Bill, I wasn't aware you had one of those Misco Bold North tweeters. In fact you are the only person I know that does. Being what it is, you could likely xover lower, even 1st order electrical. Compared to the LD25X, which you like more?
The complication with a 1st order theme would be 1st order electrical or acoustical and how many parts can be used. check out the image I found for a Thiel 3.6 crossover. Is it real? I don't know, but they were famously complex.
@Wolf said:
If i were to go that route, the goal would be a 6dB acoustic slopes, not 6dB electrical slopes. I have thus far managed to do it maybe twice myself.
This confuses me, I thought a slope was as slope, which may or may not equate to electrical order.
Acoustically there are slopes 6, 12, 18dB, etc, which are 1, 2, 3rd order etc.
For electrical, I thought a 6db slope, no matter how you got there electrically, would still be called a 6dB electrical slope, irrespective of the electrical order. No?
From what I gathered the electrical orders tend to go hand in hand with the basic part count (caps n coils).
The issue being that the driver impedance changes will alter the effectiveness of those parts depending on frequency. Often an issue with low passed drivers as their impedance tends to rise with frequency, having a nullifying effect on the low pass. Some of the can o worms I referred to was in relation to this...
Would a zobel then be ok to flatten the impedance rise? If so then what is uniquely being learned or gained by sticking to 1st order with extra added parts to make it work? Possibly there are some old electrical loopholes to dust off?(that would be exciting).
A lot of times, 6dB acoustic filters take a LOT of conjugate filters to make the circuit work as intended. Looking at some old Thiel, etc filters will show this.
As we learn the hobby, I find there is a huge disconnect in this area for the uninitiated. Most newbs, and even some long time builders, focus on electrical filters and equate them with the actual slope, not knowing the electrical and acoustic are 2 very different things. The acoustic slopes are what we should be focused on, which is a combination of the acoustic response of the drivers and the electrical transfer function the xover provides. For example, a tweeter is a sealed acoustic system with its chamber(s). This means the acoustic rolloff is inherently 12dB/Oct or 2nd order. Once you apply a 12dB electrical filter, this becomes 24dB/Oct or 4th order acoustic slope in most cases. It is really difficult to find a wide enough bandwidth tweeter that is even able to rolloff at 6dB acoustic because of this. Most of the designs I've seen of this type have been forced to xover higher than the average higher order acoustic design simply because the xover has to occur where the rolloff has not yet started. Besides stark contouring, you really can't remove a slope that is already present. You have to work around it by going far enough away from it.
More often than not in xover design, the 'zobel' as it's known, is not really required to get summations that align well. This area would be very difficult to understand or clarify what the filter is actually doing and whether or not it is singularly or a combination of; impedance compensation, part of the rolloff making it truly a damped 12dB or higher, or just a contour to make the drivers' acoustic response behave. For example, a true zobel intended for fixing the impedance usually also is committed to helping the rolloff to a degree. In the Prazise build I helped Psycoacoustics with over on PE some odd years ago, the zobel was not of the true form, and was made of much smaller than average values just to tweak the response. To also note, this design was first order acoustic "through the xover region", and then tapered to 3rd order into their relative stopbands. It had good phase alignment with crossing at -6dB for flat summation. This is an example of a good 1st order acoustic design.
No. Ellipticals involve notching outside of the passband to steepen the rolloff. These can also yield an amount of Ripple should it be done. Look up the Neville Thiel method, that is elliptical.
This is the example 1st order simulation (3) of the project I mentioned prior. Drivers are TB 27-1743 BeCu tweeter and Motus 5.5".
Notice how the rolloffs start more than an octave from the xover point, at 6dB/oct, hit the xover point, and then taper to 3rd order. The -6dB join for a flat summation. This is what is meant by having to 'spread' 1st order filters. The transfer function even shows 6dB slopes.
That's my kind of stuff right there. I was wondering whether or not it was a one octave rule or maybe more than that. But how can it be considered first order if they are 6 Db. down at XO?
@Eggguy said:
That's my kind of stuff right there. I was wondering whether or not it was a one octave rule or maybe more than that. But how can it be considered first order if they are 6 Db. down at XO?
I don't understand your question. They can't be -3dB at summation because that will cause a peak in response. It is not a one-octave rule necessarily, but they are usually spread quite far to make the summation work. Summations at -6dB while in-phase sum to flat response.
EDIT; maybe I understand your logic...
The slope of the rolloff is approximately -6dB/oct. to the xover point. Under textbook situations that would align perfectly with a one-octave rule to either side of the xover. There are some minor variations in actual real slopes, but due to spectral balance, actual driver responses, and physical offsets, this may shift a little and not be so precise.
Considering it seems most drivers roll off natively at a 2nd order achieving a 1st order acoustic is very difficult without some complex networks. I have tried before and the juice isn't worth the squeeze in my opinion.
I have had parts in the cart for a long while, but just can't hit the buy button. They are not cheap...
Wolf, I purchased a pair of each style of 6.5" XBL^2 woofers a while back and have subsequently done nothing with them. I can dig them out of storage send them to you for testing, if you would like to try before you buy. My tentative plan was a 2-way for the 'wideband' variant and the anchor of a 3-way for the other one. Misco also offered to make me an 8" prototype using the 6.5" motor and off the shelf parts for frame and cone and such. I think they're still investigating that particular product.
It's not really a hesitancy, or that I wouldn't feel safe that my money purchased a worthwhile product, but I thank you for your generous offer. I probably will hit buy at some point. It just seems to me they are a bit costly, but likely worth the price. Thanks.
Edit: I would feel better about purchasing Bold North over the equivalent price/size Ciare drivers for example.
No problem! If it helps at all, in person they look and feel quite well made. However, I didn't even run DATS on them before moving them to storage, so I have no numbers to back up that statement.
I'd prefer using a hybrid trans-tube with a tubed input stage, or a tubed line driver/buffer in the case of a themed event structure. This way the tubes don't get disconnected without a load, and we don't have to wait on tube temperature every time a swap is made.
However, the 'tube' premise can just be omitted too....
I have had parts in the cart for a long while, but just can't hit the buy button. They are not cheap...
Wolf, I purchased a pair of each style of 6.5" XBL^2 woofers a while back and have subsequently done nothing with them. I can dig them out of storage send them to you for testing, if you would like to try before you buy. My tentative plan was a 2-way for the 'wideband' variant and the anchor of a 3-way for the other one. Misco also offered to make me an 8" prototype using the 6.5" motor and off the shelf parts for frame and cone and such. I think they're still investigating that particular product.
If you ever pull those out of storage I'd be interested in your results. Love me some XBL motors. Have me seriously pondering hitting them up for a pair of bespoke 8's....
How about, "To Infinity, and Beyond!", must be sealed woofer with dome midrange and ribbon/amt tweeter of some sort, looking like a vintage Infinity speaker.
(I did make this suggestion to Ben previously, I just love the look of those red oak speakers... )
The Infinity builds would be super cool. Always like those old Infinity speakers. My brother had a pair of RS3000s. Couldn't afford the 3 way. His also had the Polycell Tweeter instead of the EMIT.
Comments
Bill, I wasn't aware you had one of those Misco Bold North tweeters. In fact you are the only person I know that does. Being what it is, you could likely xover lower, even 1st order electrical. Compared to the LD25X, which you like more?
InDIYana Event Website
(If referencing 1st order Xover) notch filter(s) would need to be limited- and/or if used, used in lieu of a HP or LP
The complication with a 1st order theme would be 1st order electrical or acoustical and how many parts can be used. check out the image I found for a Thiel 3.6 crossover. Is it real? I don't know, but they were famously complex.
If i were to go that route, the goal would be a 6dB acoustic slopes, not 6dB electrical slopes. I have thus far managed to do it maybe twice myself.
InDIYana Event Website
This confuses me, I thought a slope was as slope, which may or may not equate to electrical order.
Acoustically there are slopes 6, 12, 18dB, etc, which are 1, 2, 3rd order etc.
For electrical, I thought a 6db slope, no matter how you got there electrically, would still be called a 6dB electrical slope, irrespective of the electrical order. No?
Heck, make the challenge that MidWest audio drivers are required in some aspect.
From what I gathered the electrical orders tend to go hand in hand with the basic part count (caps n coils).
The issue being that the driver impedance changes will alter the effectiveness of those parts depending on frequency. Often an issue with low passed drivers as their impedance tends to rise with frequency, having a nullifying effect on the low pass. Some of the can o worms I referred to was in relation to this...
Would a zobel then be ok to flatten the impedance rise? If so then what is uniquely being learned or gained by sticking to 1st order with extra added parts to make it work? Possibly there are some old electrical loopholes to dust off?(that would be exciting).
Then there is baffle step to contend with.
A lot of times, 6dB acoustic filters take a LOT of conjugate filters to make the circuit work as intended. Looking at some old Thiel, etc filters will show this.
As we learn the hobby, I find there is a huge disconnect in this area for the uninitiated. Most newbs, and even some long time builders, focus on electrical filters and equate them with the actual slope, not knowing the electrical and acoustic are 2 very different things. The acoustic slopes are what we should be focused on, which is a combination of the acoustic response of the drivers and the electrical transfer function the xover provides. For example, a tweeter is a sealed acoustic system with its chamber(s). This means the acoustic rolloff is inherently 12dB/Oct or 2nd order. Once you apply a 12dB electrical filter, this becomes 24dB/Oct or 4th order acoustic slope in most cases. It is really difficult to find a wide enough bandwidth tweeter that is even able to rolloff at 6dB acoustic because of this. Most of the designs I've seen of this type have been forced to xover higher than the average higher order acoustic design simply because the xover has to occur where the rolloff has not yet started. Besides stark contouring, you really can't remove a slope that is already present. You have to work around it by going far enough away from it.
More often than not in xover design, the 'zobel' as it's known, is not really required to get summations that align well. This area would be very difficult to understand or clarify what the filter is actually doing and whether or not it is singularly or a combination of; impedance compensation, part of the rolloff making it truly a damped 12dB or higher, or just a contour to make the drivers' acoustic response behave. For example, a true zobel intended for fixing the impedance usually also is committed to helping the rolloff to a degree. In the Prazise build I helped Psycoacoustics with over on PE some odd years ago, the zobel was not of the true form, and was made of much smaller than average values just to tweak the response. To also note, this design was first order acoustic "through the xover region", and then tapered to 3rd order into their relative stopbands. It had good phase alignment with crossing at -6dB for flat summation. This is an example of a good 1st order acoustic design.
InDIYana Event Website
Would what you described be considered an elliptical filter?
No. Ellipticals involve notching outside of the passband to steepen the rolloff. These can also yield an amount of Ripple should it be done. Look up the Neville Thiel method, that is elliptical.
InDIYana Event Website
This is the example 1st order simulation (3) of the project I mentioned prior. Drivers are TB 27-1743 BeCu tweeter and Motus 5.5".
Notice how the rolloffs start more than an octave from the xover point, at 6dB/oct, hit the xover point, and then taper to 3rd order. The -6dB join for a flat summation. This is what is meant by having to 'spread' 1st order filters. The transfer function even shows 6dB slopes.
InDIYana Event Website
That's my kind of stuff right there. I was wondering whether or not it was a one octave rule or maybe more than that. But how can it be considered first order if they are 6 Db. down at XO?
This is a series xover, but it is also Elliptical via NTM method:
https://techtalk.parts-express.com/forum/speaker-project-gallery/63037-seriatim-from-indiyana-and-mwaf
InDIYana Event Website
I don't understand your question. They can't be -3dB at summation because that will cause a peak in response. It is not a one-octave rule necessarily, but they are usually spread quite far to make the summation work. Summations at -6dB while in-phase sum to flat response.
EDIT; maybe I understand your logic...
The slope of the rolloff is approximately -6dB/oct. to the xover point. Under textbook situations that would align perfectly with a one-octave rule to either side of the xover. There are some minor variations in actual real slopes, but due to spectral balance, actual driver responses, and physical offsets, this may shift a little and not be so precise.
InDIYana Event Website
Thanks, and thanks again for understanding what you generously stated as logic.
Considering it seems most drivers roll off natively at a 2nd order achieving a 1st order acoustic is very difficult without some complex networks. I have tried before and the juice isn't worth the squeeze in my opinion.
The circumstances are few that it works out well, IME, as well.
InDIYana Event Website
Wolf, I purchased a pair of each style of 6.5" XBL^2 woofers a while back and have subsequently done nothing with them. I can dig them out of storage send them to you for testing, if you would like to try before you buy. My tentative plan was a 2-way for the 'wideband' variant and the anchor of a 3-way for the other one. Misco also offered to make me an 8" prototype using the 6.5" motor and off the shelf parts for frame and cone and such. I think they're still investigating that particular product.
It's not really a hesitancy, or that I wouldn't feel safe that my money purchased a worthwhile product, but I thank you for your generous offer. I probably will hit buy at some point. It just seems to me they are a bit costly, but likely worth the price. Thanks.
Edit: I would feel better about purchasing Bold North over the equivalent price/size Ciare drivers for example.
InDIYana Event Website
No problem! If it helps at all, in person they look and feel quite well made. However, I didn't even run DATS on them before moving them to storage, so I have no numbers to back up that statement.
possible Indie theme: _bring a little something horny _
front, rear, and/or lense minimum one driver so employed
"Monkey Business" - Make a speaker akin to Devore/Orangutan aesthetics.
InDIYana Event Website
and power with a very low watt tube amp
"FlatulANT" (ant farts)
I'd prefer using a hybrid trans-tube with a tubed input stage, or a tubed line driver/buffer in the case of a themed event structure. This way the tubes don't get disconnected without a load, and we don't have to wait on tube temperature every time a swap is made.
However, the 'tube' premise can just be omitted too....
InDIYana Event Website
If you ever pull those out of storage I'd be interested in your results. Love me some XBL motors. Have me seriously pondering hitting them up for a pair of bespoke 8's....
How about, "To Infinity, and Beyond!", must be sealed woofer with dome midrange and ribbon/amt tweeter of some sort, looking like a vintage Infinity speaker.
(I did make this suggestion to Ben previously, I just love the look of those red oak speakers...
)
^with reasonable size and cost restrictions, that would be some pretty cool builds.
I'm a sucker for horns (if you couldn't tell). Been daydreaming another horn topped build.. even though I probably already have 5 builds lined up.
The Infinity builds would be super cool. Always like those old Infinity speakers. My brother had a pair of RS3000s. Couldn't afford the 3 way. His also had the Polycell Tweeter instead of the EMIT.