Everyone who have heard one of these designs comment on the firm, punchy and clean bass. The ROAR12 seems to sound somewhat similar to a large front loaded horn.
I will never build another tapped horn or normal size-constrained front loaded horn again.
My kids and their friends laugh scornfully at bass reflex boxes, band pass and closed boxes after comparing the same driver in several different enclosures.
There is an avalanche of TPQWR builds in Sweden right now. Both for home-theater, car audio competition and for professional use.
Regards,
Johannes
I will never build another tapped horn or normal size-constrained front loaded horn again.
My kids and their friends laugh scornfully at bass reflex boxes, band pass and closed boxes after comparing the same driver in several different enclosures.
There is an avalanche of TPQWR builds in Sweden right now. Both for home-theater, car audio competition and for professional use.
Regards,
Johannes
Do you have any idea why the road doesn’t perform as the sim in the lower region? Do you reckon adding 45 degree plates in every 90 degree corner would help avoid sudden rises in horn-path-size would increase efficiency in the lower region?
http://www.diyaudio.com/forums/newreply.php?do=newreply&p=5294733
I’ll add handles, wheels and inputs on the sketchup file some time
It seems that this type of cabinet decreases the movement of the woofer, meaning it can handle more wattage than normally before it hits xmax. I'm pretty sure that the placement of the driver gives plenty of cooling as well. So basically you can drive more watts into the cab!
And according to the simulated cab (If we can get the same results in the low-frequency-region), you get MUCH lower response. Comparing the Roar18 to the 1850, you'll have 100+ db response all the way to 37hz, where as the 1850 goes to 47hz (100+ db).
The 1850 does have a flatter response, at 105dB, where the ROAR18 varies between 102.5 and 105.5. So the 1850 might win sensitivity by a small margin.
Lastly it seems that BesPav was able to use just one cab to satisfy ALOT of people. So it's capable of being used as singles. Front Loaded horns like the 1850 are usually best in packs. Preferably 4 or more.
Not really. I believe there is some strange coupling between the large front resonator via the driver and into the tapped pipe. In every ROAR or TPQWR I have seen measurements from the measured response is much smoother and has a much lower Q then the simulation predicts. I guess this also explains the discrepancy in cone excursion between measurements and simulation. My ROAR12 behaves like the tapped pipe section was at least 50% smaller in cross section then it actually is.
Today my son bought some particleboard to build a small prototype TPQWR for his GAS Comp 15D1.
An externally hosted image should be here but it was not working when we last tested it.
Not really. I believe there is some strange coupling between the large front resonator via the driver and into the tapped pipe. In every ROAR or TPQWR I have seen measurements from the measured response is much smoother and has a much lower Q then the simulation predicts. I guess this also explains the discrepancy in cone excursion between measurements and simulation. My ROAR12 behaves like the tapped pipe section was at least 50% smaller in cross section then it actually is.
Today my son bought some particleboard to build a small prototype TPQWR for his GAS Comp 15D1.
An externally hosted image should be here but it was not working when we last tested it.
Keep us updated on the process! You must be proud of your son!
Good to have someone as enthusiastic around I would think!I looked at my old comments and found that the RCF LF18G401 delivers a far more flat response. Only downside I see about the driver is the lower xmax, but from what I've seen in your videos, xmax shouldn't really be a concern with this cab. So I like to know if you could point out any other weakness with this driver? I am no shark with speaker parameters, and know very little about how they influence the performance. And perhaps you could say a bit more about the driver in this cab, since you've been playing around with this type of cab for some time!
An externally hosted image should be here but it was not working when we last tested it.
The reason why I'd like a flatter response is from what BesPav have said. That the cab didn't sound as well without eq. I know some have already pointed out that one shouldn't care so much about a couple of dBs, but in my opinion, a flat response is a very good start with any cab, and I'll care about it anyway
I would really love to help develop this cab. I feel like it has so much potential, but don't really know how to join in. And also. I'm focusing on the Roar18, and the thing is just so expensive to make multiple of (If only for testing and evaluating) How do you do it?
Also - If we just go for best possible sound quality and frequency response (Especially in the lower frequencies. Do you reckon any of these 45 degree panels would help? The ones I am particularly interested in are the ones that are next to the baffle at the far back of the cab, and the other is when the horns join up.
Also - If we just go for best possible sound quality and frequency response (Especially in the lower frequencies. Do you reckon any of these 45 degree panels would help? The ones I am particularly interested in are the ones that are next to the baffle at the far back of the cab, and the other is when the horns join up.
An externally hosted image should be here but it was not working when we last tested it.
No, I don't think it would help. My experience points to the opposite. You don't want to loose volume in the bends.
Build something and measure and listen. I am tempted to build a very small TPQWR with one of my small cute SB-Acoustics 4 inch SB12PAC25-4 drivers 4" SB12PAC25-4 :: SB Acoustics
It is a really cheap but nice little driver with nice specs and a stiff aluminum cone.
Suitable for a small TPQWR for a computer sound system or kitchen stereo.
I have been simulating, building, testing and fooling around with this type of design for two years now, but it actually just gets me more confused and dumbfounded for each build. It does not really behave like I expected it to. I am pleasantly surprised by better then expected performance.
Regards,
Johannes
I like simplicity of that design. I see laying the enclosure in it's current position for car audio and standing it up for home or pro audio. I like the mounting access too.
So, the ROAR12 in-car peaks just below 60 Hz, is about 5dB down at 50 Hz and about 16 or 17dB down at 40 Hz?
What's the lowest resonant frequency of the ROAR12?
Looking at the most recent measurement - the blue line - I think it is much more correct to say that my car has a cabin gain with a 20 dB peak centered at 60 Hz and a lot of leaky losses below that frequency. I am having a hard time breaking in this driver. Every time I play music with more then 10 watts through my ROAR12, the rest of my family and my neighbors complain. The blue line is with a few hours of 10 watts or so and maybe two hours of more then 100 watts. The driver is still way to stiff and new.
But the 12 dB drop between 40 Hz and 80 Hz is part of what baffles me with this design. The simulation does not show this at all.
The ROAR12 behaves like the tapped pipe is much longer than it physically is, but only in the lower part of the passband. It seems like the front resonator couples much more aggressively to the tapped pipe then the simulation predicts. This would also explain the large discrepancy between the simulated cone excursion and the measured cone excursion.
Regards,
Johannes
Thanks!!
It is much easier to build then the "normal" split path ROAR.
We hope the Panzar-GAS Comp15D1 can take the load without problems. Even at very large cone excursions and lots of power. It is build like a tank.
I prefer the split path ROAR design for normal pro audio drivers. The symmetrical load on the cone easies my mind at high power levels.
I have been wanting to do this for a long time, but between my ME/CFS, all the day to day responsibilities of being a father and a husband and having two kids who constantly wants help with all their wild ideas, I just can't seem to find energy or time to do all these things I want to. I have the energy and stamina of an 80 year old, so I have to prioritize hard. This is why there are long periods of time where I am not very active here and with my speaker projects.
Regards,
Johannes
YouTube
YouTube
Two short crappy videos of us fooling around with the GAS Alpha 12D2.
Impressive spl and midbass kick from such a cheap car audio driver in a unbraced TPQWR in cheap 22 particle board. It is the cheapest possible prototype build, and the box is not made for this driver. I hope this new TPQWR for the GAS Comp15D1 will have a much better performance and sound much better.
YouTube
Two short crappy videos of us fooling around with the GAS Alpha 12D2.
Impressive spl and midbass kick from such a cheap car audio driver in a unbraced TPQWR in cheap 22 particle board. It is the cheapest possible prototype build, and the box is not made for this driver. I hope this new TPQWR for the GAS Comp15D1 will have a much better performance and sound much better.
Have you taken an impedance curve of the build? That's probably the most accurate way to confirm whether or not the sim matches the build, or vice-versa. It's normally the first test I do, not only to confirm the build matches the sim, but also to check for leaks or excessive panel flex.
The frequency response graphs, well at least the green and purple ones, suggest an inflexion point around 42~43 Hz, which seems to match the sim you published in the Hornresp thread. The peak around 100 Hz is a bit off, which suggests that the sim may need some adjustment, or it could be that the driver's t/s parameters are not as published. Overall, the FR in the passband is a bit different, but reasonably close (going by the purple graph), if you consider that Hornresp does not include the impact of box losses in the sim, and the higher the "Q" of the resonances at low frequencies, the greater the difference between what Hornresp predicts and what's actually measured. And then there's the question of the driver's t/s parameters perhaps being not as published, which would impact the FR as well.
The in-car response is curious. That must be a very lossy cabin if the response from 50 Hz down is decreasing *faster* than it does in free air, as suggested by the graphs. Are you sure there was no high-pass filtering engaged?
Some interesting things can happen when you put higher-order designs in a car's cabin. When I tried out my POC3 in my car (it's an SUV, so the cabin includes the trunk), the impedance curve showed the middle impedance peak was almost completely damped away, suggesting significantly reduced excursion because the cabin was "loading" the TH at that frequency. The end result was that I could actually feed it a bit more power in-car before it started hitting its excursion limits (the flip side of that being I had to make sure and turn the volume down before anyone opened the trunk!).
What's the net box volume and overall box volume of the ROAR12?
An externally hosted image should be here but it was not working when we last tested it.
Yesterday we gave my 12PS100 a 2 hour workout with a 15 Hz tone at xmax.
The suspensions feels softer.
I am redoing the driver mounting to assure that there is no leaks.
I hope I can take some impedance measurements soon.
The three curves outdoors, in car and indoors a few posts above this is done without any smoothing.
The big monster-TPQWR is taking all the time and energy I have at the moment. We have rented a small PA with some 1 kw/channel/8ohm amps this Wednesday. I hope we can take some better measurements then.
Cheers,
Johannes
Nice!
I hope you will like the ROAR 18 as much as I like my ROAR12.Please keep us updated on your project!
The RCF LF18G401 seems like a great driver. If you don't push it too hard it will work great.
Since it is an older generation driver with less Mms and more Vas then the current crop of drivers, you should keep the amp below 1200 watts or so, and highpass the ROAR18 at 30 Hz. The driver might take much more power in a ROAR18, but I would prefer some caution.
With a nice low Qts of 0,29, a low Mms and a reasonably high Vas, I guess based on my experience that it will have a wicked midbass punch and attack and a great efficiency. You will probably not need a lot of power before it will be painful to come close to the mouth of the ROAR.
I am trying to get better measurements of my ROAR12. It seems like something strange is going on. I hope it is not due to air leaks or a faulty driver (seems unlikely due to the sheer spl, efficiency and the SQ i get from my ROAR12) but it is not behaving like we expect it to.
I hope I can give you more information about this issue later next week. We are trying to complete the large TPQWR-monster so we can test it to. I need more experience of this strange beast.
Cheers,
Johannes
An externally hosted image should be here but it was not working when we last tested it.
A Hornresp sim of the RCF LF18G401 in a stock ROAR 18.
I am afraid that the sim does not reflect the real performance, so take this sim as a very crude first indication of the passband and the efficiency.
An externally hosted image should be here but it was not working when we last tested it.
No smoothing. The microphone 50 cm from the mouth.
I upgraded the mounting of the driver with 6 mm bolts and T-nuts, sanded the mounting surface smooth etc. I want to avoid any potential leaks and have a firm mounting of the driver to better handle the immense pressures Hornresp predicts at high power levels.
I guess there was no leaks before. The response did not change noticeably. The measurement is taken in a slightly different place in my room due to stuff standing in my way. We are in a complete reorganization of our large attic bedroom, so there is stuff standing all over the place.