Looks good Iron SF. One can se the characteristics from the pdf.
https://www.morelhifi.com/wp-content/uploads/2019/12/TSCM634.pdf
Maybe this midrange is very picky on angles or else it running the staircase.
https://www.morelhifi.com/wp-content/uploads/2019/12/TSCM634.pdf
Maybe this midrange is very picky on angles or else it running the staircase.
No, better closer than that. You don't need to get very low frequencies, not near the crossover, so you can gate. This way you only see the varying effect of the cross and don't look at the room.
You can do this other ways, I've simply noticed that you are using measurements so this way will give you the answer you need.
You can do this other ways, I've simply noticed that you are using measurements so this way will give you the answer you need.
Well, the nearfield measurement compared with the measurement on 1m still don’t rule out a weird kind of baffle step function, combined with a too fast decay above 2k of this specific unit.
I’d measure the driver on a big baffle, preferably IEC dimensions, so one should read data sheet values. If that’s the case, design another baffle of your speaker, as virtually every critical distance is in the range of the wavelength associated with 2k. But I suspect the driver not to be fully OK.
I’d measure the driver on a big baffle, preferably IEC dimensions, so one should read data sheet values. If that’s the case, design another baffle of your speaker, as virtually every critical distance is in the range of the wavelength associated with 2k. But I suspect the driver not to be fully OK.
As long as the mid alone has this almost step function at 2k, trying to design a crossover is pointless. Just my 2ct.
Btw Just noticed the data sheet shows a little bump in the impedance curve. Where? At 2k! Did you measure your impedance curves?
Last edited:
Attached are the impedance measurements:
Attachments
Last edited:
IEC should be 1m x 1m. I'd try that driver on anything you have handy that isn't the same size of your finished baffle, just to get a handle on what this is. The impedance wrinkle seems too small to have that big of a result in frequency response. Almost seems like the driver has a mechanical issue like non-linear suspension compliance.
Last edited:
It is interesting to look at the breakup modes of this woofer. It is not completely clear which of these is setting the reasonable upper limit to the usability of the driver, are we assuming at this point that it can be used up until general directivity runs aground?
That assumption suggests that cone resonances are to be equalised.
In addition, it may be cone breakup that is causing the impedance blips, do you think?
It could be cone breakup (the 2k dip). But once that occurs, irregularities in the response rule. This measurement shows a dip, followed by a rather clean response higher up the frequency range. Strange indeed. The fact that the dip persists off-axis would indicate it is not diffraction-induced cancelling.
@ OP, how is the mid enclosure constructed? Which internal dimensions, what damping (acoustic stuffing) is applied? Are you able to produce CSD or BD plots? Were the impedance measurements done in free air or in the enclosure?
@ OP, how is the mid enclosure constructed? Which internal dimensions, what damping (acoustic stuffing) is applied? Are you able to produce CSD or BD plots? Were the impedance measurements done in free air or in the enclosure?
I'm as yet uncertain that breakup can be summarily dismissed based on the current information, but willing to be shown otherwise.
If the extant problem is only response irregularities, then can't we equalise it?
Ran, try put the naked driver on the bench and run a measurement, if the dip still exists then it's most probably related to the driver itself and other variables can be ruled out.
Well, I don't rule it out either. But this graph keeps me puzzled. Why would a grille (cloth on a wooden frame) be maleficial to a cone breakup? Waiting for the CSD/BD-plots!