous across the tweeter and woofer.

Can you explain how I would model phase response with the FRD tools, at this point I'm not sure how to do that.

re your setup, I think I know the "bump" you're talking about. This bump is part of the reason I have the notch in there. The combination of the bump and the baffle step hitting @ the same freq causes the final response to be almost 8dB high there! :-o

How low are your crossing over?? Lowering my XO point would help out on multiple levels.... But I'm not sure how much lower I can go with this tweeter.

Have you looked into the response of the XT18WO, that bump is lower than the WH series (from the larger magnet added damping was my guess).
Also, do you think the smaller magnet would be better than the large for an open baffle speaker?


I would love to see whatever material you have in regard to your speaker design. I wanted to make the Linkwitz open baffle, but with that ultra-complex XO and the $$$ Excel drivers I decided to try using the config I'm working on right now.
 

Attachments

  • bds.zip
    490 bytes · Views: 44
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.