Weird after crossover frequency estimation

[Mario Pankov]

Hello,

Farfield measurement, 1m distance from woofer, used JustMLS. Looks fine when into the optimiser ( upper graph ) but once a simple 2nd order electrical crossover is attached, there`s a weird spike at 4.5Khz. Measured with the crossover, the woofer exibits no such spike. What is the reason LSPCad to interpret the posterior response like it is shown? It makes a simple crossover construction a nightmare and can`t get rid of it, what am I doing wrong?
Also, once all steps have been taken ( following the manual ) the Start button for the automatic optimisator does not do anything. Any ideas why?

An externally hosted image should be here but it was not working when we last tested it.

Uploaded with ImageShack.us

 

[sreten]

Hi, I'd guess a silly number entered as the driver's inductance, rgds, sreten.

 

[Mario Pankov]

Hi Sreten,

This is done through LSPCad 6.2, no option to enter T/S there, at least I couldn`t find one As far as I can tell, it only required frequency response/phase + impedance plot. No issues with tweeter simulation.

 

[DrBoar]

Is there any kind of smoothing in the graphs? If so there could be a one of numercal value in response or impedance plot that is an order of magnitude larger or smaller than all other. Such an outlier could throw of the calculations without showing up if there is smoothing.

So if you look at the textfile of numerical data how does it look around 4800 Hz?

 

[Mario Pankov]

Went back to using Holm impulse and completed ( almost ) the project already I looked at this and I think it is some crossover high Q peak combined with the frequency bump of the driver which does not happen in reality ( when measured ). No idea why simulated this way, have checked the values multiple times one by one.