Crossover design - critical resistor values - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

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 15th September 2012, 10:31 AM   #1
diyAudio Member
 
Join Date: May 2011
Default Crossover design - critical resistor values

My latest speaker design is well underway. The drive units, SEAS L18 and SEAS 22TAFG are in temporary cabinets fully run in. The crossover is a 2nd order (electrical) with equalisation for some room induced midrange lift and a notch for the L18. The whole idea for me behind home speaker design is to get speakers to sound good in my listening room - commercial speakers tend to sound terrible... Anyway, the BSC inductor is an oversized 3.5mH in parallel with a resistor - sorts out BSC as well as the room 'bump' giving as flat as an in room response as I could hope for. Well, almost...

The problem I have is the parallel resistor. Using standard values, 12R gives a sound which is a shade too bright, and 15R is a shade too warm. Only when I choose a mid point of 13R5 is the sound 'right'.

It isn't difficult to get this size but I am sort of suprised at how delicate the balance is between the various sizes tried. Still, I have a great sounding speaker now so if a bit of fine tweaking gets the best out of them, that's all that matters
  Reply With Quote
Old 15th September 2012, 10:54 AM   #2
tinitus is offline tinitus  Europe
diyAudio Moderator R.I.P.
 
Join Date: Nov 2005
could be your inductor is too big, since your resistors are so small, and 'touchy'

using a paralel resistor on the inductor does work, but is a compromise introducing minor errors
it works, but compromised
I guess thats your struggle right now
  Reply With Quote
Old 15th September 2012, 01:34 PM   #3
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
You can get exactly correct resistors by using multiples in series or parallel. As this is for BSC, you can even use a power wire-wound adjustable resistor. A little inductance here won't hurt. The range you are talking is about where I would expect. In the 4 to 5 dB range.

Is your BSC in series with just the woofer so you have to adjust the teeter independently, or is it isn series with both drivers? If both, then it is really touchy as you are adjusting both the depth of the step for the woofer and the pad of the tweeter at once. I put the BSC only on the woofer and adjust till the hump goes away. Then I set the tweeter level.

The mid-range hump is not the room. It is as you have discovered, getting the BSC correct in both first pole and in depth. It can be quite sensitive. The larger the edge radius on the box the easier.

I finally succeed in combining a third order LP with the BSC by increasing the first inductor. It wound up about 20% higher than calculated. In the end, I was within 1 dB 100 to 18K. ( Dayton RS and Seas soft dome)

You have the right idea, measure to close and for hints, then voice by ear in place. So many people forget this.

As those are quite nice drivers, would you mind publishing your final crossover and baffle dimensions?
  Reply With Quote
Old 15th September 2012, 04:16 PM   #4
diyAudio Member
 
Join Date: May 2011
Yes, I'm afraid the mid hump is from the room. My previous speakers include an original pair of Musical Fidelity MC4s which required strong equalisation (-6db at 500hz on an equaliser) and currently I have a second (low-fi) system in another room using KEF iQ5s - they sound awful without any mid EQ (and not so good with it tbh).

My previous build was using the MC4 cabinets (my cabinet making skills were poor back then) with a SEAS L22 RNX/P, Morel EM1308 and SEAS 22TAFG. They sounded great with loads of detail, but the mid was a bit too polite. The new build is a smaller floor standing speaker. Deep bass isn't an issue as I have a Peerless XXLS 12" powered by a BK300 amp.

Any other size of inductor (I've tried a few) starts the EQ at the wrong frequency. The combination of inductor and resistor has to flatten the mid hump, provide BSC and provide the correct roll off in combination with the 15mfd cap. Fiddling with the cap spoils the roll-off. Tricky, which is why I'm stuck with 13R5...

I could use a mid notch to provide the EQ, but that would mean re-designing the xo from scratch. I'm not happy with this solution.

Using electronics (equaliser) would work with a lesser system, but destroys the dynamics of my Linn Karik/Naim NAC72/NAP250 as well as not providing a good match for the 'bump'.

Listening to them, the differences in resistor sizes are only slight - I suspect most would be happy with either 12R or 15R depending on personal taste.

The front baffle is 850mm x236mm, and the cabinet is 250mm deep. The cabinet is double skinned so internally it's only around 20ltr. Crossovers are kept separate and the speaker is bi-wired. 2nd order wired in phase - bass is 3.5mH in parallel with 13R5 and the cap is 15mfd; tweeter has 12R on the input side, with a 10mfd cap and 0.3mH inductor. Roll-offs are nice and clean with a xo at 1850hz.

I'll be documenting the build fully once I get going, along with my failed attempt at using SEAS 27TBFCGs. I'm sure some are wondering why I'm using the 22TAFG so low when the L18/27TBFCG combination is so popular It's all to do with measured v subjective qualities of the two tweeters.
  Reply With Quote
Old 16th September 2012, 01:31 AM   #5
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
Hmmm. Never seen a room produce a mid hump like that. I have some Q1's and their midrange sounds awful even with eq. Are the baffle edges radiusd? 500Hz could also be the initial bump in the diffraction.

Love the peerless subs. I run two in 60L sealed.
Just built a pair using the 27TBFC. Easier to use than I expected. Did another pair with the 27TDC. Also good but the metal dome does have some detail the cloth is missing. My crossovers are about 1850 and 2K or there about. I used the ER18RNX in one pair and Dayton RS150 on the other. I really like the ER18. My wife liked the TBFC and I preferred the TDC.

You did not say where in the crossover you placed the BSC network.
  Reply With Quote
Old 16th September 2012, 09:45 AM   #6
diyAudio Member
 
Join Date: May 2011
I think the room problems are down to the material the walls are made of. It's a concrete type full of large pebbles. I get a bump between 400 and 600hz, and the other bane, a very narrow but massive dip at 60hz. These happen regardless of what speaker I'm using. You can see now why I could never use a commercial speaker or someone else's design.

What the inductor/resistor does is start BSC an octave early but at a shallower slope. This introduces a dip in the mid response to compensate for the room hump whilst keeping the upper mid at the right level. What I have now is very little trace of the hump when listening to music and a very even upper mid (with the right size resistor). Apart from moving house, this is a compromise I have to make. Speaker design is not easy...

When I started this design the 27TBFCG looked perfect - low distortion and an ability to go well below 2k. On listening I found strings a little wirey, cymbals 'flat' sounding and very little depth or ambience. Detail was better than a pair of Morel MDT30s I have though.

By comparison the 22TAFG has very sweet sounding strings. Cymbals are more forward, but clean and delicate without any treble emphasis. The sound stage opens up with an ability the hear the space between instruments. Despite tests showing this tweeter has an increase in distortion below 2k, in practice the lower treble is very clean sounding with a very natural sound to piano and vocals. However, given the size of the dome, I wouldn't use these at disco levels - I don't listen loud so power handling is not an issue for me.

Overall I have a very nice speaker which doesn't get in the way of listening to music.
  Reply With Quote

Reply


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
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Please comment on my design -and a few resistor values.. Miniwatt Tubes / Valves 13 16th July 2010 10:14 PM
no resistor values........ pointy Tubes / Valves 19 19th December 2009 12:23 AM
Critical Q design theory buggsson Multi-Way 6 28th January 2008 01:24 AM
leap design crossover values scottericsonon Multi-Way 9 26th November 2007 09:58 PM
Resistor values Stocker Chip Amps 3 4th May 2004 04:10 PM


New To Site? Need Help?

All times are GMT. The time now is 01:18 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2