Introducing Directiva - An ASR open source platform speaker project

[Rick Sykora]

The SF pics just show a pack of high power resistors.

Seems like those would be overkill, but may just depend on the motor in the woofer...

I have the dual ES180 box that I could readily use to try and see how feasible this may be with generally available drivers.

 

[dansan69]

Apparently the resistor based system is also referred to as a “Cadawas” system, I found a patent from 1975 describing the operation:
https://patents.google.com/patent/US4145571A/en?oq=4,145,571

 

[MrPeabody]

The resistor will change (increase) the amount of damping on the PR than without (open circuit). The LCR suggested will apply damping in a small freq range than over the entire bandwidth, and this could kill a resonance in the PR.

A driver with voice coil but left open circuit produces no damping since there is no current. If a simple wire is used to close the circuit, this allows current to flow and damping occurs in association with the Ohmic heating of the voice coil. The R in the LCR filter will contribute to damping but only for frequencies for which current flows, which now becomes a tricky question because the coil itself is LR in series and you are adding another LCR (series or parallel?) into the voice coil LR circuit. On the surface is seems trivial but is probably less trivial than it seems.

 

[MrPeabody]

Let us all remind ourselves of that saying about what happens when there are too many cooks in the kitchen. When Rick and ctrl are encountering a bit of difficulty with some particular, clearly identifiable aspect of what they are trying to do, they will no doubt appreciate helpful input. But some of the suggestions that have been made amount to proposals for a major overhaul of the design. Rick is responding to all of it in a perfectly diplomatic manner. I'm just saying ...

 

[Wolf]

A driver with voice coil but left open circuit produces no damping since there is no current. If a simple wire is used to close the circuit, this allows current to flow and damping occurs in association with the Ohmic heating of the voice coil. The R in the LCR filter will contribute to damping but only for frequencies for which current flows, which now becomes a tricky question because the coil itself is LR in series and you are adding another LCR (series or parallel?) into the voice coil LR circuit. On the surface is seems trivial but is probably less trivial than it seems.

I thought I laid this out rather clearly. Not really sure this was more than redundant.
Best regards,

 

[MrPeabody]

I thought I laid this out rather clearly. Not really sure this was more than redundant.
Best regards,

I am really, really, really sure that what I wrote was not redundant. You wrote two consecutive sentences that weren't grammatically correct. Sometimes it is possible to figure out exactly what someone was trying to say in spite of grammatical problems, and I did in fact make a sincere effort to do that with both of your sentences, but I eventually gave up. Here's what you wrote:

The resistor will change (increase) the amount of damping on the PR than without (open circuit). The LCR suggested will apply damping in a small freq range than over the entire bandwidth, and this could kill a resonance in the PR.

Now why would I have difficulty figuring out exactly what it was that you were trying to say (and that you believe that you had said)? It is manifest that there is no damping with an open circuit. It would be meaningful to point out that damping will be greater with a low resistance value than with a large resistance value, and that damping will be even greater using a simple jumper wire compared to any resistor. You evidently think that what you wrote actually managed to say this little amount. You even think, evidently, that what you wrote said this little amount "rather clearly". Not only did it not say this "rather clearly", it did not come close to saying this little amount.

I had a perfectly good reason for writing what I wrote, and since you did not understand the reason ("Not really sure this was more than redundant"), I decided that it was appropriate for me to explain the reason in a way that you might be able to understand. Also, obviously phony politeness ("Best regards") following something that was surly only increases the surliness.

 

[MrPeabody]

P.S. Having now given this a moment of two of thought, the correct sort of circuit to connect across the driver terminals, to damp a particular range of frequency, is a band pass filter.

 

[Rick Sykora]

I appreciate everyone's contributions! I think the filtered PR proposal is worth pursuing if it shows enough value.

Won't know until I test and still have to finish up my nephew's Xbox and my own NC502MP work. So let's not get too far along until I investigate a bit more. Thanks!

 

[Wolf]

P.S. Having now given this a moment of two of thought, the correct sort of circuit to connect across the driver terminals, to damp a particular range of frequency, is a band pass filter.

Which is of course- a (series wired) LCR! I stated that. The fact that the wire bridge would increase the braking or damping of the PR was what I thought was a given. Maybe that was an assumption I should not have made. Resistors will of course be less damping as there is resistance present rather than a short.

I really had no intention of starting a war with words, hence my 'BR' at the end, but I did feel what I stated omitted the obvious given information. Sometimes I think something is a given, when apparently it's not to everyone else.

 

[Rick Sykora]

So, just a quick update on Directiva. @ctrl and I are working on the initial design in VituixCAD. We have both worked on crossover design with different approaches. Since only one can go to Amir, we are hashing it out. This has taken some time as we are both busy with other projects but expect will post the project this weekend.

 

[Rick Sykora]

Still working on crossover design. @ctrl is using Hypex dsp vs my minidsp and differences are much more than I expected in VCAD modeling.

 

[tktran303]

Btw, How much baffle step compensation are you shooting for?

 

[Trdat]

Still working on crossover design. @ctrl is using Hypex dsp vs my minidsp and differences are much more than I expected in VCAD modeling.

Can you just give me a real basic idea of how VituixCad is used for crossover design if your using DSP such as Hypex.

I mean, the end result is the values of crossover components right? Then do we take that and use it in the DSP or is it something I am missing?

 

[Rick Sykora]

Can you just give me a real basic idea of how VituixCad is used for crossover design if your using DSP such as Hypex.

I mean, the end result is the values of crossover components right? Then do we take that and use it in the DSP or is it something I am missing?

I have a simple version of my crossover design with the minidsp HD. The details of how to do this in VituixCAD are in the help here: VituixCAD help 2.0

After measuring each driver in the cabinet, you add each driver to the project and link it to its measurements. Next you add the crossover components and see you can see how they affect frequency response, phase, etc.

Here is a simple example using only the on-axis measurements:

In this case, I am including leveling the driver outputs and the time delay I measured between the drivers. The rest is simply the crossover filters and equalizations to even out the response. As I showed in earlier posts, I can test how well the drivers are integrated by inverting the tweeter and seeing how deep a notch I get in the frequency response...

I have done some off-axis measurements and @ctrl has integrated them into his model and has done his (much more sophisticated) approach to the design. This will get refined further by Amir's measurements in the near future.

As soon as we resolve the differences due to the dsp choice, will post our initial design and the VCAD project files.

 

[headshake]

I have a simple version of my crossover design with the minidsp HD. The details of how to do this in VituixCAD are in the help here: VituixCAD help 2.0

After measuring each driver in the cabinet, you add each driver to the project and link it to its measurements. Next you add the crossover components and see you can see how they affect frequency response, phase, etc.

Here is a simple example using only the on-axis measurements:

View attachment 126431

In this case, I am including leveling the driver outputs and the time delay I measured between the drivers. The rest is simply the crossover filters and equalizations to even out the response. As I showed in earlier posts, I can test how well the drivers are integrated by inverting the tweeter and seeing how deep a notch I get in the frequency response...
View attachment 126433

I have done some off-axis measurements and @ctrl has integrated them into his model and has done his (much more sophisticated) approach to the design. This will get refined further by Amir's measurements in the near future.

As soon as we resolve the differences due to the dsp choice, will post our initial design and the VCAD project files.

I think the hypex plate amps calculate the gain for each amp based on the wattage or something wonky. Were you finding a match between the virtuixcad model and the hypex plate amp gain for each driver? I had to forget it and just make things flat within the DSP separately.

 

[Trdat]

I have a simple version of my crossover design with the minidsp HD. The details of how to do this in VituixCAD are in the help here: VituixCAD help 2.0

As soon as we resolve the differences due to the dsp choice, will post our initial design and the VCAD project files.

Great, that helps a lot. Just a few questions...

So for off axis you again measure off axis but if you got what you want on axis with the crossover how do you tweak the off axis? Basicaly is there a way to integrate the off axis measurements or how is the overall goal achieved with trying to match the off and on axis measurements in Vituix cad?

And what is the LR24 IIR F 2200hz? What is it trying to mitigate in the response? Sorry I can't figure this one out.

Lastly, the link doesn't work. Is it just the Vituixcad help or a thread on ASR?

 

[Rick Sykora]

Great, that helps a lot. Just a few questions...

So for off axis you again measure off axis but if you got what you want on axis with the crossover how do you tweak the off axis? Basicaly is there a way to integrate the off axis measurements or how is the overall goal achieved with trying to match the off and on axis measurements in Vituix cad?

And what is the LR24 IIR F 2200hz? What is it trying to mitigate in the response? Sorry I can't figure this one out.

Lastly, the link doesn't work. Is it just the Vituixcad help or a thread on ASR?

The link is fixed. Spend some time there and then can address more questions later. Will spend some time in this thread comparing our initial approach and what adjustments we make after Amir’s more precise measurements.

Finally, the LR24 is the crossover filter.

 

[Trdat]

The link is fixed. Spend some time there and then can address more questions later. Will spend some time in this thread comparing our initial approach and what adjustments we make after Amir’s more precise measurements.

Finally, the LR24 is the crossover filter.

Great, I will look into it. I am desinging a center speaker emulating my right and left speakers but with an added waveguide really would like to get the hang of it.

Yes, it is the crossover filter I was just confused because it was in the middle of the filters used but its technically at the end. And the last one must be delay or something.

Thanks again.

 

[Rick Sykora]

Am attaching the VCAD project candidate for Amir to measure. This is just a draft and will change based on Amir's measurements.

This design is from @ctrl and he uses a Hypex FA dsp. Before it goes to Amir, I will verify it with minidsp HD instead. Weather has been exceptional here for April (80F today). So have been doing outside projects, Will rain later this week and will do the measurements with the candidate design then .

 

[Rick Sykora]

Here is a look for those who do not have VituixCAD installed,,,