When compared to directly driving a driver with a flat voltage, the relative level presented across the speaker terminals after being crossed will represent the relative acoustic level, passive or active.
The voltage presented after a passive crossover has the current taken in to account. How would the impedance change by any significant amount?
you're not going to do any measurements?? I think you need to get what is required (cheap really) but then you gotta learn it I guess, not so cheap timewise.
Now I understand you're not going to measure, the question makes more sense! However, in those cases of not measuring..perhaps that is why they say 'copy an established design'.
In your case (making it active whilst trying to duplicate the passive) you won't even be able to check to see if you duplicated it!
And, if you could measure, then why duplicate passive with active? Why limit what you can do to the limitations of passive? (as but one example, you can lower the tweeter by steeper slopes-all time aligned etc-which then stops the need of the woofer to go higher simply to protect the tweeter yet introduces off axis anomalies) You don't HAVE to lower the xover point, twas just an example.
No measuring gear? Then you can't even take advantage of eq in the bass region (after the speakers have been built, done in room).
I'd urge you to consider being able to take some sort of measurements, even (at worst) some sort of RTA mobile phone app. You can at least find out how much to pad the tweeter down in minidsp that way.
You need some sort of basic measurements. Reckon the guys who designed the passive did it by ear?
Now I understand you're not going to measure, the question makes more sense! However, in those cases of not measuring..perhaps that is why they say 'copy an established design'.
In your case (making it active whilst trying to duplicate the passive) you won't even be able to check to see if you duplicated it!
And, if you could measure, then why duplicate passive with active? Why limit what you can do to the limitations of passive? (as but one example, you can lower the tweeter by steeper slopes-all time aligned etc-which then stops the need of the woofer to go higher simply to protect the tweeter yet introduces off axis anomalies) You don't HAVE to lower the xover point, twas just an example.
No measuring gear? Then you can't even take advantage of eq in the bass region (after the speakers have been built, done in room).
I'd urge you to consider being able to take some sort of measurements, even (at worst) some sort of RTA mobile phone app. You can at least find out how much to pad the tweeter down in minidsp that way.
You need some sort of basic measurements. Reckon the guys who designed the passive did it by ear?
No, of course not. I believe that's why many of you in the DIY speaker community advice beginners to start with a proven design. Because there are experienced people out there who knows what and how to measure and simulate, and based on this can design a crossover that integrates instead of ruins an otherwise good driver combo. I get what you're all saying. And now I realize that active XO apparently isn't the best option for newbies since it doesn't allow for building most of the proven designs out there.
It would be nice though if designers started to include an "active filter suggestion" in their XO designs in the shape of e.g. a biquad filter. Class-D amps with integrated DSP are becoming more and more popular, and I think we'll only see more active XOs in the future of DIY speaker building.
BTW I do have a measurement microphone, but so far I've only used it for (sub)woofer room-EQ. I guess it's time to get learning
Last edited:
When you do a DIY, often you'll end up with wanting the best sound. It is just too hard and expensive to get the best from class-D and active XO.
Imagine a $100 MiniDSP. Do you think I want to use the DAC and opamp inside of a $100 machine?? Some others build their own software but mostly this requires computer as part of the music system, which is not yet my preference.
I have been more than a year stopping building passive system. But I'm still in the analog domain, not yet going digital. I have been building popular analog active crossovers from the internet while designing my own. So far, I have decided that I will reject any opamp based systems. Now I'm going with discrete. There is a chance for good sound but not as flexible as digital system I guess. An analog active system specialized for certain speaker drivers (combined with passive element)
For sure, all the measuring and confirmation has already been done.
I am not sure I am quite saying that. All I was saying is that as you are trying to DUPLICATE an existing passive with an active where there is nothing to translate it for you (the whole point of your thread) then it does devolve to (hopefully) intelligent guesswork, at least initially.
To CONFIRM that you have duplicated the passive actively, then you need a means to do that. Measurement.
I was most certainly NOT saying active is not an option for newbies. If you are going active, then 'forget the passive' and just start active. You can bypass all these transference problems, forget also having to learn the arcane art of passive design (I could not design a passive network if my daughters life depended on it) and just cut to the chase.
Yes, you will have to learn some programs (and why not rew), and just buy the same drivers (if that is what your heart is set upon) and start a new build thread. Get the help that way.
It would be nice though if designers started to include an "active filter suggestion" in their XO designs in the shape of e.g. a biquad filter. Class-D amps with integrated DSP are becoming more and more popular, and I think we'll only see more active XOs in the future of DIY speaker building.
BTW I do have a measurement microphone, but so far I've only used it for (sub)woofer room-EQ. I guess it's time to get learning
I'd imagine it would be a rare thing for many to do what you have proposed. Most would either build the published design or build an active. Yours would be unique I'd suspect. In other words, very little call for them to do what you have proposed. You have extrapolated your unique desire into a general thing! No, I suspect the truth is just that, yours is a unique request.
Yep, you can never lose from learning!
Geez Terry, it's not that bad
OK, the load is not a resistance like with a conventional active filter. It's an impedance, and this can be a lot to wrap ones head around but all other aspects of a crossover design are the same. Although an active filter can be like a computer in a way...easy to make big mistakes at a fast rate
ha!
maybe you can teach me when you turn up at the next gtg...that will be me in the corner, nodding off with a confused look on his face.
Anyway, my point, if there was one, is that if you are getting into building your own speakers, either you learn how to do it passively (if you will end up passive) or learn to do it actively (if the final product will be active).
I can go thru the motions actively, passively-forget it. None of this is saying one approach is better than the other, simply that which ever way you go you gotta learn it.
See you in oct allen??
maybe you can teach me when you turn up at the next gtg...that will be me in the corner, nodding off with a confused look on his face.
Anyway, my point, if there was one, is that if you are getting into building your own speakers, either you learn how to do it passively (if you will end up passive) or learn to do it actively (if the final product will be active).
I can go thru the motions actively, passively-forget it. None of this is saying one approach is better than the other, simply that which ever way you go you gotta learn it.
See you in oct allen??
Yes, I think this is what I'll do. Start with a proven passive design and then see if I can make it as good, or even better, with an active XO.AllenB said:
My initial question in this thread has been answered. Active and passive XO designs work differently and each have their own advantages and drawbacks, I see that now. So duplicating a passive design with an active doesn't make much sense I guess.terry j said:
But shouldn't we begin to see more active proven designs then? I mean, there will always be DIY'ers out there with superior measurement equipment and XO designer skills, who will make great filter designs, passive or active. And if more people in the DIY speaker community are going active, then active proven designs should start to show up, no? (Maybe they already are and I just haven't noticedterry j said:
)I will for sure experiment with some bookshelf TM and definitely try the active route. I'll probably start a new thread when that time comes. Perhaps this can be the beginning of a "proven" active XO design
hey, dunno. good question, never thought about it.
heck, just had a minutes thought, no good answers popped up! History? 'the way things have always gone'?
OR, just take a design that has already worked out a driver selection that would work together, get your measurement gear out and put in the filters that give you what you want? Flat on axis, whatever.
anyway, I think others might have some ideas on that. good question.
From the point of view of this thread, yes... but from mine there is so much else to designing a crossover which is common ground. The question of passive vs active is just one of implementation, like deciding which brand of nails to use when you're designing a house (so I guess MiniDSP is like a gas fired nail gun
)I hope so. Could you give me a bell then? (I'm subscribed to your thread)
Not sure which thread you mean allen, but yeah I should remember when we start adding up 'who is coming this year?'. A dim spark should ignite from somewhere.
we usually end up 'testing something' each year, next was supposed to be 'can we really hear cap differences in a power amp', were going to do a blind test with some of the more vocal 'I can hear differences between wire types in an amp' but they seem to have spat the dummy and it's looking tenuous. I'm not that upset to be frank, the idea of having ten golden ears walking around kinda gave me the heebie jeebies.
SO, long story short, if you have any ideas of what might make an interesting thing to investigate this year throw em into the pot. I am thinking 'why not just a good old dac comparison'...but done blind via instant switching from the LP.
That might throw a spanner into the works of the 'oh my god, every dac is sooo different' and the true relative differences start to emerge.
we usually end up 'testing something' each year, next was supposed to be 'can we really hear cap differences in a power amp', were going to do a blind test with some of the more vocal 'I can hear differences between wire types in an amp' but they seem to have spat the dummy and it's looking tenuous. I'm not that upset to be frank, the idea of having ten golden ears walking around kinda gave me the heebie jeebies.
SO, long story short, if you have any ideas of what might make an interesting thing to investigate this year throw em into the pot. I am thinking 'why not just a good old dac comparison'...but done blind via instant switching from the LP.
That might throw a spanner into the works of the 'oh my god, every dac is sooo different' and the true relative differences start to emerge.
A very real problem with active filters is that using different types of DSP results in differing output response even when all the crossover types, slopes, frequencies, and parametric filters are set the same.But shouldn't we begin to see more active proven designs then? I mean, there will always be DIY'ers out there with superior measurement equipment and XO designer skills, who will make great filter designs, passive or active. And if more people in the DIY speaker community are going active, then active proven designs should start to show up, no? (Maybe they already are and I just haven't noticed
As Bennett Prescott puts it, "the DSP tower of Babel".
Simply put, if I design a crossover using "X" DSP, and a DIY uses those settings on "Y" DSP, they will not measure or sound the same.
The differences may be subtle or gross, the more filters employed the farther off the end results will be.
Even different models under the same parent company often don't match up.
In addition to "Q" variations, there are plain stupid design errors- for instance the DBX DriverackPa inverts polarity between input and output, no big deal when used alone, but the DriveRackPA+ inverts only the midrange outputs.
If someone correctly designed a crossover on that unit, the midrange would be reversed polarity using the same settings on any other "normal" DSP.
Art
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.