It all depends on the speed setting .........33.3 or 45 rpm
'Old blue eyes' never liked the sound of his voice ...... he apparently said.
C.M
I can see where you are coming from. There are many considerations leading to more ways or less ways. I'm more of a three way kind of guy, however I think at the end it needs to be done well for all the right reasons.
Hello,
I am new here on this fantastic thread.
I am going to build a two way speaker crossed over at 1920Hz. For my 4 ohm woofer this should give me a 0.33 mH inductor.
However, to accommodate a test enclosure I should run a 0.47 ohm resistor in series before the woofer. The inductor's Rdc wil be around 0,18 ohm. Which gives a total of 0,65 ohm.
Now my question is; will this raise my woofer's impedance to 4,65 ohm which will give me
a 0.39 mH inductor? Is this the right way to deal with this (still learning
I am new here on this fantastic thread.
I am going to build a two way speaker crossed over at 1920Hz. For my 4 ohm woofer this should give me a 0.33 mH inductor.
However, to accommodate a test enclosure I should run a 0.47 ohm resistor in series before the woofer. The inductor's Rdc wil be around 0,18 ohm. Which gives a total of 0,65 ohm.
Now my question is; will this raise my woofer's impedance to 4,65 ohm which will give me
a 0.39 mH inductor? Is this the right way to deal with this (still learning
Your thinking is reasonable, the higher value might be the better choice.
Maybe though, there isn't much difference. Especially if you later discover that balancing the bass around the cabinet size makes a larger value useful. I think it is good to have pairs of inductors of different values spare in any case.
Maybe though, there isn't much difference. Especially if you later discover that balancing the bass around the cabinet size makes a larger value useful. I think it is good to have pairs of inductors of different values spare in any case.
WOW Just to confirm that I understand correctly.?
A first order (inductor only) woofer and a second order (cap and inductor ) are 180 degrees out of phase. This makes more sense for a simple design. ?
I must admit I have been running my two way Volt 2208 and Morel tweeter this way for quite a while.. This also gave me the purest cleanest sonics after slowly and surely working out a xover point by ear as the Volt gets very edgy above 3k I guess.
45 degrees. Hmm many thanks and this will explain one of the problems consistantly experienced whilst playing around making speakers but hardly ever highlighted. ? I am not sure even some well known speaker designers always remember or know this.
It could be said that it could make a mockery of simple theoretical computerised designs ?. when they do not work in practice. oooops.
My first real hifi speaker was the Mission 770 and that was a 1st order/ 2nd order.
Can you comment as to whether a notch filter applied to the higher end above the xover frequency is going to effect this? It is just one of the options I am considering to smooth out a weird anomoly clearly measurable and audible from the Volt woofer. 4k ish, especially at high volume.
I also must admit that over 50 years of "playing around" using my ears as arbitrator I came to the conclusion that this simple design always seemed too have a weaker midrange near that crossover point despite much better imaging than 2nd order and higher. The bigger the woofer the more obvious in general.? Even allowing for claimed higher frequency performance by the speaker . Greater than the 3k hz rolloff on the Volt units. Just an observation. ?
Dealing with woofers near the crossover region doesn't have to be difficult. They tend to get more directional as the cone approaches a wavelength (345 metres divided by the frequency of interest). It is best to stop using them at this point.
There are two reasons for this. For one, the cone can bend. Sound can be produced with inconsistent directionality, and with inconsistent response. The large peak at higher woofer frequencies can be an issue especially when it is not brought down far enough. How far is enough? Aim for -20dB and possibly -30dB.
It really is better if the speaker is crossed and done before that frequency because the response peak can be mostly focused in front of the speaker. Turning it and listening from one side by 15-30 degrees may show this. Notching the peak hard may take too much away if a woofer is used too high.
The second reason is that as a wavelength gets smaller than the cone, the directivity narrows because at wider angles (at one given point) the sound meets out of phase from each side of the cone.
What you get is less substance from the woofer at that frequency. In other words from the front they sound right but after the sound escapes into the room and returns, it is missing something. This narrow beam of sound amounts to less power in the reverberant field.
There are two reasons for this. For one, the cone can bend. Sound can be produced with inconsistent directionality, and with inconsistent response. The large peak at higher woofer frequencies can be an issue especially when it is not brought down far enough. How far is enough? Aim for -20dB and possibly -30dB.
It really is better if the speaker is crossed and done before that frequency because the response peak can be mostly focused in front of the speaker. Turning it and listening from one side by 15-30 degrees may show this. Notching the peak hard may take too much away if a woofer is used too high.
The second reason is that as a wavelength gets smaller than the cone, the directivity narrows because at wider angles (at one given point) the sound meets out of phase from each side of the cone.
What you get is less substance from the woofer at that frequency. In other words from the front they sound right but after the sound escapes into the room and returns, it is missing something. This narrow beam of sound amounts to less power in the reverberant field.
Last edited:
Once again a WOW and the reply is a perfect description with clarity of explanation that goes into different areas of problems which people and myself will also find helpful to explain why simply moving from one room size |(especially width and height) to a larger wider room can never get back to the sound they created in the first place.
I think this also explains why the long held view that if the sound is ok from listening from outside the room is ok that it must be something else. (quote )
So, in my case pushing the boundaries by experimentation and going close or past the limitations of design then THIS is really important.
Once again many thanks and I will have another go at lowering the xover point.
The idea of notch filtering seems counter to my sonic Nivarna. So that is a relief (I hope)
Dave
I think this also explains why the long held view that if the sound is ok from listening from outside the room is ok that it must be something else. (quote )
So, in my case pushing the boundaries by experimentation and going close or past the limitations of design then THIS is really important.
Once again many thanks and I will have another go at lowering the xover point.
The idea of notch filtering seems counter to my sonic Nivarna. So that is a relief (I hope)
Dave
Notch filters aren't so bad..
Say the peak is in the middle of the usable band so it can't be avoided. This is a region where the cone doesn't bend and the dispersion is consistent. It might be due to a suspension resonance. This changes the response a little around one frequency.
The ideal is to fix the response. This is also shown in the phase response, which wiggles with the peak but will be drawn smooth again with the filter.
As usual with passive crossovers, be aware of component interactions. A side effect of reducing the power using passive components is that the impedance is raised around the peak.
Say the peak is in the middle of the usable band so it can't be avoided. This is a region where the cone doesn't bend and the dispersion is consistent. It might be due to a suspension resonance. This changes the response a little around one frequency.
The ideal is to fix the response. This is also shown in the phase response, which wiggles with the peak but will be drawn smooth again with the filter.
As usual with passive crossovers, be aware of component interactions. A side effect of reducing the power using passive components is that the impedance is raised around the peak.
Last edited:
Thanks, I have a pair of seas t17rex speakers, which I would like to use instead of t18rex in one of these projects.
The first project I saw is the seas loki, in reflex, here you can find it SEAS Loki MkIII
The second project is in a closed box, aw7, here it is Haku - Hifitalo[1].pdf
Both use t18rex H1353-08/06 T18REX/XFC
http://www.seas.no/images/stories/prestige/pdfdatasheet/H1333_T18RE_XFCTV2_Datasheet.pdf
I would like to create the aw7 project, but using the speakers I have, the t17rex seas, mine is a 4 ohm woofer version http://www.seas.no/images/stories/vintage/pdfdataheet/h0723_t17rex_coaxf.pdf
The question is: if I use the aw7 project crossover, changing the value of L1 to 0.9 mH, can it work well?
Will I not have problems with the tweeter that is different from the t18rex used in the project?
Or do you think the loki project crossover is better with some modifications?
Thank you.
I cannot see this link, maybe it requires registration.
Next, the tweeter on the T17 has a resonance about an octave higher than the T18. This is a larger difference, compared with the difference between 17 and 18cm. I think you will have to run the woofer higher.
For the woofer it is OK to go up to 2kHz. I'm not sure about 4kHz it might begin to sound thin. If you try using a lower order filter to spread the effect, it will reduce power handling. If you use a lower crossover it will reduce power handling also.
But, maybe the cone gives a waveguide effect to the tweeter. This makes me think you should cross at least 2k5Hz, because in this case, the narrowing directivity is wanted, but you can't push this too far unless the woofer is well behaved.
The tweeter is different in sensitivity, in response, and has a different impedance peak.
I guess the change in crossover will involve changing more than one component. I think the T18 crossover can be a starting point but you may also end up adding new components or taking them away.
I have looked at the two different drivers. The first thing I notice is the T17 wants a smaller enclosure.
Next, the tweeter on the T17 has a resonance about an octave higher than the T18. This is a larger difference, compared with the difference between 17 and 18cm. I think you will have to run the woofer higher.
For the woofer it is OK to go up to 2kHz. I'm not sure about 4kHz it might begin to sound thin. If you try using a lower order filter to spread the effect, it will reduce power handling. If you use a lower crossover it will reduce power handling also.
But, maybe the cone gives a waveguide effect to the tweeter. This makes me think you should cross at least 2k5Hz, because in this case, the narrowing directivity is wanted, but you can't push this too far unless the woofer is well behaved.
The tweeter is different in sensitivity, in response, and has a different impedance peak.
I guess the change in crossover will involve changing more than one component. I think the T18 crossover can be a starting point but you may also end up adding new components or taking them away.
Here the aw7 pdf that I used as a crossover reference, as you can see on the woofer there is only L1 which I carry at 0.9 mH because the woofer I have has 4 ohms.
The box I have has about 9 liters and I think it's important to say I'm going to help the speakers with a subwoofer (2.1).
I am ignorant in the field of crossover, but I would like to use the seas, because they have been given to me and they are new.
Thank you.
The box I have has about 9 liters and I think it's important to say I'm going to help the speakers with a subwoofer (2.1).
I am ignorant in the field of crossover, but I would like to use the seas, because they have been given to me and they are new.
Thank you.
At first I thought that this is a rather simple crossover, but then I looked at the whole design. The plot of excess group delay is promising and makes me think this can sound good.
C1 behaves as a first order filter.
L2 and C3 probably take the impedance peak, and turn it into an impedance dip because there is no resistor here. This would decrease the response by making C1 and the resistors more effective at this frequency.
R2 reduces the level of the tweeter, while R1 and C2 give some level back only at higher frequencies.
L1 might be less than a first order filter if the woofer impedance is rising. The effect depends on both the component and the load impedance.
C1 behaves as a first order filter.
L2 and C3 probably take the impedance peak, and turn it into an impedance dip because there is no resistor here. This would decrease the response by making C1 and the resistors more effective at this frequency.
R2 reduces the level of the tweeter, while R1 and C2 give some level back only at higher frequencies.
L1 might be less than a first order filter if the woofer impedance is rising. The effect depends on both the component and the load impedance.
First though is yes, this is a good idea.
Be careful, sometimes a 4 ohm woofer is different to the 8 ohm version of the same woofer.. more than just the impedance.
Is the tweeter a 4 ohm version as well?