Yeah , nice curve but the knee at the bottom end is still too high ...
TB, Is it possible that there might be some path length or volume that was somehow left out in that Akabak model ? Can you post a displacement or impedance curve so we can see where FB is?
Have fun at the party
If you are lucky they wont have you dress up as a clown , or SpongeBob or some such thing ..... Batman would be cool though , or maybe a pirate! 
Hi
Thank you for the explanations. The whole point was to get a sub that would nicely complement a bookshelf at 40-80 hz. Since efficiency wouldn't have to be that high, I thought that a 6.5 or 8" driver would be sufficient. But, as we can't beat laws of physics, the line would still have to be quite long in order to get to 40 hz. Otherwise, I thought about using a 6.5 car bass driver (high Qts) in a negative tapered line and force it below its Fs,but then it would be different from what's being discussed here...
Regards,
Vix
Thank you for the explanations. The whole point was to get a sub that would nicely complement a bookshelf at 40-80 hz. Since efficiency wouldn't have to be that high, I thought that a 6.5 or 8" driver would be sufficient. But, as we can't beat laws of physics, the line would still have to be quite long in order to get to 40 hz. Otherwise, I thought about using a 6.5 car bass driver (high Qts) in a negative tapered line and force it below its Fs,but then it would be different from what's being discussed here...
Regards,
Vix
I don't have any particular model on my mind. Let's choose one that is easily available, inexpensive so many people can replicate it. Imagine it: Mini Karlsonator on top of dual 6.5" bass Karlsonator
Error Correction: Post #460
Hi Matthew Morgan J,
Post #462: "...Is it possible that there might be some path length or volume that was somehow left out..."
Yes, you are correct, one section had the wrong length. Here is the error correction: changed waveguide 'W4' Len=2.625in to Len=10.297in.
In the attached pictures the original one is on the top, and the corrected one is on the bottom (the ones where everything is a little lower in frequency).
It's interesting how much of a difference those 7.672in make. Looks like to get lower the overall duct length will have to be increased. Nothing new there.
Regards,
Hi Matthew Morgan J,
Post #462: "...Is it possible that there might be some path length or volume that was somehow left out..."
Yes, you are correct, one section had the wrong length. Here is the error correction: changed waveguide 'W4' Len=2.625in to Len=10.297in.
In the attached pictures the original one is on the top, and the corrected one is on the bottom (the ones where everything is a little lower in frequency).
It's interesting how much of a difference those 7.672in make. Looks like to get lower the overall duct length will have to be increased. Nothing new there
.Regards,
Attachments
Hi Matthew Morgan J,
Post #462: "...Is it possible that there might be some path length or volume that was somehow left out..."
Yes, you are correct, one section had the wrong length. Here is the error correction: changed waveguide 'W4' Len=2.625in to Len=10.297in.
In the attached pictures the original one is on the top, and the corrected one is on the bottom (the ones where everything is a little lower in frequency).
It's interesting how much of a difference those 7.672in make. Looks like to get lower the overall duct length will have to be increased. Nothing new there
Regards,
TB46 ,
Fantastic! Looks like you are getting a lot closer to what HR was giving me
Only about 5hz away from the 35hz target now ..... Thank you for attempting this in Akabak ..... If in the end it turns out that there just simply isn't enough path length in that configuration to get the FB down to around 35 then i can make some changes to the HR model in order to fix that... I think i am about to drop the idea of using flares around the constriction altogether anyway... Since my new goal is to keep the constriction to between 33% and 50% of the main path's area i really don't think that the flares will be necessary (at the m/s rates that will be generated with these drivers) .. ...
TB46,
Thanks for trying out and modifying the script. You are not bad with Akabak yourself.
You may find it surprising but I find HR hard for me to use once I am familiar with how to do it in Akabak. I struggle just getting the T/S params in because they are always given by manufacturer as Re, Le, fs, Qms, Qes, Mms, Vas,... Then I always forget what the names of all the boxes mean because it is abbreviated. However, HR's slider controls for the geometry are way cool.TB46,
Thanks for trying out and modifying the script. You are not bad with Akabak yourself.
This is how i do it. Find SD, which they basically always give you. Calculate cms, calculate mmd, enter re calculate bl calculate rms enter le if given or estimate based on like drivers if have reference.
That's a lot of hoo-ha for something that can be typed in "as-is" in Akabak... Although Cms and Rms are the true lumped-element T/S parameter representing the system capacitance and resistance, respectively. The Qes, Qms, and fs make more physical sense to us though than Cms and Rms values.
I use David's wonderful Hornresponse to calculate missing parameters .... As long as you have SD , FS , VAS , RE , QTS , QMS , LE the rest can be calculated by HR ... Takes a little fiddling sometimes by adjusting the BL to dial in the QTS , but be sure to calculate MMD first by clicking on the MMD field and entering the FS there ..
Saba , QES is electrical damping (Q electrical) and has to do with the strength and qualities of the motor (magnet & coil) in relation to things like moving mass (MMS or MMD) and VAS (stiffness of suspension) .... QES is also the biggest influence upon QTS (total Q) ....
A heavier moving mass tends to drive up the QES , and so will a very tight suspension (low VAS figure) ... It takes a stronger motor and lots of gauss in the gap to drive the QES back down ...
These Alpines for example have a very tight suspension, and moderate moving mass for 10s and 12s .... Alpine had to put a strong magnet on these to keep the QES and QTS from going too high .....
QMS (mechanical Q) has much less influence upon QTS , especially if the QMS is high.... I tend to like the way higher QMS drivers perform when simming my designs if all the other parameters are in the right ranges of course, a high QMS is a bonus! ...
BY THE WAY Saba , i have been working on a few things for your SWS12 , i think i might have something you will like ... I just finished my final design for the Lab15 special and as soon as i can finish sketching that up and posting it i will get your SWS12 stuff together to post as well..
... Takes a little fiddling sometimes by adjusting the BL to dial in the QTS , but be sure to calculate MMD first by clicking on the MMD field and entering the FS there .. Saba , QES is electrical damping (Q electrical) and has to do with the strength and qualities of the motor (magnet & coil) in relation to things like moving mass (MMS or MMD) and VAS (stiffness of suspension) .... QES is also the biggest influence upon QTS (total Q) ....
A heavier moving mass tends to drive up the QES , and so will a very tight suspension (low VAS figure) ... It takes a stronger motor and lots of gauss in the gap to drive the QES back down ...
These Alpines for example have a very tight suspension, and moderate moving mass for 10s and 12s .... Alpine had to put a strong magnet on these to keep the QES and QTS from going too high .....
QMS (mechanical Q) has much less influence upon QTS , especially if the QMS is high.... I tend to like the way higher QMS drivers perform when simming my designs if all the other parameters are in the right ranges of course, a high QMS is a bonus! ...
BY THE WAY Saba , i have been working on a few things for your SWS12 , i think i might have something you will like ... I just finished my final design for the Lab15 special and as soon as i can finish sketching that up and posting it i will get your SWS12 stuff together to post as well..
Last edited:
Hi MMJ,
You are correct - it was done intentionally, to simplify things.
It never occurred to me that it could prove to be a problem in practice
. It will be fixed in the next release.Many thanks for the feedback.
Kind regards,
David
UPDATED (AND LIKELY FINAL) LAB15(special) ML-TRANSFLEX!
Ok , here we go
I am pretty proud of this one
Version 3
Still a very simple design, and this one is 100% compatible with HR's segments, chambers and ports so this curve should be close enough to what you will get in real life Someone out there could still try to Akabak it if they want, it should be a lot easier with this one because it is relatively straightforward compared to the last version i uploaded ...
I plan to make this my final version of the ML-Transflex for the LAB15 special..
I will get a sketch drawn up for this thing tomorrow , dimensions are very similar, i just have to extend one dimension to allow for a few more liters of internal volume ...
Ok , here we go
I am pretty proud of this one
Version 3
Still a very simple design, and this one is 100% compatible with HR's segments, chambers and ports so this curve should be close enough to what you will get in real life
Someone out there could still try to Akabak it if they want, it should be a lot easier with this one because it is relatively straightforward compared to the last version i uploaded ... I plan to make this my final version of the ML-Transflex for the LAB15 special..
- Great output
- Low group delay
- Excursion stays below 13mm at a full 58v (600w)
- driver is easily accessible without the use of a removable panel
- efficient use of cabinet volume, no wasted space
- simple single fold design, easy build, all straight panels, no angles
- acoustically shortened path by use of a mass loading constriction, which is a long vent at 50% of the main pipe's area, velocity should not exceed 20 m/s in worst case scenario in the real world.(calculated by multiplying the velocity at mouth (*2 for the 2:1 mouth/port area ratio)
- Very affordable driver $134 shipped, get em while they last!
- Woofer's motor is located in the mouth (the high velocity end of the pipe) which maximizes cooling of the voicecoil ! Which equates to reduced thermal compression when used at high SPLs for extended periods of time.. Longer life, and more output!
- well balanced bass output, just the right amount of damping.
I will get a sketch drawn up for this thing tomorrow , dimensions are very similar, i just have to extend one dimension to allow for a few more liters of internal volume ...
Attachments
Just to clarify - Cms and Rms are not T/S (Thiele-Small) parameters. Thiele-Small parameters are:
fs
Vas
Qes
Qms
Qts
Cms and Rms are loudspeaker driver mechanical parameters:
Cms is the driver diaphragm suspension mechanical compliance
Rms is the driver diaphragm suspension mechanical resistance
Qes is the driver resonant electrical Q at the driver resonance frequency, due solely to inherent electromagnetic damping.
Qms is the driver resonant mechanical Q at the driver resonance frequency, due solely to internal mechanical damping.
Hi MMJ,
You are correct - it was done intentionally, to simplify things.
It never occurred to me that it could prove to be a problem in practice
Many thanks for the feedback.
Kind regards,
David
Thank you David, you are AWESOME! The DIY community owes you immensely for making HR available, available for free even! Your karmic credit account must be fully astounding in the eyes of gods!
Hi xrk971,
Post #469: "... HR's slider controls for the geometry are way cool."
Agreed. There are so many neat things in Hornresp, one is the AkAbak file export capability. That was my starting point into AkAbak. I'm still not through the AkAbak manual though.
For me the T/S parameter entry into Hornresp has become very natural; you have to remember to double-click on Cms, Mmd, Rms and Bl to get to the entry boxes for Vas, fs, Qms and Qes.
One point that does not get mentioned very often is that after data entry you should double-click on Sd, and check the driver T/S parameter values Hornresp will use against the values you may have, e.g.: from measurements.
I prefer this a lot over WinISD's: "I'm not going to work as there is an error somewhere, and I won't tell you what it is."
Regards,
Post #469: "... HR's slider controls for the geometry are way cool."
Agreed. There are so many neat things in Hornresp, one is the AkAbak file export capability. That was my starting point into AkAbak. I'm still not through the AkAbak manual though.
For me the T/S parameter entry into Hornresp has become very natural; you have to remember to double-click on Cms, Mmd, Rms and Bl to get to the entry boxes for Vas, fs, Qms and Qes.
One point that does not get mentioned very often is that after data entry you should double-click on Sd, and check the driver T/S parameter values Hornresp will use against the values you may have, e.g.: from measurements.
I prefer this a lot over WinISD's: "I'm not going to work as there is an error somewhere, and I won't tell you what it is."
Regards,