You are jumping to grade 7 from kindergarten. @temp25 kinda tries to explain it to you.
You have to sort out the question if you want to risk building that speaker. 8 + 1 inch rarely works. If it works - it works borderline with some drawbacks and on some very specific conditions. Physics is in the way, not diyaudio community. Directivity and breakup modes of woofer are the keyword
Here is the simple tool, i calculated for RS225. You can see that SPL drops at higher freq.
https://www.tonestack.net/software/web/speaker-directivity-modeler.html
Here's your baffle. I selected the height of the woofer to get a smooth response. It changes as you move the woofer but it really doesn't get horrible. It's adding 2dB at 600hz, and then smoothly falling off. The bass at 100hz has dropped 5dB. The left edge of the graph is 100hz. Everything above 1k is virtually untouched.
Attachments
Last edited:
Then i have to review my steps in baffle simulator again. I can see now cone breakup higher in spectrum, smaller woofer? As you said RS 180?
Where is the dip at 1.5k coming from? The spec sheet shows flat in that range.
https://www.parts-express.com/pedocs/specs/295-356--rs225-8-spec-sheet.pdf
https://www.parts-express.com/pedocs/specs/295-356--rs225-8-spec-sheet.pdf
If you mount your speaker in-wall flush, in the woofer box as Dayton recommends:
Vituix is a 2D program, it does not calculate sound around your speaker cabinet. Also those wiggles in RS225 graph are breakup nodes. As you can see with the classic 2800Hz you will have theoretically very very good in-wall speaker, but it will have a nasty breakup nodes and unbearable sound
Now if we try to go as low as possible with tweeter:
You get ridiculous components and not so much nice curves. You still need some extra filters. Also this is just theoretically possible with this tweeter, but I don't know, as I don't have any of these drivers.
Both "designs" are for flush in-wall speakers
Vituix is a 2D program, it does not calculate sound around your speaker cabinet. Also those wiggles in RS225 graph are breakup nodes. As you can see with the classic 2800Hz you will have theoretically very very good in-wall speaker, but it will have a nasty breakup nodes and unbearable sound
Now if we try to go as low as possible with tweeter:
You get ridiculous components and not so much nice curves. You still need some extra filters. Also this is just theoretically possible with this tweeter, but I don't know, as I don't have any of these drivers.
Both "designs" are for flush in-wall speakers
I will describe how i get into sim "workflow" so you can comment what i do wrong.
1. I open enclosure tool and when i am done with box i export SPL and Z traces. I make shure im set to 2,83 V as reference level.
2. I open diffraction tool, draw my baffle, place the woofer, set mic distance to 1m, set mic in centre of speaker, enter cone SD..
3. Dayton audio gives me frd files for 0, 15, 30 and 45 degree angles, so i uncheck directivity and manually open every one of those files as half space response, first 0 degres and export that as full space response. I do this for all 0,15,30 and 45 angles.
4. I open merger tool. As low frequency part i select SPL form step 1. I set to spherical baffle step and use formula 115/(width of baffle in m)=115/0,3=383 Hz. For high frequency part i select what i manually saved one by one for 0,15,30 and 45 degree angles from step 3. I merge the output to FRD and use that in frequency response files for woofer input.
5. For tweeter i open diffraction tool and do same as step 3. for every angle i save files. Here is example on 45 degrees.
6. Again in merger tool i repeat step 4, only difference being for low frequency part i use manufacturer data and for high frequency i use data form step 5. I use output files and manufacturer impedance file ZMA for simulation input.
If any of this is good i dont know. When i started i read the document published on how to simulate with Vituix without measurements and it didnt work, the steps were different from this. This process i used now is from a video and is probably wrong so please correct me. I hope i traced steps well enough for you to see what "stupidness" i am up to.
1. I open enclosure tool and when i am done with box i export SPL and Z traces. I make shure im set to 2,83 V as reference level.
2. I open diffraction tool, draw my baffle, place the woofer, set mic distance to 1m, set mic in centre of speaker, enter cone SD..
3. Dayton audio gives me frd files for 0, 15, 30 and 45 degree angles, so i uncheck directivity and manually open every one of those files as half space response, first 0 degres and export that as full space response. I do this for all 0,15,30 and 45 angles.
4. I open merger tool. As low frequency part i select SPL form step 1. I set to spherical baffle step and use formula 115/(width of baffle in m)=115/0,3=383 Hz. For high frequency part i select what i manually saved one by one for 0,15,30 and 45 degree angles from step 3. I merge the output to FRD and use that in frequency response files for woofer input.
5. For tweeter i open diffraction tool and do same as step 3. for every angle i save files. Here is example on 45 degrees.
6. Again in merger tool i repeat step 4, only difference being for low frequency part i use manufacturer data and for high frequency i use data form step 5. I use output files and manufacturer impedance file ZMA for simulation input.
If any of this is good i dont know. When i started i read the document published on how to simulate with Vituix without measurements and it didnt work, the steps were different from this. This process i used now is from a video and is probably wrong so please correct me. I hope i traced steps well enough for you to see what "stupidness" i am up to.
I know, but i dont want to buy drivers if i know they are terrible to work together and dont fit in my box. As i see from box simulator, RS225 fits in box but as you said it doesnt cover upper mids good because cone breakup. Maybe another tweeter that goes lower?
Do you recomend any other driver combo that would give better results in box dimensions as posted?
Correct only one factory file for baffle step. The on axis response.
If you are convinced that you need to do otherwise, you are apparently far beyond me, and I can't help you.
Off axis is important and we'll get there eventually. Maybe your software does everything at once. I'm old school. I even make x-overs by trial and error.
If you are convinced that you need to do otherwise, you are apparently far beyond me, and I can't help you.
Off axis is important and we'll get there eventually. Maybe your software does everything at once. I'm old school. I even make x-overs by trial and error.
i skimmed through the thread.
regarding baffle step - it loses output in the lows, which means you have to reduce mid and highs to the same level, which in your case would be to about 80 db / watt. however i would ignore the baffle step completely because you can always EQ that later using a graphic equalizer and in fact because bass is affected by room it can never be predicted ahead of time and is best to equalize later using EQ that way you correct both baffle and room in one step.
just ignore baffle step completely. my 1st speaker i never implemented baffle step compensation and it was fine. remember baffle step is only on-axis. it doesn't exist in power response ( averaged space response ). just ignore it.
regarding driver choice 8" is too big for 2-way unless the tweeter is in a waveguide. Dayton is not the best brand - it's a house brand of PartsExpress the best i can tell, and no matter what driver you search for on PartsExpress you get Dayton in results because their search algorithm is programmed to always show Dayton drivers on top because it's their own brand ( i think ). They also have their own forum where you probably can't criticize Dayton. My advice is try to use something that isn't Dayton. I only had one Dayton driver and it was pretty poor.
whatever you do - DO NOT SEAL THE CROSSOVER INTO THE CABINET. make the crossover externally accessible somehow so you can tweak it later.
choose the right drivers and make crossover external. then you can always fix it later once you have accurate measurements.
regarding baffle step - it loses output in the lows, which means you have to reduce mid and highs to the same level, which in your case would be to about 80 db / watt. however i would ignore the baffle step completely because you can always EQ that later using a graphic equalizer and in fact because bass is affected by room it can never be predicted ahead of time and is best to equalize later using EQ that way you correct both baffle and room in one step.
just ignore baffle step completely. my 1st speaker i never implemented baffle step compensation and it was fine. remember baffle step is only on-axis. it doesn't exist in power response ( averaged space response ). just ignore it.
regarding driver choice 8" is too big for 2-way unless the tweeter is in a waveguide. Dayton is not the best brand - it's a house brand of PartsExpress the best i can tell, and no matter what driver you search for on PartsExpress you get Dayton in results because their search algorithm is programmed to always show Dayton drivers on top because it's their own brand ( i think ). They also have their own forum where you probably can't criticize Dayton. My advice is try to use something that isn't Dayton. I only had one Dayton driver and it was pretty poor.
whatever you do - DO NOT SEAL THE CROSSOVER INTO THE CABINET. make the crossover externally accessible somehow so you can tweak it later.
choose the right drivers and make crossover external. then you can always fix it later once you have accurate measurements.
I already gave you the guidelines, but you are the designer, so the choices are yours. Right now I cannot think any reasonable tweeter that can be paired with 8 inch woofer.
If I were you, my starting point would be to get info about the bigger tweeters
https://en.toutlehautparleur.com/sp...tml?hp_diam_bobine_a=9478,9481,9482,9485,9436
AND
Try to get info about the waveguides as they are strengthening the low end SPL
If you make a revised, baffle corrected frd for the woofer, please post it here, a long with the zma.
i know it is generally not advisable to have 3-way as your first speaker but if you can't do a 3-way then you should probably use a smaller woofer like 6.5" and in that case you may not get the performance you want.
with a 3-way it is much easier to find a good combination of drivers, and even though the crossover will be harder you can fix it later if you don't bury it inside the cabinet.
use the cheapest inductors and electrolytic crossover capacitors for the first crossover. once you are happy with the final crossover you can upgrade those capacitors later.
and even further down the line you can go to fully active DSP crossover.
only put the speaker drivers in the cabinet and have a separate set of terminals for each driver. do not put the crossover inside.
if doing a 2-way try to stick to 6.5" woofer and use a tweeter with a waveguide like this: https://www.parts-express.com/Dayto...-Neodymium-Tweeter-with-Wa-275-051?quantity=1
if doing a 3-way use anything from a 3" dome to a 5" cone for a midrange, and a 10" or a 12" woofer or a pair of woofers.
you can also do a 2.5 way. this is when you have two 6.5" woofers but one of them only plays below baffle step frequency. the crossover for 2.5 way is cheaper than for a 3-way and the efficiency is higher than for 2-way because you don't have to pad the woofer down for baffle step. google it.
you could of course also do a 3.5 way so like 1" tweeter, 3" dome midrange and two 8" woofers. in fact that's what i would do.
also i haven't looked into passive crossovers in about 20 years but i don't think there is any problem with using different impedances for different drivers. most good amps can handle 2 ohm or higher. as long as your overall impedance never dips below about 1.5 ohms you're good.
you could of course also do a 3.5 way so like 1" tweeter, 3" dome midrange and two 8" woofers. in fact that's what i would do.
also i haven't looked into passive crossovers in about 20 years but i don't think there is any problem with using different impedances for different drivers. most good amps can handle 2 ohm or higher. as long as your overall impedance never dips below about 1.5 ohms you're good.
The RS225 (aluminum version) needs to be crossed over acoustically 4th order or higher at 1,400Hz or lower due to the cone breakup.
Google RS Duet by Paul Kittinger.
Here is one that doesn't make sense from a cost perspective but proves it can be done (PDF is at the bottom of post #1)
Google RS Duet by Paul Kittinger.
Here is one that doesn't make sense from a cost perspective but proves it can be done (PDF is at the bottom of post #1)