D’Apollito’s recent work on the THOR TL speaker describes his preference for even order crossovers, even on MTM. For the THOR, he used 4th order electro-acoustic crossovers and I think that might be something I will consider.
THOR measured XO response:
I usually aim for 4th order BW on the woofer for the Harsch and 2nd order for the tweeter. A shallow waveguide on the tweeter might be needed and will give a true Harsch XO near 3.5kHz. However, a proper waveguide for a RAAL requires removal of the front faceplate to access the ribbon exit plane to avoid the resonance dip caused by placing the waveguide at the faceplate. The faceplate, however, is an integral part of the mechanical design of a RAAL in that it serves to clamp the ribbon tension and removing it requires the waveguide to serve as the new clamp, plus the ribbon tension adjustment must be redone. Rather than mess with that, I will try what I did earlier in this thread by accepting the lack of physical delay and flipping the polarity if the RAAL to achieve a pseudo (inverted tweeter) Harsch XO. Compromises for the sake of practical speaker building. I don’t want a tweeter that is dependent on how well the waveguide is made or clamped to it. The tweeter should be removable as a unit. We will see what the polars end up looking like, but in general, an MTM config will have a symmetric vertical polar response.
THOR measured XO response:
I usually aim for 4th order BW on the woofer for the Harsch and 2nd order for the tweeter. A shallow waveguide on the tweeter might be needed and will give a true Harsch XO near 3.5kHz. However, a proper waveguide for a RAAL requires removal of the front faceplate to access the ribbon exit plane to avoid the resonance dip caused by placing the waveguide at the faceplate. The faceplate, however, is an integral part of the mechanical design of a RAAL in that it serves to clamp the ribbon tension and removing it requires the waveguide to serve as the new clamp, plus the ribbon tension adjustment must be redone. Rather than mess with that, I will try what I did earlier in this thread by accepting the lack of physical delay and flipping the polarity if the RAAL to achieve a pseudo (inverted tweeter) Harsch XO. Compromises for the sake of practical speaker building. I don’t want a tweeter that is dependent on how well the waveguide is made or clamped to it. The tweeter should be removable as a unit. We will see what the polars end up looking like, but in general, an MTM config will have a symmetric vertical polar response.
Last edited:
No worries. There is no such thing as a dumb question in DIYA forums - we are all here to learn and I learned something by today because you prompted me into digging into the recent work that D’Apollito did on the THOR. A fine looking MTM TL design. One could in fact, use a more traditional dome tweeter flush mounted and use a similar 4th order XO. The Purifi’s are so predictably flat and wide band with almost no resonance breakup peaks that almost any XO is easier to implement than a driver with quirks.
I got a request to see how this TL would work with dual PTT6.5-4 (4 ohm drivers). The person requesting this says that they have a 2ohm capable amp.
It still works quite well with no changes to the TL design. The sensitivity will be about 89.5dB at 2.83v (nominal 3.4ohms and minimum of 2ohms at 200Hz) with -3dB at 35Hz and -6dB at 29Hz. Max SPL of 105dB (1m) at Xmax of 10mm and 18Vrms.
Here is predicted freq response at 2.83v:
Here is predicted impedance:
It still works quite well with no changes to the TL design. The sensitivity will be about 89.5dB at 2.83v (nominal 3.4ohms and minimum of 2ohms at 200Hz) with -3dB at 35Hz and -6dB at 29Hz. Max SPL of 105dB (1m) at Xmax of 10mm and 18Vrms.
Here is predicted freq response at 2.83v:
Here is predicted impedance:
Attachments
Last edited:
There was a thread on his TL design. Was claimed that he goofed on the enclosure size so a redesign thread was started. I built one for a friend who originally purchased the Madisound TL kit. The new enclosure sounded much better.
I just redid the simulation (same cabinet dimensions and plans) with the PTT6.5X08-NFA-01 Driver. It apparently is tuned a bit lower and got some really good results. The sensitivity is still 86dB but F3 is now 33Hz and F6 is 24Hz.
I would say that this is the driver to go for.
Predicted Freq Response for 2.83v at 1m:
I would say that this is the driver to go for.
Predicted Freq Response for 2.83v at 1m:
Attachments
Purifi are not making it easy to know which is what. Apparently they have juggled the naming convention and if you don't have an X series, you may still have and X series if you have a W series. They have W drivers listed AKA X, 
But I'm glad I have the W series after your findings about the X series-
But I'm glad I have the W series after your findings about the X series-
