An externally hosted image should be here but it was not working when we last tested it.
My son bought this GAS Competition 15D1 and wanted to build a TPQWR for it.
It is a nice driver. Built like a tank and weights like a tank.
With a low Vas of 50 liters, a Fs of 30,9 and a Qts of 0,21 - it should work great in a TPQWR.
An externally hosted image should be here but it was not working when we last tested it.
And it sure works great in a TPQWR.
Here is a simple indoor measurement taken in our living room compared with the Hornresp simulation (thanks Josh Ricci for the tip!).
The driver is brand new. It needs a kilowatt or two for ten hours before it is broken in. The suspensions is really hard and stiff.
And the bass!!! OMG!!!
I am overwhelmed by the sheer power and brutality. I have never heard anything like this before. I will write more tomorrow. I hope we get some nice weather soon. I want to take it outdoors for some measurements, but it might take a while considering the prevailing weather here in Sweden this time of the year.
Regards,
Johannes
An externally hosted image should be here but it was not working when we last tested it.
I woke early this morning and had some uninterrupted quality time with Hornresp and REW.
Here is the actual in room measurement compared with a simple "dumb" Hornresp sim with the exact measurements from the physical dimensions of the TPQWR, and with a optimized Hornresp sim that clearly differs a lot from the physical dimensions from inside the box - but matches the measured response much better.
If I measure the front resonator it is 620 mm deep. If I try to get the Hornresp model to match the real world measurements the front resonator suddenly becomes 450 mm deep.
I cant find 450 mm anywhere.... The bottom of the resonator has a lot more area then the driver and the tapped pipe exit.
???
The engineers at GAS did not spare any Le when they designed the COMP15D1...
It has a plentiful Le of 7,5 mH according to my findings comparing the simulation, the driver model and the measured response. Since it is not an actual measured value it should not be taken as an absolute in any way.
It has a plentiful Le of 7,5 mH according to my findings comparing the simulation, the driver model and the measured response. Since it is not an actual measured value it should not be taken as an absolute in any way.
An externally hosted image should be here but it was not working when we last tested it.
Simulated GD (green) and measured GD (red). Remember that this is an indoor measurement of a raw and brand new driver.
An externally hosted image should be here but it was not working when we last tested it.
Phase.
The measured phase response (red) is much better then the simulated phase response (green) with no sudden phase shifts in the middle of the passband.
Looking at the GD, Phase and spl I would say that this TPQWR is useful between 30 and 120 Hz, using a low pass of 48 dB/octave at 100 Hz.
It has an extremely physical and tactile character. Once I hear the bass above the fan noise from the amp, the bass permeates the whole house in a way that makes it totally useless in a normal family home where not everyone wants to hear bass at the same time.
Perfect in a brutal home theater setup where the physical and tactile character really adds a physical dimension to the home theater experience.
Perfect for live audio where once again the physical and tactile character adds a new level of realism to whatever is being amplified.
If you saved those traces, then you already have the distortion results. Just click the REW button to see them.
What was the distance of your mic from the speaker?
B.
What's the distance from the mouth to the entrance of the other section of the horn?
Despite the physical distance being 62cm, acoustically it is acting like it is 45cm deep. Green simulation is MUCH closer to the performance / behavior of the system. Why is it acting shorter is the question
. I do not have a lot of information about this cab but I would bet that assumptions made about the path length and "flow" out of the final section are off a bit. HR is a powerful tool, but it is a simplified approximation of the acoustic properties of what may be a complicated system. There are a limited number of inputs to approximate the system so they need to be the best fit in a broad sense. Sometimes the parameters picked for the initial simulation are not necessarily the best and will need some tweaking to match the system measurements. Splitting the difference may be required for some parameters. Sometimes the required adjusting of the parameters will make sense and other times they may not at first. There is a lot to be learned doing this type of comparison.
I measured it today. It is 495 mm from the edge of the panel (making up the side of the resonator and at the same time the end-section of the tapped pipe leading to the tapped pipe exit into the front resonator) to the mouth.
25 cm. I wanted to minimize the acoustic contributions from the room.
An externally hosted image should be here but it was not working when we last tested it.
The microphone was 30 cm from the mouth in this measurement. This is still indoors. It has been windy and very cold outside today. No outdoor measurements yet.
It is quite clear that this driver is a pure spl competition driver. No shorting rings on the pole piece. I don't think the Tamp E800 likes the 2 ohm voice coil resistance, so this fact might effect the THD measurement a little.
I would like to test this driver with a negative output impedance amp, but first we are going to demolish the BRL-store in Gothenburg (where we bought this driver) with this acoustic weapon of mass destruction.
Last edited:
Ok, so we're getting a little closer. What's the distance between the mouth and the top plate of the driver's motor structure?
Yes. This is a very humbling experience. I have been using Hornresp for many years, and I usually view myself as quite proficient in designing horns and other types of enclosures for bass reproduction. I am used to a quite good correlation between my simulations and the real world results given the empirical knowledge and experience I have gained throughout the last 30 years of designing and building my different kind of bass loudspeakers.
This TPQWR (ROAR) has really thrown me a curve ball.
I hope I can measure this beast at much higher power levels soon. At 20 watts indoors our guinea-pigs squeal in sheer terror, and my wife scream "shut it down now!!!" with a very angry voice. It is totally useless in a normal domestic environment, as everything starts to rattle and buzz at very low spl.
Regards,
Johannes
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
After some more quality time with Hornresp and REW I got this.
Quite good correlation. I am happy I got this far.
The microphone is exactly at the mouth. This does not correlate with the one meter that Hornresp uses. If I move the microphone one meter back the room acoustics wreck havoc with the frequency response.
Cheers,
Johannes
Last edited:
So a number of things here: (1) the "resonator" does not really extend the full depth of the box, as the bottom section is not closed off - it leads to the rest of the box. The maximum depth of the "resonator" is actually going to be 495mm or less, based on previous information you provided. And secondly that driver is a bit large to consider equivalent to a "flat piston" that Hornresp assumes for its sims, particularly considering the way that it is physically mounted in the enclosure, with one side in line with the path and the other orthogonal to it. And then there's the case of the rapid expansion at the last bend in the cabinet. The path-mapping methods (centerline, advanced centerline) are likely to be less accurate in those circumstances.
Josh has suggested the best approach - try to make the sim fit what you've built. It won't be the same, but it should come close. I suggest comparing both FR and impedance curves in the process.
I believe it is the closest to the way it is placed in my living room.
It does not stand in a corner and it is not places on a large open free field of some kind.
What radiation angle do you propose I use in my sim?
This should depend on frequency and the phase relationship between the tapped pipe exit and the driver diaphragm.
At some frequencies the tapped pipe exit and the driver diaphragm should work like two driver spaced 40 cm apart and within a small range of phase to each other. Here the front resonator should be almost 62 cm deep. Where there is 90 degree phase or more between the exit and the driver, the resonator should behave like it is 49,5 cm deep. Not 43 cm.
The tappad pipe opening in the bottom of the resonator is 600 cm2 in cross section.
600/3000 = 20% of the resonator.
I will try to close half of the tapped pipe exit into the front resonator and see if there is any difference.
I don't fully believe in the "open bottom theory", even though it could partly explain some of what is going on in the TPQWR-monster.
Cheers,
Johannes
As a comparison - you would not say that the port in a BR box shorts out the acoustic output from th front of the driver at the box-port resonant frequency. And you would not claim that the first port in a series tuned 6th order BP shorts out the driver output in the second chamber at its resonant frequency. It is all phase dependent.
It is basically a series tuned 6th order BP with quarter wave resonators instead of Helmholtz resonators. But without the small radiating surfaces of a port, you avoid port compression and get a better coupling of the acoustic energy (much higher efficiency) through a much larger radiating surface and lower air particle velocity.
Regards,
Johannes
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.