full range + woofer OR full range + tweeter ?

[zman01]

Originally Posted by Bigun

(all tube, SE for the full range, PP for the woofer).

Even SS would be ok for the woofer I guess, specially given that it takes more watts to produce LF in a smaller box. There are good options in Class D for cheap.

[chris661]

I'd insist on SS for the woofer. Output transformers give valves higher output impedances, which affects Qes of the speaker, in turn messing up any nicely tuned cabinet you might've designed. The added effective series resistance also adds output around impedance peaks.

Chris

EDIT - that said, I've heard some very pretty sounds from valves. I just feel the design of a speaker to be driven by valves requires more involved design.

[planet10]

Originally Posted by chris661

Make it 7kHz or higher - the critical part, so far as I can tell from Fletcher-Munson etc, ends around 5kHz.

+1, it is usually done with XO only on the tweeter, so the extension of the FR determines the frequency, usually above 10k.

My preference is the FR + Woofer.

Why it has the potential to work better is down to the wavelength at the XO point. One gets rid of a very significant downside of XOs if one can space the drivers such that they are within a 1/4 wavelength (or less) of each other. If one can do this then the drivers are essentially co-incident -- to do the same with a helper tweeter it would need to be co-axial with an appropriate time delay on one driver.

Further, there are LOTs of decent midbasses to choose from that extend to 1kHz+, and a growing number of FRs that will hit 100 Hz or less and have extension to 20k+. This engenders a simple 1st order XO which means retention of time/phase. It is also an ideal place (if one takes care with driver selection) to do a series XO. The very nature of the series XO also means that the responses of the 2 drivers can automatically compensate for anomalies inherent in the drivers. This is one place where a passive XO can have advantages over active XOs.

Third, careful choice of XO frequency (amps & impedances too) can get you inherent BSC (doesn't apply to OBs, MJK has good coverage of that in his passive baffle articla)

Forth, back to the wavelengths... when down low, there are no issues with lobing of the dispersion because, due to 3, driver response is approaching omnidirectional.

And a comment on item 2. As the cost, sophistication and choice of DSP continues to plummet I can see some serious advantages to using a 1st order series XO to get into the ballpark, and then DSP EQ to tidy up the details.

It should be noted that a FAST is actually closer to a "real" 2-way than a helper tweeter, as the bass driver handles 2-4 octaves (using 80-320 Hz XO point) whereas a helper tweeter is more like a single octave (XO @ 10k) so you get more of the advantage out of job sharing.

dave

[vinylkid58]

Originally Posted by Bigun

although I haven't found the right application for my FE127Es - they are sitting unused until I figure out how to deal with their treble peakiness.

That's easy, sell them and buy FX120's.

jeff

[planet10]

Originally Posted by Bigun

(all tube, SE for the full range, PP for the woofer).

Originally Posted by chris661

I'd insist on SS for the woofer. Output transformers give valves higher output impedances, which affects Qes of the speaker, in turn messing up any nicely tuned cabinet you might've designed. The added effective series resistance also adds output around impedance peaks.

Althou SS (including class D) is the usualy 1st choice for the woofer, it is very possible to make a tube amplified woofer that works very well. To date, out 3.2 w Class A EL84 PP (within its power limitations) has seriously good bass, cleaning the clock of many a SS amp.

And the whole Qes argument has holes you can drive a truck thru... althou not common, with careful design, the Qm of the system can be made to be <1 and then one can start thinking of taking advantage of what current drive brings in terms of dealing with backEMF, compression due to VC heating and not using the impedance of the driver as a voltage to current converter.

dave

[chrisb]

Originally Posted by Bigun

chrisb - yes I've had some good experiences so far with 3" (Fountek) and 4.5" (Fostex, CSS) full range - although I haven't found the right application for my FE127Es - they are sitting unused until I figure out how to deal with their treble peakiness.

pardon the brain-fart question ( i.e. I should know this) - are they treated to any degree? EQ can sometimes help, but I've enough experience with careful treatment to be inclined for a "mechanical" (permanent) fix

Originally Posted by vinylkid58

That's easy, sell them and buy FX120's.

jeff

well, that'd work too

Cogitech - I noticed the issue with recording quality when I got my first hi-fi, it's a 2-way floorstander PMC speaker with Byrston amp and YBA CD and Magnum Dynalab tuner. I ended up throwing away half my CD collection and finding only a few FM channels that are up to the challenge. Moving to full range speakers did make the difference more acute. I don't have any vinyl - and perhaps that is something I should change in the future....


Thanks for the replies from everyone - I see a rather clear preference emerging so far and this helps me in thinking about the amplifier design that will be needed (all tube, SE for the full range, PP for the woofer).

as to the class of amp for woofers, there will be many opinions and much advice proffered on that subject, don't look for the same degree of consensus as to what type of drivers and XO points to select

consensus on a DIY forum, what was I thinking?

[chris661]

Originally Posted by vinylkid58

That's easy, sell them and buy FX120's.

jeff

The notch filter I use on my 126s (still a little forward for young ears) was 3uF, 4r7 and 0.3mH, all in parallel, in series with the speaker. Puts a ~4dB notch centered around 7kHz, which is where I found the most noticable bit was.

They're a more relaxed listen now, which suits me nicely.

That said, I can understand why people would prefer the more forward sound.

We should probably get back on topic

Chris

[picowallspeaker]

Originally Posted by planet10

It should be noted that a FAST is actually closer to a "real" 2-way than a helper tweeter, as the bass driver handles 2-4 octaves (using 80-320 Hz XO point) whereas a helper tweeter is more like a single octave (XO @ 10k) so you get more of the advantage out of job sharing.

Indeed , it is easier to put together a 3" and a woofer (or two in isobaric like I did ) .
It's design choice , also budget , seeing that there are little FRs for 200$...
The tweeter choice would otherwise look at smaller than 1" domes , I guess..

Tony

[chrisb]

Originally Posted by chris661

The notch filter I use on my 126s (still a little forward for young ears) was 3uF, 4r7 and 0.3mH, all in parallel, in series with the speaker. Puts a ~4dB notch centered around 7kHz, which is where I found the most noticable bit was.

They're a more relaxed listen now, which suits me nicely.

That said, I can understand why people would prefer the more forward sound.

We should probably get back on topic

Chris

per my recollection the FE126E's peaks were even more pronounced and objectionable than FE127E, so exact values of notch filtering might need juggling

I found the new FE126En to be quite better in all regards

[tuxedocivic]

Definetely small full range with bass support.