Home Theater Forum and Systems banner

1 - 3 of 3 Posts

·
Registered
Joined
·
10 Posts
Discussion Starter #1 (Edited)
I have been working with DIY audio for years. I believe my understanding is somewhere in between intermediate to advance in terms of theory. I built my own home theater system and the goal was to stay moderately cheap. I used Dayton reference series for all my drivers. I'm happy with the overall turnout of the project except for a portion of the response which I'll get in to below. My main consideration when designing the system was the RMS power output of my Denon AVR-X4200. I chose the drivers I did because in my 3-way(left and right) and the 2-way(center) need approximately what the Denon puts out. The other factor was cosmetics and materials, I wanted dark gray or black and tend to lean towards the sounds speakers produce when made of paper, Kevlar or carbon fiber. I don't like polyurethane or metals. With tweeters, I like textile only. I designed the crossovers using HT audio's X-over 3 pro. This software takes the drivers TS parameters in to consideration in the design. Attached shows the "theoretical" amplitude response I was supposed to get as well as the phase response. Also in the image is the actual response I got using my calibrated mic and REW.

I attached a picture of the system, a pic of the crossovers themselves, the manuals to the drivers used, and all the graphs from REW and Xoverpro.

Yes, i know i used non-polar electrolytic caps but they were a quarter of the price. Maybe once i fix this issue ill swap them out to see what all the fuss is about. I tried to stick to air core inductors, but for the 6mH an air core was like $35 vs the iron core which was about $18. Honestly the single most expensive component in this build was the assembled crossovers, even with the cheaper components.

The sub i built a few months ago using a different batch of LEDs, i have swapped them out since the attached pic so now they are an exact match.

Now if you look at the graphs, according to X-over pro i should of gotten a theoretical flat overall response but the measurement from REW says different. If you look, from about 1K to 6K there is a dip which is where the crossover between the mid and the tweeter are in both the center and the L&R which utilize different crossover types. Its not a problem with phase, I thought it might be because the sensitivity was lower with the mid but the crossover contains L-pads to accommodate this, plus this issue doesn't exist with the crossover point between the woofer and the mid in the L&R speakers. Im a loss on how to resolve this....Any Thoughts/other software to try?

Also the roll off of the tweeters after 10K is because of the tweeters themselves. I don't really hear anything above 16-17k so its not a big deal to me. Not to mention i usually turn them down (treble) anyways. I may eventually swap them out with better performing tweeters.

I know i can fix this with the Denon's Audysssy features but i want them to have a decently flat response with without processing.
 

Attachments

·
Registered
Joined
·
10 Posts
Discussion Starter #2
Figured it out, it has to do with where the mid and the tweeter cuts off. I overlapped them by 5-6K which is about the bandwidth of the dip. I'm surprised the software doesn't warn me or predict the results.... Back to the drawing board.
 
1 - 3 of 3 Posts
Top