I have several pairs of Dunlavy speakers.
I am a huge fan, my budget finally allows what I drooled over in the ‘90’s
So I own a pair of SM-1’s and built boxes with downfiring Scanspeak 25w8565-00 to emulate Dunlavy models SM-IIIA, Corinthian, Aletha, Athena etc… all similar. I have read John Dunlavy state that there is no crossover between the Viva 6.5’s and the 10” I was under the assumption that there was only an inductor on the 10” downfiring Scanspeak. But recently a gentleman I met
On FB showed me a drawing of his crosssover for only the Scanspeak 10”
This is what I was given as it’s crossover
The DCR OF THE Inductor seems high.
I am a huge fan, my budget finally allows what I drooled over in the ‘90’s
So I own a pair of SM-1’s and built boxes with downfiring Scanspeak 25w8565-00 to emulate Dunlavy models SM-IIIA, Corinthian, Aletha, Athena etc… all similar. I have read John Dunlavy state that there is no crossover between the Viva 6.5’s and the 10” I was under the assumption that there was only an inductor on the 10” downfiring Scanspeak. But recently a gentleman I met
On FB showed me a drawing of his crosssover for only the Scanspeak 10”
This is what I was given as it’s crossover
The DCR OF THE Inductor seems high.
As this is DIY audio, a person could load that driver and schematic into a crossover simulator like Xsim, boxsim, VituixCAD or other software and see what the effect of the coil is. From experience I expect that the high resistance will lower the effective Qe of the driver and raise the Qbox of the driver, providing a bit of boost to the low end. That large inductance will serve to lower the upper bass to essentially give preference to the lower bass. In modern designs the best way to extend the bass response woofers ins sealed boxes is to employ a Linkwitz transform filter (asymetric 2nd order shelf filter) using DSP or line level equalization. Of course you want a woofer with a relatively large Xmax to be effective. In the absence of line level equalization, these large inductors were used to produce a first order filter that could work well but had limited effectiveness.
The loss of power in the coil resistance alone is a problem. But the damping factor also takes a nosedive.
Yes, but the power wasted in the resistance of the inductor is unusual.
A disadvantage of air core.
A disadvantage of air core.
1.8 pretty normal for this gauge air core.
People will start crying when you use I core.
Never bothered me.
Because DCR at this gauge will be more than half as less.
Depends really on the actual power or voltage
for real listening levels.
Since the gauge can also be lowered to lower DCR.
By ditching the typical huge magical inductors for 100 watt
or less systems.
Since 18 gauge air core will go well into actual 300 + watts.
and a I core 270 watts before it thinks of saturation
Anyways, if the speaker is close to 5 or 6 ohms
relatively low crossover point.
And likely a lot of Q can happen to the real world filter depending on the impedance curve
So a little resistance inline with the cap for compensation.
Nothing too surprising for 2nd order woofer circuit
People will start crying when you use I core.
Never bothered me.
Because DCR at this gauge will be more than half as less.
Depends really on the actual power or voltage
for real listening levels.
Since the gauge can also be lowered to lower DCR.
By ditching the typical huge magical inductors for 100 watt
or less systems.
Since 18 gauge air core will go well into actual 300 + watts.
and a I core 270 watts before it thinks of saturation
Anyways, if the speaker is close to 5 or 6 ohms
relatively low crossover point.
And likely a lot of Q can happen to the real world filter depending on the impedance curve
So a little resistance inline with the cap for compensation.
Nothing too surprising for 2nd order woofer circuit
Last edited:
Unfortunately you did not include any dimension. But to get an idea for orders or magnitude.
https://www.circuits.dk/calculator_multi_layer_aircore.htm
gives 71 meters of 0.8 mm^2 wire. Wire of 0.8 mm^2 has 1.25 ohm of resistance per 60 meters.
That seems to be in the ball park. It is a consequence of passive crossover.
Don't laugh on people who use 0.01 ohms cables for $1000/meter to keep the resistance path to the speakers as low as possible.
https://www.circuits.dk/calculator_multi_layer_aircore.htm
gives 71 meters of 0.8 mm^2 wire. Wire of 0.8 mm^2 has 1.25 ohm of resistance per 60 meters.
That seems to be in the ball park. It is a consequence of passive crossover.
Don't laugh on people who use 0.01 ohms cables for $1000/meter to keep the resistance path to the speakers as low as possible.
Lots of talk, talk, talk based on guessing and no calculations. So I loaded the TS parameters into boxsim. Here's the response with the series coil.
The first graph is the coil resistance 2.55 DCR ohms. See the nice relatively flat extended bass response.
Below is the response with the coil DCR set to zero, giving that very high damping factor.
With the zero ohm coil the bass response is not as good, with an emphasis at 80 Hz.
All this desire for zero ohm cables and high damping factors are just audiophile salesperson non-sense. When you want to do bass equalization in a passive crossover series resistance can help. For comparison look at the impedance of the woofer at 32 Hz. It is 50 ohms for crying out loud. You don't need 0.1 ohm cables to make bass. The power lost in that resistance is not important at all.
Go plug in the TS numbers, coil parameters and try it yourself with your favorite crossover design software. I used a 134 liter box as recommended by Madisound. That huge capacitor rolled off the response pretty rapidly, so I left that out. https://www.madisoundspeakerstore.com/approx-10-woofers/scanspeak-classic-25w/8565-00-10-woofer/
The first graph is the coil resistance 2.55 DCR ohms. See the nice relatively flat extended bass response.
Below is the response with the coil DCR set to zero, giving that very high damping factor.
With the zero ohm coil the bass response is not as good, with an emphasis at 80 Hz.
All this desire for zero ohm cables and high damping factors are just audiophile salesperson non-sense. When you want to do bass equalization in a passive crossover series resistance can help. For comparison look at the impedance of the woofer at 32 Hz. It is 50 ohms for crying out loud. You don't need 0.1 ohm cables to make bass. The power lost in that resistance is not important at all.
Go plug in the TS numbers, coil parameters and try it yourself with your favorite crossover design software. I used a 134 liter box as recommended by Madisound. That huge capacitor rolled off the response pretty rapidly, so I left that out. https://www.madisoundspeakerstore.com/approx-10-woofers/scanspeak-classic-25w/8565-00-10-woofer/
In some or many cases. Yes for sure
18 gauge inductor is more than good for 300 watts
Which exceeds the 100 to 200 Watt thermal rating of the
actual driver itself.
As mentioned real world listening levels
could be 30 to 60 watts at most.
1 to 1.8 ohms is pretty normal for 2.5 or 3 way woofer filter.
And yup indeed if crossover is lower. The impedance roller coaster
of a normal everyday Reflex enclosure . Is yes 30 to 60 ohms
Series resistance will Q up the response and often make a useful
peak.
What is great about sim, seen plenty of designs where large
inductors were used. And impedance compensation or different
cap values could even lower the coil value for same crossover point.
Or the system will easily sum with less filter, or wider baffle.
Darn I must have been really tired yesterday night when I tried to estimate the length of the wire (post #9). The OP gives the weight and the cross section which is all you need. Shame on me!
570 grams. Copper has a density of 8.9 g/cm3. Volume: 570/8.9 = 64 cm3=64000mm3
Cross section 0.8mm. Hence length 64000/0.8=80,000mm = 80m
Resistivity 0.017 ohm/meter/mm2: 80*0.017/0.8 = 1.7 ohms.
So yes, the DCR is correct.
And IMHO if the coil resistance is a quarter to a fifth of the speaker resistance I expect the effect to be non-negligible.
Now most of the box simulation programs are not able to include DC resistance in the model. Which means the response can be significantly off, right? Then active filtering and separate amplifiers must be (a) closer to the simulation response and (b) easier to design.
I don't say "better", because I can imagine (not sure) it is possible to design an enclosure such that changed characteristics of speaker and filter still produce a satisfactory result.
570 grams. Copper has a density of 8.9 g/cm3. Volume: 570/8.9 = 64 cm3=64000mm3
Cross section 0.8mm. Hence length 64000/0.8=80,000mm = 80m
Resistivity 0.017 ohm/meter/mm2: 80*0.017/0.8 = 1.7 ohms.
So yes, the DCR is correct.
And IMHO if the coil resistance is a quarter to a fifth of the speaker resistance I expect the effect to be non-negligible.
Now most of the box simulation programs are not able to include DC resistance in the model. Which means the response can be significantly off, right? Then active filtering and separate amplifiers must be (a) closer to the simulation response and (b) easier to design.
I don't say "better", because I can imagine (not sure) it is possible to design an enclosure such that changed characteristics of speaker and filter still produce a satisfactory result.
No you weren't tired, not until just now. It was me msg#2 with the Jantzen, not OP. I was trying to illustrate Dunlavy probably used a slightly smaller than 18 gauge aircore inductor. Sorry to have confused you, but please tell us the wire gauge thanks.
In any case, I was pleasantly surprised the first time I used aircore for bass LPF, and never looked back. Sometimes even added/guesstimated R deliberately (now I know why more visually than before thanks @olsond3).
Hahaha yes, now I see. Both your icons are yellow.
I would not be able to tell the wire gauge. Inductance is very dependent on coil geometry. If you have a very flat and large diameter coil the number of turns may be vastly different from a more compact coil. And then again because the diameter is large the wire length is very different.
So knowing weight and AWG you can calculate the length and the resistance. But for calculating AWG you really need to know the weight once the R is known.