I've got a pair of Advent Maestro speakers that were given to me today. They are in good condition and the woofers were refoamed a few years ago.
I already ordered new caps for the crossover.
Two things I've noticed.
1. The speaker wire terminals on both speakers has separated from the terminal cup with the crossover on it at the bottom. What should I use to secure it back down?
2. I've read where the woofer doesn't play high enough causing a hole in the frequency response near the crossover point of the woofer and midrange which is 900Hz. Listening to them, I believe I actually hear that, but will wait until the film caps come in and are installed to see if that fixes it. If I still hear the issue, what is the best way to eliminate the issue? Normally I'd just use a different inductor value for the woofer, but the inductor is part of the dish where the fuse, speaker terminals and other crossover components are and I'd have to add an inductor which would require mounting it somewhere in the cabinet.
What I should do is disconnect a midrange and test it using my DATS V3 speaker tester so I can see what the impedance of it is as in order for the 14uF cap value to be correct the midrange impedance would have to be 12.5 ohms provided the specs I found are correct. That's not exactly possible though given the speakers are at work now.
Specs of the speaker.
https://audio-database.com/ADVENT/speaker/a-1012.html
Also I will run these with a powered sub.
The response is +/- 3dB 42-23kHz
I currently have the powered sub crossover set to 40Hz, but am not that happy with it.
I can put two 160uF electrolytic caps in parallel to make a 320uF cap which based on the rated 6 ohms impedance of the speaker will give me a high pass of 82Hz.
I might try that tomorrow with the caps between the + speaker lead and + speaker terminal. If it sounds better then I may install the caps between the + speaker lead and woofer inductor (when i install the film caps) so that I don't run the risk of the caps affecting the midrange and treble.
EDIT:
Here's the crossover schematic I found.
However the response graph I found seems to suggest there is no hole in the frequency response.
I already ordered new caps for the crossover.
Two things I've noticed.
1. The speaker wire terminals on both speakers has separated from the terminal cup with the crossover on it at the bottom. What should I use to secure it back down?
2. I've read where the woofer doesn't play high enough causing a hole in the frequency response near the crossover point of the woofer and midrange which is 900Hz. Listening to them, I believe I actually hear that, but will wait until the film caps come in and are installed to see if that fixes it. If I still hear the issue, what is the best way to eliminate the issue? Normally I'd just use a different inductor value for the woofer, but the inductor is part of the dish where the fuse, speaker terminals and other crossover components are and I'd have to add an inductor which would require mounting it somewhere in the cabinet.
What I should do is disconnect a midrange and test it using my DATS V3 speaker tester so I can see what the impedance of it is as in order for the 14uF cap value to be correct the midrange impedance would have to be 12.5 ohms provided the specs I found are correct. That's not exactly possible though given the speakers are at work now.
Specs of the speaker.
https://audio-database.com/ADVENT/speaker/a-1012.html
Also I will run these with a powered sub.
The response is +/- 3dB 42-23kHz
I currently have the powered sub crossover set to 40Hz, but am not that happy with it.
I can put two 160uF electrolytic caps in parallel to make a 320uF cap which based on the rated 6 ohms impedance of the speaker will give me a high pass of 82Hz.
I might try that tomorrow with the caps between the + speaker lead and + speaker terminal. If it sounds better then I may install the caps between the + speaker lead and woofer inductor (when i install the film caps) so that I don't run the risk of the caps affecting the midrange and treble.
EDIT:
Here's the crossover schematic I found.
However the response graph I found seems to suggest there is no hole in the frequency response.
Last edited:
I'm not sure what you're describing. A picture might help.
Without measuring the result, you're flying a bit blind. Tuning can be done that way, but it tends to be more difficult. Even a simple real time analyzer app can make a big difference.
Does your subwoofer have a line level high-pass crossover? Those tend to be more accurate and easier to adjust at low frequencies. A passive speaker level crossover there can be a little harder due to the rising impedance of the speaker in the range where the crossover is trying to operate.
Unfortunately I cannot take any photos of the issue as I cannot have cameras at work where the issue is.
The best way to describe it is the terminal block on the rear of the terminal cup looks like it was either molded into the terminal cup or it was glued in.
The part at the bottom is separated from the plastic.
I'm thinking maybe some plastic epoxy would work with something to hold the terminal block down while the epoxy sets.
I've had good luck in the past tweaking by ear, however this is a speaker worthy of doing it right.
Now I can make an active 24dB/octave 70Hz two way crossover if need be as I had built one for another stereo system or given that system no longer needs the higher impedance input that crossover has (the output of a TV headphone jack drives it now) I could swap in a pro audio crossover and use that crossover for these speakers.
The best way to describe it is the terminal block on the rear of the terminal cup looks like it was either molded into the terminal cup or it was glued in.
The part at the bottom is separated from the plastic.
I'm thinking maybe some plastic epoxy would work with something to hold the terminal block down while the epoxy sets.
I've had good luck in the past tweaking by ear, however this is a speaker worthy of doing it right.
Now I can make an active 24dB/octave 70Hz two way crossover if need be as I had built one for another stereo system or given that system no longer needs the higher impedance input that crossover has (the output of a TV headphone jack drives it now) I could swap in a pro audio crossover and use that crossover for these speakers.
Looked at the crossover schematic again and saw the midrange and tweeter were wired out of phase with the woofer.
I wired them in phase with the woofer and the issue I was hearing is gone.
Not sure why they were wired like they were? Maybe Advent was trying for a certain sound?
I also installed the active crossover and that made it sound better as well. I do not have to set a crossover point and only have to tweak the subwoofer level.
I wired them in phase with the woofer and the issue I was hearing is gone.
Not sure why they were wired like they were? Maybe Advent was trying for a certain sound?
I also installed the active crossover and that made it sound better as well. I do not have to set a crossover point and only have to tweak the subwoofer level.
Last edited:
That's a pretty common result when the speaker has a little too much midrange/treble. That can make the speaker sound more detailed, but on the "right" song or at higher playback levels it can be annoying to some listeners. For me this is often an issue in the 3-4 kHz range, where the posted frequency response plot shows a bit of excess energy.
A lot of this comes down to personal preference. It's why some brands sound more laid back, while others are "ruthlessly revealing" or downright bright.
This is typically done to deal with Z-offsets in the physical driver alignment, to compensate for phase changes induced by the crossover, or to steer the lobe in the vertical directivity of the speaker. Without measurements, it's hard to say what your change did.
The change made the speaker sound better and the hole in the frequency response around the woofer and midrange crossover frequency is no longer there.
Some of the issue I may be hearing could be the stock crossover caps as the non-polar caps sometimes increase in capacitance as they age which will lower the crossover point and that could cause the tweeter to reproduce more of the midrange's range. I've got new film caps on order.
I could add something like this in the signal chain to give me some control over the sound besides just bass and treble controls.
https://www.schiit.com/products/lokius
I use a McIntosh C-24 preamp to drive an Adcom GFA-545II amp to power the speakers.
Some of the issue I may be hearing could be the stock crossover caps as the non-polar caps sometimes increase in capacitance as they age which will lower the crossover point and that could cause the tweeter to reproduce more of the midrange's range. I've got new film caps on order.
I could add something like this in the signal chain to give me some control over the sound besides just bass and treble controls.
https://www.schiit.com/products/lokius
I use a McIntosh C-24 preamp to drive an Adcom GFA-545II amp to power the speakers.
In properly functioning average equipment, there's normally far more variability in speaker frequency response than in any other part of the system. The room they are playing in also makes a significant difference. Therefore, I typically suspect any significant frequency balance issues are in the speakers/room.
People obviously still have preferences for all kinds of different components though, and distortion characteristics and other specifics vary from one circuit to another. If you're dealing with tweaky components, things can have more interactions also (low powered tube amps, very high output impedances, very high or very low sensitivity speakers, low speaker impedance, etc.). Those are outliers to me, but some spend a lot of time pursuing synergies in everything down to power cords and cable elevators. To each their own.
People obviously still have preferences for all kinds of different components though, and distortion characteristics and other specifics vary from one circuit to another. If you're dealing with tweaky components, things can have more interactions also (low powered tube amps, very high output impedances, very high or very low sensitivity speakers, low speaker impedance, etc.). Those are outliers to me, but some spend a lot of time pursuing synergies in everything down to power cords and cable elevators. To each their own.
What I cannot figure out is why on some songs the powered sub I'm using sounds perfectly balanced, on some songs it sounds too loud and on some songs it sounds too quiet.
I know it's set up right.
That said I read somewhere that the cabinet wasn't necessarily 100% completely and totally sealed (maybe the foam on the baffle sealed the drivers but due to it deteriorating is not doing that function) so perhaps that's the issue.
I might use some RTV to seal each driver to the cabinet. Probably will do the tweeter in one just to see if there's any difference.
Now if the speakers were at my house I'd feed a 20Hz signal to them and see if I hear or feel any air leaks.
Given the woofers were refoamed, I'm going to assume the woofers are sealed properly to the cabinet. At least when I tried to remove one it didn't seem to want to come out easily.
I know it's set up right.
That said I read somewhere that the cabinet wasn't necessarily 100% completely and totally sealed (maybe the foam on the baffle sealed the drivers but due to it deteriorating is not doing that function) so perhaps that's the issue.
I might use some RTV to seal each driver to the cabinet. Probably will do the tweeter in one just to see if there's any difference.
Now if the speakers were at my house I'd feed a 20Hz signal to them and see if I hear or feel any air leaks.
Given the woofers were refoamed, I'm going to assume the woofers are sealed properly to the cabinet. At least when I tried to remove one it didn't seem to want to come out easily.
In my opinion it does sound like the tweeter is playing lower than it is spec'd for which could be caused by the 4uF cap in series with it.
The good thing is given it's just a simple 6 dB/octave crossover, it's easier to play around with.
What I should have done is take one speaker home and used my DATS V3 speaker tester to get the impedance of each driver. I then could have verified if the stock crossover parts were correct for the listed crossover frequencies.
The good thing is given it's just a simple 6 dB/octave crossover, it's easier to play around with.
What I should have done is take one speaker home and used my DATS V3 speaker tester to get the impedance of each driver. I then could have verified if the stock crossover parts were correct for the listed crossover frequencies.
What does this mean though? Have you measured the frequency response at the listening position? Are you using a sub with automatic room correction? If you haven't done either of those, it's very likely that you have the typical lumpy in-room bass response that most systems have. And you likely have it differently for the subwoofers and the main speakers. That can easily make some songs sound fine and some sound exaggerated. Similarly, if the sub and main speaker integration are not very good, it's easy to have dips in response as well. On many songs the subwoofer isn't really required for the music to sound right so how do you judge those?
I know people do things solely by ear all the time, but as I said before if you're working without acoustic measurements it makes everything much harder. Your ear can latch onto different frequency bands as their "reference" for different songs. It makes for shifting frequency balance perceptions from song to song. It can also make for a lot of tail chasing.
At a minimum, I would suggest having a reference speaker or reference headphones with known neutral character to compare to. Bouncing back and forth instantly between those and the speaker you are working on can often highlight issues. Listening to pink noise the same way can be even more revealing at times.
Small leaks tend to have negligible affects on overall response. Consider all the woofers out there with porous dust caps and vented poles. Every one of them is likely far leakier than what you have going on (assuming nothing major is wrong).
One of the annoying things about leaks is they sometimes sound like rattles. That's typically far more egregious than any output changes.
I often prototype with boxes I use over and over again. Those cabinets have multiple size driver holes in them, and the ones I'm not using are "sealed" with the original cutouts held in with hastily applied masking tape. That imperfect sealing makes no difference in the grand scheme of things.
I would not. It makes them nearly impossible to remove in the future without doing damage.
Far as the sub being set up right, it's in the same exact location as it was with the other speakers I was using.
The other speakers were smaller, but used a 6dB/octave crossover like these speakers do.
The only other thing I can think of is that I use an external active two way crossover and according to the manual for the plate amp I used in the sub in order for it to work with a LFE input (without the internal crossover affecting anything) it says to set the built in crossover to its highest frequency. I wonder if that is causing an issue given the crossover is not being bypassed?
The other speakers were smaller, but used a 6dB/octave crossover like these speakers do.
The only other thing I can think of is that I use an external active two way crossover and according to the manual for the plate amp I used in the sub in order for it to work with a LFE input (without the internal crossover affecting anything) it says to set the built in crossover to its highest frequency. I wonder if that is causing an issue given the crossover is not being bypassed?
The cap for the tweeter is close to 4uF.
The cap for the midrange is 15.62uF.
Does anyone know the impedances of the individual drivers?
I'd like to use the component values and driver impedances to determine the crossover frequencies.
EDIT:
I did a crude impedance test which is only valid at one frequency for each driver though. I used the amp connected to a signal generator fed 100Hz to the woofer and measured the voltage. I then inserted a 100 ohm pot in series and adjusted it until the voltage across the driver wa exactly 1/2 what was originally measured then I measured the pot resistance.
I did the same for the midrange using 1kHz and the tweeter using 6kHz.
What I got were
Tweeter 5.3 ohms
Midrange 7.6 ohms
Woofer 5.8 ohms
The cap for the midrange is 15.62uF.
Does anyone know the impedances of the individual drivers?
I'd like to use the component values and driver impedances to determine the crossover frequencies.
EDIT:
I did a crude impedance test which is only valid at one frequency for each driver though. I used the amp connected to a signal generator fed 100Hz to the woofer and measured the voltage. I then inserted a 100 ohm pot in series and adjusted it until the voltage across the driver wa exactly 1/2 what was originally measured then I measured the pot resistance.
I did the same for the midrange using 1kHz and the tweeter using 6kHz.
What I got were
Tweeter 5.3 ohms
Midrange 7.6 ohms
Woofer 5.8 ohms
Last edited:
Now either the inductors are wrong on the schematic or my impedance measurements are way off or there was a certain design goal in mind.
Are there any three way crossover calculators that do not require a 3 octave spread between the two crossover frequencies?
EDIT:
I may pull one set of drivers tomorrow and take them home to measure them on my DATS V3 speaker tester.
That way I know their impedance for sure.
Once I do that maybe I can figure out the crossover as some of the values don't make sense when I use a crossover calculator with the crudely measured impedances..
Are there any three way crossover calculators that do not require a 3 octave spread between the two crossover frequencies?
EDIT:
I may pull one set of drivers tomorrow and take them home to measure them on my DATS V3 speaker tester.
That way I know their impedance for sure.
Once I do that maybe I can figure out the crossover as some of the values don't make sense when I use a crossover calculator with the crudely measured impedances..
Last edited:
Real-world crossover values are rarely those predicted by basic equations. Natural driver roll-offs, response anomalies, impedance variations, enclosure effects, and listening tests can all push them away from textbook values.
https://www.diyaudio.com/community/threads/why-is-crossover-calculator-spread-always-fixed.407805/
Comments here:
https://www.diyaudio.com/community/threads/why-is-crossover-calculator-spread-always-fixed.407805/
Last edited:
I found one that seems to do what I want.
https://soundcertified.com/speaker-crossover-calculator/#3-way_crossover_details_for_the_calculator
I also can add a zobel and series notch filter to the drivers as well.
https://soundcertified.com/zobel-network-series-notch-filter-calculator/
EDIT:
Using that calculator and the impedances I got with my crude measurement method, I get these component values.
Tweeter 6.67uF
Midrange .19mH 33.15uF
Woofer 1.03mH
An LC bandpass calculator https://www.the12volt.com/caraudio/narrow-band-pass-calculator.asp
gives me .234mH and 26.6uF for the midrange driver.
I'm still confused though.
Either my calculations are way off or the designer of the speakers wanted a specific sound from the speakers for a specific music.
They seem to sound real good with jazz and oldies.
EDIT:
Now if I were to adjust the frequencies for the component values using the bandpass calculator I get a low pass of 2.75kHz and a high pass of 1.2kHz.
Using this calculator for the tweeter https://www.the12volt.com/caraudio/passive-crossover-calculator.asp
I get a high pass of 7.5kHz.
If I adjust to the low pass frequency of the midrange I get a cap value of 10.92uF.
https://soundcertified.com/speaker-crossover-calculator/#3-way_crossover_details_for_the_calculator
I also can add a zobel and series notch filter to the drivers as well.
https://soundcertified.com/zobel-network-series-notch-filter-calculator/
EDIT:
Using that calculator and the impedances I got with my crude measurement method, I get these component values.
Tweeter 6.67uF
Midrange .19mH 33.15uF
Woofer 1.03mH
An LC bandpass calculator https://www.the12volt.com/caraudio/narrow-band-pass-calculator.asp
gives me .234mH and 26.6uF for the midrange driver.
I'm still confused though.
Either my calculations are way off or the designer of the speakers wanted a specific sound from the speakers for a specific music.
They seem to sound real good with jazz and oldies.
EDIT:
Now if I were to adjust the frequencies for the component values using the bandpass calculator I get a low pass of 2.75kHz and a high pass of 1.2kHz.
Using this calculator for the tweeter https://www.the12volt.com/caraudio/passive-crossover-calculator.asp
I get a high pass of 7.5kHz.
If I adjust to the low pass frequency of the midrange I get a cap value of 10.92uF.
Last edited: