But there looks like slight increase in distortion (not much though) at some other places too..
Maybe I am trying the wrong notch filter for this driver. Maybe I should target the 1st breakup peak at the start of 3kHz instead of somewhere in the middle of that breakup shelf.
Distortion measurements are a challenge to do.
If you are able to control for factors such as background hum, wind and extraneous sources of noise pollution the data will be easier for decipher. I can see that your noise floor under 100Hz is up to 60dB, perhaps something is the causing low level hum. In domestic environments watch out for sources like air conditioning, refrigeration, fans etc that operate around 50-60Hz etc. With a little investigation it may be possible get the noise floor a further 20dB, down to approximately 30-40dB.
As for the notch you want to aim for something like this:
which is trivial to do with DSP, but with passive it needs a bit of work.
In the completed speaker, you may not need to aim for perfection- the driver will be played with a HF unit, which may mask the distortion. But how low down is enough? Some data suggests 40 or 50dB for H2/H3. Beware that if measure your distortion to be below your noise floor, it may not be because the distortion is LOW.
It may be because your noise is too high.
You need more signal to noise!
I find it essential to record the absolute drive levels (V) at the terminals? eg. 2V, 2.83V, 4V etc.
What was your drive level for the WO24P?
Hificompass shows that it well composed up to about 11.2-16V for frequencies between 50Hz and 1Khz.
Last edited:
Here are some screenshots showing common problems/mistakes in distortion measurements:
Wow! No distortion?
Maybe. But the drive level too low- thus not discerning enough.
75dB is a walk in the park for this drive unit.
"Odd resonance - driver problem?"
No. Driver(s) resonating in the cabinet. Once gasket between frame and cabinet and screws and frame:
Wow! No distortion?
Maybe. But the drive level too low- thus not discerning enough.
75dB is a walk in the park for this drive unit.
"Odd resonance - driver problem?"
No. Driver(s) resonating in the cabinet. Once gasket between frame and cabinet and screws and frame:
Thanks a lot @tktran303 for these suggestions..
I will definitely try them out next time I do measurements.
I will also measure my soundcard and check its own distortion profile. I had done this in the past for a pair of input and output through loopback but forgot the absolute values. Will do that again and save the cal files. Just to get a reference.
Ambient noise is a real issue at home. Something or the other causes noise all the time. There is no air conditioning at the place I live. I even switch off fans and refrigerator while doing these measurements. But still some noise comes up in between from outside. Anyway. I will try again.
Another issue is I had done all the above distortion measurements with a MiniDSP UMIK-1. REW doesn't allow me to turn up the volume on the amp connected to this driver to more than 90dB SPL with mic at 20mm from the driver saying that it causes clipping.
I will try and use my Dayton EMM-6 to see if I can get better results.
Next time I try this distortion measurement exercise, I will consider all these and see if I can get more reliable measurements.
Currently I am waiting for my real speaker cabinets and the below thing to come home so that I can print some waveguides and small enclosures for drivers..
Regards
Vineeth
I will definitely try them out next time I do measurements.
I will also measure my soundcard and check its own distortion profile. I had done this in the past for a pair of input and output through loopback but forgot the absolute values. Will do that again and save the cal files. Just to get a reference.
Ambient noise is a real issue at home. Something or the other causes noise all the time. There is no air conditioning at the place I live. I even switch off fans and refrigerator while doing these measurements. But still some noise comes up in between from outside. Anyway. I will try again.
Another issue is I had done all the above distortion measurements with a MiniDSP UMIK-1. REW doesn't allow me to turn up the volume on the amp connected to this driver to more than 90dB SPL with mic at 20mm from the driver saying that it causes clipping.
I will try and use my Dayton EMM-6 to see if I can get better results.
Next time I try this distortion measurement exercise, I will consider all these and see if I can get more reliable measurements.
Currently I am waiting for my real speaker cabinets and the below thing to come home so that I can print some waveguides and small enclosures for drivers..
Regards
Vineeth
This is a 2 way.
Is that a dip around 2-3KHz?
Is that a problem, or even a "design feature"? (eg. 'BBC dip')
Unhelpful measurement, borderline useless, because it's measured at LP.
Measuring at the LP and drawing conclusions is common amongst first timers. Who complain on the Internet.
Re-measure in what some may call semi-nearfield (contains baffle step loss but also some room influence).
Hmm is that a midwoofer problem, or a tweeter problem?
Turns out neither woofer or tweeter faulty. But a faulty (but new) capacitor in the crossover.
Capacitor replaced:
Is that a dip around 2-3KHz?
Is that a problem, or even a "design feature"? (eg. 'BBC dip')
Unhelpful measurement, borderline useless, because it's measured at LP.
Measuring at the LP and drawing conclusions is common amongst first timers. Who complain on the Internet.
Re-measure in what some may call semi-nearfield (contains baffle step loss but also some room influence).
Hmm is that a midwoofer problem, or a tweeter problem?
Turns out neither woofer or tweeter faulty. But a faulty (but new) capacitor in the crossover.
Capacitor replaced:
Last edited:
There is a gain switch inside the UMIK to give more headroom during loud testing.
Hi fluid,
I cannot see any switches on the body of the UMIK-1. So I have to remove the back cover or something and then adjust the switch?
Thanks
Vineeth
Yes you do, there is a bank of DIP switches on the PCB inside, the legend to set them is on the back of the PCB
https://www.avsforum.com/threads/massively-clipped-umik-1.1797489/
https://www.avsforum.com/threads/massively-clipped-umik-1.1797489/
It is 35ish litre volume for each driver (each driver is in a separate sealed compartment in the bigger box). Probably slightly under 0.7 Qtc for each driver
Thank you
I tried some different drivers. Played with 3 x 18W 8545 in parallel. Now trying with 2 different boxes - 2 x Purifi 6,5 and 2 x SB23NRX. The SB seems to have more "kick" to them. This might just be the more surface area - even though they measure almost the same. The SB goes deeper - maybe because they get 3o liters each and the Purifi only get around 12. So I was very much looking at WO24, because it's only 6 mm bigger than SB23NRX, and has that smooth clean FR for a woofer in a 3 way front, where I also use 4 subwoofers. I don't need super high SPL, but I kinda feel that it seems like an exponential requirement, the lower you go in frequency... you really need a lot more surface area.But 70 liters.... is starting to become an "adult" box, that can be a little tricky to make elegant
Please also look at Wavecor WF223 driver. Will help with more svelte cabinets..Thank you
But 70 liters.... is starting to become an "adult" box, that can be a little tricky to make elegant
Interesting.... just seems like the WO24 is still the most "clean" of them all.Please also look at Wavecor WF223 driver. Will help with more svelte cabinets..
I chose (and bought) these drivers in the same way as most beginners do I think, Impulsive buy..
I have always wanted to build a 3 way speaker. I had seen this driver being discussed in different forums. It looked like having great specs.
One day, I saw it being offered at a really good discount at my SB dealer. Very close to the price of the 8inch SB23 drivers.
So I jumped in and bought it..
.Then we had interesting discussions about directivity with the help of simulations done by fluid.
The sims showed that in the Rockport style box I had initially set out to build, having 2 of these drivers per side was better than one. So I went ahead and bought one more pair of these drivers (at a much higher price, though)
Then I saw discussions about the SB15/17CAC drivers and SB26CDC tweeter being used in hifijim's project with great results. So I jumped in and bought those too..
Dimensions-wise, the current twin Satori bass cabinet width is about 30cm (if not slightly less). Its height is about 88cm. Depth is 35ish cm. I am not saying these values precisely since these were in the design file I gave to the person who made the cabinet (about 600km away from my place). But he might have made his own minor modifications.. So I need to get the cabinet and measure dimensions on my own to know the exact values (will do all this in a week or two)..
Anyway, after having had a little bit of experience playing around with these drivers now, I find that these are great drivers (to my ears) in limited SPL applications like at my home
. It nicely sits in that transition region between big drivers and smaller drivers. I think it can be crossed over anywhere till 1kHz in different applications.So, in the works/plans are the following top modules (with this bass module)
:1) Something like below, but with the SB15 mid and waveguided SB26CDC
I might change the 2 way top module cabinet shape like below (but a bit more simplified) compared to the above top module (due to my love for rockport style cabinets
)2) An ASR directiva R2 style cardioid mid + waveguided tweeter
3) Something like below, with an ATH waveguide
4) A coaxial mid+tweeter (using Sica 5.5/6.5 inch coax or similar ) in a spherical ikea cabinet like below
Hopefully, I will get to do at least options 1 and 3 above if not all..
In the meanwhile, my 6 channel amp for all this DSP active crossover being made by a friend
Last edited:
Hi @tktran303
Here is the frequency response curve of the ESI U86XT soundcard (for one input nd one output)
Here is the distortion curve
Regards
Vineeth
Here is the frequency response curve of the ESI U86XT soundcard (for one input nd one output)
Here is the distortion curve
Regards
Vineeth
Here are the results for 88kHz sampling rate
Plots look about the same.
I have been implementing the DSP at 44.1kHz sampling since most of my music has that sampling rate. For measurement 88kHz may also be used but for implementing DSP, whichever sample rate one's music is having might be better. Otherwise one may need to do some sort of resampling of the music file for DSP and I don't know the complexities involved..
Small update and thinking about what all to do next for the Satori boxes
So my Satori woofer boxes have been built and packed for shipping. Each box weighs about 27kgs (for just the boxes alone and not with any damping material or other treatments inside ( also each Satori driver weighs about 4kg. So with drivers and box, it will weigh 35kgs per box after I fit the drivers in the box). Anyway the shipping part is stuck in a limbo at the moment while the guys who made the cabinet are figuring out which service to use for shipping. Hopefully the boxes will reach me in a week or two.
Now I am thinking about the treatments to do inside the cabinet for fixing the sealed box alignment for each woofer and pondering about things to purchase for the same. I have the following doubts:
1) What to use for treating the insides of the cabinets and filling it?
I already have the following material at home:
a) 2mm thick car sound damping sheets (https://www.amazon.in/STP-iSilver-Deadening-Damping-2-00mm/dp/B09S33TDSL). Are these kind of sheets used to add mass to the cabinet walls? or is there another purpose too
b) Plenty of 10mm thick acoustic felt that looks like below:
c) 25mm thick melamine acoustic foam:
This one: https://www.auralexchange.com/produ...-2ft-x-2ft-basotect-g-melamine-acoustic-foam/
d) Plenty of dacron polyfill: Mostly the type used for stuffing pillows and mattresses.
I was wondering about what to use where/do i really need to go and buy sheep wool (which is very very hard to get here) or something for filling the cabinet etc.. Then I came across this document where a study is performed to measure the effectivness of stuffing material and seems like there is not much difference: https://drive.google.com/file/d/16j...pW0T4LCJ9LgoPYvxHjSrr1epr8MaG3PzujwzIOvHH8Dwk
The I am trying to go through fluid's build posts here to study his observations: https://www.diyaudio.com/community/threads/full-range-tc9-line-array-cnc-cabinet.303417/post-5053920
I am a little confused about the whole thing..
2) What should I as cabinet feet?
I have tiled floor at my house. I am clueless about the advantages and disadvantages of spikes (with/without spike feet) vs rubber feet or something else.
What are the principles that govern the selection of feet for a woofer cabinet in my scenario?
Requesting help to get some clarity about the whole thing..
So my Satori woofer boxes have been built and packed for shipping. Each box weighs about 27kgs (for just the boxes alone and not with any damping material or other treatments inside ( also each Satori driver weighs about 4kg. So with drivers and box, it will weigh 35kgs per box after I fit the drivers in the box). Anyway the shipping part is stuck in a limbo at the moment while the guys who made the cabinet are figuring out which service to use for shipping. Hopefully the boxes will reach me in a week or two.
Now I am thinking about the treatments to do inside the cabinet for fixing the sealed box alignment for each woofer and pondering about things to purchase for the same. I have the following doubts:
1) What to use for treating the insides of the cabinets and filling it?
I already have the following material at home:
a) 2mm thick car sound damping sheets (https://www.amazon.in/STP-iSilver-Deadening-Damping-2-00mm/dp/B09S33TDSL). Are these kind of sheets used to add mass to the cabinet walls? or is there another purpose too
b) Plenty of 10mm thick acoustic felt that looks like below:
c) 25mm thick melamine acoustic foam:
This one: https://www.auralexchange.com/produ...-2ft-x-2ft-basotect-g-melamine-acoustic-foam/
d) Plenty of dacron polyfill: Mostly the type used for stuffing pillows and mattresses.
I was wondering about what to use where/do i really need to go and buy sheep wool (which is very very hard to get here) or something for filling the cabinet etc.. Then I came across this document where a study is performed to measure the effectivness of stuffing material and seems like there is not much difference: https://drive.google.com/file/d/16j...pW0T4LCJ9LgoPYvxHjSrr1epr8MaG3PzujwzIOvHH8Dwk
The I am trying to go through fluid's build posts here to study his observations: https://www.diyaudio.com/community/threads/full-range-tc9-line-array-cnc-cabinet.303417/post-5053920
I am a little confused about the whole thing..
2) What should I as cabinet feet?
I have tiled floor at my house. I am clueless about the advantages and disadvantages of spikes (with/without spike feet) vs rubber feet or something else.
What are the principles that govern the selection of feet for a woofer cabinet in my scenario?
Requesting help to get some clarity about the whole thing..
Vented and sealed boxes need to be looked at differently. A sealed box can take much more stuffing as you don't have to worry about killing the Helmholz resonance.I am a little confused about the whole thing..
If you can run impedance measurements, (it is very simple and a valuable source of information) then you can compare for yourself in your box.
It will also highlight where the trouble frequencies are and which strategies are likely to be effective.
The felt is a very good damping material for stopping internal standing waves and generally containing the back wave. Fibreglass as a bulk stuffing can add to the felt and also will increase the apparent enclosure size. Foam is similar but not as effective. Sheep's wool is good but not necessary if it is hard to get. Poly fibre is quite variable. The kind used for insulation works much like fibreglass, basic pillow stuffing is next to useless for controlling cabinet based resonances but it will increase the apparent volume.
Use only as much as you need to control the resonances. I would think a layer of felt at the walls with a layer of foam or fibreglass in front should be sufficient. Using a thin layer of polyfill behind the driver can help to keep any fibres from getting in the driver motors and helps with a gradual impedance change by not being very effective in every other way
For a hard floor I would use some form of rubber feet. You can get sorbothane half spheres or similar but they can be expensive and might be difficult for you to source.