It's probably been covered before, but I am trying to alter a Troels Gravesen design, as I am unhappy with the treble. Specifically I would like to model the crossover changes needed to use Fountek CD3 ribbons. Jenzen-NEXT
But my problem with XSim is that I don't know how to load driver parameters. I have simmed the existing crossover with the generic values as set in XSim, but can't see a way to load in the actual parameters as given by SEAS for the existing drivers. So obviously I can't make changes.
The reason for trying the ribbon is that I have a hearing peak around 6-8kHz, and have always liked the way ribbons are kind to my ears.
Oh, and I am not very computer savvy, so what may seem simple to you probably isn't to me! And LT Spice is worse!
But my problem with XSim is that I don't know how to load driver parameters. I have simmed the existing crossover with the generic values as set in XSim, but can't see a way to load in the actual parameters as given by SEAS for the existing drivers. So obviously I can't make changes.
The reason for trying the ribbon is that I have a hearing peak around 6-8kHz, and have always liked the way ribbons are kind to my ears.
Oh, and I am not very computer savvy, so what may seem simple to you probably isn't to me! And LT Spice is worse!
Xsim accepts any frd and zma files but you have to manipulate these with another program. Bagby's spreadsheets is what I use for simplicity. Response Modeler adds baffle difraction and box response (so called hard splice method of TSP) to an infinite baffle response of the driver. Vital is to supply the realistic data, otherwise it's no good.
Last edited:
Hi Awkward!
Ahhhh, well I see the confusion! You are trying to use XSim as a cabinet simulator, but it's not. XSim doesn't know what to do with Thiele/Small parameters at all.
To use XSim properly you will need to measure the frequency response, impedance, and acoustic offset.
This is done with OmniMic/DATS or Room EQ Wizard with the right jig, or similar tooling.
Best,
E
Ahhhh, well I see the confusion!
You are trying to use XSim as a cabinet simulator, but it's not. XSim doesn't know what to do with Thiele/Small parameters at all. To use XSim properly you will need to measure the frequency response, impedance, and acoustic offset.
This is done with OmniMic/DATS or Room EQ Wizard with the right jig, or similar tooling.
Best,
E
Thanks Erik.
Actually I thought I was simulating the crossover, which is what I need. The cabinet has only a minimal effect on the MF/HF crossover point, AFAIK.
However, if XSim won't do the job, what will? Once I am in the ballpark I can measure the response; I have a Behringer ECM8000 mic and EQ Wizard.
This is what I managed, but it doesn't help much!
Actually I thought I was simulating the crossover, which is what I need. The cabinet has only a minimal effect on the MF/HF crossover point, AFAIK.
However, if XSim won't do the job, what will? Once I am in the ballpark I can measure the response; I have a Behringer ECM8000 mic and EQ Wizard.
This is what I managed, but it doesn't help much!
Last edited:
Beatiful, you have everything, I mean rew and mic. Measure each driver impedance in a box (no XO) , and then raw spl response in a box at the same spot. There is a white paper by Bagby on how to measure in a room a quasi anechoic response down to 10 Hz. There is also another paper on how to determine acoustic offset of the units which you will need in XSim. Slowly, we might pull this one off.
Thank you Lozjek, but that is not what I need. It's the crossover details to change from a SEAS dome tweeter to a ribbon.
And I won't be taking my speakers apart to measure the drivers' responses!
What I need is the theoretical crossover values, based on driver impedance and sensitivity. Which are values I am unable to input into XSim.
And I won't be taking my speakers apart to measure the drivers' responses!
An externally hosted image should be here but it was not working when we last tested it.
What I need is the theoretical crossover values, based on driver impedance and sensitivity. Which are values I am unable to input into XSim.
In order to create a text file ( zma or frd) one ought to use first a tracing software ( FP Graph tracer, SPL copy...etc) and find a graph that represents what you seek. Fountek cd 3.0 offers only FR graphs, on and off-axis, no impedance plots, as far as I know. You can spot at Zaph Audio website the measurements of Fountek cd 2.0 and on Audio Excite website Fountek cd 3.5H, a ribbon in a horn. Those should track the impedance of your unit more or less well. After that you'd only want the minimum phase calculated by a program (Response Modeler does it) and you're done, beside guesstimating acoustic offset between mid and tw (maybe 10-20mm).
There is my problem; too much software and programmes to get to grips with. If I could do that I wouldn't be asking for help.
I guess it might be worth asking Troels for some advice, but generally he states that alternatives are not what he wants to get involved with.
Oh, and Fountek say the impedance is a flat 7 ohms, more or less. But it's the other drivers I need to model better.
I guess it might be worth asking Troels for some advice, but generally he states that alternatives are not what he wants to get involved with.
Oh, and Fountek say the impedance is a flat 7 ohms, more or less. But it's the other drivers I need to model better.
Thanks Lojzek. I would like to be able to do it myself, but your replies have made me realise it isn't as straightforward as I had hoped. And I am not too proud to ask someone cleverer than I am to help. After all, without huge assistance I wouldn't have been able to build my type 26 DHT preamp!
The current crossover is here, 2.8kHz MF to HF. Jenzen-NEXT
Richard.
After all, without huge assistance I wouldn't have been able to build my type 26 DHT preamp!The current crossover is here, 2.8kHz MF to HF. Jenzen-NEXT
Richard.
Have you tried reducing the tweeter attenuation resistor (R1011) from 2.7 ohms to something lower, like 2.0, 1.5 or 1.0 ohms to boost the treble? That may be all you need to do. The current tweeter crossover is at 3 KHz, LR2 which most ribbons can't handle. So changing the tweeter to a ribbon will most likely require re-doing both the mid and tweeter crossovers. In addition, the acoustic offset offset of the ribbon will be a lot different from the dome too. Try the resistor change before you dive off into the deep end.
I've been down to 2.0 ohms, currently at 2.2. It's not the amount of treble, but the lack of clarity and the sibilance. Actually I have just changed the tweeter capacitors back to the original Janzen Superior Z-Silvers, and it is better than the Russian caps I was trying, but it's simply back to how it was in the beginning. I used to run an active system with an analogue crossover and Jordan drivers which didn't exhibit these characteristics, but I can't go back to that.
Troels does use the NeoCD3 with a 4th order, but I don't know what the crossover frequency is; he does say not to go lower than 2.5kHz though. So maybe it's possible to use it down to 2.8kHz in my speakers. Monitor W15NeoSEAS W15CY001 +
Troels does use the NeoCD3 with a 4th order, but I don't know what the crossover frequency is; he does say not to go lower than 2.5kHz though. So maybe it's possible to use it down to 2.8kHz in my speakers. Monitor W15NeoSEAS W15CY001 +
The midbass FR was borrowed from Troels measurement ( first graph under T25 measurements) and the impedance simulated as a sealed box. You wouldn't simulate woofer and mid so it's insignificant.
For the Fountek I used 15 deg horizontal off-axis manufacturer graph and impedance measurement of Neo 3.5H made by Göran of Audio Excite. Baffle diffraction for tweeter was not included for it made a barely visible difference at 15 deg horizontally.
Hopefully I didn't mess up something important.
For the Fountek I used 15 deg horizontal off-axis manufacturer graph and impedance measurement of Neo 3.5H made by Göran of Audio Excite. Baffle diffraction for tweeter was not included for it made a barely visible difference at 15 deg horizontally.
Hopefully I didn't mess up something important.
Thanks for that. Now I have to figure out how to get them into the simulation.
And I emailed Troels. He can't help with the Fountek, but suggested I change the notch filter on the Seas T25CF001. That won't cost me anything so I'll play with that .
And I emailed Troels. He can't help with the Fountek, but suggested I change the notch filter on the Seas T25CF001. That won't cost me anything so I'll play with that .
An externally hosted image should be here but it was not working when we last tested it.
to this An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.