The S2 "driver only" suggestion was for the Frequency Response window primarily. If you're going that route, you can set the crossover components back to normal, go to the frequency response window and turn on the S1, S2, and S3 curves; and then also turn on the S2 (driver only) curve.
I can't see the left hand axis of the impedance graph, so I can't tell if there's something going on with the S2 (driver only) impedance also or just a low scale selected. Please expand the range so the entire impedance fits in the graph and get the axis labels visible.
I can't see the left hand axis of the impedance graph, so I can't tell if there's something going on with the S2 (driver only) impedance also or just a low scale selected. Please expand the range so the entire impedance fits in the graph and get the axis labels visible.
Also, when I want to match the db do I add a lpad to mid and woofer to round up to tweeter, or can I have the tweeter match down to the woofer and mid.
Yes. That gives a better idea about some other things that are going on.
Your midrange impedance file looks out of whack also. According to the spec sheet, it should look approximately like this:
Your file has a much steeper rise into the high frequencies, and is still off the graph even though it should top out at a bit under 40 ohms at 20 kHz.
On your frequency response graph, also go into Scale>Max frequency, and set it to 20 kHz.
In passive crossovers, you have to go down to match levels.
Your midrange impedance file looks out of whack also. According to the spec sheet, it should look approximately like this:
Your file has a much steeper rise into the high frequencies, and is still off the graph even though it should top out at a bit under 40 ohms at 20 kHz.
On your frequency response graph, also go into Scale>Max frequency, and set it to 20 kHz.
In passive crossovers, you have to go down to match levels.
Last edited:
So I would have to bring the tweeter down? And how does this look, I adjusted the zma file for the wavcor?
Looking better. Please rescale the impedance graph to the 10 ohm setting now.
And set the Frequency Response graph so it goes up to 20 kHz. We already included some padding to bring the tweeter level down roughly 5 dB, but I can't see its impact with the restricted frequency range of the graph.
May as well turn on the S1 (driver only) frequency response also, so we can look at that.
And set the Frequency Response graph so it goes up to 20 kHz. We already included some padding to bring the tweeter level down roughly 5 dB, but I can't see its impact with the restricted frequency range of the graph.
May as well turn on the S1 (driver only) frequency response also, so we can look at that.
OK, your tweeter and midrange raw frequency responses look about right, so S1 (driver only) and S2 (driver only) frequency response can be turned off.
Tweeter level still looks pretty hot, so bringing it down another 4 or 5 dB is probably in order. Right click on the tweeter padding resistor and use "tune" to increase resistance until the overall frequency balance for Sys looks like what you want. The up and down buttons are good for this kind of adjustment.
Tweeter level still looks pretty hot, so bringing it down another 4 or 5 dB is probably in order. Right click on the tweeter padding resistor and use "tune" to increase resistance until the overall frequency balance for Sys looks like what you want. The up and down buttons are good for this kind of adjustment.
Some adjustments that may get you closer....
Cut L1 in half. It's allowing the tweeter to extend to low. 0.1-0.5 range.
Cut L3 in half. This value is huge, more typical is 1-4mH.
Because L3 was off, so is overly huge L4, and I would try 2.5mH to start with. Toggle down from there.
Bump C4 up double.
So, are you using manufacturer curves? If so, the the curves are optimized for a HUGE IEC baffle and not your intended box. It'd be like making in-walls if the files are not adapted for the baffle size and driver placement intended here. The instruction I and others have given you are for the files you have provided. While this is good for familiarity of the program and usage, as well as practical values, it still may not be optimal for your intended drivers in their final installation.
Cut L1 in half. It's allowing the tweeter to extend to low. 0.1-0.5 range.
Cut L3 in half. This value is huge, more typical is 1-4mH.
Because L3 was off, so is overly huge L4, and I would try 2.5mH to start with. Toggle down from there.
Bump C4 up double.
So, are you using manufacturer curves? If so, the the curves are optimized for a HUGE IEC baffle and not your intended box. It'd be like making in-walls if the files are not adapted for the baffle size and driver placement intended here. The instruction I and others have given you are for the files you have provided. While this is good for familiarity of the program and usage, as well as practical values, it still may not be optimal for your intended drivers in their final installation.
Overall, yes. The simulation seems to be behaving now and system response is basically flat, so fine tuning can be done from here if you desire.
As wolf_teeth said, there are lots of things that can be done to decrease inductor size (which decreases cost also), try to get closer to target curves, etc. How much of that you want to do is up to you. Given the low resonance of your tweeter and its amount of padding, as long as power levels are moderate you'll probably be OK for basic driver protection/function with the circuit as is. If you want to learn more and optimize, some tweaks can be done.
Ok I will try this out. Thank you. I am using the manufacture curves but putting them in an enclosure. Will this crossover design make them sound much worse. I am planning on having a pretty large port as well if that may help.
Thanks again for all the help. Really I just want to be able to make my own crossover and have the speakers sound decent. So if you think this crossover design will work and be listenable I’ll pull the trigger and get to work.
As long as this design looks pretty decent that is perfectly fine with me. Really I just want to build my own crossover that can accommodate the drivers I have installed into my enclosure. If this design looks good I’m probably going to build it then report back.
Originally I had a passive radiator, the wave core and the tweeter. I installed a prebuilt crossover but the bass was lacking and the highs were very sharp. So if this crossover will incorporate the woofer i bought and dampen the highs that’s all i really want. Thanks!
The reason the bass was lacking was due to baffle step. This is a function of the width of the baffle, and relates to a -6dB rolloff below the wavelength associated with the width dimension. If the Wavecor had been on/in a wall enclosure, this would have been much better because the xover did not compensate for it. The off the shelf xover is for ideal drivers that don't and can't exist: flat impedance, flat frequency response, concidently mounted, and in-wall speakers. The smaller box is the reason you have the bass droop in the first place, because of baffle step. You likely did not need the woofer, but a more optimal xover.
As to whether it looks good?! Well- garbage in equals garbage out in this case. We do not know how well the mfr curves will reflect the response of them in your boxes, as they have not been adapted for this task. To say it would look good would be dishonest, because we really have no way of knowing for certain. What has been presented is information that may or may not have been best to use for your simulation. I have not seen a resim since you looked at my suggestions. If the response looks flat, and these will be near wall, as in 1' in front of, then it might be okay, and just only okay, and not likely fantastic.
As Matt said- how deep down the rabbit hole do you want to go? How much work do you want to do? This is just the beginning of a design phase, and you will have the voicing stage and possible adjustments to make to make them better than a haphazard simulation such as this.
Don't let me dissuade you, as that is not my intent. My point is that you have only scratched the surface, and don't know what you don't know yet. Venture on, and if disappointed, regroup, learn more, and alter and improve.
Have fun, and good luck!
As to whether it looks good?! Well- garbage in equals garbage out in this case. We do not know how well the mfr curves will reflect the response of them in your boxes, as they have not been adapted for this task. To say it would look good would be dishonest, because we really have no way of knowing for certain. What has been presented is information that may or may not have been best to use for your simulation. I have not seen a resim since you looked at my suggestions. If the response looks flat, and these will be near wall, as in 1' in front of, then it might be okay, and just only okay, and not likely fantastic.
As Matt said- how deep down the rabbit hole do you want to go? How much work do you want to do? This is just the beginning of a design phase, and you will have the voicing stage and possible adjustments to make to make them better than a haphazard simulation such as this.
Don't let me dissuade you, as that is not my intent. My point is that you have only scratched the surface, and don't know what you don't know yet. Venture on, and if disappointed, regroup, learn more, and alter and improve.
Have fun, and good luck!
These two phenomena tend to mimic each other. If the treble is very hot, bass will seem lacking. Adding the woofer should give you easier flexibility to balance the system later, so that can be seen as a benefit.
Your tweeter is definitely much more sensitive than the other drivers, so that's real and should be addressed more adequately in the new crossover as modeled.
But it's not the best it can be right now, even given the limitations. As a basic comparison, here's the frequency response of your crossover with flat impedance and flat frequency response used:
The individual driver curves are a bit away from the result you've achieved so far. With some basic adjustments, you can get closer to the above target curves. Again, you don't have to do it, but you're in pretty far, so adjusting a bit more seems like the logical thing to do.
There's still the uncertainty of the manufacturer frequency response curves vs reality, but many people design the way you have and are happy with the result. Taking accurate measurements is a whole other endeavor that can be pretty hard for some people to get their head around if the ideas aren't already familiar. I don't think most people pick up all the nuances quickly.
Another couple points.
The overlap you have between the woofer and midrange is not the textbook way to do things given your stated crossover points, but it's also giving you some bass/midbass boost that you might want.
You can try wolf_teeth's suggestions individually and keep the ones you like.
And again, it's your speaker, so do what makes you happy. That's what the hobby is about.
The overlap you have between the woofer and midrange is not the textbook way to do things given your stated crossover points, but it's also giving you some bass/midbass boost that you might want.
You can try wolf_teeth's suggestions individually and keep the ones you like.
And again, it's your speaker, so do what makes you happy. That's what the hobby is about.
Thank you. I think taking accurate measurements is something I could get into in the future but this is my first time building a crossover and my own speakers. If I want to get closer to the target curves, how should I tweak the design? Just play with the values?