Which DIY speaker designs work best on low volumes?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
...is it not a contradiction to use a small, inefficient
driver to reproduce nuances at low listening levels?

Hi,

Where is the contradiction ?

Small high quality speakers have very low levels of cabinet
colouration compared to far bigger higher efficiency speakers
with the same bass extension capability.

They are more revealing within their useful dynamic range.

The contradiction is somehow insisting you need high efficiency
speakers for low level listening, you simply don't unless you
have a flea powered amplifier.

High quality low distortion small drivers with good bass capability
(hence low efficiency) built into high quality cabinets will wipe the
floor with anything else (built to the same price) at modest levels,
as many blind listening tests have shown.

It is much easier to build a top quality small speaker than a big one.
Advanced cabinet construction is an art in its own right and e.g.
for the ZD5 its probably best to look at other high quality
designs for cabinet ideas if you want to push the boat out.

rgds, sreten.
 
Last edited:
I really don't know if it's a contradiction or not...

High-eff speakers IMO usually work better on low SPL, but they come with other issues: magnified noise etc. Also: some low-eff speakers work far better than other low-eff speakers, so efficiency alone is not the deciding factor. Or is it, I don't know, it would make sense...

Then some say good speakers sound good regardless of SPL. Also, in my limited experience, Scanspeak drivers work better than most on low SPL, despite their low efficiency. And they are one of the best around, so in that train of thought something like ZD5 or ZRT makes sense to me as well...
 
Hi,

Many high efficiency speakers sound best at low SPL simply because
they can't do high SPL in the bass at all. Many have very limited bass
power handling due to low driver excursion capability and won't go
anywhere near the maximum bass levels or as deep as similar sized
less efficient speakers.

Just because many only sound good at modest levels doesn't mean
they sound better than "normal" speakers at modest volumes, as
many blind listening tests have shown, they simply don't.

They do suit flea powered amplifiers though ......

rgds, sreten.
 
p

ok. Maybe i wasnt clear in my question. At small signal level, does BL/mms influence the he attack decay envelope of transients? Ie heavy cone, more inertia, requires higher BL to accelerate equally? Im not generally a HE fan, it just occurred to me that it may make more difference at low amplitude depending on suspension linearity.
 
As for electrostats, I love the sound but the size is just unacceptable for our apartment. Also they usually are super directional; I somehow don’t feel comfortable with that.
That's why multi-way electrostats are implemented with varying strip widths. Still line arrays, though. Constructing point source speakers is considerably easier with magnetostats as suitable drivers are available off the shelf.

Heavy cone, more inertia, requires higher BL to accelerate equally?
Provided the amount of mechanical damping is held constant, yes. HE drivers usually also have higher Qms and lower Rms in addition to lower Mms.
 
Hi,

BL and Mms .....

So high efficiency tweeters sound faster than less efficient ones ?
They have more detail at all levels ? Attenuating resistors are a
disaster as they reduce effective BL ? etc .... your disappearing
up a number ratio based cul-de-sac, its pretty meaningless.

rgds, sreten.
 
from what i've made; smaller drivers seem to sound more full than the larger ones at lower SPL.

seems like you can get the drivers moving properly and still not be puttingout too much sound.

i don't know about the science stuff as much as these other guys but just reading your posts i would look at these:

ELLAM-25

good drivers, high WAF, very similar to the ZD5 with a boost on the low end if you want to turn it up
 
The carefully designed ones with quality components. Flippant answer? No. A good design works well. A poor design may have more limitations. I have heard speakers that need more power to "open up". I consider that a flaw and look elsewhere.

Lets get down to earth. No need theoretical debate. Here is I think the secret to achieve what Pekkakala wants to achieve:

1. The key point is you want the midrange frequency (thus the crossover frequency) is produced perfectly in phase by the drivers when it is a multi-way. If it is not perfect, it will sound okay only at high SPL because it relies on flat summed response (and steep crossover), like most speakers out there.

2. High sensitivity fullrange drivers might achieve this easily but they have other problems such as distortion (even at low SPL!) and no low extension.

3. When the midrange is produced in phase by 2 drivers, the sound will sound as if coming from a single plane like electrostatic, but with better dispersion.

4. To get the best performance, you want a not too sensitive woofer (to allow for lower bass extention and a good sound at high SPL) but not too steep roll-off (because the signal overlaps is the key to this project). Third order is the maximum, but second order is the best.

5. For shallow second order filter, choose paper cone to avoid distortion. Low bass extension is of very critical if you want a good bass from shallow filter.

6. I need to go to work, but it seems ZD5 has it all. Go for it.
 
Administrator
Joined 2004
Paid Member
Nah, it looks like it's making progress now. I'm a big Hi-Eff fan, but I don't think they are needed here.

If it weren't DIY, I'd use a pair of John Blue JB3 and a small sub. The best tiny speaker I've ever heard. If you don't see them, you'd never believe how tiny they are. They will not play loud, but at low volumes they sound great. For me, they would fit the bill of apartment dwelling, low level listening perfectly.
 
Hi,

I'm getting bored. Its a 25 metre square room which is not small.
Its big enough to have some serious low frequency main modes,
5m x 5m is bad mode wise but ~ 8m x ~ 3m is very good, ~
6m x ~ 4m isn't too bad either, ~ 7m x ~ 3.5m is not the best.

At low volume what you need is a speaker that goes low enough
to drive the room modes effectively rather than just giving up.

6m givers a room cutoff of 29Hz, 7m 25Hz, and 8m 22 Hz.
In such a room the ZD5 placed away from walls for what it is
will sound immense, but again I must stress at modest volume.
(But again stress it can do bass way beyond a typical 5"/1",
the fact bass is extended means it must be at modest volume.)

Having said all that the ZD5 is a relatively small speaker.
The ZRT on Zaphs site has the same recommended power rating,
(With the same bass power handling limitation of about 30 watts),
but will go louder as its ~ 5dB more efficient and in a bigger box it
doesn't dig quite as deep, but it will sound like it does at its higher
volume limits (due to F+M curves), some compromise is involved.

The ZRT is less compromised, its a bigger speaker. But if you
really want WAF and the sonic advantages of a smaller cabinet
and don't need its extra 5dB level capability, then the ZD5.

rgds, sreten.
 
Last edited:
@ Sreten: Easy. It was an innocent question. Maybe its a crackpot idea, I dont know. I just thought id ask someones opinion, someone i respect; and I get shot down.

@twest:

Thanks, all else being equal, which it rarely is. Almost to the point of being moot. Ok. That answered it nicely :D

@TVRgeek: would a small fullranger, <3" and a 5/6" woofer achieve the same, as the driver/s you suggest? In your opinion I mean lol

Ive used the Tangband W3-1285SG with Visaton AL130, and after I wasnt sure I wanted to go back to the Dome tweeter :(
 
Last edited:
I'd just wanted to chime in some rubbish after a few beers.

I love my system playing low SPL. I dream of it. I don't have DIY gear but it is a real joy to hear my Duntech Princess (90db) powered by my Krell KSA150 amp. The system noise floor is non existent.

My biggest trouble and something that a lot of renovated terrace house owners will understand is it is now fashionable to have the kitchen and living room in one. I have to wait for my fridge compressor to become silent before I can really enjoy a late night quite session.

Why anyone would want to sit in the same room as their fridge/dishwasher/clothes washer is beyond me.
 
To get the best performance, you want a not too sensitive woofer (to allow for lower bass extention and a good sound at high SPL) but not too steep roll-off (because the signal overlaps is the key to this project). Third order is the maximum, but second order is the best.
Paper lacks the strong cone resonances of metal but it's still desirable to avoid having the drivers emit much SPL at breakup. I've tried every crossover alignment from LR2 to Butterworth7 at some point and would say LR4 and LR6 are the most useful---LR2 isn't steep enough to avoid driver issues and the lobing from B3, B5, and B7 summing in quadrature sounds weird to me. Doing LR4 passive is a hassle but there's plenty of active pro audio LR4 gear one can grab and it's only a couple Sallen-Keys if you're rolling your own active crossover---like most pro audio defaults minimum phase LR4 is pretty good tradeoff. DSP makes implementing LR4 or LR6 in either minimum or linear phase pretty much trivial (LR8 too, though one hits diminishing returns).

An alternate approach for woofer selection is to use a larger part and obtain the necessary volume displacement via Sd rather than xmax. In the bass this favors pro audio rather than hi fi drivers and I would go so far as to say that the pro stuff sounds a bit better provided you use the EQ it was intended to be used with (also trivial with DSP). If you're not going to EQ then the alternative is to rely on the greater damping of hi fi drivers to flatten out the SPL.

Thanks, all else being equal, which it rarely is. Almost to the point of being moot. Ok. That answered it nicely :D
You're welcome. ;) mige0 has hypothesized high Qms/low Rms drivers would tend to have cleaner impulse responses. I suspect he's right but the necessary measurement survey of drivers to support or refute hasn't been done. It's also a function of crossover, power amplifier, and enclosure implementation as those govern the amount of control the amp has over the cone and the amount of back wave energy the amplifier has to absorb to keep the cone moving as desired. Plus the Thiele-Small model is hard to apply to magnetostats as they don't have a well defined mechanical resonance to use for parameter extraction.

Oh man, I use to have a living room system that was right next to the kitchen. The fridge ran all the time. Really annoying, but not much I could do about it.
Yeah, I have that problem in my current place. I just turn the fridge off for the evening's listening session. Works great, except for when I forget to turn it back on before going to bed. :whacko: Maybe my next project will be to DIY an audiophile grade refrigerator, followed thereafter by an audio grade furnace.
 
who said open plan was great? Probably the guy with a 100 sqm living area kitchenette! When the washing machine spins i get a lovely 25hz earthquake to mask any bass i may have had...fridge is annoying too. If i had my dream home, id try that high qms low rms approach with some real woofers, for now im stuck with the 5inch audax, which arent bad in that respect.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.