Introduction to designing crossovers without measurement

Got a question for you AllenB I have a JBL 2225 woofer and a JBL LE8 fullrenge. I'd like to use the LE8 as midrange only. The issue I have is the 2225 is 96db and the LE8 is 89db. I was hoping that the placing the LE8 in a small compartment and rolling it off near 400Hz it would allow it to be more efficient and play loader... be a better match to the 2225 woofer... possible?
 
The following is literature found on the JBL spec. sheet for the LE8T-H models:

"2 The sensitivity rating of JBL full range loudspeakers is based on a signal swept from 500 Hz to 2 5 kHz, rather than the conventional I kHz single frequency test signal Usable sensitivity of the LE8T-H may be substantially greeter than that of loudspeakers with higher published ratings"
 
Moderator
Joined 2008
Paid Member
Putting it in a smaller box isn't going to raise the sensitivity. Sometimes you can get a boost over a small region this way but that's probably not what you're looking for. Sometimes you can achieve a bandpass gain over a wider region by using a high and low crossover that resonate together, but this is difficult to do without testing.

This doesn't have to be a problem. If you don't cross below your baffle step, you should be able to tilt the woofer response down to meet the mid, and use this like a bass boost. This is not a large amount to be a problem if you keep the boost at low enough frequencies by tilting the woofer the right way. Considering the way rooms interact at bass frequencies you may use positioning to make it right, or maybe some EQ.
 
Putting it in a smaller box isn't going to raise the sensitivity. Sometimes you can get a boost over a small region this way but that's probably not what you're looking for. Sometimes you can achieve a bandpass gain over a wider region by using a high and low crossover that resonate together, but this is difficult to do without testing.

This doesn't have to be a problem. If you don't cross below your baffle step, you should be able to tilt the woofer response down to meet the mid, and use this like a bass boost. This is not a large amount to be a problem if you keep the boost at low enough frequencies by tilting the woofer the right way. Considering the way rooms interact at bass frequencies you may use positioning to make it right, or maybe some EQ.
Ok AllenB, I do understand the Baffle step process and goal. I will most likely try this Baffle s. if to gain a better understanding. I do have another question though, I was thinking of a hybrid combo along with the Baffle step. With O.B. design there is a dispersion issue that gives the notion of db. loss. Combined with a full range as mid Fr. duty only ... would this idea "in theory" closely balanced the output of the more efficient woofer to the lesser full range 15"woofer being 96db and full range listed at 89db?
Kingfisher
 
Moderator
Joined 2008
Paid Member
Since an open baffle's response tilts, you can't say what level you'll get out of it until you decide where you'll cross it, and whether you accept the low end cutoff which your crossover has a hand in fixing.

I consider proper dipole design to be an advanced topic. It's fine to experiment with, even without measurement... it will give you a feel for the rear radiation which some have a preference for, but there's a learning curve. I suppose in general I can say you should expect quite some reduction in level, a fair amount of balancing to match the differing directivity, and a limit to the output level you can achieve if you don't use a large enough woofer.
 
  • Like
Reactions: 1 user
Not sure am i aksing in a right place or ot but Is there any case when crossover not needed inside speakers? And speakers is connected directly to the amplifier? Have technics sc-dv170 and 3 way speaker is directly connected to the low, mid and high seperated output. Is there a chance crossover is builted in inside amp or is it ex holder of system removed crossover from speakers?
 
Moderator
Joined 2008
Paid Member
A woofer doesn't produce the highest frequencies so sometimes no components are used on the woofer. It works, but it is a bad idea because the woofer goes through cone breakup around its upper limit. Sometimes this also means that resonances create too much midrange.

Piezo tweeters are made differently to ordinary dynamic tweeters (I don't know what is used on the sc-dv170). They don't need a crossover because they naturally block the lower frequencies. This doesn't mean we don't use a crossover anyway when trying to improve their performance.
 
Member
Joined 2023
Paid Member
Hello gents,....I'm reasonable new to diyaudio and refurbishing vintage equipment. Main focus has been refurbishing Harman Kardon amplifiers from teh '80 which I kinda like to do:)...I jus stumbled upn this quite interesting thread and I've pretty much read and understood the first few posts. I have a set of Kef Reference Model Ones from the '90 which I'm driving with a H/K PM645VXi which I fully refurbished last year. Although i hink it all sounds great there is always room for improvement I'd say and the corssovers are obviously also almost 30 years old.

My question is,...is there any benefit of recapping, refurbishing the cross-overs and/or change the crossover frequencies for that matter? I have inserted the speaker specs but have not found tweeter, midrange or woofer specific specs....yet.

Thx in advan for any tips

reg

willem
kef-specs.PNG
 
Moderator
Joined 2008
Paid Member
A simple crossover can often be improved by designing one that is more specific, especially if you are able to measure. However there are times when you can make simple improvements as well. Apart from choosing values that you find improve the tone, sometimes a manufacturer skips using a resistor on a tweeter due to cost.. not that a resistor is always needed.

Another example is when they don't use an inductor on the woofer and the upper midrange is too bright. If you make changes like this then you'll probably want to look at the tweeter cross next to help balance the midrange. Rarely you might find the tweeter doesn't want to blend, or doesn't want to be crossed lower, and you have to go deeper into the design.

In a lot of cases you don't need to change old capacitors. Electrolytics might vary with use over time and it would help if you could measure them to see their value is near original. When unsure, you can change them.
 
  • Like
Reactions: 1 users
Moderator
Joined 2008
Paid Member
That filter is the low-pass version of a second order high-pass, equivalent to the capacitor/inductor tweeter filter from this thread. Since capacitors alone don't resonate the way a capacitor does with an inductor, this circuit uses the gain of the op-amp to push up the response. This is why it is also known as an active filter.
 
  • Like
Reactions: 1 user
Moderator
Joined 2008
Paid Member
I would look at the upper and lower ends of the band as separate issues. Each driver/baffle has it's own limitations which are different at each end.

After that it is usually better to stay wide, as this reduces the number of crossovers. Reasons for introducing more crossovers tends to get into an advanced topic.

Another concern is our hearing sensitivity around 3-4kHz. You can cross there but if it is not right it may be easier to hear.
 
  • Like
Reactions: 1 users