DIY vs brand speaker

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This is a compromise, if you want to play really loud with these speakers then the max long term power would be improved by moving the crossover frequency up. However as you do this the off axis polar response will get worse and worse. Also woofers tend to distort in this region as the cone has stopped moving as one and breaks up with different sections of the cone moving at different time. As the cone goes more and more into cone break up they resonate and the cone starts to ring like a bell this causes energy storage and poor trasient response.

The tweeter you have chosen seems to have been designed with running at relatively low frequencies in mind, hence the 2k power bandwidth measurement. Moving it another 200Hz lower is unlikely to cause any power problems as the tweeter is quite sensitive and if you look at the filter electrical responses you can see the filter is already attenuating quite alot at the Xover frequency. Combining this with the extra order of filtering should give you the full power response. What is much more difficult to predict is how the distortion performance will behave. As you increase cone excursion on the tweeter it will distort more and the lower the frequencies you run into it will increase the excursion. I usually find for home use using as much of the range of the tweeter as possible gives a lower distortion, better transient response design which sounds better, typically better sound than running the woofer higher. The trade off is that it might not behave as well at very very high power levels, but as this design is quite sensitive anyway at about 92dB, you will get alot of sound for the input power.

Its my best guess, I can’t promise it will work well in all circumstances but I think it will give you a good compromise between sound quality, transient response, polar response, power response and distortion. If I were making this design I would go with the more complex crossover.

I will look to see if you can shift the crossover up to 2K with this design by only changing the caps (as this it is typically cheaper and easier to get hold of capacitors) which would give you the option of trying both however I think I found the phase was getting poor.
I don't have words to thank you for all your help. I am learning a lot with your explanations and also having the possibility of building the speakers with a solid help. As I said before, I had the idea that designing a two way speaker it was better to crossover it as high as possible to limit the amount of mid range going to the tweeter. I see your experience is different for home use. In any case, playing with the crossover, we can change the way the drivers are set up, but we need to have range with both drivers. I was looking to Part Express and found they have a line of speakers, Dayton Audio, where they have woofers going as high as 7000 Hz. They aren't very expensive but I don't have any idea about the quality of those drivers. I think that having a woofer that responds as high as 4K Hz or more and a tweeter that can be set as low as 2k Hz, we have enough range to change the crossover in case it be necessary. I mean more possibility of adjusting the design after listening to it. However, if the selected woofer has a top of 2K Hz, it forces you to crossover at that point or lower and there is not margin to play. Anyway, I am very excited and hoping to begin the project soon.
 
Last edited:
This speaker seems to have reviewed well but I have some concerns about it. The plots have been smoothed, it always a bit misleading to compair unsmoothed plots with smoothed ones. There could be short peaks in the response that we cant see and the peaks we can see might be worse for short frequecy spans

The cone break up at 1khz looks like it is resonating quite strongly, notice the disturbance in the impedance plot at the same frequency. This is indicative of resonance in the cone and thus energy storage causing poor transient response. You always get a bit of this but it seems quite high in this driver. It might sound quite musical depending on the behaviour of the resonance but it is not accurate and usually smears the image slightly. See the regions I have circled in the plots.

If it is especially cheap it might be worth putting up with these faults and designing to minimise the effects but they would be quite hard to mask in the crossover.
 

Attachments

  • 67E1277F-75AB-4C8D-B2A2-4302734230E2.jpeg
    67E1277F-75AB-4C8D-B2A2-4302734230E2.jpeg
    918.7 KB · Views: 65
This speaker seems to have reviewed well but I have some concerns about it. The plots have been smoothed, it always a bit misleading to compair unsmoothed plots with smoothed ones. There could be short peaks in the response that we cant see and the peaks we can see might be worse for short frequecy spans

The cone break up at 1khz looks like it is resonating quite strongly, notice the disturbance in the impedance plot at the same frequency. This is indicative of resonance in the cone and thus energy storage causing poor transient response. You always get a bit of this but it seems quite high in this driver. It might sound quite musical depending on the behaviour of the resonance but it is not accurate and usually smears the image slightly. See the regions I have circled in the plots.

If it is especially cheap it might be worth putting up with these faults and designing to minimise the effects but they would be quite hard to mask in the crossover.

The good reviews and technical data called my attention but I don't have the resources neither knowledge to judge it right. The price is half of the Scanspeak. There are other Dayton drivers with good reviews. I would like to find a woofer with a frequency response higher than 2000 Hz, no matter if the crossover be designed to less than 2000 Hz. I would like to feel safe in case a live test doesn't sound the way I like. Having a woofer with more mid-range allow to increase the crossover point and to release the tweeter in any case. Maybe it isn't necessary but the possibility is there.
 
The above driver has low Qms.

I understand the low Qms doesn't have the last word. Qts is more important according to some articles I read. For instance one of them says it :


Qms will tend not to make much of a difference, since the forces arising from mechanical properties o the cone and suspension will be small compared to the loading caused by a nice enclosure and by the electrical properties.

Reading this old discussion, I am still confused :

The significance of high Qms..?
 
Last edited:
Of the two I think the DS215-8 would probably be easier to work with as its response is more controlled over a larger frequency range. However neither of these speakers has 60deg of axis plots (They stop a 45deg) so their off axis response could become very poor the further out you go. (the scan speak included 60deg)


Whilst they do offer extended on axis response the off axis response of any 8 inch driver becomes problematic for good power response in the room as you go over 2kHz. It why most manufacturers of 2 way speakers now use 6 - 6.5 inch drivers rather than 8 inch.
 
If I am understanding all this right, high Qms = softer surround and less rigid, and low Qms = harder surround material and more rigid cone (or spider) So, some people favor one, and other the opposite. My Paradigm Studio 100 V2, that were top of the line, have rigid woofers. The surround material is rigid and doesn't allow the cone moves easily. I tested them with a Classe power amp with a very high damping factor, even driven with this amp, the Paradigm's woofers didn't move a lot (even at high volume) One observation I can say is lower line Paradigm's models didn't have woofers with hard surround materials. So, Paradigm designers could have a good reason to chose low Qms drivers to build their top of the line speakers with them. I am sorry if what I am reasoning isn't correct. I am learning.!
 
Of the two I think the DS215-8 would probably be easier to work with as its response is more controlled over a larger frequency range. However neither of these speakers has 60deg of axis plots (They stop a 45deg) so their off axis response could become very poor the further out you go. (the scan speak included 60deg)


Whilst they do offer extended on axis response the off axis response of any 8 inch driver becomes problematic for good power response in the room as you go over 2kHz. It why most manufacturers of 2 way speakers now use 6 - 6.5 inch drivers rather than 8 inch.
In a few words, the Scanspeak is still the winner ??
 
frugal-phile™
Joined 2001
Paid Member
A low Qms means the driver is mechanically providing the necessary damping. Most speakers today instead depend on the amplifier to supply the damping (Qes). One cannot make any judgements unless the entire system is considered. In the end the total of th emechanical damping & the electrical damping is what is important (Qt)

One thing a low Qms driver allows is an amplifier with higher output impedance to work well.

dave
 
A low Qms means the driver is mechanically providing the necessary damping. Most speakers today instead depend on the amplifier to supply the damping (Qes). One cannot make any judgements unless the entire system is considered. In the end the total of th emechanical damping & the electrical damping is what is important (Qt)

One thing a low Qms driver allows is an amplifier with higher output impedance to work well.

dave

Yes, that is what I understand with my poor knowledge. The driver is "playing the role" of a high damping factor amplifier, or at least, helping the poor damping of most modern home receivers and integrated to control the sound. Not sure if I am right.!
 
Last edited:
frugal-phile™
Joined 2001
Paid Member
Most modern SS amplifiers, including commercial receievers & integrated amps, have low enuff output impedance that they are essentially the same (ie damping factor of something like 20+ are essentially the same).

A high damping factor is in no way a measure ofamplifier quality, just a clue as to which speakers will work best with it.

dave
 
Most modern SS amplifiers, including commercial receievers & integrated amps, have low enuff output impedance that they are essentially the same (ie damping factor of something like 20+ are essentially the same).

A high damping factor is in no way a measure ofamplifier quality, just a clue as to which speakers will work best with it.

dave
Yes, there are speakers demanding a high damping factor amplifier. It all depends of the drivers. However, I understand a high damping factor is a resource that offer better speaker control, a better way to drive the speakers without distorting.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.