I'm afraid I'm not following you, miero. I do in fact specify the use of the report function:
Regardless, are you saying adding this specification is no longer required in the latest version of ATH? Even without the spec, it will generate a default report? I'll give it a try.
Code:
Report = {
Title = "Demo"
NormAngle = 10
Width = 1024
Height = 768
SPL_Range = 50
MaxRadius = 90
PolarData = "SPL"
}Regardless, are you saying adding this specification is no longer required in the latest version of ATH? Even without the spec, it will generate a default report? I'll give it a try.
Hello,
I have dismonted one of my T34B and I can confirm that the T34B has a convex surround, please see picture.
Hope this help
Attachments
Unfortunately there are a few occasions where something else is at play and the report function does not generate all of the text files needed to complete. Without the report section in the config nothing happens at all and you get a different error.
I've moved on to demo7. I added the same report specification to the script. Then I tried to generate the ATH report, but again it gave me the error message referring to the static.txt. I then added the static.txt fluid made to the results folder and I ran it again. It worked, kind of. The results don't look plausible, quite different from VACS.
Your example dies not include one parameter that @miero had given in his solution, MaxAnglePM =
I often received that promt with definition scripts which included multiple sets of reports: for vertical, horizontal, tweeter, woofer etc. pp. In some cases it seemed to me as if ABEC was not properly parsing/selecting the title of a report, but if @miero is right, this would explain the illegible output too.
@miero I had misunderstood your previous post. I tried it today by adding in the MaxAnglePM and MaxRadius with the same problem config and it made no difference. I did notice that I had a static.txt file in some results directories where I had not actually run the report function so they seem to be created by Ath in the first instance. These were much more generic than the previous one and I was easily able to edit the dimensions and distance to correspond to the correct ones. Putting this in the directory manually and running with -r then worked with sensible output just like VACS. Attached in case it helps anyone.
Yes that is the problem, something causes the coords and static files not to be created on first run, only the static causes the basic report not to generate though. I only came across it while solving someone else's old style config file, the same problem had been mentioned but without a clear resolution. Certainly for me it does not matter in normal use.
@keyser -
I used to run into that exact same error. For about six months I couldn't produce a report properly. I was never able to solve the issue.
Then one day, it just went away.
I have three different workstations with ATH / ABEC installed, and my best guess is that when I upgraded one of the dependencies, it fixed the issue. I never made any real changes to the config files that I used to make my waveguides. What DID change was that I've upgraded the dependencies a few times.
So it was either an upgrade to gnuplot or an upgrade to ATH that's fixed it. I've been using the same (old) version of ABEC for years.
The machine I'm typing this on uses ATH 4.9 beta, gmsh 4.8.4, gnuplot 5.4 patchlevel 3
Looks like my gnuplot is pretty old, from 2021
I used to run into that exact same error. For about six months I couldn't produce a report properly. I was never able to solve the issue.
Then one day, it just went away.
I have three different workstations with ATH / ABEC installed, and my best guess is that when I upgraded one of the dependencies, it fixed the issue. I never made any real changes to the config files that I used to make my waveguides. What DID change was that I've upgraded the dependencies a few times.
So it was either an upgrade to gnuplot or an upgrade to ATH that's fixed it. I've been using the same (old) version of ABEC for years.
The machine I'm typing this on uses ATH 4.9 beta, gmsh 4.8.4, gnuplot 5.4 patchlevel 3
Looks like my gnuplot is pretty old, from 2021
Look interesting. For open baffle design?
the SB18SW800 looked also great.
BTW here is an interesting article form the same guy that did compression driver comparison a while ago.
The interesting bit is the TAD 16003 data (as well as 18s 15LW2400 and BMS 15N850v3).
https://www.justdiyit.com/comparatif-15-pouces-pro/3/
It is a bit off-topic for the woofer
but finding good drivers is always a big part of any design...
For those interested I did have a look at the three driver via simple T/S simulations with WINISD.
Note: I am only talking about active system, the 18s look the best to me because of the directivity and the possibility to Xover at 800Hz or lower (if the gating does not smooth out too much of the response) and you can actually buy it.
The differences are
1. The 480Hz trough on the 18s but not a deal breaker and higher distortion above 100Hz whatever that means, as Vance Dickason says “Interpreting the subjective value of THD distortion curves is difficult and opinions vary considerably in the industry. However, looking at the relationship of the second to third harmonic distortion curves is of value”. The directivity is more import though…
2. The 18s seem displacement limited and the TAD is probably Thermal limited -> only in Pro (PA and Monitoring) applications this could be an issue with SPL 115dB+.
3. The 18s could be happily used in a much smaller enclosure
4. The 18s have more Xmax and/less distortion under 100Hz
For the T/S simulations, to compare apples to apples, you can use the factory parameters derived box volume for the TAD which is more of less the recommendation for the EBS 6dB for the 18s…
Hereafter are the “signal” values and “filters” are designed to maximize the bandwidth and the Max SPL, LEAP simulation would be necessary to check what is actually happening at high SPL though and a Klippel report to actually know the xMax but let's think that these are honest manufacturers...
The signals are set to reach xMax for each driver i.e. max SPL that is guaranteed.
Here are the parameters for the box common for all drivers:
165L/28Hz with a vent 14x18x45cm (air speed < 5% Mach at max SPL so no port shuffling at even loud listening levels).
For the 18s (based on your measurement) you’ll need the following filters:
1. Parametric EQ 35Hz/Q=0.6/gain = 5.5dB
2. HighPass Butter n = 4 / fc = 23Hz / Q = 0.707
3. Set the signal to 122.9W
-> f-3dB = 29Hz / f-6dB = 26Hz
For the 18s (based on Factory) you’ll need the following filters:
1. Parametric EQ 35Hz/Q=0.6/gain = 6dB
2. HighPass SOS n = 4 / fc = 27Hz / Q = 0.707
3. Set the signal to 163.1W
-> f-3dB = 28Hz / f-6dB = 25Hz
For the TAD (Both factory and your measurement) you’ll need the following filters:
1. HighPass Butter n= 4 fc = 22.5Hz
2. Parametric EQ 30Hz/Q=2.0/gain = 4.5dB
3. Set the signal to 120W; 200W is the max continuous power recommended by Pioneer and you'll need 300W to have Xmax so only for short term 400W short term is supported)
-> f-3dB = 28Hz / f-6dB = 26Hz (factory)
Use the Transfer function magnitude Plot to see that the 18s and TAD are neck and neck in terms of BW
Use the SPL Plot to see that the 18s provide 2dB higher SPL
Use the Group Delay plot to see that there is not much difference above 30Hz
Use the Maximum SPL Plot to see that the 18s provides significantly highSPL (10dB @200Hz and @30Hz) if required (with dynamic compressors).
1. The 480Hz trough on the 18s but not a deal breaker and higher distortion above 100Hz whatever that means, as Vance Dickason says “Interpreting the subjective value of THD distortion curves is difficult and opinions vary considerably in the industry. However, looking at the relationship of the second to third harmonic distortion curves is of value”. The directivity is more import though…
2. The 18s seem displacement limited and the TAD is probably Thermal limited -> only in Pro (PA and Monitoring) applications this could be an issue with SPL 115dB+.
3. The 18s could be happily used in a much smaller enclosure
4. The 18s have more Xmax and/less distortion under 100Hz
For the T/S simulations, to compare apples to apples, you can use the factory parameters derived box volume for the TAD which is more of less the recommendation for the EBS 6dB for the 18s…
Hereafter are the “signal” values and “filters” are designed to maximize the bandwidth and the Max SPL, LEAP simulation would be necessary to check what is actually happening at high SPL though and a Klippel report to actually know the xMax but let's think that these are honest manufacturers...
The signals are set to reach xMax for each driver i.e. max SPL that is guaranteed.
Here are the parameters for the box common for all drivers:
165L/28Hz with a vent 14x18x45cm (air speed < 5% Mach at max SPL so no port shuffling at even loud listening levels).
For the 18s (based on your measurement) you’ll need the following filters:
1. Parametric EQ 35Hz/Q=0.6/gain = 5.5dB
2. HighPass Butter n = 4 / fc = 23Hz / Q = 0.707
3. Set the signal to 122.9W
-> f-3dB = 29Hz / f-6dB = 26Hz
For the 18s (based on Factory) you’ll need the following filters:
1. Parametric EQ 35Hz/Q=0.6/gain = 6dB
2. HighPass SOS n = 4 / fc = 27Hz / Q = 0.707
3. Set the signal to 163.1W
-> f-3dB = 28Hz / f-6dB = 25Hz
For the TAD (Both factory and your measurement) you’ll need the following filters:
1. HighPass Butter n= 4 fc = 22.5Hz
2. Parametric EQ 30Hz/Q=2.0/gain = 4.5dB
3. Set the signal to 120W; 200W is the max continuous power recommended by Pioneer and you'll need 300W to have Xmax so only for short term 400W short term is supported)
-> f-3dB = 28Hz / f-6dB = 26Hz (factory)
Use the Transfer function magnitude Plot to see that the 18s and TAD are neck and neck in terms of BW
Use the SPL Plot to see that the 18s provide 2dB higher SPL
Use the Group Delay plot to see that there is not much difference above 30Hz
Use the Maximum SPL Plot to see that the 18s provides significantly highSPL (10dB @200Hz and @30Hz) if required (with dynamic compressors).
Other interesting drivers:
8inch
https://en.toutlehautparleur.com/speaker-celestion-tf0818-8-ohm-8-19-inch.html
http://www.troelsgravesen.dk/midrange-drivers.htm
https://audioxpress.com/article/test-bench-wavecor-sw223bd03-8-75-inch-high-xmax-subwoofer
10inch
https://www.bcspeakers.com/en/products/lf-driver/10-0/8/10cl51-8
https://mereni.prodance.cz/files/storage/1/1704/bc_10CL51.pdf
https://audioxpress.com/article/tes...covery-26w-4558t00-10-aluminum-cone-subwoofer
Larger
https://audioxpress.com/article/test-bench-scan-speak-32w-4878t00-subwoofer
https://audioxpress.com/article/test-bench-b-c-speakers-14na100-8-14-pro-sound-woofer
https://audioxpress.com/article/Test-Bench-B-C-Speakers-12NDL88-12-Woofer
https://mereni.prodance.cz/files/storage/10/19252/BC_Speakers_12NDL88.pdf
Attachments
Last edited:
I downloaded ATH 4.9 221115 and now generating the ATH report works again. Thanks!
Hmm... the impedance does look implausible though. This is demo7.
EDIT: Apparently I celebrated too soon. I moved on to demo8 and again I got the static.txt error message. I pasted fluid's static.txt file into the results folder and then the ATH report script did run successfully. I'll give ATH 4.9 230518 pre-release a go now.
EDIT: Apparently I celebrated too soon. I moved on to demo8 and again I got the static.txt error message. I pasted fluid's static.txt file into the results folder and then the ATH report script did run successfully. I'll give ATH 4.9 230518 pre-release a go now.
Last edited:
Yeah, there's a catch with the sources compound of more than one driving groups - what is actually plotted is the impedance of one of the surround rings, I assume (in ABEC / VACS you can see it clearly - you will see all the parts). I have to look at what's needed to make the whole tweeter appear as a single source (I think it already used to be like that, I just must have changed that at some point...).
Maybe it would be best to leave the old demos altogether...
Last edited:
That makes total sense.
I'm going through the demo versions and it seems like demo8C, OSSE with roll-back, is no longer supported. Also, demo10 gives me the following error when running ATH:
Yes, open baffle, active. The 18SW500 seemed as good enough for what I need, but I haven't made any tests yet.