You don't have to merge the tweeter (and mids) FR because they are dome drivers and there is no cabinet/LF part.
I agree with the "you don't have to merge tweeter FR".
The reason however is that usual situations in living spaces allow gated measurements up to around 5 ms. That allows "anechoic" tweeter measurements, but usually no accurate mid or woofer measurements.
It looks like you altered the woofer response, when you should have left it alone. You want to add a tweeter response, and add them like a typical x-over sim.
From a few posts back....This should be your woofer response. Your graph looks like MAYBE the baffle correction has been applied twice.
Last edited:
that's not an answer. there was a 10 decibel dip at 3 khz - where did it go ?
what is the schematic for a circuit that accomplishes all this stuff ? why is there a schematic for the tweeter but not the woofer ? the woofer is the one to which 90% of response shaping is applied.
and how did the woofer magically gain efficiency around 100 hz ?
this i believe:
https://www.diyaudio.com/community/attachments/1741317031420-png.1431801/
this i do not:
https://www.diyaudio.com/community/attachments/1741351845732-png.1431958/
somehow crossover moved from 1.4 khz to 2.3 without any negative effects from breakup
and somehow the system gained 10 decibels in efficiency
HOW?
Last edited:
@Dissident Sound , it is hard to answer all the questions, because this thread has more signs of mental exercise than real development.
VituixCAD is not 3D tool. It is 2D tool which works under specific conditions. I can make zillion simulations with manufacturer curves, processed through the internal VituixCAD tools - still it will differ a lot from the simulations done from measurement data.
VituixCAD is not 3D tool. It is 2D tool which works under specific conditions. I can make zillion simulations with manufacturer curves, processed through the internal VituixCAD tools - still it will differ a lot from the simulations done from measurement data.
This is the filter that was used. This response was later corrected for the baffle, so the bass became flatter on the graph. That corrected graph was entered in the sim above, so no filter is applied. The tweeter filter the OP made, is made to match the previously filtered woofer. We still need to play around with offsets. If you look closely, you can see a reddish line that is a LR4 2.3k target. You may need to zoom in to see it clearly. I made the filter to match that ARBITRARY target. This sim was in my computer recycling bin. We also need to see if we are below 6 ohms. I didn't even look. We'll get to that. I'm trying to get there one step at a time. The factory data shows 87dB at 100hz. We are now at 83dB after baffle step losses. We did not gain 10db.
Odds are, we will want to cross lower than this. The breakup may still be audible. I don't want to design the final filter. I think the OP should. This is more of an exercise. Might move up to a 3-way. Who knows. Just trying to show him how I would use software. His software is foreign to me. So it's hard for me to tell him exactly what to do. I use PCD, as you can see below.
Odds are, we will want to cross lower than this. The breakup may still be audible. I don't want to design the final filter. I think the OP should. This is more of an exercise. Might move up to a 3-way. Who knows. Just trying to show him how I would use software. His software is foreign to me. So it's hard for me to tell him exactly what to do. I use PCD, as you can see below.
Last edited:
Follow the link in post #40 - the PDF writeup is 22 pages long - or just go HERE. It discusses the cone breaking, acoustic slopes, baffle diffraction bump, notch filter, etc. And also how the $170 Satori tweeter doesn't necessarily make sense 🙂
I almost started believe in miracles.
That is very high level trolling, my friend 😉
Too much of stuff on -22dB
you guys are doing something wrong. it is not physically possible.
how did a 87 db driver lose 6 db to baffle step at 150 hz and end up at 88 db ?
likewise the crossover transfer function is definitely realizable in DSP but i wouldn't know how to design it passively so i would like to see the schematic for it.
i mean without a schematic - how are you going to build it ?
The writeup link in post #131 goes into detail of pretty much everything being discussed. Also, it is clear with the actual measurements, that after 6dB of baffle step loss you end up with...6dB of baffle step loss. (Absolute SPL levels are not accurate except in Figure 14)
Actual measurement, with filter:
Schematic:
Actual measurement, with filter:
Schematic:
Haha. This exercise makes a point, and -22dB is a workable minimum value. You can push it down any amount but below a point you won't hear the difference. You may need more components to push further down.
By the way, breakup power is usually considerably less than an axial response plot would suggest.
I think that is more like -30dB and less for me, at least if I trust VituixCAD sim and virtual crossovers played through EQ APO. -22dB will still mess with my brains.
Why breakup power is less than plot shows? Is there some rule, that simulations are more harsh?
Here's the baffle sim. It shows about 4dB loss at 150hz. I don't have the computer on right now, but there's a good chance that the 4th order filter humped up the 100hz to 200hz range a couple dB.
I can sketch a schematic if you need it. It's just 2 coils, and 2 caps. Typical 4th order layout. The values are in the PCD sim shown in post # 130 above. The little numbers are coil resistance.
Last edited:
OK i will again excuse myself with a warning to @electronicsman to not buy anything until you can go through the entire design process on your own without relying on any files generated by somebody else.
they could have made a mistake and you will end up eating the cost of it unless you can return the driver and crossover parts to the store.
By contrast @A4eaudio did actually build this speaker and made it work but with a 1.4 khz crossover which is an extremely low crossover point for a 1" dome tweeter.
My guess is that if you want to do a 2-way using this 8" driver you will have to go with that low 1.4 khz crossover point, which will put a lot of stress on the tweeter.
If your goal is to simply build something then you could go through with building something similar to what @A4eaudio has built except using a cheaper dome tweeter.
but if you want a high performance system to actually enjoy for at least a few years then this 8" Dayton won't work in a 2-way.
but frankly unless you're willing to build your own box, do a 3-way and use drivers other than Dayton you can't have very high expectations anyway.