Hi all,
Please answer me some questions.
I read somewere that can be some gain in db output in the midrange driver when I design a 3 way passive crossover.
1.-The quiantity in db gain is dependant of the frequency range of the passband?
2.-How much gain can I expect in the midrange if I use a 3 octave passband?
Thanks
Please answer me some questions.
I read somewere that can be some gain in db output in the midrange driver when I design a 3 way passive crossover.
1.-The quiantity in db gain is dependant of the frequency range of the passband?
2.-How much gain can I expect in the midrange if I use a 3 octave passband?
Thanks
1) It is dependent on the frequency range as well as the type of crossover.
2) that can not be specifically answered. In addition to the answer above, it also depends on how well the woofer and tweeter roll off match the ideal acoustic targets for them in the stop band.
In theory band pass gain is the result of the output of the woofer and tweeter adding to the midrange driver. For example, using 1st order crossovers at 300 and 1800, the woofer and tweeter sum to about -11dB at the center of the midrange. This happens to be in phase with the midrange and thus sums to the output yielding a gain of about 2dB. In reality, it is probably significantly less because most likely the woofer and tweeter won't actually roll off following a 1st order response in both amplitude and phase, all the way into the midrange. And in fact, you probably would not want them to. Using 2nd or 3rd order x-os the resulting ideal BP gain would be less. Ultimately you just have to design the 3-way for the response you want.
2) that can not be specifically answered. In addition to the answer above, it also depends on how well the woofer and tweeter roll off match the ideal acoustic targets for them in the stop band.
In theory band pass gain is the result of the output of the woofer and tweeter adding to the midrange driver. For example, using 1st order crossovers at 300 and 1800, the woofer and tweeter sum to about -11dB at the center of the midrange. This happens to be in phase with the midrange and thus sums to the output yielding a gain of about 2dB. In reality, it is probably significantly less because most likely the woofer and tweeter won't actually roll off following a 1st order response in both amplitude and phase, all the way into the midrange. And in fact, you probably would not want them to. Using 2nd or 3rd order x-os the resulting ideal BP gain would be less. Ultimately you just have to design the 3-way for the response you want.
Thanks John K, I star to see some light now.
If I undederstand well. If I use a 4 order LR crossover may be I have less gain in the passband due the roll off of th LR slopes?
Best ragards
If I undederstand well. If I use a 4 order LR crossover may be I have less gain in the passband due the roll off of th LR slopes?
Best ragards
I think that Robert Bullock's findings and papers published by the AES and also in Speaker Builder magazine are what you need to read. Article: "Passive Crossover Networks: Parts 1–3" by Robert M. Bullock III
Magazine: Issues One/85–Three/85 of Speaker Builder
He showed how the midrange gain was affected by not only the octave range covered but also the type of bandpass crossover used on the midrange. I couldn't find any details posted online but I didn't do an extensive search.
Magazine: Issues One/85–Three/85 of Speaker Builder
He showed how the midrange gain was affected by not only the octave range covered but also the type of bandpass crossover used on the midrange. I couldn't find any details posted online but I didn't do an extensive search.
Yes, that is correct. Actually, if you can run DOS programs on your PC go to this website,Bullock and White's Home Page and download the DOS Boxmodels (link at top center of page) and the DOS PXO exe will simulate the ideal crossover filters with component values. The component values are really useful but you can play around with the code to see how x-o points and slopes affect the bandpass gain. This software is based on the work DaveL mentioned.
Thanks John K and David L, I apreciate your kind help.
I will try to search for those papers of Robert Bullock.
I will try to search for those papers of Robert Bullock.
Very late correction 😱 the issue of Speaker Builder with the bandpass circuits is issue 4 of 1987. Very sorry for the error.
You can simulate bandpass gain with speaker workshop - you need to make an "ideal driver" with flat frequency and impedance and apply the standard crossover values to them. This is all academic anyway because real crossovers with real drivers are often not much like cookbook ones and you often need to do a lot of response shaping in the crossover to account for these things.
Yes also Calsod can do a lot more, allowing you to import the FRD and IMP files and modeling any type of crossover you want. The point is Mr Bullock came up with other types of bandpass crossovers that allow us to consider not only the frequency response but also the impedance variations typical of a C-Bandpass The T-Bandpass crossover layout is better when your frequency spread causes excessive dips in the impedance using the C-Bandpass.
- Status
- Not open for further replies.