Rerurn to Romy the Cat's Site

Audio Discussions
Topic: Comfy crossing [Re: Prototyping crossovers.]

Page 1 of 1 (9 items)


Posted by Ronnie on 07-23-2006

I too want a monitor speaker to play with, especially for amplifier experimentation.
I put another 4" Fostex in a sealed "box" (globe actually) and crossed it over at 150hz to one SS 25w/8565 on the side. Wishful thinking? -Yes.

About 88-90dB sensitive 2-way and lovely with jazz. Smile
3rd movement of Tchaikovsky 4th sounds like distant humming moskitos. Sad

Time to think "upper bass channel" I suppose, and make it a 3-way speaker.
It needs to have only 90dB and cover 100-1000Hz, but 150-500 might be fine as long as it makes big orchestras sound more like big orchestras!
Any suggestions about drivers (current production and preferrably <$100 each) and boxes?
I've picked up these ideas from around your site:

Driver X in ported cab? (without port-helped bass)
Celestion SL600 woofers
SEAS 6.5" (Wilson CUB)
JBL P650, GTO603, LE5
Focal Utopia 165A
Scan-Speak
15W-8530K
I think those drivers may be unnecessarily bass-capable since your LF towers are crossed low at 63Hz, while I can easily turn mine up a bit?

Technically, I have no idea what to look for in an upper bass driver/box combo.
Grateful for any little push in an interesting direction!


Posted by Romy the Cat on 07-23-2006

Ronny, it is funny because I was juts about to carve my further observations about the Sound of my monitors. Last night I was listening Rigoletto by Callas and Gobbi on my monitor.  I come to an interesting observation that I would like Monitors to sound good but I never thought to have them to sound “serious”. With all beauty that I am getting now from the haled SL600 I would do not compare their capacity to Makondo.  Let be perfectly real about it. In order to make SL600 do not have that very pleasant but still alien tone at upper bass they should be crossed at ~150Hz. At that frequency my LF section (one 21W/8555 driver) does not work well and I end up with no real upper bass. Now I crossing SL600 at ~65Hz and the LF spread to 100Hz and still there is no upper bass in the room. I mean it is very good sound all together but if you flick the switch and let Macondo to kick in then it become kind of funny…

The thickness of the upper bass with SL600++ is there and the only known to me way to deal with it is to make a 2-chanal LF section – separate for upperbass and separate for the lower bass. I know that I will not be doing it and therefore I do not even try to get from SL600 more then it does. You see I was looking for monitors do not do bad things, not necessary to do the things very-well. Frankly speaking the SL600 have already surpass my expectations (after buying into 5 well operating but crappie sounding tweeters!)

The 4" Fostex in a sealed "box" with 25w/8565 should not be a wishful thinking of you cross the Fostex at ~ 700 cycles and add an upperbass channel, so you are on the very correct direction. I do not have a lot of experience making box speakers and I would not be able to advise you. You might try the Scan-Speak drivers, they always were very nice.  The SS 18W-8535 was use in ProAs 2.5 and it did well. My only concern that the rubber suspended SS have very freakish transients at higher frequencies. I would probably go for more “ringy” drivers – some kind of miniature version of Altec 515… I really do not know what to suggest. Since you in Europe you might try to fish in German ebay a pair of 5”-8” original Görlich drivers. They have their cone that it not suppressed by the suspension but do not let them to go all the way down and kill them at 100Hz. 

The Focal Utopia 165A was mystery. They did phenomenally in my car but they did very poor in my monitor. You might also try catching on Ebay an older version of JBL LE8. it was well sounding driver. You might work with it dumb aluminium cone and have difficulties to low-pass it but I would defiantly try it. The Danish company AudioTechnology does very good drivers. I think Bruno for Verity then them and few others. I would defiantly dive in the pound of their drivers if I do an boxed upperbass.

The Scan-Speak 15W-8530K might be very interesting chose but they are around $500. I always say that if they are good and useable then they might cost 10 times as much but if they are not good for your application then they are not worth even $10. Can you try them without paying $500?

I’m mentally searching the collection of my driver in my storage and I realise that I have no direct radiators for box solutions. I have large coalition of LF transducers and compression driver but in the lower MF I never looked at anything else besides the horns… My Fane 8M would feet your fill but it is not good as direct radiator. Although it has a very good tone as direct radiator but it severely compress sound with being loaded into a horn…What is boils down: the question: what commercial box speaker has good upperbass… Well, if I know an answer to this question then I would not do horns…

I think if I was doing “fundamental channel” ($100-$1000 for monitors then I would go for a sealed array of 4 X 6”-8” drivers trying to spread them as tall as possible. The larger radiation space would be the better they would load room. Still I would not go too high as it would crate time discrepancies at HF, I would say 60” high no more (however, it would create a huge problem, perhaps irresolvable with a “half ass project”, for the implementation of the enclosures…). I do not think that Large Sound might be emanated from a single driver, in addition the arrays have own advantages….  Well, now you understand why I am not wiling to go into this direction with my SL600 project. Perhaps is you find any “simple” solution then I would replicate it…

Sorry, I posted a long replay but I feel that I am completely no equipped to discuss this subject. The Tanglewood broadcast of Requiem is too boring and I entertain myself by replying to you during the performance…. How bad should this play be!?

Rgs,
The caT


Posted by Ronnie on 08-04-2006

Do you know if there is a gadget similar to this on the market?
Line-level and 6dB/oct. Passive or Active.

sketch_3-ch_xo.jpg


Posted by Romy the Cat on 08-04-2006
Ronnie,

I do not know any. But the major question is why such a crossover is necessary. Obviously it requires 3 amplifiers and if so then the filters might be built in the amps. It one needs it for system prototyping purpose then if is fine but the degradation in the crossover should be taken under consideration. For my prototypes, in fact it was what I used for Macondo I have Marchand XM26 crossover (slightly updated) and many-many preset card. I have a few prototype cards into which I solder any slopes and any frequencies I want. Marchand is not transparent enough as a final solution and frankly I have difficulties with it even as a prototype. Still it is flexible… For $300 a digital crossover is a good solution that would help to find the cut offs and then implement the cut off by more noble means….

It would be not big deal to make this crossover but it would be necessary to have an absolutely transparent buffer. The Guy Hammel’s buffers are absolutely transparent but very expensive and ….your would need at least 6 of them + 6 very high quality stepped attenuators…. Not to mention anything else. So, the project might be expensive and complex but how much do you agree to invest money and efforts into a solution that you will be using for a couple weekends? Still, if you make this thing to be absently transparent and responsibly inexpensive then let everyone know as a LOT of people would willing to buy it from you….

It is kind of strange: in the beginning I proposed that this crossover would not be necessary and then I propose that it will have a large demands? Well, ironically I feel that I am correct in the both proposals…. -Smile

Rgs,
Romy the caT

Posted by Ronnie on 08-05-2006

Yes. For prototyping.
I found out that Luxman built an 8-tube unit exactly like it.
I borrowed a picture from the net: www.webbtjanst.se/temp/2003.jpe

However I think it might be interesting to build one, with decent pots, 6dB filters and 6 buffers after the filters. Looking for buffers in the tubes box is depressing.
Anyway. It's for prototyping and I can allow it to have some gain to waste. Something like 4 6BX7 perhaps. Rather wish I knew anything about solid state... and that Guy would suddenly post all his schematics in the appropriate asylum ;-)


Posted by Romy the Cat on 08-05-2006

Well, I never felt that for prototyping I need more then one-channels crossover as by use of the poor sounding prototyping crossovers I never look at the system in it’s entirely but rather for the “technical listing” of  just two channels that I need to bind.  Therefore for me do “a channel a time” worked always well and after I found the necessary crossover points I always trued to go line level. Two things I have to mention though.  My high pass for MF channel is not implemented on my system at line level and I still use the speaker level 3uF cap. I was not able to make the amplifier to sound good when the amp did not pass the lower frequency harmonics through itself. Also, do not forget that I use always fort order and therefore the precision of crossover point is not crazily critical.

Also, despite I kind of used the active crossovers but I personally more tend to do prototyping on the speakers level. I have a large box of all imaginable filter caps and coils and I can easy build a simple filter. After I find a crossover point, measure it, listen it and got satisfaction then I implement the same filter at line level and listen if I get everything right. I sincerely feel that a good quality speaker level filter is WAY more reasonable solution then a poor performing line level active crossover. Also I feel the drivers better what I do them with passive components at line level…. the drivers response to the changes more initiatively...

There many active ready to do crossovers out there. Audio Research did one, Briston did, Pioneer did, the same Marchand has many of them. Mark Levonson did one big and expansive; a zillion pro companies did…. They all good enough for prototype and I in past did rented some horrible sounding but superbly flexible pro crossovers for some experiments. It is important to clearly know what you need to hear and then you will be able to cancel out everything else… Well, of course some times it does not work… BTW, if you wish to know how it “might be” I re he was getting member a few years back Bob Crump during SES was trying to setup his room with some kind of speakers that he did not know but needed to promote (I think it was Wisdom Audio but I do not remember correctly). The speaker come with own LF section and own second order low pass active filter. Crump hardly demonstrated any interesting sound but the sonic horror that he was getting in his room that year was too much even for him to bear. So, he decided to blame this speakers and that active crossover. How big was his surprise when he opened up that crossover and found on the circuit board of the second order low pass filler 300 op-amps (read three hundreds!!!). I did not know if he actually counted them but he insisted that it was the number. It is not surprise that the speaker cost $15K and I am sure that they had a wonderful press in audio publications….

Anyhow, I think nowadays, is you do not wiling for whatever reasons to go speaker-level prototyping, a digital crossover might be a comfy solution, with a price tag of $300 it might be a direction to go…

The caT


Posted by Ronnie on 08-05-2006

Very interesting read. I like comfy!

Now for ultimate comfiness: would you recommend an actual unit for prototyping?
...A digital crossover that works OK, with >=2 analog unbalanced inputs and outputs, 2*3 channels output, possible 6db slopes and not needing a higher-than-line-level input (I think Thorsten used some Behringer after having amplified the incoming signal a few dB?). And <$400.

I have no idea if such things are common or unobtainable!

Heeh heeh. I'm asking you to RECOMMEND a DIGITAL CROSSOVER! =)


Posted by Romy the Cat on 08-05-2006

I am really not an expert in digital crossovers. I rented in past a few, inducing the Behringer DCX2496 that Thorsten mentioned. I less was concern at that time about the unit as prototyping crossover but rather was trying to learn if the digital crossovers are as good as some people claimed. I know that some folks who visit this site (I think slowmotion does it) use digital crossovers and they might be more qualified advisers then me. Still, if need to do something with speakers (and I will start next week to play with my “Fundamental Channel”) then I would go speaker-level as I feel that line level active unit introduces more methodological variables and more uncontrollable ingredients.

The caT


Posted by zanon on 12-02-2009
fiogf49gjkf0d
Behringer has a 3 channel digital crossover just like what you want.

Don't remember the model, but it's on their site.

Page 1 of 1 (9 items)