Hey diyAudio Friends.
I am designing and building my first set of desktop speakers. Casting them from concrete! See below.
Here are some specifications:
Internal Volume 3.1 L
Woofer: Dayton Audio TCP-115 8Ohm
Tweeter: Dayton Audio PTMini-6 Planar Tweeter 6 Ohm
I am currently designing the crossover. I am using VituixCAD to do the modelling, but I am a total newbie.
Could anyone give some input. I am especially confused about the power supply settings.
Any input is appreciated!
Thank
Sam
I am designing and building my first set of desktop speakers. Casting them from concrete! See below.
Here are some specifications:
Internal Volume 3.1 L
Woofer: Dayton Audio TCP-115 8Ohm
Tweeter: Dayton Audio PTMini-6 Planar Tweeter 6 Ohm
I am currently designing the crossover. I am using VituixCAD to do the modelling, but I am a total newbie.
Could anyone give some input. I am especially confused about the power supply settings.
Any input is appreciated!
Thank
Sam
Hi and welcome to the forum!
There has been no response as your request really needs to be in the 'Loudspeakers: Multi-Way' section of the forum.
Click on the 'three black circles' symbol at the bottom right of your post, then click on 'Report'. There you can ask a moderator to move your thread to the appropriate location.
I'm not a crossover design expert, but I can tell you that there is something awry with your crossover implementation. A 180uF capacitor is much too large a value for a tweeter filter. A series capacitor in the order of a few microfarads would be more like the thing.
There has been no response as your request really needs to be in the 'Loudspeakers: Multi-Way' section of the forum.
Click on the 'three black circles' symbol at the bottom right of your post, then click on 'Report'. There you can ask a moderator to move your thread to the appropriate location.
I'm not a crossover design expert, but I can tell you that there is something awry with your crossover implementation. A 180uF capacitor is much too large a value for a tweeter filter. A series capacitor in the order of a few microfarads would be more like the thing.
We're in the correct place now, Sam! 😎
Here's the spec sheet for the PTMini-6 tweeter: https://www.daytonaudio.com/images/resources/275-083--dayton-audio-ptmini-6-specifications.pdf
It's resonant frequency is quoted as 4,461 Hz with a 'useable frequence range' of 2,900 - 25,000 Hz.
The spec sheet says that the tweeter requires a '4,000 Hz crossover, 12dB/octave'. A simple 6dB/octave crossover layout could destroy this tweeter unless its frequency were exceedingly high.
I hope the experts will help you out with the required crossover design.
Here's the spec sheet for the PTMini-6 tweeter: https://www.daytonaudio.com/images/resources/275-083--dayton-audio-ptmini-6-specifications.pdf
It's resonant frequency is quoted as 4,461 Hz with a 'useable frequence range' of 2,900 - 25,000 Hz.
The spec sheet says that the tweeter requires a '4,000 Hz crossover, 12dB/octave'. A simple 6dB/octave crossover layout could destroy this tweeter unless its frequency were exceedingly high.
I hope the experts will help you out with the required crossover design.
Here's the TCP115-8 spec sheet for good measure:
https://www.daytonaudio.com/images/resources/295-416--dayton-audio-tcp115-8-specification-sheet.pdf
https://www.daytonaudio.com/images/resources/295-416--dayton-audio-tcp115-8-specification-sheet.pdf
Don't worry about the generator settings, the effect will be common to both speakers but it's the differences you're interested in.
Have you modelled the cabinet? It's not a deal breaker if you haven't it's just helpful to know. I wonder whether 4kHz would be good for both of these.. you may find yourself experimenting along those lines. Second order for both looks interesting at first glance.
Have you modelled the cabinet? It's not a deal breaker if you haven't it's just helpful to know. I wonder whether 4kHz would be good for both of these.. you may find yourself experimenting along those lines. Second order for both looks interesting at first glance.
Hi Galu!
Thanks for the info.
I have since read more on the whole design process and after following this guide got a second order crossover that just needs confirmation that it will work or at least is a decent start.
I have attached the new design below:
How do I model my enclosure?
I calculated the necessary internal volume and port, but how would I convert the inclosure into its electrical equivalent? I plan on casting it from concrete. I guess this would have to be accounted for in the estimation? Is there a good guide somewhere I can follow?
Cheers,
Sam
Thanks for the info.
I have since read more on the whole design process and after following this guide got a second order crossover that just needs confirmation that it will work or at least is a decent start.
I have attached the new design below:
How do I model my enclosure?
I calculated the necessary internal volume and port, but how would I convert the inclosure into its electrical equivalent? I plan on casting it from concrete. I guess this would have to be accounted for in the estimation? Is there a good guide somewhere I can follow?
Cheers,
Sam
Attachments
I believe you have the z offset on the woofer incorrect. Vituixcad convention is for positive values in z to be away from the listener therefore it needs to be +29mm. The impedance curve looks a bit too varied, if you can flatten this out it will help the amplifier load. Also if you can get the phase plots to be coincident an octave above and below the crossover point that would also be nice.
Crossover simulation is not my field of interest. So I know less than you, Sam, about using crossover software. However, that won't stop me commenting on your crossover schematic!
All I can do is compare your component values with those obtained from a 😈😱crossover calculator😱😈. I used a crossover frequency of 4,000 Hz, 8 ohm for woofer, 6 ohm for tweeter and a 2nd order Linkwitz-Riley alignment.
Now, before the crossover experts start berating me, I am well aware of the reasons why one should not use a crossover calculator! However it does provide me, a non-expert, with some means of comparison.
My tweeter components work out as 3.3uF and 0.47mH while the woofer components work out as 0.60mH and 2.2uF (translated to standard available component values) - just different enough from yours, Sam, to give me pause for thought.
So experts, what say you about Sam's component values?
A "decent start" or not?
More inportant than bass reflex tuning, which only affects lower bass extension (and can be more or less considered independent from crossover) is the influence of baffle step and baffle diffraction. As far as i know vituixcad can simulate this.
I hope you are learning from the experts, Sam.
As an aside, may I say I like your moniker. My friend Tam the Bam was the 'lighting technician' during my embryonic 70s disco days.
Tam was placed in charge of a row of three toggle switches with which he could independently operate the red, blue and green disco lights.
He put on a fantastic sound to light show - one not to be bettered by the electronic sound to light unit that eventually replaced him.
Sadly, Tam is no longer with us, but I can still see that intense look on his face as his fingers flew across those switches.
Happy days.
As an aside, may I say I like your moniker. My friend Tam the Bam was the 'lighting technician' during my embryonic 70s disco days.
Tam was placed in charge of a row of three toggle switches with which he could independently operate the red, blue and green disco lights.
He put on a fantastic sound to light show - one not to be bettered by the electronic sound to light unit that eventually replaced him.
Sadly, Tam is no longer with us, but I can still see that intense look on his face as his fingers flew across those switches.
Happy days.
Hi Galu et all.
Thank you for all of your input. I have tried to incorporate your advice. I added the Z-offset as Ugg10 suggested.
I also managed to flatten the impedance curve a bit.
What do you guys think? Can I go ahead and order the parts?
Baffle Step and baffle diffraction is still something I would need to read up on. It would mean adjusting the enclosure, right?
Thanks everyone.
PS: Glad my username brought up some fond memories, Galu. Tam the Bam sounds like quite the legend.
Thank you for all of your input. I have tried to incorporate your advice. I added the Z-offset as Ugg10 suggested.
I also managed to flatten the impedance curve a bit.
What do you guys think? Can I go ahead and order the parts?
Baffle Step and baffle diffraction is still something I would need to read up on. It would mean adjusting the enclosure, right?
Thanks everyone.
PS: Glad my username brought up some fond memories, Galu. Tam the Bam sounds like quite the legend.
Your impedance is getting a little low. Was there a reason for doing this as there's no sense in trying to flatten the peak otherwise?
You could apply the downward trend (shelf) by eye.
Got the impedance up above 4ohms again. No real reason, other than it being to high in the beginning and trying to adjust for that.
Not sure what you mean by this
Here is the intial plot after following the guide metioned above. The inductors are quite a lot smaller as you mentioned.
I tried to tweak this schematic to get rid of the valley around 4K and flatten the response overall. This is what I came up with:
Hello samthebaam,
no, it will hugely influence the response (in your case mainly the midwoofer) and will require a complete redesign of crossover.
So you will have to adapt the x-over including effects of baffle step and diffraction.
You may also need to adjust the baffle, but i think your desugn is already quite ideal (narrow with chamfers).
A shelf is lilke a step. Above and below some frequency it is flat. One end is higher. Between them there is a gradual slope. Looking at this plot, say you reduced the region near 2-3kHz below the pink line and did the same with the region above 6kHz. You now have raised bass level. When you compensate the baffle this way you try to find a good frequency and the right amout for the level to fall.
That should have been "design", of course!
In addition and for better understanding:
I suppose you are working with the " infitine baffle" responses provided by dayton. That only makes sense if you actually use an infinite baffle.
If you however include the effect of your small baffle you will see two effects:
- lower bass response level, theoretically -6dB below the baffle step, which is where the loudspeaker directivity transitions from half space to full space. Frequency depends on baffle size, maybe ~500 Hz.
- several higher frequency peaks and dips created by edge diffraction depending on distance from driver to baffle edges and on chamfers/roundovers, may not be so pronounced for your baffle (edit: actually true as i see in your baffle simulation!)
You should consider those when designing the preliminary crossover. You can simulate baffle effects in vituixcad. And it is advisable to take measurements of drivers in the enclosure for final simulation of crossover.
Last edited:
That will be difficult unless you plan to have an active speaker.
The baffle effect will be slightly compensated by the response of the tcp115. And for the rest you will need to cut higher frequencies, e.g. by using a bigger inductor for midwoofer and L-pad or a resistor for the tweeter.
Just include the baffle effect to your woofer response and continue developing the crossover. I am no vituixcad user yet so i cannot advise on how to include baffle effects.
Last edited: