seas thor crossover redesign help

[jimangie1973]

Hey Bill,

What did you end up for values of the tweeter resistors and shunt cap?

Jim

[Bill Fuss]

Hi Jim,
The main series res is 1.22, the 2nd series res is 2.2, and the shunt is 8 ohms with a 4uf cap, just as you simmed. I haven't considered any changes since the fix except possibly slightly less BSC but I would need some directions for that. I do have the capability to measure inductance if that would be a help.
Bill

[tktran]

Sorry to ask, but what's a right angle drill, sawsall or sabre saw?

regards.

[Bill Fuss]

Hi tktran,
A right angle drill has a 90* gearbox instead of an inline gearbox so it will fit into much smaller spaces. A sawsall is a contractors reciprocating saw that can use a blade as long as 10 or 12 inches [30 cm]. A sabre saw is a compact right angle version of a reciprocating saw, also called a jig saw. They are all quite common, if you GOOGLE them you will get thousands of references.
I must say though, using a sawsall with a 12" blade in it isn't for the timid, just take your time and don't get frustrated.

Bill

[JohnyHiFi]

Sweet!

Finally, I had a chance to assemble and install the new crossovers! I've let them 'burn in' for the past couple of weeks over which time the sound has definitely smoothed out a little (or perhaps it has just been my ears adjusting). Regardless, I really like the tuning of this new crossover. The first thing I noticed was an increase in the precision of the imaging. I thought my Thors were already quite good at creating a wide and precise soundstage so the increase was quite the surprise. I also noticed how much smoother and non-fatiguing my Thors are now. I dont know if Id say that the new crossovers have increased the bass response other than the fact that with the more forgiving tonal balance, the midbass is no longer over-shadowed by the midrange. What Ive also found interesting is how much more exciting my speakers now sound but without sounding over-etched. These crossovers will definitely be staying! Thanks Jim for sharing your design!

Now if I can only find the time to build a set of small thor cabs :cry:

[Renron]

I'm a regular over at the other DIY (audio) site, and have built a pair or curved small thor' s which sound quite detailed but a bit hot on the top end which is made worse by my bare wooden floors. Hence, I've decided to rebuild my existing original X-overs.

A couple of questions to those who have built Jim's new X-overs:
1) Which Capacitors (brand and model) did you use?
1a) And are you happy with them?
1b) Anybody use the Obbligato films?
2) On the woofer (midbass) circuit, the high value caps are expensive, since they are not directly in series with the drivers (they go to ground) is their quality as important ( as the tweeter's caps which are in series?

Thank you for your efforts, we have awesome speakers don't we?
Ron

[tktran]

Hi Jim,

Did you receive my last few emails from my Yahoo account?

best regards,
Thanh.

[Bill Fuss]

Hi Renron,
I'm reading that you are going to try Jim's XOs. They are a totally different sound, and so tweakable on the top end that you might want to keep access to them by mounting them in a separate box. All the tweaks are listed in this thread, and are worth exploring. I'm not going to get into the merits of different brands of caps, just use good quality polypropyline caps.
I'm pretty happy with my ported cabs, I imagine the Smalls kick some serious butt.

Bill

[JCC]

Lots of Questions

I have been lurking here for a while now and find this an interesting thread. I built Thors a while back. They are a bit bass shy, which I think can be cured with a better cabinet design, but my biggest problem is the sharpness in the lower treble/upper midrange. I don't have measuring equipment, but I suspect this has to do with the woofer's resonance in the 4.8 khz region as others have suggested. After reading about several ways to tackle this problem I have a number of questions before I spend what it will cost to completely redo the crossovers.

Jim's new crossover looks like it addresses the 4.8khz problem but his original plots also look like there is decreased output in the 200-1000 hz region compared with the original crossovers with a pronounced dip at about 350 hz. Can anyone comment on this?

One of Jim's earlier posts also said that in addition to not having a deep enough notch, the original crossover had the notch at 5.5 khz instead of 4.8khz. (Post 41). But in an even earlier post, Jim's graphs look like the notch in both crossovers are at exactly the same frequency although the notch created by Jim's crossovers is much deeper and starts rolling off sooner. See post 24, second graph. What am I missing?

Also, the original audio express article indicates that the depth of the notch in the woofer's response is controlled by the value of R1 in the shunt in the woofer circuit. Has anyone tried just reducing the value of that resistor to deepen the notch? That seems like the most obvious first thing to try, but I have not seem anyone suggest it so I must be missing something.

Finally, most of the other suggestions for dealing with the speaker's sharpness seem to focus on altering the high pass filter to modify the tweeter's output, which I don't really understand if the fundamental problem is with the woofer's resonance peak at 4.8 khz. For example, I have collected the following suggestions (quoted from other posts):

"I did put a 1ohm resistor between the caps and the inductor on the high pass filter which helped tame the midrange a little (thanks for the advice Dennis)."

"I have finally managed to tame the Thor a little and get it to stop 'shouting'. Here is what I've done so far (all changes were done to the HPF):
- removed the 2.2uf cap
- added a 1ohm resistor in front of the inductor
- changed the l-pad values to 2.4ohm in series and an 8.2 ohm in parallel."

"He said something about a tweeter which doesn't work properly in the area around 4000 Hertz and changed two capacitors (C2 8,2mF to 5,6mF and C3 18mF to 15mF)
According to him the Thor was working well after that."

"my mod is quite cheap only consisting of one capacitor , coil and resistor . you add these up in series with each other the resistor on the end , The values I have at the moment is 5,6 µF capacitor and 0,22mH coil the resistor is atm 12.1 ohm I had it set to 4,5 khz before but changed it to 3,8 khz wich gives more holographic soundstage yeat smooth.
You connect this circuit in paralell with the tweeter ,
Make sure to put the mod after the orginal XO directly to the tweeter ( in paralell) .
If you have som small Capacitors laying around you can try to paralell it to the 5,6 µf and you will then hear that the damping will move to higher frequency 8,6µf will make the filter set to 4,45 khz , Adapt the level with the resistor I found 12-12-4 ohm something to be quite good."

Can anyone comment on these other proposed modifications, how well they work and why they focus on the tweeter?

Thanks.

[jimangie1973]

Jim's new crossover looks like it addresses the 4.8khz problem but his original plots also look like there is decreased output in the 200-1000 hz region compared with the original crossovers with a pronounced dip at about 350 hz. Can anyone comment on this?

Ignore the frequency response plots below 400 Hz or so. I don’t have an anechoic chamber so I have to use a gated noise method to eliminate room reflections from affecting the measurements. The crossover was designed with full 6dB baffle step compensation so there will definitely be enough low end, maybe too much if the speakers are too close to walls.

One of Jim's earlier posts also said that in addition to not having a deep enough notch, the original crossover had the notch at 5.5 khz instead of 4.8khz. (Post 41). But in an even earlier post, Jim's graphs look like the notch in both crossovers are at exactly the same frequency although the notch created by Jim's crossovers is much deeper and starts rolling off sooner. See post 24, second graph. What am I missing?

I wasn’t talking about the Thor crossover in that post. That was the original crossover in the commercial speakers I bought used. I didn’t like the sound, so I tried the Thor crossover, didn’t like that either, so I redesigned a crossover myself.

Also, the original audio express article indicates that the depth of the notch in the woofer's response is controlled by the value of R1 in the shunt in the woofer circuit. Has anyone tried just reducing the value of that resistor to deepen the notch? That seems like the most obvious first thing to try, but I have not seem anyone suggest it so I must be missing something.

The problem with that is when you change the R1 value, the slope of the notch changes so that the frequency response below the 4.8 kHz notch frequency is adversely affected. Specifically, the notch becomes deeper and narrower.

Finally, most of the other suggestions for dealing with the speaker's sharpness seem to focus on altering the high pass filter to modify the tweeter's output, which I don't really understand if the fundamental problem is with the woofer's resonance peak at 4.8 khz. For example, I have collected the following suggestions (quoted from other posts): ""

The fundamental problem isn't the woofer resonance. That is an issue but the uneven power response due to the tweeter circuit is much more audible. Fixing the tweeter circuit to reduce the lower treble energy is the easiest and cheapest way to get substantially better sound. The easiest thing to do is remove the 2.2 uF cap in the tweeter circuit. If it’s still too harsh, then lower the 8.2 uF cap. Try 6.2uf, 6.8uf, 7.5uf, etc. That will help substantially.

Edit: After thinking about it some more, I would not lower that 8.0uF cap because it will push the 2.5 kHz crossover frequency even higher. So many problems with that crossover...

The other major problem I had with the Thor crossover was there wasn’t nearly enough baffle step compensation. The 1mH inductor is just too low IMO.

[KarlXII]

Hi Jim,

one year coming up soon, listening to the Thors.
I'm still very happy and just wanted to extend a thanks to you for your work. :T

[Renron]

Time to add something new to this old thread.
Sometimes it takes longer than we think to change something as simple as a crossover.
I'ts been 1 1/2 years now, no sense rushing into things. :roll:
I had been using a DIY cat 5e speaker cable (7 braided) but wanted to try something new. I bought an AudioQuest Gibraltar which has separate solid core wires for bass and tweeters. So I split the crossover into 2 crossovers which only share the negative line at the amplifier, not at the crossover.
I've only got one side finished now, but I'll report my review after they both are installed. The caps were burnt in for over 1 week each.
There is a Lexan cover on the box that is very hard to see because it is so clear, but it's there.
Ron

[Renron]

Finished both today and had my first listen.
First impression is full sound from instruments and good voicing of female vocalists. The speed and attack of instruments has increased but that could be due (in part) to changing speaker cables at the same time. They are now single to bi-wired AQ Gibraltars. Different gauges of wire for Bass and Treble. They are NOT broken in yet.
The high tartness/sharpness that was there with the original (premium component) crossovers is gone and a crisp high end has taken it's place. Siblance is lessened somewhat as well.
Bass is full without being boomy, this could be due to the addition of ~1/4 cubic foot of Dacron stuffing. (pillow guts) or the new brand of capacitor in the crossover. Only time will tell with the further break-in of the caps/crossovers. But I like the new sound better.
Midrange sparkle and presence which was the hallmark of the original crossovers is gone! What the H3LL happened??? Has anyone else experienced this? It could be the driver tubes in my Preamp as they are new to me and I don't know their sound. (RCA 1950s grey glass 6SN7s) Tomorrow I'll change out those tube for a different set and see if they are the thief that stole my midrange......I may just eliminate the preamp and use my reciever as a pre-out. PITA to take my equipment out and shuffle components around. Ya know, you hate doing it too.....

In my instance I had to change many items at once so it is difficult, at best, to be certain which component has changed the sound. I know that this is not the way to do comparissions but it was unavoidable. I changed the CDP to the new OPPO Blue-ray BDP-83. Very cool player.

I'll change some stuff back and test then report later. I'm open to ideas about the midrange if anyone has any....

Ron

[Johnloudb]

Crossover components have a tolerance for error, like +/- 10 percent on caps and 1-5% on resistors. So it's likely to sound different. Could be that you wired something wrong? Maybe your inductors don't have the right DCR value? Did you change all the crossover components or just the caps?

[Renron]

John, thank you for responding.
I measured the Obbligato capacitors prior to building the new crossover as designed by Jmangie. The Obbligatos were shocking close to their claimed value, within +/- .1 uF! It certainly could be the caps have not yet burnt in completely. I used a DIY cap burner I found on DiyAudio forums........I have no idea if it worked as intended.
Resistors were measured also and were Mills wirewound non-inductive, know for their close tolerances also.
Inductors were also measured----------but here's the kicker, I ordered the .05mH coils from Madisound and they sent 2 marked as .05mH, but they measured .025mH . In my bleak understanding this small difference should not make much of a difference in the sound. JMangie, what do you think?

No doubt I could have wired something incorrectly, I've done it before! I'll double check my wiring today.

I eliminated the Preamp today and the midrange is still MIA.
Another way of explaining is the midrange is set MUCH deeper into the soundstage, recessed if you will, and not "in your face" as with the D'Appolito crossover. The overall sound is mellow rather than detailed. Metal cone speakers are know for being detailed........... I wish I understood the magic of crossovers.........
I'll change out the CDP today and see if that is the midrange thief.

Ron

[Renron]

One more thing I noticed as a difference between the two crossovers, I lost about quite a bit of volume (loudness) when I changed to the newer crossover. Anybody else notice the difference? Volume knob used to be at 1/4 for normal stereo listening, now I have to set it at ~ 7/16 for the same volume (db). What am I missing.

Ron

To answer my own question.....I re-read Jim's post #102..........
"The downside is slightly more amplifier power is needed to get the same volume level."

So much information, so much to remember.....

[Renron]

Jim,
If I change the 4uF // cap in the tweeter circuit to a 3.3uF, will that increase the tweeter output energy (db)? I think I read that in one of your posts on the second page #36.
Also, you advised Karl to try different size resistors in that same // position, would it be preferable to change / experiment with resistors or cap values?

Xovers are a mystery to me.

Thanks for the help
Ron

[Renron]

:hb arty: :hb :banana: :2fingers: :beer: :clap: :dance: ;h :dothewave: :dancenana: :^x


I did some surgery on the external crossovers after reading through this entire thread twice today.
1st, I changed the 4 uF parallel cap in the tweeter circuit to a 3.3 uF. That helped quite a bit.
2nd, I added a .6 ohm resistor to the 8 ohm in the same // location, making it 8.6 ohms.

This brought up the treble to the same level as the bass. I have the small Thors which already has improved bass response over the original cabinets, along with the new crossovers which also increase bass response, the combo was overwhelming and overshadowed the midrange and tweeter.

After I changed the above RC values everything fell into place and the overall sound became full, clear and lush. Very much like adding a tube preamp would.
I have my midrange and tweeter back! And without the shouting that was there previously with the original premium Xovers.
These speakers have risen to a new level of premium performance and have far exceeded my hopes for a DIY project.

This Crossover along with the redesigned cabinets make this a completely different (and superior) speaker than was designed by D'Appolito.

It deserves a new name to honor those who have changed this into one of the finest DIY speakers available today.
Planet 10 / ScottMoose / Jmangie.............My hat is off to you all and many thanks for your hard work which I enjoy daily,

Respectfully crossover tweaked, ;x(

Ron

[Johnloudb]

Glad you got it worked out. Congratulations!! :T

[hhildebr]

Once again: coil values

Hi all,

old thread, old question, but still..

I am currently building the Small Thor (already crashed the old boxes to put up the pressure ) and just hit this XO thread. Sounds great, one question though:

Jim, the coil values over here in Europe follow the numbering 0.27, 0.33 and 0.56, 0.68 - it's difficult for me to get hold of the 0.3 and 0.6 values in the tweeter branch. Could I use one of "ours"? Which? Also, would I choose one lower and one higher (eg 0.27 and 0.68)? Also, you usually get 0,71mm (~22 AWG) and 1,0mm (~18 AWG) over here. I guess the important bit is the DRC value - how much variation is tolerable?

It's always the same - you help some people, and they keep asking for more :W ...

Thanks so much for your help,
Hauke

[jimangie1973]

Use the LR4 crossover shown here:

http://www.diyaudio.com/forums/showthread.php?t=142615&highlight=short+thor

The schematic is shown about 5 posts down from the top.

Also, just use a 4 Ohm in series with the tweeter instead of the 8 parallel 9.

Jim

[hhildebr]

Hi Jim,

thanks for the fast response! I might try this new XO. However, looking at the schematic (I don't know a lot about XOs...) the topology looks pretty similar to Hobby Hifi's Evo04 variant (which I already have). The values are of course somewhat different, as is the crossover frequency (around 2500Hz with Evo04, it seems to around 2kHz with your XO). Have you ever heard the Evo04 XO? Can you compare the two?

Evo04's values are:

- 5,6 µF and 15 µF instead of 5,6 and 18 in series with the tweeter, 2,2Ohm instead of 4 and 0,27mH parallel to the tweeter.

- just one 1,0mH coils in series with the woofers, with an RLC of 0,68Ohm-0,15mH-8,2µF in parallel.

Thanks so much,
Hauke

[Dennis H]

Copying Jim's latest crossover here for continuity.

[jimangie1973]

Thanks Dennis for the copy.

Have you ever heard the Evo04 XO? Can you compare the two?

Evo04's values are:

- 5,6 µF and 15 µF instead of 5,6 and 18 in series with the tweeter, 2,2Ohm instead of 4 and 0,27mH parallel to the tweeter.

- just one 1,0mH coils in series with the woofers, with an RLC of 0,68Ohm-0,15mH-8,2µF in parallel.

No, I've never heard the Evo04, but it is fairly similar to the original Thor/Odin crossover. The woofer circuit is identical and the tweeter circuit has a couple mods to reduce the low treble.

My LR4 crossover has substantially more baffle step correction. The second coil in series with the woofer does nothing more than shape the baffle step to the optimal curve. The LC notch I use is different as well to optimize the response. On the tweeter side, my LR4 has a higher value shunt inductor which brings the tweeter response lower in frequency. The XO point is just below 2 kHz on-axis, and slightly lower off axis.

My LR4 will not sound at all like the Evo04. It sounds night and day different from the original Thor/Odin XO. I really would go with the my LR4 over my original LR6 steep slope design.

Jim

[hhildebr]

OK, the LR4 XO it be. One final question, though (I promise ):

Since I've already got the Evo04 with a 15µF Mundorf Supreme cap I would prefer to reuse them by adding a 2,7 or 3,3µ (maybe a Clarity SA). Would that difference hurt? Better go for 17,7 or 18,3? Alternatively I have two spare 10µFs which I could combine with an 8,2µF.

Hauke

[jimangie1973]

OK, the LR4 XO it be. One final question, though (I promise ):

Since I've already got the Evo04 with a 15µF Mundorf Supreme cap I would prefer to reuse them by adding a 2,7 or 3,3µ (maybe a Clarity SA). Would that difference hurt? Better go for 17,7 or 18,3? Alternatively I have two spare 10µFs which I could combine with an 8,2µF.

Hauke

The response is not overly sensitive to the value of that cap. I'd go with the 17.7uF value. Also, the response in the lower treble is changes audibly with small changes to the 5.7uF tweeter cap. Try to get a few combinations between 5.6 and 6 uF and use what you like best. The higher cap value will raise the lower treble (2-3 kHz region) slightly.

One final thing, you're baffle is 9" wide, correct? My crossover was developed with a 10" baffle. I will determine what the slight difference is in the baffle step and we can modify the 2.5Ohm resister if necessary. This is not a big deal. Have a 2 Ohm and 3 Ohm on hand if necessary.

Jim

[hhildebr]

Actually, I'm at 9,5 inch - I am lining up the side panels with ceramic tiles on the inside to increase mass, so I had to make the front a little wider. Then again, I was planning to round off the corners, so the effective baffle width might be closer to 9 than ten.

Now that you mention resistors... we have values of 2,2 or 2,7 Ohm.

Btw, I have decided to leave the 15µFs in the old XO and use the spare 10µF Mundorfs for the new one. That way I will have the chance to make an A-B comparison between the two XOs, which will be interesting!

Hauke

[Renron]

Hauke,
I have used Clarity SA caps after reading the reviews from "the net", and they aer very good caps IMO. The mix of Mundorf and Clarity might be great!

I have used the Obbligatos on this new XO build, and have to say I like them better then Clarity Caps. Their film and foil are very rich and detailed.

Good luck with your build and show us some photos when your finished

Ron

[avddreamr]

Curiously, now that the "problems" with the original thor, and odin have been dealt with to a degree that one would never expect given the nature of a published and "regarded" hifi-diy design, has anyone given thought as to adapting to one of these crossovers for the odin for center channel use?
I would guess that making the tweeter offset, and even sealing the odin would be a good start, before crossover tweaking would begin. Since there seems to be alot of builders that used the odin on its side as a center channel with some degree of success. Or even going MT in the center with a w15 varient and flank it with the w18s.... but at that point you may even be inclined to ditch the millennium for either a less expensive tweeter that provides similar level of performance, or even something in the same price range that reminds you that the millennium tweeter is coming up on a decade since it was designed.

My own experience with using the as designed kits, was pretty good, except for the improper use of the odin as a center. I was sadly, never satisfied with the bass output of the original thors, but using them above 70hz was more than satisfactory.

I have been lurking in these threads since inception, since I really would like build thors and odin again.. but this time with the improved designs. I had a flood some years ago that damaged the cabinets of the thors, so I ended up going back to a commercial speaker that I never got around to selling. But my interest is peaked again, and the speakers need to do something other than occupy space in my closet.

[benthe8track]

Curiously, now that the "problems" with the original thor, and odin have been dealt with to a degree that one would never expect given the nature of a published and "regarded" hifi-diy design, has anyone given thought as to adapting to one of these crossovers for the odin for center channel use?
I would guess that making the tweeter offset, and even sealing the odin would be a good start, before crossover tweaking would begin. Since there seems to be alot of builders that used the odin on its side as a center channel with some degree of success. Or even going MT in the center with a w15 varient and flank it with the w18s.... but at that point you may even be inclined to ditch the millennium for either a less expensive tweeter that provides similar level of performance, or even something in the same price range that reminds you that the millennium tweeter is coming up on a decade since it was designed.

My own experience with using the as designed kits, was pretty good, except for the improper use of the odin as a center. I was sadly, never satisfied with the bass output of the original thors, but using them above 70hz was more than satisfactory.

I have been lurking in these threads since inception, since I really would like build thors and odin again.. but this time with the improved designs. I had a flood some years ago that damaged the cabinets of the thors, so I ended up going back to a commercial speaker that I never got around to selling. But my interest is peaked again, and the speakers need to do something other than occupy space in my closet.

Sell me the drivers if they are just collecting dust

I've been planning on building Jim's Thor Xover in the "small" box for almost 2 years now. Things just keep coming up that take priority. I too would love to see a center option for the Thors, however, my impression is that most guys with Thors used them exclusively for 2 channel and had another setup for their home theaters.

[avddreamr]

Sell me the drivers if they are just collecting dust

I've been planning on building Jim's Thor Xover in the "small" box for almost 2 years now. Things just keep coming up that take priority. I too would love to see a center option for the Thors, however, my impression is that most guys with Thors used them exclusively for 2 channel and had another setup for their home theaters.

I'll probably sell them if i don't get around to doing something with them over winter break. I just don't have the time to build cabinets right now. That and I sold my table saw. I have to either find someone to build the cabs, or buy some more tools.

Regarding the center channel, maybe we can pool a bunch of builders to pay for the center channel design.

How many people do you think that built the thors? I mean if we each pay 20- 50 dollars, I think we can easily get enough money to compensate a competent xover designer for his time. I wonder if this is feasible.

Hmm...

[benthe8track]

I'll probably sell them if i don't get around to doing something with them over winter break. I just don't have the time to build cabinets right now. That and I sold my table saw. I have to either find someone to build the cabs, or buy some more tools.

Regarding the center channel, maybe we can pool a bunch of builders to pay for the center channel design.

How many people do you think that built the thors? I mean if we each pay 20- 50 dollars, I think we can easily get enough money to compensate a competent xover designer for his time. I wonder if this is feasible.

Hmm...

Yeah let me know, I'll take them off your hands for sure.

I'd chip in some cash if someone reputable was interested in taking on that project. I actually have everything one needs to build the crossovers including an anechoic chamber to do the testing in but that's too high end of a design to practice on.

[grakz]

I have a built pair of Odins with the original Xovers and cabinet. They are a little harsh on the treble and the effeciency is a bit to good. Seems this xover design might help with that. Where have you ordered the parts from? Doesnt seem like i can get the correct parts here in Norway...

[jimangie1973]

Which parts can't you get? Also, what is the closest you can get. Let me know and I'll tell you what to do. Jim

[JCC]

LR4 crossover

I just built Jim's LR4 crossover and have been comparing it to the original Thor crossover. It is a very different sound. The original had thin bass and a harshness in the 4-5000 hz region. However, it was very clear and articulate. The new crossover seems to decrease the sensitivity of the speaker quite a bit. It cures the harshness and there is subjectively more bass. However, the bass now seems bloated and the clarity and articulation of the original is missing. It seems to fix the problems with the original but goes too far the other direction and is really muffled.

I could use some tuning advice. I have some slight differences in component values from what Jim specified. Instead of a 4 ohm resistor I am using a 3.9. and instead of a 2.5 ohm resistor I am using a 2.7. Also for the .7 inductor I have a .68.

Suggestions?

[grakz]

Have ordered parts for the xover from Parts-express.com.

[jimangie1973]

JC,

How old are the drivers you're using? Mine are from about 2006 with the rubber boot. I have no idea how the drivers have changed over the years although Zaph has mentioned that the breakup of the W18 has drifted.

Also, as Ron mentioned, how close to the back wall are they set up? The crossover implements close to a full baffle step so if you're getting reinforcement from the walls, it will sound bass heavy. Try plugging up the port (or play with the stuffing amount) of the TL and see how that effects the bass. I run mine sealed and there is no bass bloat whatsoever. I even run with a 12 AWG 1.2 mH inductor which slightly increases the relative bass level.

If you want more midrange presence, raise the 5.7uF cap to 6.2 or even 6.8 uF. You can also reduce the 2.5 Ohm resistor in parallel with the 0.7 mH inductor to reduce it's influence on the baffle step. You could try 2.2 or even 2 Ohm. Finally, you can reduce the 4 Ohm resistor slightly more to raise the treble level (if you wish). These mods have minimal impact on the driver integration.

Good luck,
Jim

[JCC]

Thanks for the reply Jim. My drivers are about the same vintage as yours.

I have ordered the parts to make changes and will let you know what happens. More stuffing might help. Mine are fairly lightly stuffed because with the original cabinet design too much stuffing attenuated the bass when the speakers were already a bit bass shy.

As I understand the baffle step compensation, the first inductor rolls off the high frequencies, the second inductor rolls them off more and the resistor in parallel reduces the effect of the second inductor, thereby compensating for the fact that the lower frequencies are more omnidirectional and do not reflect off the baffle. Is that basically correct? What frequency range will be affected by reducing the resistor?

I suspect increasing the 5.7 uf capacitor may help. Not sure I need to lower the 4 ohm resistor as the upper treble (cymbals and such) seem bright enough.

The rear of the speakers are about 2 ft from the wall right now but I don't think my wife will let me keep them there.

Are you using the original cabinet design or one of the small/short/fat versions?

[JCC]

Mea Culpa

I can't believe I did this but I had the .05 and the .45 inductors swapped. No wonder the new crossovers didn't sound right.

Now that I have them assembled correctly they are a definite improvement over the originals. Bass is much improved, as is the upper midrange. They also sound more dynamic, probably because of the different tonal balance. They have a mellower, more laid back feel than the originals which had a very etched upper midrange that would make your ears bleed (well, that's a bit of an exaggeration, but you get the idea). The sound is now more "analogue."

I have a 2.7 ohm resistor in the woofer circuit and I have tried putting an 8 ohm in parallel, which drops it to about 2 ohms total. This reduces the baffle step a bit; maybe too much. I have also played with adding a .47 uf capacitor in parallel wwith the 5.7 uf cap. This gives vocals a little more texture. They sound more etched. Whether that is a good thing seems to depend on the music being played. My inclination right now is to leave it off.

Jim thanks for designing this new crossover. I hope no one shies away from it because I assembled it wrong the first time. It is far better than the original.

[grakz]

Have assembled Jim`s new xover, comparing it to the original Odins xover now. The treble is not near as "upfront" as on the original. Seems to be a bit more bass output also, but by much. My baffle element alignement is different from the original, but same baffle width.

[grakz]

A few pics of one of the assembled xovers. Does the coils polarity have any difference? Have used the outer end as plus, and inner end as minus.

[jimangie1973]

A few pics of one of the assembled xovers. Does the coils polarity have any difference? Have used the outer end as plus, and inner end as minus.

No, the coil direction does not matter. Your assembly looks fine.

What are the resistor values? Looks like you are doing the 8 parallel 9. If there's not enough treble, you can lower this to 4 or slightly lower.

[grakz]

the resistors are 8 in paralell with 9 (4+5). Will get to test with both of the new xovers later tonight

[grakz]

There is too much treble, think i need to lower the treble with a few db. The midrange seems very good. If i raise the resistor to 5 ohm...?

[grakz]

Damped the treble with a resistor in series and another in paralell with the loudspeaker cable. Now the treble level is playing well with the midbass.
Damped the treble with 6db. A 3,3 ohm R in serie, and 5,6 ohm R i paralell.

Used this scheme: