My Morel MTM Project

[pheonix358]

Nicely done!

 

[AllenB]

I think there is something suspect with the phase,

It looks like group delay, maybe 1 to 2cm in the measuring distance.

On Love Over Gold I can't go past Telegraph Road, myself.

 

[wintermute]

Thanks Terry!

Allen yes it would only be about a cm I would guess, I should have marked where the feet were, to get an exact match but wasn't thinking

edit: I love telegraph road as well, but I just can't go past the strings and drums in private investigations. Piano is great too.

Tony.

 

[tcpip]

Hey, what's the current picture? Are you still planning to add the woofer at 300Hz?

 

[wintermute]

Hi tcpip, yes, but it may be 200Hz. I have to get my active crossover project back on the boil. I took some in room measurements of the MTM's and the 10" current cabinets last week with the idea that I would tailer the active crossover to give me the best response in room. I've put the circuit diagram into tinycad and need to do the board layout (verro board). I might need to revise to third order electrical on the bass though I suspect (only addition of one resistor to my circuit though I think).

It's a very slow process with me unfortunately

Tony.

 

[tcpip]

Hi tcpip, yes, but it may be 200Hz.

I have some thoughts on my own 3-way project which may have some bearing on this. Hope you get around to reading this page which has all my thoughts and rationale.

I have been thinking of this 3-way for a long time. There is this strange belief among many speaker builders that the human voice is between 300-3000Hz. The reality is that male voices can go much lower than 300Hz, maybe down to 150. And the 300-3000Hz range is the range of maximum sensitivity of our ears, but not the range of the human voices. Since we all want to get fantastic realism with the human voices, I don't want to entrust a woofer with the job. The driver which reproduces the thump-thump of bass percussion with large Xmax should not be asked to reproduce any part of the human voice range, I felt.

Therefore, I was thinking of crossing over my woofer at 100Hz. This is also the reason why I want a conventional midbass driver (at least 6", with an Fs below 80Hz) for my midrange. I can't use a typical ultra-low-Xmax ultra-sensitive 4" midrange driver.

There is another reason (very attractive to a lazy designer like me) to do this. If you push down the Fc lower and lower, it becomes less and less critical getting phase alignment between woofer and midrange at Fc. Think about it. At 100Hz, the wavelength is about about 11-12 feet. This means that woofer and midrange will need to be displaced by 3 feet to be out of phase by even 90-degrees. This gives the designer a much larger margin for error than, say, at 200Hz or 300Hz.

Therefore, for my 3-way, I was thinking of simply building an electrical LR4 LP for the woofer at 100Hz (which would also pretty much give me acoustic LR4), and an acoustic LR4 for the midrange in its sealed enclosure. I could do this by using LT to get 2nd order at 100Hz, and then adding another electrical LR2 to make it fourth order. I have a suspicion that textbook filters will do the job, and no careful phase matching will be needed.

These were the two reasons why I suggest that you consider moving your Fc down. You'll get the active filters built on breadboard in no time, and no measurements (other than impedance measurement of the midrange in its enclosure for the LT) will be needed.

Does this make sense?

 

[AllenB]

an acoustic LR4 for the midrange in its sealed enclosure. I could do this by using LT to get 2nd order at 100Hz, and then adding another electrical LR2 to make it fourth order.

Would you need the Linkwitz transform if you built the box for the 100Hz response?

 

[tcpip]

Would you need the Linkwitz transform if you built the box for the 100Hz response?

I guess you are referring to the midrange box, right?

In that case, if you have accurate T/S parameters of the driver and can size your enclosure to get 0.7 Qts in the box, then you can skip the LT. I was intending to size and shape the box as per my convenience, and then use LT to get the rolloff to fit. Gives me more flexibility in box dimensions.

 

[pheonix358]

I have some thoughts on my own 3-way project which may have some bearing on this. Hope you get around to reading this page which has all my thoughts and rationale.


snip

These were the two reasons why I suggest that you consider moving your Fc down. You'll get the active filters built on breadboard in no time, and no measurements (other than impedance measurement of the midrange in its enclosure for the LT) will be needed.

Does this make sense?

Please consider the following.

Most 6" drivers will happily go down to 120Hz in a sealed box. You get a better midbass performance.

I would never cross to a sub any lower than 150Hz and any sub should handle this easily. There is no stereo info at or below 150 Hz, ther can be at higher Fs.

Just my two cents worth.

P

 

[tcpip]

I would never cross to a sub any lower than 150Hz and any sub should handle this easily. There is no stereo info at or below 150 Hz, ther can be at higher Fs.

We are not talking about a sub here. We are talking about a woofer in a 3-way design. The fact of whether there is or isn't any stereo separation is not very relevant here, right, once we have already decided to put one woofer driver per channel?

Therefore can you please explain your thoughts a bit here? Why not cross over below 150Hz?

 

[pheonix358]

We are not talking about a sub here. We are talking about a woofer in a 3-way design. The fact of whether there is or isn't any stereo separation is not very relevant here, right, once we have already decided to put one woofer driver per channel?

Therefore can you please explain your thoughts a bit here? Why not cross over below 150Hz?

OK, I got that wrong! Sorry!

In that case with a woofer each side you don't set the frequency until you know what drivers you want to use. You have options.

You can run the mids sealed and allow them to be used full range. Great for passive x-overs because you avoid using large and expensive caps.

You can choose to cross via an active crossover, by far the best option.

Many modern sub woofers have problems in the mid bass zone. Yes, you can correct for these in the x-over, but why add the complexity when you can just roll them off prior to the problem area.

Choose your drivers and then decide where to cross them. A cheap variable electronic crossover is a great tool to help with that decision. Once you decide then you can make an outstanding circuit.

Basically, the less bass the mid handles the better is will sound. The method I use is to use the mid where it does not have xmax issues at full system power.

In the end, there are competing considerations and you as the designer get to make the decision that you like the most.

Lastly, Phasing issues are not much of a problem under about 1K because the wavelengths are so long that a few cm makes no discernible difference. (Well, as long as both are operating in phase )

Sorry again for my stuffed up answer above.

P

 

[wintermute]

Hi Terry, I haven't seen you post for ages, nice to see you are still lurking

TCPIP yes one of the reasons that I considered 200Hz now, is that this is pretty much the frequency where the mids are starting to roll off, in their 5L sealed enclosure (it may be a bit lower but some experimentation will tell me).

I figured with my 2nd order active combined with the natural 12db rolloff of the sealed enclosure I'd get close to my 4th order accoustic response. I'm hoping that 3rd order on the woofer will be enough (because to do third order I only need to add one resistor to my circuit, but to do 4th order I have to double up on my circuit, (one of the interesting things about an FDNR low pass). If it is not then I'm going to have to rethink the low pass part of the crossover, or moving the crossover point higher again.

Tony.

 

[tcpip]

Thanks a lot for your response. If I try to understand your pointers, they can probably be summarised in this line:

Choose your drivers and then decide where to cross them.

I am wondering why I need to choose carefully. I am choosing a midbass driver which people use in 2-way systems -- in other words, it gives "acceptable" performance down to the bottom octaves. (What is "acceptable" for me may not be acceptable for a more uncompromising designer, but you get the drift.)

And I am using a woofer which will happily go two octaves above my cut-off point, i.e. it will operate till above 500Hz.

In that case, why is any care needed to select the crossover point? The range of overlap is so large that I can choose the Fc based on factors other than driver characteristics, can't I?

Basically, the less bass the mid handles the better is will sound. The method I use is to use the mid where it does not have xmax issues at full system power.

True, I agree, but in the case of a proper purpose-built midrange driver. If I use a 6" or 7" midbass driver as my "mid", it's not really a "mid" in construction. In that case, I can assign it fairly wide-band duties, can't I? It will at worst be no worse than its performance in a 2-way.

 

[tcpip]

TCPIP yes one of the reasons that I considered 200Hz now, is that this is pretty much the frequency where the mids are starting to roll off, in their 5L sealed enclosure...

Understood. Yes, in that case, I too would not want to push the SPL curve of the midbass unit lower by applying LT.

In my planning, I was intending to have an enclosure at least 15-20 litres for the midbass driver, therefore getting a natural acoustic rolloff at about 100Hz or so, whatever the Qts may be. After that, I was thinking of applying LT to get it to behave the way I need. If your acoustic rolloff is at 200Hz, then it's best that you cross over to the woofer at that point.

 

[pheonix358]

Thanks a lot for your response. If I try to understand your pointers, they can probably be summarised in this line:


I am wondering why I need to choose carefully. I am choosing a midbass driver which people use in 2-way systems -- in other words, it gives "acceptable" performance down to the bottom octaves. (What is "acceptable" for me may not be acceptable for a more uncompromising designer, but you get the drift.)

And I am using a woofer which will happily go two octaves above my cut-off point, i.e. it will operate till above 500Hz.

In that case, why is any care needed to select the crossover point? The range of overlap is so large that I can choose the Fc based on factors other than driver characteristics, can't I?



True, I agree, but in the case of a proper purpose-built midrange driver. If I use a 6" or 7" midbass driver as my "mid", it's not really a "mid" in construction. In that case, I can assign it fairly wide-band duties, can't I? It will at worst be no worse than its performance in a 2-way.

If you are perfectly happy then go for it. In that configuration your max power will be limited by the mid. The mid will be as good as a 2 way and no better.

Your 3 way will not reach its max sonic performance. The idea of the 3 way is to limit the mid so that it is not ever nearing its x-max and will as a result will have a much greater power handling to match the woofer. As a benefit the mid frequencies will be cleaner. There will be no cone break up issues etc etc.

When all is said and done it is your choice!

Terry

 

[tcpip]

If you are perfectly happy then go for it. In that configuration your max power will be limited by the mid. The mid will be as good as a 2 way and no better.

"Will be as good as..." is a loaded phrase. I guess there are different meanings for the term "better".

Your 3 way will not reach its max sonic performance.

I guess there are different meanings for the term "sonic performance". I feel anyone who is combining the whump-whump of deep percussion and the delicate nuances of the male voice in the same driver's cone is compromising sonic performance. This is what happens when one crosses over at, say, 300Hz or higher.

The idea of the 3 way is to limit the mid so that it is not ever nearing its x-max and will as a result will have a much greater power handling to match the woofer. As a benefit the mid frequencies will be cleaner. There will be no cone break up issues etc etc.

I guess "the idea" of the three-way is different for different designers. My idea of the three-way is to use a wide-range mid which effortlessly stretches from the point where the tweeter leaves off till the fundamental frequencies of a deep male voice, but excludes the high-excursion area of percussion instruments. As a benefit, the lower-mid frequencies will be cleaner, without the detriment of the mid-mids or higher-mids.

Your "idea" does not match with my "idea." I feel your approach will result in muddy reproduction of lower-mid frequencies which you are choosing to delegate to the woofer. But then, what do I know?

When all is said and done it is your choice!

True.

 

[AllenB]

I feel anyone who is combining the whump-whump of deep percussion and the delicate nuances of the male voice in the same driver's cone is compromising sonic performance. This is what happens when one crosses over at, say, 300Hz or higher.

The same might be said about curtailing bandwidth on a small driver. What do you have against woofers?

 

[Jay]

I feel your approach will result in muddy reproduction of lower-mid frequencies which you are choosing to delegate to the woofer. But then, what do I know?

So it depends on the drivers used. You want each drivers to do what it is better to do. Crossing higher will generally be better (than the common subwoofer concept) but then you will need a high quality woofer to take care the lower mid.

 

[tcpip]

The same might be said about curtailing bandwidth on a small driver.

Am completely confused here. Which small driver's bandwidth am I curtailing?

What do you have against woofers?

Strange choice of words. Why should I or anyone have anything for or against a speaker driver?

Phoenix358 prefers to allow the woofers to operate up to, say 300Hz, so that the midrange driver is stressed less and can be confined to a narrower frequency band. I prefer to stop the woofer at something like 100Hz because I want the midrange (which is a 6" or 7" midbass driver in my case) to handle the entire range of frequencies of the human voice. Both of these appear sensible trade-offs to me -- they both lose something and get something in exchange.

Where did you get the impression that either of us has anything for or against a class of speaker driver?

 

[tcpip]

So it depends on the drivers used.

Yes. This is always true.

You want each drivers to do what it is better to do. Crossing higher will generally be better (than the common subwoofer concept) but then you will need a high quality woofer to take care the lower mid.

Both Phoenix358's approach and mine will be better than the common subwoofer concept. Having two separate woofers for two channels is at least as good as, and almost always is better than, a common subwoofer concept.

Our debate is about where should we cross over between woofer and mid. In his approach, he prefers to load his mid driver as lightly as he can, and hands over the lower-mid duties to the woofer. Yes, as you said, this will mean that a high quality woofer is needed.

I prefer to confine the woofer to the lowest two octaves, and use a large (by midrange standards), powerful midbass driver to handle the entire mids, starting from 100-120Hz or thereabouts. With my approach, the woofer quality will be a bit less important, but the midbass quality and ability to handle frequencies down to 100Hz or lower, will be very important. (That's why I cannot use the classic 4" low-Xmax midrange drivers; I must use a midbass driver of the kind used in 2-way designs.) I was in fact thinking of using wide-range drivers like the Jordan JX92S for midbass duty.