Hi,
re wiring schematic.
the top row show the left speaker on +ve phasing.
The bottom row show the right speaker on +ve phasing.
So the +ve phase speaker in each combination will need to be on the same side of the box i.e. each +ve on the INside and each -ve on the OUTside. This also matches the conventional phasing for a single speaker with +ve phasing.
The overall impedance of the set of eight speakers will be twice a single speaker impedance.
re wiring schematic.
the top row show the left speaker on +ve phasing.
The bottom row show the right speaker on +ve phasing.
So the +ve phase speaker in each combination will need to be on the same side of the box i.e. each +ve on the INside and each -ve on the OUTside. This also matches the conventional phasing for a single speaker with +ve phasing.
The overall impedance of the set of eight speakers will be twice a single speaker impedance.
AndrewT said:
Hi Andrew,
Thanks, yes, I see that I got that slightly wrong (I was tired drawing it in bed last night, then possibly dreaming of speakers).
I will knock up a fresh one, which I will follow very closely when I come to make the connections for real. It's not very intuitive to wire something like this correctly, for me. It needs putting to paper first!
Hi Simon,
Some things I have found when measuring drivers: you must hold them vertically and away from reflecting surfaces when measuring parameters. The results won't be accurate if you lie them flat on the floor or table and measure them
Also you want the weight used to measure VAS accurate as possible (to 0.1g or better). Also slight variations in your calculated Sd will affect VAS so it's important to calculate this accurately.
I have a permenant test rig set up now (Speaker Workshop/Wallin Jig/accurate weighing scales/speaker clamp). If you still want 'em measured then you're welcome to send a driver or two down to me for testing. (It'll cost you delivery both ways of course).
Some things I have found when measuring drivers: you must hold them vertically and away from reflecting surfaces when measuring parameters. The results won't be accurate if you lie them flat on the floor or table and measure them
Also you want the weight used to measure VAS accurate as possible (to 0.1g or better). Also slight variations in your calculated Sd will affect VAS so it's important to calculate this accurately.I have a permenant test rig set up now (Speaker Workshop/Wallin Jig/accurate weighing scales/speaker clamp). If you still want 'em measured then you're welcome to send a driver or two down to me for testing. (It'll cost you delivery both ways of course).
SimontY said:
Hmmm.........
An inelegant solution to an ill defined problem.
Sorry to enter the discussion at such a late stage but.....
I'm going to play devils advocate and assume that in fact there is
nothing wrong with the Seas drivers, so therefore there must be
something wrong with the "current system", I think there is.
Frequency response
Transient response
Distortion
I can see possible problems with all three.
1) FR - not enough BSC
A maximally flat reflex design with full BSC will sound bass heavy
due to the influence of room gain. If the bass level is set by the
level of BSC applied the end result will be an lack of upper bass.
i.e. by setting the level of bass below 100Hz with BSC, whilst
BSC should apply to ~ 150Hz to ~1.5Khz you cause a dip.
2)distortion
Lots. By definition any bass your sub is producing the speakers
are also not producing as they are operating below the reflex
cutoff frequency. What they are producing though is lots of
distortion. Reflex power handling below cuttoff is very poor.
3)transient response
Any arguments such as sealed vs. reflex go out of the window if
you introduce a 4th order LR c/o at 50Hz. The resultant group
delay of adjacent 4th order high and low filters would give most
subwoofer designers nightmares.
Possible solutions
The speakers reflex alignment should be severely detuned to
give power handling down to the lowest frequencies.
Usually this means a port frequency of 26 to 32 Hz.
Alternatively they could be sealed and and a 2nd order electrical
high pass applied to give an overdamped 4th order overall.
Either of the above strategies should aim for overdamped bass
which should allow full BSC to be applied. The room gain below
100Hz should kick in to give reasonable bass extension.
The sub should have an overdamped low pass roll-off. I'd suggest
adjustable frequency 2nd order Bessell. The frequency and level is
adjusted to suit the speakers response in room placement.
For optimum transient response the subs high pass alignment
should also be somewhat overdamped, though at 20Hz its
arguable that transient capability is really an issue.
A minor point to the above is I believe MTM 2 ways are fine for
home theatre but flawed and very difficult to get right for hifi.
Using full BSC would allow them to be 2.5 ways.
I hope this helps...........
/sreten.
Vikash said:Hi Simon,
Some things I have found when measuring drivers: you must hold them vertically and away from reflecting surfaces when measuring parameters. The results won't be accurate if you lie them flat on the floor or table and measure them
I have a permenant test rig set up now (Speaker Workshop/Wallin Jig/accurate weighing scales/speaker clamp). If you still want 'em measured then you're welcome to send a driver or two down to me for testing. (It'll cost you delivery both ways of course).
Hello Mr Vikash,
That's a kind offer, on which I may have to take you up, could you e-mail your address pls?
I performed my measurements with the driver sat on an armchair foot rest. I don't know if you read the rest of the thread but I used pound coins for weight with estimated blue tack weight.
It really would be fascinating to see how close my result is to an undoubtedly vastly more accurate one made by you (I have just looked at your website, very nice).
I will post you two drivers (maybe there'll be some variation) and I won't need them back (still have loads) and I'm poor so I'd rather not pay the postage back anyway. You can give them away or whatever you like afterwards. They might even make a cheap 2 or 3 way speaker...but may lack the midrange.
I wish I had the accurate measurement before building anything, but as this box is only a prototype it doesn't matter much. I just hope I'm not waaay far out.
I'm a latecomer to the thread as well but I'll say that I think that in your situation I would seal the current MTM boxes, add a tailored 2nd-order high-pass filter, and run the Tempest sub up a bit higher with a complementary 4th-order low-pass filter. I know you think it might be localisable, but you would be surprised how well subs can hide when integrated properly. IMO your current new proposal has too many crossover points (so phase shifts) in too small a bandwidth, and 2x 6 inch drivers are not up to proper bass reinforcement duty, and you have a lovely Tempest having half of it's capabilities wasted. Plus you are putting in a vented box with associated group delay in a fairly critical region of sound (50Hz) although sreten's suggestions will go some way to addressing this.
However in spite of these criticisms I think you should go ahead because
1) you have already cut wood and the cost in carrying on is minimal
2) it will be a good learning experience and you can always try my suggesion later and compare (and comparison is the best way IMO to learn what's best in practice)
However in spite of these criticisms I think you should go ahead because
1) you have already cut wood and the cost in carrying on is minimal
2) it will be a good learning experience and you can always try my suggesion later and compare (and comparison is the best way IMO to learn what's best in practice)
sreten said:
And what a nice way to enter it.
I must disagree. The problem with the Seas drivers is they can't play bass well enough for my needs.
The speakers seem to have been designed (Seas Pontus is the design) with no baffle step correction. Obviously, this didn't suit my ears. And can I just say that we can't buy our first speaker kit or design our first speakers knowing everything about speaker design. When you pay good money an assumption is made that the design will be sound. It wasn't.
You are correct in your assumption that the balance of bass is that of a lack of midbass and moreso upper bass.
This sentence doesn't quite make sense (I am tired though).
So you think they also designed the speakers to be distorted. I'm not sure what your point of reference is here, when you say "lots". Most speakers are not controlled below port tuning, are they?
The implied assumption here is not borne out in reality. This system *needs* 4th order lowpass and not 2nd order to integrate the sub and speakers. I'm not saying it can't be better, but there is a reason for everything, Streten, including my putting a x-over there.
I don't like this solution. I can't imagine it would provide the correct bass impact. Two low xmax 6" drivers seem to be insufficient. The speakers would also overlap with the sub's output, and there'd be a weak midbass area, or so it would seem.
Ok, I see the logic when you combine this with the above, but with this shallow rolloff I'll be hearing a little too much of that big subwoofer again.
High pass, on the sub? Is that a joke? It is tuned to 20hz. There is no need for a hi-pass filter on it.
Perhaps a passive 2.5 way x-over will get the BSC sounding right. This is something I'm yet to try, but I don't know that it's appropriate if I'm adding a bass only section.
I hope this helps...........
Well it certainly has opened up deeper thoughts on the subject.
I do wonder why, when you want help, some people only contribute when it's already somewhat too late.
richie00boy said:I'm a latecomer to the thread as well but I'll say that I think that in your situation I would seal the current MTM boxes, add a tailored 2nd-order high-pass filter, and run the Tempest sub up a bit higher with a complementary 4th-order low-pass filter. I know you think it might be localisable, but you would be surprised how well subs can hide when integrated properly. IMO your current new proposal has too many crossover points (so phase shifts) in too small a bandwidth, and 2x 6 inch drivers are not up to proper bass reinforcement duty, and you have a lovely Tempest having half of it's capabilities wasted. Plus you are putting in a vented box with associated group delay in a fairly critical region of sound (50Hz) although sreten's suggestions will go some way to addressing this.
However in spite of these criticisms I think you should go ahead because
1) you have already cut wood and the cost in carrying on is minimal
2) it will be a good learning experience and you can always try my suggesion later and compare (and comparison is the best way IMO to learn what's best in practice)
Hello Rich,
Firstly, thank you for contributing. About your points. These Seas units are not suitable for sealed loading. The Tempest would have to be brought up way too high, and you're damn right I feel it will be localisable. No doubt it's cabinet ringing and harmonic distortion, but when you x-over any higher than 50hz you know where the bass is coming from. Please just believe me on this one. I may not have the designing ability, but I do have ears.
Next point - the drivers are more like 7" (now that I actually measured them). Also, I'd be using 4 compound pairs. With sufficient power they are capable of about 120db, if you believe WINISD's possibly crude simulations. This is louder than the Tempest will play its own range I think so it's approprate in that respect I feel.
Next point - the vent is tuned to 40hz, not 50, so group delay is pushed down a little. In the Seas speakers the vent may be tuned to about 50hz. So this is something I am moving *away* from.
I'm pleased you have spotted some key factors as to why I am doing this - minimal cost (I paid very little for these drivers), learning/experience (a lot gained already!), and comparing more systems to one another. Further reasons - immense coolness factor of using many drivers, massive output potential - useful for house parties, adding to the dearth of well documented push-pull isobaric bass projects (at least that I've stumbled across), and finding a home for some of my many "displaced" drivers.
At least my thread has got some interest now, even if it's just something to ridicule.
SimontY said:
Hi,
all subs have an acoustic highpass (followed by lowpass) response.
Otherwise they would play down to DC.
I'm sorry that it took about a week of mulling it over before I thought
that I could explain most of your problems without the Seas drivers
being at fault.
Edit : and that a reflex speaker with a 50Hz acoustic L/R alignment
would have serious distortion issues below 50Hz if driven full range.
The inelegant comment referred to using two 6.5" in large boxes
for the range 300Hz upwards, when a single 5" could do the job.
It wasn't intended to be derogatory.
/sreten.
sreten said:
Ok, well considering I don't intend to play music containing DC I'm not going to start worrying about distortion in that range.
You make it sound like your sole purpose was to exonerate those drivers, lol
I feel touched that you mulled over my problems.
Well, I've not worked out the tuning frequency of the port, but I can measure the length and width if you'd like to know. Again, are you suggesting almost all commercial speakers have distortion issues below tuning? (it would make sense I suppose - our ears are less sensitive there, and it's well down in level normally.)
The inelegant comment referred to using two 6.5" in large boxes
for the range 300Hz upwards, when a single 5" could do the job.
It wasn't intended to be derogatory.
Thank you for clarifying that. I should add at this point that a part of this approach I am taking or at least toying with, is to have a modular system that can be taken apart later. I do not want to physically alter the current Seas Pontus speakers at any cost.
Also, it seems counter intuitive to my way of thinking that a 5" driver could provide the level of realism given by 2 x 6" drivers. I do like to listen loud sometimes, and I absolutely hate soft or harsh sound - both of which can happen when pushing drivers beyond their comfortable limits in my experience.
SimontY said:
Do you have any T/S parameters for the Seas drivers? I'd still argue that in the application I suggested they would function extremely well.
I don't doubt your listening abilities at all, but all I can say is that maybe there is an unaddressed issue which is giving you problems with localisation above 50Hz. For instance I thought I had a localisation issue and it turned out to be room interaction. What filters are you using on your sub at present? How are you driving the sub?
SimontY said:
Yes that sounds fine then.
SimontY said:
Yes, but my point was that the group delay is being added to whatever is already there, and is very close in terms of tonal distribution to the point that the whole bass end may sound a little detached or slow.
SimontY said:
Fair play to you. If I hadn't built loads of what are now looked back upon as quite frankly crap efforts I wouldn't have got where I am today. It's all part of the learning process, and the fun. As you have loads of these drivers and enjoy DIY then go for it, do what you want but just be aware of advice from the sidelines and use this to shape your next attempts
As for using using the MTM rather than a 5 inch, as was suggested, I fully understand where you are coming from. I also like the big sound and I find that this just doesn't happen as well with a smaller speaker. With the MTM you are also getting the benefit of point source alignment, albeit at a very defined and limited spot, but it's there nonetheless.
Although not using your driver collection, an alternative approach could also be using a peaky high-pass filter on your MTM cabinets which could give the midbass that extra punch, yet retain and possibly even give back a more controlled sound as the heavily distorting and SPL eating below-tuning region is severely curtailed, ready for the Tempest to blend in without venturing into your feared >50Hz.
richie00boy said:
Search for Seas H571 or look further back in the thread, but the published specs don't tally with the way they're used for mine really. I put their numbers in and they model badly - they roll off really early. I wonder if the Audiocom W17ppi version is slightly different. I have obviously not measured the params myself, yet.
Yeh, you could well be right. We can never be too sure. The subs feed comes from the pre-amp, then into a diy 2nd order active xover, variable from ~30hz+, normally set to 45hz. It then trails round the room to the plate amp/sub box (RCM Akustik 150w/4), which uses a 2nd order active, dialled to 50hz. The coils are wired in parallel to present 4 ohms to the plate amp.
It seems you may have misunderstood my intended application - I will be crossing the Seas drivers out from a couple of hundred hertz, so their bottom part will not interfere at all in the 40-50hz trouble zone. I will not be playing these together with the other ones, and I see no immediate need to high pass them particurly, depending on how well they 'naturally' work with the sub.
Fair play to you. If I hadn't built loads of what are now looked back upon as quite frankly crap efforts I wouldn't have got where I am today. It's all part of the learning process, and the fun. As you have loads of these drivers and enjoy DIY then go for it, do what you want but just be aware of advice from the sidelines and use this to shape your next attempts
It is *supposed* to be fun and learning, yes! I am trying to absorb ALL of the advice presented here, and I very much appreciate helpful, friendly and informative advice. You will notice I try to reply to every point raised.
Yeh, a big sound is important. It's what real music sounds like. Even one single small instrument sounds massive compared to a lot of 'hi-fi', not excluding my own humble kit.
Interesting idea - I had not considered this 'peaky' option, but I certainly had thought about rolling off the sub 50hz to give the Seas midbasses an easier ride.
Phew, this thread is consuming my office time!
SimontY said:
Sorry, yes I did look at your flow chart ( systems diagram I would call it
) but only briefly then forgot by the time I'd read more of your replies and written a few on my own in the meantime.SimontY said:
The downside to this approach is a worsening of group delay over even a vented box, but in some cases it can work really well. Worth a try in any case, if you want a bash sometime.
I do appreciate your search for a more slamming midbass, it's something so few systems do well. Funnily enough one of the best systems for this is my mate in Barnsley's bedroom system with little (9 litres or so, 5 inch Polish/Czech made driver) Opera Duetto loudspeakers, you wouldn't think to look at it and certainly not from looking at the modelled responses, but to hear it it knocks your socks off.
Interesting, they look to have very long throw little drivers! (those Operas)
The best bass slam/impact/definition/impressiveness/weight (all those things) I heard from a hi-fi speaker came from the B&W DM702, and this in a poor dem room where most speakers sounded terrible in the bass.
http://www.bwspeakers.com/index.cfm/fuseaction/products.models/label/MODEL 703
Maybe I'd not be so bothered by my 'ok' bass if I'd not heard these speakers (most I've heard are nothing like them).
The best bass slam/impact/definition/impressiveness/weight (all those things) I heard from a hi-fi speaker came from the B&W DM702, and this in a poor dem room where most speakers sounded terrible in the bass.
http://www.bwspeakers.com/index.cfm/fuseaction/products.models/label/MODEL 703
Maybe I'd not be so bothered by my 'ok' bass if I'd not heard these speakers (most I've heard are nothing like them).
Thanks, interesting to see. I wonder if it's the very same driver.
Certainly my ~25wpc chipamp can send the cones to where they become uncomfortable.
However, when I had nearly 200wpc (Roksan Kandy) I was much more happy with the lould listening bass slam. Just a shame the sound was crap in other ways (harsh, and generally grey and useless).
edit: have I just made it sound like I'm looking in the wrong place for the lack of bass impact? Well...even with that amp it wasn't close to what 703s can do with even a poor home cinema amp!!!
Certainly my ~25wpc chipamp can send the cones to where they become uncomfortable.
However, when I had nearly 200wpc (Roksan Kandy) I was much more happy with the lould listening bass slam. Just a shame the sound was crap in other ways (harsh, and generally grey and useless).
edit: have I just made it sound like I'm looking in the wrong place for the lack of bass impact? Well...even with that amp it wasn't close to what 703s can do with even a poor home cinema amp!!!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.