I don’t have enough confidence to go and say PE guy did a bad job, but there’s a possibility that It was put together rather in a hurry. I wouldn’t mix together different types of crossovers. Here’s a 3rd order Butterworth for MTM. 13 kHz bump can be notched out with LCR all though the inductor is going to be huge.
Attachments
R-Carpenter said:
The impedance curve of the lower woofer will match the upper. A Zobel could be put on the lower one to trim the peak at resonance slightly, but I don't think it's needed for rising impedance since this driver will be crossed below any serious impedance rise.
Like I said, some BSC. Cross the woofer at 1700 and it's no longer a 2.5.
Adding a coil and hoping for the best is what I do. I have designed several crossovers with software and have always had to change at least one thing. These (in my view) are just a good starting point.
R-Carpenter said:
Using different crossover alignments are done all the time. I think the PE design is good to start with. Trying to improve it without hearing it is not the best approach.
Fantfool, you'd be doing yourself a favour by learning everything you can about crossovers and how to tweak them to get the sound you want.
EDIT: I wouldn't worry about that 13k bump. Simulation, remember.
R-Carpenter said:
No.
It is the overall acoustic response and phase which has to meet the target. The electronics used to get there are almost irrelevant in the end (if you take my meaning).
My first thoughts on the PE crossover (without having tried it) is that it appears to have been done by an amateur.
In my humble opinion, simulations are very accurate then it gets higher frequencies. As you go lower, more problems will arise with BSC, box resonance, standing waves and room response. I agree that simulation is a good starting point and in the end, the crossover may have to be tuned up by ear. I have had success with just simulations then the speaker would sound great, no tweaking necessary. Then I also did some tweaking in other instances.
13khz bump may be audible, perhaps not, my suggestion would be in this case to notch it, but again it has to be listened to or measured. LCR can be added after x-over is together, all I am saying, it could be fixed.
To my knowledge, the orders (not the alignments) of crossovers are mixed all the time, which may cause phase more phase variations.
Don’t forget that PE guy never heard this drivers, he just WT3 it.
I guess Fant has picked the way he wants to go in prime, MTM or 2.5.
It would be interesting to do both and compare. His center pretty much has to be MTM to get moderately good dispersion and sides could be 2.5
13khz bump may be audible, perhaps not, my suggestion would be in this case to notch it, but again it has to be listened to or measured. LCR can be added after x-over is together, all I am saying, it could be fixed.
To my knowledge, the orders (not the alignments) of crossovers are mixed all the time, which may cause phase more phase variations.
Don’t forget that PE guy never heard this drivers, he just WT3 it.
I guess Fant has picked the way he wants to go in prime, MTM or 2.5.
It would be interesting to do both and compare. His center pretty much has to be MTM to get moderately good dispersion and sides could be 2.5
Yes I would really like to try the mains both 2-way and 2.5 way and see which sounds better. It would be an interesting experiment. But, if I go 2.5 way on the mains, and 2-way on the center, am I sacrificing the HT aspect of the set up?
Plus, I end up with more crossovers sitting around. I already have 3 now from the Zetag guy, and I will be replacing the 3 existing ones so that makes 6. Looks like lots of spare parts!
R-Carpenter said:
Somehow I missed the fact that the highpass is a Chebychev. I wouldn't use that alignment. Also this tweeter will cross lower than 2.5K. 2k would be better I think.
I don't have a problem with mixed alignments and slopes. Do what works best, keeping an eye on phase. Still, I think I'd try a 2nd order LR for the midwoofer and tweeter, 1st order BW on the woofer (if going with 2.5 way).
fantfool said:
Do you really think you could differentiate the difference audibly between 2.5 way mains and 2 way centre? You're one up on me if you can.
My centre speaker uses an entirely different driver than my mains. Plus my mains are 3 way active and my centre is 2 way passive. Sounds splendid.
It's a good thing to have some spare crossover parts laying around. How else will you tweak?
jnb said:
Does the D'Appolito completely eliminate lobing? Just a third order BW, no? Centre MTM could also be configured 2.5way without problems.
I have never built an MTM before. I'm not really attracted to the way they look and the ones I've heard haven't knocked my socks off.
I used a Seas 7" coax for my centre, and I can say it has good dispersion, good off axis response and, of course, no lobing.
New curve from PE guy. I talked to him and pointed out I thought he was using the wrong tweeter curve and that the 27TDFC could be crossed over lower. He re-did the curve and this is the result. Definitely better, I kind of think this is a starting point I can live with. I still wish the lower part of the woofer curve could be flatter but we'll see how this sounds to start with.
I did talk to Madisound today, they said they only do that kind of testing if you buy the drivers from them. He basically said forget sending him the woofer because it would be far too expensive for them to test it the way they do. He said the testing process is expensive "as well it should be". I told him the tweeters I have they sell, but he just said "by the time you pay us to do the test, you could buy all new woofers". But I did talk to him about the Seas tweeter and he said it was definately one of their best.
I asked the PE guy why he chose that allignment and he said because it allowed him to get the response the flattest. I read up on it a little this afternoon and the reviews seemed mixed, kind of what someone else said earlier about whatever allignment works. There was some talk about it being a "ripple effect" but I don't see that in the curve. I do think it looks better, defininately in the 1K area, and with the new tweeter curve it is better in the high end as well. You know somebody asked me earlier this evening about the box type and it occurred to me that the curve I am looking at doesn't take into account the low end I will get from the box, so it may be better then I first thought.
I asked the PE guy why he chose that allignment and he said because it allowed him to get the response the flattest. I read up on it a little this afternoon and the reviews seemed mixed, kind of what someone else said earlier about whatever allignment works. There was some talk about it being a "ripple effect" but I don't see that in the curve. I do think it looks better, defininately in the 1K area, and with the new tweeter curve it is better in the high end as well. You know somebody asked me earlier this evening about the box type and it occurred to me that the curve I am looking at doesn't take into account the low end I will get from the box, so it may be better then I first thought.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.