Even if the non-linearities would try to make the excursion go beyond its prediction in reality at Xmax (+/- 70% BL) a PA driver already suffers from 2dB-3dB power compression. Adding another few dB's in power and it will rise to 4dB-6dB. That means the excursion will be kept under control. But temperature in the voicecoil can start to rise rapidly from this point and that is something you should worry about.
Last edited:
Jim; Did you have any weighing in your tests? I did mine flat and will be retesting soon. I have A, B, and flat. Witch should I use? Andy
Andy,
Jim used flat, as you should.
You can see the difference in weighting in the chart in post 886, using "A" or "B" weighting would make your cabinet look like it has very little bass.
If you check excursion while you are testing at 28.3 volts, you will find out if your cabinet's low corner is actually lower than the standard SS15.
As I mentioned before, a lower corner will be accompanied by less output in the same volume box.
Art
Last edited:
I don't have examples of excursion being higher than predicted, in fact suspension rigidity will make for less excursion below Fb than predicted.
That is why I suggest measurement of the actual speaker, rather than relying on a sim for proper setting of the HP filter.
VAS is a measure of suspension rigidity.
Reduced excursion would be due to a non linear suspension that becomes less compliant with travel, i.e. more rigid.
Revised Low reading of the pyro sub box
My test tone CD was messed up. I converted the MP3s to WAVs that could be read by the player on a USB stick. I used the smallest amp to set the 2.83VRMS.
and stood the meter up so I could read it better from the side.
Not the best day for testing but I knew the box would go lower and boy did it.
97db at 20Hz
102db at 25Hz
105.5 at 30Hz
106.5 at 35Hz
105.5 at 40Hz
106 at 45Hz
106 at 50Hz
Thats all I did for now. I will try and rerun the full range agin soon.
Andy
My test tone CD was messed up. I converted the MP3s to WAVs that could be read by the player on a USB stick. I used the smallest amp to set the 2.83VRMS.
and stood the meter up so I could read it better from the side.
Not the best day for testing but I knew the box would go lower and boy did it.
97db at 20Hz
102db at 25Hz
105.5 at 30Hz
106.5 at 35Hz
105.5 at 40Hz
106 at 45Hz
106 at 50Hz
Thats all I did for now. I will try and rerun the full range agin soon.
Andy
As most suspensions are, fortunate for those who don't employ proper HP filters
.Okay Art, for a minute I thought you suggested higher excursions did occur.
As far predictions and low cut filters, as long the 1/3 wavelength of the design (HornResp prediction) aligns perfectly to real life SPL measurements like Jbell’s there is no problem to use calculated low cut settings. In other words, Oliver has proven already his latest HornResp from March aligns perfectly to Jbell’s SS15. Therefore it’s absolutely safe to use predictions as guideline for setting out the Low-Cut filter. So it’s not necessary to measure the excursion physically for that.
And for those who ;-) still believe excursions can be higher then predicted:
Power compression: Since most PA drivers have less then 5% efficiency almost all of the rest of the input power is converted into heat. The more input power the more this heat will rise and therefore the resistance of the voice coil.
Dynamic compression: When a PA speaker cone reaches the limit of its excursion the mechanical 'elastic' resistance becomes higher. In the latest generation low frequency PA drivers this dynamic compression is minimised by the extension of the (linear) physical excursion and therefore its Xlim.
That looks really good there Andy, can't wait to see your 28,3V measurements...
Andy,
Whether wind made for the higher readings, or you managed to set levels incorrectly, or the file conversion increased the LF content, or you changed the mic or cabinet position your current readings would defy the laws of physics for a half space outdoor measurement.
I think you will find your previous readings, which correlate well with Jim's, are correct.
Not to play the Hoffman's Iron Law record out, but when a lower frequency response is obtained from the same size box, it will be at a reduction of output level, not an increased level.
Cheers,
Art
Attachments
Horns also have significant radiation resistance - and the portion of input power that goes into this is converted to *sound*. Look at the impedance rise (above the AC free air resistance), and this is an indication of how much is converted to sound and not heat. Power compression still sets in, it just takes more power to do it.
Which is why most woofers (and instrument speakers) survive at all. The harder you it it, the more it takes to move it. But when this happens with a horn, you very quickly lose some of the radiation resistance and the temps skyrocket at these excursion levels. There's a sweet spot where power compression is low, efficiency is high, and you're not losing much to suspension nonlinearity. Below that, horns rock. Above that, things fall apart quickly.
Sorry Wg_ski if I made it sound for drivers in general cause it was meant for power compression levels in TH's.
The point here is not where efficiency is the highest and excursion the lowest in a Tapped Horn. Here it’s about the point where excursion reaches its maximum. At this point power compression levels are similar to basreflex. You can also read that from the differences between Jbell's 2,83V and 28,3V measurements.
Most low frequency PA drivers have efficiency between 2 and 3%. At the point in TH’s where excursion is max (around 1/3 wavelength) you have an efficiency gain around 3dB. This means you end with 5%-6% efficiency.
I hope this cleared things up...
Last edited:
Hm... I think it would have to be a LOT bigger for the output to change significantly at low frequencies..
FR measurements can be a b*tch to get right. That's why I prefer to look at the impedance response and compare it to the HornResp predictions. Impedance response measurements are a lot easier to get right, and if they are a very close match to what HornResp predicts, it's likely that HornResp's FR predictions would be a close match as well.
Hi Y'all,
I went through the drawings, and Hornresp simulations for the 4pyros modified SS15, and the original SS15 in quite some detail earlier in this thread. And I tried repeatedly to maximize the Hornresp output for both enclosures. The results in Post #907 do not seem possible, as Brian Steele says in Post #918: "...it would have to be a LOT bigger...", and I for one would put more stock into the Hornresp results.
Regards,
I went through the drawings, and Hornresp simulations for the 4pyros modified SS15, and the original SS15 in quite some detail earlier in this thread. And I tried repeatedly to maximize the Hornresp output for both enclosures. The results in Post #907 do not seem possible, as Brian Steele says in Post #918: "...it would have to be a LOT bigger...", and I for one would put more stock into the Hornresp results.
Regards,
