Hi to all after a long absence I'm back. In the meantime I have received my ECM 8000 mike but before make any gated tests as suggested in post #4 by Znu, I would like to follow the advice of DougL as per post #9 and review all the distances between the drivers:
(1) Move the mid nearest to the tweeter,
(2) Can any one suggest the ideal distance between the woofer and the mid and finally the ideal distance of the vent from the woofer or from floor.tks
(1) Move the mid nearest to the tweeter,
(2) Can any one suggest the ideal distance between the woofer and the mid and finally the ideal distance of the vent from the woofer or from floor.tks
Ideal distance between drivers is within 1/4 wavelength of crossover frequency for least lobing. Within 1/2 wl is ok it seems from forum postings. If longer distance, sound changes depending on your listening position (due to lobing). Might be a problem or not.
About the port: all you want is the low output from the port to extend your bass response. Anything above that is not needed or wanted. Most of the time there seems to be strong output from the port in the low midrange depending on your enclosure and port dimensions. If you got high crossover frequency from woofer to mid driver the port mid range "leakage" might mess up the sound. How severely, you have to try and measure and listen
Anyway its interesting subject and I've been doing some investigation lately. Based on hornresp model and experiments it seems that internal standing waves of the enclosure and rear output of the woofer that could come through the port can be minimized by woofer and port placement and by stuffing the box + some magic trickery.
Here are few rule of thumbs I've found, not a definitive list but will help a little at least
- Put the woofer mid height of the enclosure, it will suppress the lowest standing wave within the enclosure.
- It seems that worst place for the port is at the end of the enclosure where standing wave pressure nodes are. Put the port to an antinode of the second lowest standing wave (~1/3 - 1/5 height of the enclosure) to prevent that one leaking out.
- Experiment with stuffing, don't put stuffing between woofer and the port inlet. Too much stuffing or at wrong place and you lose the low end output which is the only reason for the port.
- one could put the port to some other panel than the front.
Magic tricks I've yet to try
- folding the port to suppress mid frequency leakage
- try to dampen lowest port resonance by drilling holes mid length of the port and add layer of stuffing around (think engine exhaust muffler)
Sorry, no comment on how the port location relative to the woofer affects floor bounce.
Good luck
About the port: all you want is the low output from the port to extend your bass response. Anything above that is not needed or wanted. Most of the time there seems to be strong output from the port in the low midrange depending on your enclosure and port dimensions. If you got high crossover frequency from woofer to mid driver the port mid range "leakage" might mess up the sound. How severely, you have to try and measure and listen
Anyway its interesting subject and I've been doing some investigation lately. Based on hornresp model and experiments it seems that internal standing waves of the enclosure and rear output of the woofer that could come through the port can be minimized by woofer and port placement and by stuffing the box + some magic trickery.
Here are few rule of thumbs I've found, not a definitive list but will help a little at least
- Put the woofer mid height of the enclosure, it will suppress the lowest standing wave within the enclosure.
- It seems that worst place for the port is at the end of the enclosure where standing wave pressure nodes are. Put the port to an antinode of the second lowest standing wave (~1/3 - 1/5 height of the enclosure) to prevent that one leaking out.
- Experiment with stuffing, don't put stuffing between woofer and the port inlet. Too much stuffing or at wrong place and you lose the low end output which is the only reason for the port.
- one could put the port to some other panel than the front.
Magic tricks I've yet to try
- folding the port to suppress mid frequency leakage
- try to dampen lowest port resonance by drilling holes mid length of the port and add layer of stuffing around (think engine exhaust muffler)
Sorry, no comment on how the port location relative to the woofer affects floor bounce.
Good luck
Last edited:
Thanks for your precious advice. Encl is the photo of the enclosure.
.
To recap: in order to achieve the 1/4 wave length, I must move up the midrange driver close to the tweeter. As regards the woofer it's height seems ok. Shal I move the port nearer to the woofer.
What diameter and how many holes must be drilled halfway inside the port? tks
The frequency cut-off curves are at approx 700 HZ 6db and 6000Hz 12db for the tweeter
. To recap: in order to achieve the 1/4 wave length, I must move up the midrange driver close to the tweeter. As regards the woofer it's height seems ok. Shal I move the port nearer to the woofer.
What diameter and how many holes must be drilled halfway inside the port? tks
The frequency cut-off curves are at approx 700 HZ 6db and 6000Hz 12db for the tweeter
Last edited:
Don't move the port yet, measure it first and adjust with enclosure stuffing and see if there is a problem.
No clue about the muffler/port holes yet, see this thread and experiment. I'm gonna experiemnt at some point in near future. https://www.diyaudio.com/forums/multi-way/94342-port-midrange-leakage.html#post1109354
You can push the port resonance higher by using smaller shorter port. try to find out smallest possible port that doesn't do the chuffing by simulating port air velocity in winisd for example. Many say keep velocity under 20m/s. I don't know where the chuffing comes audible, so experiment this as well
No clue about the muffler/port holes yet, see this thread and experiment. I'm gonna experiemnt at some point in near future. https://www.diyaudio.com/forums/multi-way/94342-port-midrange-leakage.html#post1109354
You can push the port resonance higher by using smaller shorter port. try to find out smallest possible port that doesn't do the chuffing by simulating port air velocity in winisd for example. Many say keep velocity under 20m/s. I don't know where the chuffing comes audible, so experiment this as well
Your woofer sensitivity is quite low for MS-130.
Your baffle width is probably 28-30cm. That means the baffle step frequency is about 380-410Hz.
In other words the driver reproducing frequences below this point must be 5-6dB more sensitive than the others.
In your case:
Option 1 Change woofer-midrange crossover point at 400Hz and attenuate MS-130 with 9dB, or you can use another amplifier for woofer.
Option 2 Add a second woofer with first order low pass filter at 400Hz in parallel (3.5 way) and attenuate the MS130 with 3dB.
I would recommend to use at least second order high pass filter for midrange with only 0.7mm xmax.
Midrange-tweeter crossover point shouldn't be higher than 3.2kHz for 5 inch driver.
Your baffle width is probably 28-30cm. That means the baffle step frequency is about 380-410Hz.
In other words the driver reproducing frequences below this point must be 5-6dB more sensitive than the others.
In your case:
Option 1 Change woofer-midrange crossover point at 400Hz and attenuate MS-130 with 9dB, or you can use another amplifier for woofer.
Option 2 Add a second woofer with first order low pass filter at 400Hz in parallel (3.5 way) and attenuate the MS130 with 3dB.
I would recommend to use at least second order high pass filter for midrange with only 0.7mm xmax.
Midrange-tweeter crossover point shouldn't be higher than 3.2kHz for 5 inch driver.
After long absence I'm back again.In the meantime I have been doing intensive reading regarding taking measurements and practising on some software namely arta and rew. As for the practical work, after much trial and error and basic experience I've been able to measure my drivers, namely Woofer monacor SPH255, mid monacor MS130 and a tweeter make Jibolu using REW from which I have exported FRD and used LIMP to export ZMA. I now realise how approximate is my crossover and my next objective is to remake the crossover all over again step by step. My immediate problem is that I am still struggling with ARTA
to merge the woofer FRD neafield, port and farfield to get an quasi-anechoic measurement .I am surely missing something somewhere. I have enclosed all the data files and am asking help from someone to merge them.thank you.AD
to merge the woofer FRD neafield, port and farfield to get an quasi-anechoic measurement .I am surely missing something somewhere. I have enclosed all the data files and am asking help from someone to merge them.thank you.AD
Read "White Paper - Accurate In-Room Frequency Response to 10Hz" by J.Bagby
edit1: your initial crossover filter and crossover frequencies are inadequate. Make sure you prepare the measurements to be used in a simulator as precise as possible, before you start designing the filter. I'd cross at about 400 Hz and 3kHz.
edit2:woofer impedance at 150 Hz would be somewhat higher than Re, so you might want to double check if everything is ok with your impedance measurement setup.
edit1: your initial crossover filter and crossover frequencies are inadequate. Make sure you prepare the measurements to be used in a simulator as precise as possible, before you start designing the filter. I'd cross at about 400 Hz and 3kHz.
edit2:woofer impedance at 150 Hz would be somewhat higher than Re, so you might want to double check if everything is ok with your impedance measurement setup.
Last edited:
Thanks Lojzek, I am currently going through the manual and familiarise myself with the excel sheet. To get it clear, in step1 line 9 please confirm to what FRD" the high frequency FRD" refers to and how to proceed blending for the port for the port FRD does not contain phase.
------------------------------------------------------------------------------
STEP 1
High Frequency Response FRD data Frequency Response Measurement
Low Frequency Response FRD data Box Model or Nearfield Measurement
Corrections to be added to LF data Diffraction / EQ / Crossover model
------------------------------------------------------------------------------
STEP 1
High Frequency Response FRD data Frequency Response Measurement
Low Frequency Response FRD data Box Model or Nearfield Measurement
Corrections to be added to LF data Diffraction / EQ / Crossover model
A Visaton publishing says otherwise. They measured two port positions, one directly at the bottom, the other close to the driver. Both measured identical up to 40Hz. They measured between 50 and 150Hz ~2dB more than at the upper position (close to the driver) and explained that's because of the better phase because of the longer distance the sound has to travel. At taller speakers the gain will appear in lower frequencies.
They also explain that the standing wave with the bottom port open can't even form because the lower end isn't completely closed.
There are no magic tricks, you have to pay the one or other way, there is no 'free lunch' in audio physics. Your idea works partly, the mid leak will be reduced. Folding the port should be avoided because that causes turbulences and losses which can lead to compression and port air noises ('chuffing'). I would never do that intentionally. If you absolutely must, avoid sharp turns if possible and round the edge generously (~5cm). The worst case are 180° turns. If mid leak is a problem at your 2-way speaker, place the port on the back. At 3-way speakers the mid-garbage isn't a problem in the most cases though.
That doesn't work the way you imagine it. Someone in a German forum tried that and measured it. With drilling the holes you indeed take out the first port resonance but you also lose port efficiency, IIRC, you lose ~2-2,5dB bass port level. The stuffing around does not work like that, that's like drilling the holes and then close them again, effectively doing exactly the opposite.
Also, a muffler does not work that way. Either they build a resonance/reflection system which partly eliminates the sound in a certain range via interference/phase shift or they do an absorption (which you were thinking of), for which they use a punched hole tube within the muffler with damping material around it. The damping effect is achieved by converting the movement energy by absorption into heat. The damping is increased by multiple reflections, the sound waves have to travel through the damping material several times. That works mainly on higher frequencies. The more often they have through the damping material, the deeper the absorbed frequencies. In your case, it travels just one time through it and disperses into the enclosure after that with very low damping unless you pack it that tight almost no airflow is possible - but that cancels the effect of the drilled holes altogether.
----------------------------------------------------------------------------
Tks gabdx. Please confirm that you need the FRD farfield 1 m but measured inside the listening room .
Please also clarify "from calibrated"? AD
I have some free time to simulate you a good XO.
To explain myself, measurements need to be done with an accurate microphone or it wont work. I can always try if it is not a calibrated microphone but it might not be first time home-run.
Yes inside the listening room I will figure the room modes, figure the xo's best points etc.
To explain myself, measurements need to be done with an accurate microphone or it wont work. I can always try if it is not a calibrated microphone but it might not be first time home-run.
Yes inside the listening room I will figure the room modes, figure the xo's best points etc.
Here are the frds you asked for. They are measured at 1 meter and ear level. They have not been smoothed.The mic used is a behringer ecm8000 and is not calibrated. I prefer not to use generic cal files found over the net. The actual distances from the drivers' centre to the mic position is : Tweeter 100cm , Mid 105 cm and woofer 114 cm. Newly measured ZMA will be attached in next post.
Did you tuned your port to 42 ?
you have a dip at 50-60 which is quite huge, maybe if you tune your port higher, reduce the length of it. A dip at 40 hz is better than one at 50 to 60.
If you are ok with the woofer playing up to 650, and the mid up to 4000, this is where it is going to be approximatively.
you have a dip at 50-60 which is quite huge, maybe if you tune your port higher, reduce the length of it. A dip at 40 hz is better than one at 50 to 60.
If you are ok with the woofer playing up to 650, and the mid up to 4000, this is where it is going to be approximatively.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.