IG,
Are you talking about de-Q'ing on the front chamber? That chamber has a gaping aperture so not much Q there. Must be rear chamber you are talking about, but how does the slot starting gap affect the rear chamber Q once it has already passed through the narrow top channel?
So, are you going to try the K-nator in triple layer FC with the FF225WK? I wish I had one, I would love to hear it reach down to 30 Hz, flat response to 250 Hz. That is some wide bandwidth on the bass shelf!
Are you talking about de-Q'ing on the front chamber? That chamber has a gaping aperture so not much Q there. Must be rear chamber you are talking about, but how does the slot starting gap affect the rear chamber Q once it has already passed through the narrow top channel?
So, are you going to try the K-nator in triple layer FC with the FF225WK? I wish I had one, I would love to hear it reach down to 30 Hz, flat response to 250 Hz. That is some wide bandwidth on the bass shelf!
Front chamber. Modeled as a Helmholtz resonator in WinISD, I usually input Ql=3 to simulate how much the K-slot de-Q's. I also think that large top-vents add on top of this, so there might be a bit of effect here that one can play with in case there is upper-bass peaking, which we have seen on various smaller than K15 models before. I agree it won't be huge though.
IG
hi X - what do you figure is the overall impact on response and dispersion of the aperture below vs a typical radial arc K-aperture? It sounds nice with my W8-1772 - but even constricted apertures sounded decent on the SK8. Carl used some formula or mix of formula to generate this curve which he suggested as the upper boundary for "openness" (my term - not his) it would be interesting if IG could try it on one of K12
this is the little "SK8" prototype - IG built a pair
this is the little "SK8" prototype - IG built a pair
An externally hosted image should be here but it was not working when we last tested it.
I've had the standard AN10 stamped frame speaker in both K15 and an X15 copy. It sounded good in K15 - I think the curved reflector of X15 did not help matters. An aperture adjustment might help it in X15. My stamped frame AN10 pair had qts quite a bit higher than published spec (and AN8 not real far off) - part of the smoothing would be the K's built-in BSC. I'll have to look up the Q tests on my A-Ns - - is the cast frame version measuring with low qts and fs?
it would not hurt to have actual parameters before building a K for AN10CF
here's the spec I got on the stamped frame AN10 - qts ~0.42, fs - 50 - IIRC the sealed box method of finding Vas was more accurate for this type than
delta mass - could be suspension non-linearities
it would not hurt to have actual parameters before building a K for AN10CF
here's the spec I got on the stamped frame AN10 - qts ~0.42, fs - 50 - IIRC the sealed box method of finding Vas was more accurate for this type than
delta mass - could be suspension non-linearities
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
Honestly, it's been a while. I think I was shooting for around 45Hz, which is about what I got using MJK's sections sheet as a kludge. IIRC, which is doubtful...
At any rate, anyone making these should try to include some provision for adjusting the internal duct vent. It could be done by using a thin piece of plywood, and taping in shims to adjust the area as desired.
Hi IG,
I am very keen on fixing all bugs found in Hornresp - I have a "zero tolerance" approach when it comes to errors in the program
.Are you using the latest release - Product Number 3220-130625?
If not, then you might like to upgrade to the latest release to see if that fixes the problem.
If the bug is still in latest release, then any information you can provide on the sequence of events leading up to the generation of the error, would be greatly appreciated.
Kind regards,
David
Hi IG,
I am very keen on fixing all bugs found in Hornresp - I have a "zero tolerance" approach when it comes to errors in the program
Are you using the latest release - Product Number 3220-130625?
If not, then you might like to upgrade to the latest release to see if that fixes the problem.
If the bug is still in latest release, then any information you can provide on the sequence of events leading up to the generation of the error, would be greatly appreciated.
Kind regards,
David
Wow, that's good attention from the publisher for free software!
Nah, I got an older version, # 2830-110101. It happend while in "Loudspeaker Wizard" mode when I was changing the "Ang" parameter and went to "0,0 x Pi" by accident. I know not to do this now, but getting the most recent version could not hurt in any case.
IG
Sounds reasonable. If you got ~45Hz in MJK's sections, xrk971 ~40Hz in Akabak and myself ~40Hz in HR, there must be something working correctly there.
IG
Re: tuning of the Karlsonator's front chamber,
I don't think there was a way to check this in MJK for GregB, so I'm not sure what his thoughts were on this. xrk971 is getting ~150Hz in Akabak and I was predicting ~120Hz with the classic form-factor method, but I think this does not apply here.
Trying to kludge something yet again in HR, I tried entering the front chamber alone as an OD horn, where S1 (area at the top) is obvious from the K'ator plan and made S3 to be the surface area of a circle with the bottom of the K-slot's width as its diameter. I adjusted S2 to have somewhat of a straight conical taper. I placed the driver somewhere close to the mouth.
Yielding:
S1: 160cm^2
S2: 750cm^2
L12: Con 50cm
S3: 1178cm^2
L23: Con 20cm
This assumes that the radial section of air adjacent to the K-slot on the exterior side is part of the effective volume. This might be completely incorrect, but I have thought of looking at it this way before, so have a couple of other K-fans around IIRC. This was originally an idea on how to view the "purer" K-Tubes. Quite literally a case of thinking outside the box.
Anyway, HR is giving me a ~140Hz tuning in this case, so probably not too far off AFAICT. Relatively in-line with xrk971's Akabak model. Too bad I can not graft both front and rear sections in this manner. The ideal way to go about it might be my method based on a TH, but I'd need a few more segments ideally.
IG
I don't think there was a way to check this in MJK for GregB, so I'm not sure what his thoughts were on this. xrk971 is getting ~150Hz in Akabak and I was predicting ~120Hz with the classic form-factor method, but I think this does not apply here.
Trying to kludge something yet again in HR, I tried entering the front chamber alone as an OD horn, where S1 (area at the top) is obvious from the K'ator plan and made S3 to be the surface area of a circle with the bottom of the K-slot's width as its diameter. I adjusted S2 to have somewhat of a straight conical taper. I placed the driver somewhere close to the mouth.
Yielding:
S1: 160cm^2
S2: 750cm^2
L12: Con 50cm
S3: 1178cm^2
L23: Con 20cm
This assumes that the radial section of air adjacent to the K-slot on the exterior side is part of the effective volume. This might be completely incorrect, but I have thought of looking at it this way before, so have a couple of other K-fans around IIRC. This was originally an idea on how to view the "purer" K-Tubes. Quite literally a case of thinking outside the box.
Anyway, HR is giving me a ~140Hz tuning in this case, so probably not too far off AFAICT. Relatively in-line with xrk971's Akabak model. Too bad I can not graft both front and rear sections in this manner. The ideal way to go about it might be my method based on a TH, but I'd need a few more segments ideally.
IG
If I understand it correctly the Karlson benefits from mating with drivers that have a relatively high Qts. What that in mind what about the Goldwood 8 inch full range. Several comments elsewhere mentioned that it does not do well in any sort of sealed enclosure. Perhaps it would find a natural home in the Karlson. It's certainly inexpensive to try and I presume the cabinet could then be used for some "driver rolling" to something perhaps better like the Betsy. Anyone have any thoughts on this?
It's the other way around usually, you want a strong motor, so that'll result in low Qts values. In any case, some high-Qts drivers have been known to work. I think that ultimately, something with a mass-corner ((2*Fs)/Qts) above the enclosure's bandwidth is needed to avoid excessive peaking within the BW.
IG
Yeah, in general, the opposite is true. The original K15 and K12 really work best with low Q drivers, which are somewhat rare these days. I should mention that the later slit vent revision of the K12 can deal with a wide variety of specs.
Part of my motivation for the Karlsonator was finding a way to make the idea work better with common modern driver specs. The TQWT on its own is fairly amenable to a variety of drivers, from low-mid to high Q. It retains this quality even after combining it with a K-coupler front load.
AFA the little Vifa drivers, I suspect a certain synergy between the lossy foamcore construction and the high Q drivers. If this was made out of plywood, it might bloom a bit, though of course you could always add damping.
And yet, when the Karlson debuted, it was designed for medium to high Qtb systems, i.e. low to medium Qt drivers coupled to a high output impedance, a point the hard core K-klan chooses to ignore/downplay, saying such alignments should be avoided to the point where one of them 'flat out' said I shouldn't be allowed to spread such BS on the forums......... Ultimately, he got banned, but there's others who are of a similar mindset.
GM
I wondered who did it. I proposed it to Freddy over a decade ago along with an explanation of how a K really worked as opposed to the somewhat BS marketing of the original and got politely told I was mistaken even though he showed me PWK’s same conclusion. Some others weren’t so polite, so left them to their delusions.
Later on, the late GregB wanted me to pursue it and I’d planned to, but 'life' continues to keep me from doing much of anything with audio beyond [re]posting stuff I’ve still got in what’s left of my technical ‘library’.
Anyway, glad to see this adaptable, better overall performing tapped ML-TQWT variant finally making some ‘waves’.
GM