x-over experiments, diff. orders etc.

[SimontY]

Hi everyone,

I would like comments on my experiences, as I have been playing around with my x-overs and I'd like someone to help give me ideas for experiments, to get better sound. I am a little confused, I'll be honest!

I've never been 100% happy with my speakers (I've whinged on the forum before, and got some ok ideas from its talented members!), so I've fiddled with the x-overs, improving the tweeter series cap and increasing it's resistor value. A recent change to Kimber cabling reduced harshness and glare hugely, I also removed the Neutrik Speakon connectors, and the x-overs sit outide the boxes. But still there is an unpleasantness to the sound...

I should say at this point the x-overs are 2-way, 2nd order, with an impedance notch filter (I think). I guess the x-over frequency is about 2.5kz. The speakers are 2-way d'apolito with Seas NoFerro 900 tweeter and H571 mid/bass. (Audiocom/Intertechnik Pontus kit)

Today I decided to remove the x-over components from the board they were intended to be used with, and wire them up point-point. I thought whilst I was at it, I might experiment, and see what the woofers sounded like with no x-over - different. I also decided to listen to the tweeters with no woofers - and there was the unplesant quality - an intrusive almost ringing effect at the lower end of the spectrum. I plugged in an old tweeter I had lying around - Audax TW034 or something, a 1.25" soft dome, and it sounded much less unpleasant. This upset me a little, as these speakers have cost me a bomb, as they were a kit.

Anyway, I fiddled around some more, and ended up running my tweeters off just the cap, but this was too bright and edgy, so I tried 4uf, this was dull and muffled, so I added a 1uf metal film and 22nf film cap - this sounds nice now!!

So, is it stupid and/or lucky that a 2-way MTM is actually sounding nice with no x-over (except notch) on the bass driver, and 5uf on the tweeter??? A short listen suggests easier to follow vocals, reduced unpleasant artificial weirdness to the sound, sweeter treble, and tighter bass... On the downside some midrange energy seems missing from somewhere, and the parts dont gel together quite as well as they did.

Please feel free to help me understand my findings, ridicule me, or suggest other things to try, now that I have some cool crocdile clip test leads! (which sparked all this off!!)

Thanks a bunch for any input guys!

[michael]

Hi simon, i remember you had been toying with the idea of removing the XO, i personally think that no XO on a woofer can be very very good, but i think you should at some point measure the fr of your speakers without the tweeter, then establish whether or not they roll off smooth enough for no XO, if not i suggest just get a single HIGH quality inductor that will roll them off at say 5Khz so that any of the higher harshness will be reduced. but you need to test. dont know if you are game for this but maybe try adding fibre pads onto the woofers, this may help with harshness up high, dont glue them though, try blu-tac, or double sided tape. Just keep on experimenting.
Regards
Michael

[SimontY]

Originally posted by michael
try adding fibre pads onto the woofers, this may help with harshness up high,

Hi Michael

Interesting idea, buuut, I'm quite convinced that the harshness/unpleasantness is only really coming thru the tweets, as I ran the woofers by themselves and the sound was very easy on my ears, despite the unnatural lack of top-end. This was the case either with the 2nd order x-over or with them running full-range, the main difference being a dulling and some change in the bass with the x-over...

Michael suggested [elsewhere] that I might try (expensive if it goes wrong!) a pair of 27ttfcs(?) as many ppl use these, and this may 'fix' things. I've never heard of anyone using the noferro900, and it isn't even listed on Seas' website (yet?) so I wonder if it's just a crap tweeter. I feel unlucky, but really I guess I'm a fool for buying a silly kit in the first place

BTW, here is the mid/bass:
http://www.seas.no/seas_line/woofers/H571.PDF

[michael]

I would not call it a silly kit, id call it more of a training tool, you have learned alot from tinkering and tweaking it. In College/university people spends heaps of money on textbooks, and they end up throwing them away, at least you have something ncie to keep in the end.
It seems that your tweeter is the only "off" part of the speakers, if you just changed it to that 27TFFC then you could try your XO, your XO with no woofer filters, and if you really wanted to try the Proteus jr XO (i can send it to ya if you want) which come in both 2.5way and MTM 2way. Im sure among these option you would find one that makes you purr like a kitten.
And then you havent wasted any money at all, you just have the tweeters you needed for your surround speakers, and you will have enough XO parts to do lots of cheap experimenting in the future.

[SimontY]

Thanks for reassuring me Michael! I can't wait to get home (from work) and experiement more with my x-over, see what I can get.

The Proteus JR x-overs are probably a nice option to try.

With the tweeters crossed very high, and 1st-order, the harshness is minimal, maybe 'average', though something is missing - it's probably not a very flat response! (interesting that I seem to need no attenuating resistor now, even though the resistance I used before was much greater than the DCR of the bass inductor)

I tell you what, I've learnt loads from this. When I design my next speakers, they will have the tweeter crossed as high as possible, and probably 1st-order too.

Using the tweeters nom. impedance of 6ohms, 5uf should make it cross at about 5.3khz. This is where the woofer is about 2db down, so not a bad guess of what to use

Might a zobel network help with the midrange here?

Is the dispersion all crappy with 1st order and FR mids in MTM?

[phase_accurate]

Did you ever check the phase of your tweeters ? This will heavily affect the FR in the crossover area (notches or humps !).

Is the dispersion all crappy with 1st order and FR mids in MTM?

One advantage of MTM designs is that the vertical dispersion is always symmetrical, independant of crossover type/order.


Regards

Charles

[SimontY]

Originally posted by phase_accurate
Did you ever check the phase of your tweeters ? This will heavily affect the FR in the crossover area (notches or humps !).

One advantage of MTM designs is that the vertical dispersion is always symmetrical, independant of crossover type/order.

Hi Charles,

I've kept the tweeters in phase all the time. I will try out of phase, to see if it's preferable.

It's pleasing to hear the MTM config is not disastrous for this x-over configuration - altho I'd probably know if it was - from the sound!

Any thoughts on zobel or the notch filter? I have the parts for it somewhere, so I'll hopefully try it tonight.

Thanks!

[Andy G]

... It looks like it would be a very good candidate for a simple first order series x-o. possibly using only 4 components !!!

I have used series x-o in MTM's and they seem to work very well.

btw, what information can you provide about the tweeter: do you have curves, Fs sensitivity etc???


http://users.tpg.com.au/users/gradds...ross-overs.htm

[SimontY]

Originally posted by Andy Graddon
... It looks like it would be a very good candidate for a simple first order series x-o. possibly using only 4 components !!!

I have used series x-o in MTM's and they seem to work very well.

btw, what information can you provide about the tweeter: do you have curves, Fs sensitivity etc???

Hi Andy, I looked at your(?) website, it makes for interesting reading, but the spreadsheet link doens't work

Some tweeter specs are:
FS - 900hz
z - 6ohm (I've not measured the R)
Sens - 92db
range is given as 1.8khz>

It's a 25mm soft dome, and as the name implies, uses no ferro-fluid - first guess as to why it can sound harsh.

Do you reckon I could stick with the tweeter's 'normal' x-over and use parallel only on the bass drivers??? Or is that just not possible? Please excuse my ignorance!

[Andy G]

I'll try emailing you.