Off-topic.
General FYI, the forum has a few display themes and fixing the colour of text takes away it's ability to adapt to the background.
General FYI, the forum has a few display themes and fixing the colour of text takes away it's ability to adapt to the background.
You might want to consider the PHL drivers available from TLHP, some version of this woofer range is what Genelec is using in the 1237A. There are a few variations.
The basket is a pain, Genelec deals with it through a trim ring.
https://en.toutlehautparleur.com/speaker-phl-audio-4531-8-ohm-12-inch.html
When I put 2 of these in my cart they are 159 EUROS each excluding tax and shipping which I think is not a bad price for what you get. I have bought plenty of stuff from them and had it shipped to Australia without any issues, well packed and arrived intact quite quickly.
Americans will probably complain as for once it would cost them more.
As an aside for context the 12PR320 used in the OSMC is 120 Euros each when buying 2. They were 108 each 3 to 4 years ago when I bought them, if only everything else in the world had only gone up that much since.
Last edited:
I frankly never understood why the 12PR320 was the preferred driver in the OSMC. It really is a good choice for a 12"-1"CD combo, smooth and with low distortions up to above 3k. But with limited Vd it isn't exactly my choice for a 3-way with serious SPL requirements.
It's not being tuned low so it can can get plenty loud. The Genelec 1237A is really quite similar although it goes a little lower, they claim a maximum SPL of 118dB, I have no idea how they work that out.
It is hard to find very high sensitivity, extended low frequency and good higher frequency behaviour all together. With a mid dome that needs a higher crossover than a cone driver could handle the better high frequency is a trade off against going low. Nothing new in that just a choice of trade off.
There is a PHL that adds mass to get a lower Fs but with reduced sensitivity. The SB34 drivers are a good compromise but they are 91dB so more like 85dB with full baffle step, the Faital being 90 to 91 with baffle step.
For my own version I have a mid dome and would like to keep the sensitivity up, I don't really want two woofers in the same box so I can understand the choice. If I want to pump air at low frequencies I can add some subwoofers. I can also see why someone would make a different choice for other reasons.
It is hard to find very high sensitivity, extended low frequency and good higher frequency behaviour all together. With a mid dome that needs a higher crossover than a cone driver could handle the better high frequency is a trade off against going low. Nothing new in that just a choice of trade off.
There is a PHL that adds mass to get a lower Fs but with reduced sensitivity. The SB34 drivers are a good compromise but they are 91dB so more like 85dB with full baffle step, the Faital being 90 to 91 with baffle step.
For my own version I have a mid dome and would like to keep the sensitivity up, I don't really want two woofers in the same box so I can understand the choice. If I want to pump air at low frequencies I can add some subwoofers. I can also see why someone would make a different choice for other reasons.
True. But if one's throwing subs into the equation (what a wise thing to do...), who needs a 12" bass-mid? Otoh music programme to me is from 40Hz upwards and if you can't deliver 40-80Hz at full strength, why again use a 12" for bass-lower mid duties?
The Genelec 118dB (short term) claim is from 100Hz upwards. We all know that fine monitor probably fails to reach 108dB at 40Hz without distortions going through the roof.
The Genelec 118dB (short term) claim is from 100Hz upwards. We all know that fine monitor probably fails to reach 108dB at 40Hz without distortions going through the roof.
The OSMC graphs are a little different than I am used to reading so I don't know exactly where it starts to roll off but being tuned at 40Hz or so, in room at any reasonable position should give 30 to 40 Hz a boost above anechoic.
I don't know that a 12" mid is needed. But every time I see the picture of (I think) @mbrennwa OSMC speakers in walnut I want some and I like the proportions.
I don't know that a 12" mid is needed. But every time I see the picture of (I think) @mbrennwa OSMC speakers in walnut I want some and I like the proportions.
This how Genelec rate their SPL:
(bold: my emphasis)
“Long term max SPL:
Maximum long term RMS acoustic output with IEC weighted noise (limited by driver unit protection circuit) @ 1m”
—> similar to what you would see in VituixCAD based on your woofer maximum long term IEC thermal power rating
“Short term max SPL:
Max. short term sine wave acoustic output on axis in half space, averaged from 100 Hz to 3 kHz, at 1m distance”
—> similar to what you would get without baffle step loss compensation, without regard to x-max limitation, and not including the limits of your tweeters
References:
https://www.genelec.com/1237a#section-technical-specifications
-> open all details
https://www.aesmelbourne.org.au/wp-content/media/AES-Dec2021-final.pdf
(bold: my emphasis)
“Long term max SPL:
Maximum long term RMS acoustic output with IEC weighted noise (limited by driver unit protection circuit) @ 1m”
—> similar to what you would see in VituixCAD based on your woofer maximum long term IEC thermal power rating
“Short term max SPL:
Max. short term sine wave acoustic output on axis in half space, averaged from 100 Hz to 3 kHz, at 1m distance”
—> similar to what you would get without baffle step loss compensation, without regard to x-max limitation, and not including the limits of your tweeters
References:
https://www.genelec.com/1237a#section-technical-specifications
-> open all details
https://www.aesmelbourne.org.au/wp-content/media/AES-Dec2021-final.pdf
Last edited:
That is so true! While we all should know better, our childhood dreams prevail and steer our emotions.
Indeed.
Sensitivity would be the obvious choice. Since it crosses to a direct radiator and leaves a range for the crossover frequency, could we reverse the question.. why not?
I'd be curious to see if there is a more recent study on that. It may have seemed a better idea when EQ wasn't so readily available.
The exponent values on the frequency axis which I always find harder to read, some have all marked some not. No big deal just makes me work harder to work out what they represent.
You probably prefer it the way you have, I find it easier to read other style graphs at a glance that is all.
You probably prefer it the way you have, I find it easier to read other style graphs at a glance that is all.
The facts : Fs is more 47 hz than 42 Hz for the 12PR320, so no miracle if vented tunned below that, it sounds less good than tunned at Fs.
SC4 seems to me better, at least theorycal, or extensed BR load as big than a sealed (too huge cabinet). I am more in the camp of tunning the BR above the Fs of the driver.
So having the 12PR320 and simmed it, it was a no-go with an F3 at 58 hz. It is surely okay in average european living room if listenn to modern music with no content below 40 hz. I don't like as well the fact it needed a so complex filter as low pass. It is definitly a good midwoofer PA ! At that time of the OSMC it was a valid choice, now with the SB34RNXL and 12RS430, no more. I miss the 12" NE-290 from Peereless silver line !
I keeped the 12PR320 for a futur open baffle or baffle less for 100 hz to 500 hz like Juhazi member did. It has the same Xmax than the Beyma he chose.
SC4 seems to me better, at least theorycal, or extensed BR load as big than a sealed (too huge cabinet). I am more in the camp of tunning the BR above the Fs of the driver.
So having the 12PR320 and simmed it, it was a no-go with an F3 at 58 hz. It is surely okay in average european living room if listenn to modern music with no content below 40 hz. I don't like as well the fact it needed a so complex filter as low pass. It is definitly a good midwoofer PA ! At that time of the OSMC it was a valid choice, now with the SB34RNXL and 12RS430, no more. I miss the 12" NE-290 from Peereless silver line !
I keeped the 12PR320 for a futur open baffle or baffle less for 100 hz to 500 hz like Juhazi member did. It has the same Xmax than the Beyma he chose.
Last edited:
Very different pets. The 12PR320 has way higher sensitivity, while the SB and 12RS have much lower resonance frequency. You can't have both at a given size.
I still don't understand how this project is supposed to work. Why does everyone suggest drivers and stuff even though we don't have a defined target to aim at? What am I missing?
Yes but at 92 dB efficienty in the upper bass I don't see nowadays the plus VS a SB34RNXL at 1dB lower there (let say 1,5 to be sure). The Faital datasheets are a little optimistic. You can still play with the room modes though to put the seat in a high pressure resonant zone...
I do understand though its attractivity for tube amps owners.
I do understand though its attractivity for tube amps owners.
I don't want this to be taken the wrong way, I am not trying to be abrasive or sarcastic
Let's assume we defined the following specs and performance targets:
Sensitivity of 95 dB/ 2.83V @ 1m
F3 of 30 Hz max, F10 of 20 Hz max
THD and IM Distortion less than 0.1% from 20 Hz to 20 kHz at 118 dB SPL
Enclosed cabinet volume of 50 liters max
70 degree constant beamwidth from 20 Hz to 20 kHz
DSP control to allow on-axis response and sound power response to be adjusted independent of each other. This would allow us to vary the DI curve to suit any room.
Target cost for all 6 drivers: $600 USD
Target cost for electronics: $200 USD
Well, that was not so hard... We just defined the specs for a system that will sound good in almost any room. There is the small problem that specifications are unrealistic, and in some cases, are physically impossible.
So what is my point?
Well, there are two points. First point is that before a set of specs or targets can be defined, we have to have a clear understanding of what is possible, given the existing selection of drivers and other components available to us. Evaluating the set of preliminary woofers that might be considered by the design team is definitely a valid activity because it helps us understand what the spec should look like... it informs our decision.
The second point is that defining a spec or target is, in itself, a subjective activity. Targets are selected on the basis of subjective judgments by the designer. For instance there is no scientific or empirical consensus that an F3 of 39 Hz will be satisfactory but an F3 of 41 Hz will not. There is no consensus on how much distortion is too much. There is no consensus on the "correct" horizontal beamwidth. Selecting any of these as a target is an exercise in personal preference.
Selecting a particular driver as part of the design spec because we simply like that driver is a subjective personal preference, but it is no more subjective than to arbitrarily define a criteria such as sensitivity or F3. Either way, it is a valid method to set a design target.
For a passive system, "what is possible" in terms of woofer implementation is defined by fundamental physics. See attached. Looking at specs of random drivers will not help much.
For an active system, "a lot more is possible" provided the woofer provides enough Xmax.
Sure! But what type of targets do we have? How do we decide on the targets?
Is the target to just build a three way monkey coffin at a given cost limit?
I sure hope there there more so we can see where to go and how to walk. At the moment most posts are just "hey, I think this driver is cool!". I say pretty much any driver is "cool" in some way. We need to understand what the targets of the project are so we can start to make sense be aiming at those targets.
The thing is that "we" (=you, me, everyone else) all have different personal preferences, and the project will not go anywhere if everyone just points out random thoughts.
Here are some suggestions of points that should be considered for narrowing down the targets for the 3-way "Monkey Coffin":
- Cost limit for the whole system?
- Size limit for the box?
- Active or passive?
- If passive: how complex can the xovers be? If active: DSP or analog filters?
- Where do we set the compromise on bass extension vs. efficiency?
- How easy/difficult do we make it to build the speaker?
(examples: do we allow non-circular drivers, which are harder to flush mount? Do we restrict the design to simple flat panels, or are 3D parts allowed?) - How important are the looks (of the drivers)?
- ...etc. There will be a lot more!$
Attachments
No one is suggesting random drivers or presenting random thoughts...
Everyone realizes that the woofer needs to be 10" or 12" and have an F3 "near" 40hz and are throwing out non-random suggestions. A random driver would be if someone suggested a 2" full-range driver as the woofer.
Hmmm, I'm not sure there are many facts there. If I simulate the 12PR320 in Vituix with 70L and SC4 the F3 is 47Hz and F10 34, that is with a very short port. Tuning a bit lower gives something that would work better in a lot of rooms, there is a reason why KEF choose that EBS style alignment in most of their speakers.
I'm not sure what you mean exactly but the SB drivers are 6dB less sensitive because they have heavier cones and lower Fs. You need two of them to be as loud as a 96/97dB driver throughout most of the passband. Maybe you forgot the baffle step or something. The one thing on the datasheet that is likely to be true is the half space response at 2.83V. The only way to fudge that is to smooth it too much.
Some may want high voltage sensitivity, some may not care. Loud, low, small. You can have two but not all three.
Last edited:
About $270 here in the US. Quite a difference! It will be hard to please both sides of the pond.
I'm unlikely to build this so take this for what it's worth:
Being budget minded, I was going to suggest the 10FE330, which actually is $128 here. Waiting for someone to object that it has no demod ring. It could work nicely with a 5FE125 mid ($50 here, does have a demod ring, ) and an SB Bianco 25CD-P ($21!) in a 3dp ATH Tritonia waveguide scaled down for a 3kHz xover. I think this hits lots of the design brief, you'd get reasonable efficiency, on size target (~73 l for the woofer), good pattern control for the tweeter, and a price that would allow a fair number of builders to participate. A serviceable second-hand 3d pinter can be had for under $100, and could be sold back off if the builder didn't have any future projects.
More thoughts:
I would consider this a 'west-coast' kind of interpretation of the ape crate, with a vented box and horn tweet. Sealed woof and dome tweeter leans more east-coast. Add a dome mid and that takes my mind to Europe.
An actual detatchable, rotatable MT panel adds a lot of complexity if the mid is open backed and needs an enclosure so I would not design for that. A design that could be built either horizontal W-T/M or vertical TMW would be much easier.
Bill
Very true, apart from SB Acoustics and Scanspeak there are few other brands that are available for directly comparable prices everywhere.
I certainly won't so my opinion is irrelevant
Scaling down a lot doesn't always work, worth checking if you haven't.