Modifying Drivers QTS with added resistance

[strangeandbouncy]

HI,


  my chum took back his LS3 5as. Even after he said I could keep them at least for comparison in development. Ho Hum.


 So I am forging ahead with my unusual nearfields! For now, I am not going to mess with transmission line, but build sealed boxes. - I need something pdq! Thanks to a member here, I have a BSS FDS360 active crossover, my amps, Monacor SBH-130 5 inch drivers, and KEF T27 tweeters. tomorrow I am seeing a cabinet maker who is going to knock up some boxes to prototype. I have calculated an internal volume of @5.7 litres, pretty much the same as the original LS3 5a in fact. Another online calculator gives a smaller volume(approx 20%), citing a need to add 20% for un-damped drivers. This same site

      http://www.lautsprechershop.de/english/index_tools.htm


   has a calculator to re-calculate the volume as Qts varies with series resistance. - Never came across this before!. Monacor SBH130 is way more efficient than T27, (89dB for 1Watt @1m against 80dB for 1 Watt @1m). I can well afford to lose some efficiency here. It seems I can get a slightly lower system cut-off by inserting a resistor in series at the expense of a bigger box. I am assuming a recommended Qtc of 0.71. Maybe I wont need to eq the bass so much, if at all, to achieve a similar response to the LS3 5a . . . Perhaps I can extend even lower if I eq as well!


  Here is a table form the caculator.

resistance           Volume       -3dB     -8dB

    0                    4.7l           99Hz     63Hz

    1 Ohm            6.7l           85Hz     55Hz

    2 Ohm            8.9l           77Hz     49Hz

    3 Ohm           11.7l          70Hz     45Hz

    4 Ohm           15.4l          64Hz     41Hz



These volumes are for so-called damped drivers, and are 20% smaller than expected. I shall allow the extra 20%, and add extra wooden blocks if necessary to reduce the volume.


  it looks like I can potentially achieve the same response as a LS3 5a(-3dB @ 70Hz) WITHOUT eq, by approximately doubling the volume and a bit more) and adding a 3 Ohm resistor in series. Size is not an issue here for me. I will also lose about 3dB. Again, not a problem. All bodes well! Yay!

  I have also managed to score an incredibly rare BSS eq card to go in the FDS360 if I need it - for a measly sum. Those puppies are rarer than rockinghorse Sh*t!


  Very excited about this. I am hoping that the Monacor will do the business. It looks good on paper, and is the flattest I could find from 100Hz to nearly 10k. Everyone is singing the praises of another Monacor 5 inch, the SPH-135AD. Looks much worse on paper, being far from flat. If this doesnt work, I am going to look at the Beyma5MP60N. It is more suited to bass-reflex, and has some nasty cone breakup like all polypropylene drivers seems to. There is also a 5inch from Coles, but I cant find any proper info yet.


  Any comments would be most gratefully received!



    ANdyP

 

[abbey road d enfer]

Apparently, your box calculators know of only one type of alignment, maximally flat.
You should use WinISD. Free download here:
http://www.linearteam.dk/default.aspx?download=winisdpro
It gives you much more choices and possibilities of tweaking.
The idea of inserting resistors to diminish electrical damping (increasing Qes) is not very attractive to me. The improvement in LF response is obtained at the detriment of transient response.

 

[strangeandbouncy]

Hi Abbey



  thank you very much for your response. I didnt realise that it would affect transient response. I would love to find out why!

  Maximally flat seemed a good idea to me . . . thanks for the link. Ill go there now!

    Thankfully, I can easily build a 12l box, and add blocks of wood to reduce. At the end of the day, it is experimentation - trial and error that is the way to go, I guess.

  Would adding active eq effect the transient response too, I wonder. You are effectively doing the same thing, arent you . . . .


    kindest regards,


      ANdyP

 

[strangeandbouncy]

    http://www.linearteam.dk/default.aspx?download=winisdpro



Oh, bugger, its PC . . . .



   

 

[abbey road d enfer]

Would adding active eq effect the transient response too, I wonder. You are effectively doing the same thing, arent you . . . .

Adding a resistor in series reduces whatever is left of the electromagnetic control exerted on the voice-coil. Some designers actually recommend the use of high Qes (little electromagnetic damping) under the fallacious concept that the loudspeaker is almost current-driven and then the voice-coil movement is an exact image of the current.
But actually, since there is very little damping, the response to a pulse is subject to more overshoot and oscillations than when properly damped.
On the contrary, using electronic EQ to achieve a specific alignment allows to retain proper damping of the movement.
The effects are not shown in a typical frequency response test, but waterfalls will be very different.

 

[strangeandbouncy]

Hi Abbey,



    thank you for that. It is interesting to note that the original LS3 5a uses a resistor to lower the efficiency and compensate for the rising response of the B110 to match the T27(actually, a resistor and industor in parallel). I guess that it will all depend on the particular system, and some compromise is always needed!

 

[abbey road d enfer]

Hi Abbey,



    thank you for that. It is interesting to note that the original LS3 5a uses a resistor to lower the efficiency and compensate for the rising response of the B110 to match the T27(actually, a resistor and industor in parallel). I guess that it will all depend on the particular system, and some compromise is always needed!

The inductors and resistors are just part of the crossover network. Anyway, an inductance in parallels with a resistor have very low resistance at LF so that would not influence the LF damping and response (a single resistor would).
As (almost) always the tweeter has higher sensitivity than the woofer, so the tweeter needs some attenuation. L3 is both the inductor in the 3rd-order HPF and used as an autotransformer to provide 2 different HF level options.
It's interesting to note that the LPF is 4th-order and the HPF is 3rd-order. But the resistors in the LPF degenerate the slope and make it closer to 2nd-order.

 

[strangeandbouncy]

Abbey,


  thank you for the insight! They are trying to boost the bass, so that makes sense. Here is a plot of the croosover output.

    Kindest regards,


    ANdyP

 

[strangeandbouncy]

OK,


  so all bits ordered, or received, cabinets being built by cabinet maker( I have nowhere to build em meself!)


  I am really excited at the prospect of this project!


  Ill keep you posted . . .


    Kindest regards,



        ANdyP