Why faital pro look a likes I think they are faital pro. And the other new have very good Q to make bass only poor Qms.
Last edited:
Hello Lars,
Imagine the path of the front wave at LF, it is a straight line between teh loudspeaker and the listener position.
Now imagine the path of the rear wave at low frequency. As the wave length is large the wave can walk around the edege of the horn mouth and they travel into the listener position.
The difference between the length of the path travelled by the rear wave and the front wave is the desired value for the diferecne of pathlength to introduce in the "combines" module of Hornresp.
See attached graphics
Best regards from Paris, France
Jean-Michel Le Cléac'h
Imagine the path of the front wave at LF, it is a straight line between teh loudspeaker and the listener position.
Now imagine the path of the rear wave at low frequency. As the wave length is large the wave can walk around the edege of the horn mouth and they travel into the listener position.
The difference between the length of the path travelled by the rear wave and the front wave is the desired value for the diferecne of pathlength to introduce in the "combines" module of Hornresp.
See attached graphics
Best regards from Paris, France
Jean-Michel Le Cléac'h
Attachments
Hello,
Helmuth wrote,
There is a classical opinion between DIYers that a series resistor has negative effects on the Impulse Response of a loudsepaker.
This is not allways the case. In fact in order to obtain the same optimal damping with a loudspeaker having a low Qts parameter value, we have several means: mechanical (and acoustical is one of them) or electrical.
In many more case than the common DIYer could imagine we can use a series resistor in order to improve the damping and to obtain the best impulse response desirable.
Here attached the impulse response of the "Horn Open Baffle" I described in message:
http://www.diyaudio.com/forums/multi-way/140190-jean-michel-lecleach-horns-110.html#post2589622
Notice how good is the Impulse response and the spectrogram. ( Compared to a TH...)
Best regards from Paris, France
Jean-Michel Le Cléac'h
Helmuth wrote,
There is a classical opinion between DIYers that a series resistor has negative effects on the Impulse Response of a loudsepaker.
This is not allways the case. In fact in order to obtain the same optimal damping with a loudspeaker having a low Qts parameter value, we have several means: mechanical (and acoustical is one of them) or electrical.
In many more case than the common DIYer could imagine we can use a series resistor in order to improve the damping and to obtain the best impulse response desirable.
Here attached the impulse response of the "Horn Open Baffle" I described in message:
http://www.diyaudio.com/forums/multi-way/140190-jean-michel-lecleach-horns-110.html#post2589622
Notice how good is the Impulse response and the spectrogram. ( Compared to a TH...)
Best regards from Paris, France
Jean-Michel Le Cléac'h
Attachments
Try the H baffle revintage. I made the front chamber smaller than the rear chamber and left an opening in the front baffle to give about 2.6:1 compression on the woofer. The path length I took as the full depth of the H baffle plus the depth of the rear open chamber of the h baffle, ie the inside of the rear because that must be part of the path surely, plus a few centimetres fudge factor to allow for the different angle of the radiation from the rear around the ouside of the horn, plus the length of the horn. As I said before I am not sure that this is correct.
How does adding a coil in series with the woofer affect the TS calculations?
In the sims that I have done using this structure there is a rise in output from the horn which needs to be knocked off a bit but this structure certainly models better than any Bass reflex/horn off a single speaker. If I am doing it right.
jamikl
How does adding a coil in series with the woofer affect the TS calculations?
In the sims that I have done using this structure there is a rise in output from the horn which needs to be knocked off a bit but this structure certainly models better than any Bass reflex/horn off a single speaker. If I am doing it right.
jamikl
Hey jamikl,
This is just a hypothetical discussion so the actual design of the baffle hasn´t discussed yet.
I find your explanation a little hard to follow, and as you talk about chambers this makes it even harder
.
Couldn´t you please show some graphics to make it easier to understand. Great if you also could show your sims.
This is just a hypothetical discussion so the actual design of the baffle hasn´t discussed yet.
I find your explanation a little hard to follow, and as you talk about chambers this makes it even harder
. Couldn´t you please show some graphics to make it easier to understand. Great if you also could show your sims.
Last edited:
Hello Lars,
In a dipolar scheme, to obtain some gain at low frequency we try to have both the front and rear waves in phase at the listener position (this means a difference of path around the half of the wavelength. If we use the common OB approach in the bass register then we need a very large baffle.
In the field of my "Horn Open Baffle" concept I prefer to use the rise of the group delay of the horn near the frequency cut-off, than to lengthen the path of the rear wave. This is what makes the concept original.
Best regards from Paris, France
Jean-Michel Le Cléac'h
In a dipolar scheme, to obtain some gain at low frequency we try to have both the front and rear waves in phase at the listener position (this means a difference of path around the half of the wavelength. If we use the common OB approach in the bass register then we need a very large baffle.
In the field of my "Horn Open Baffle" concept I prefer to use the rise of the group delay of the horn near the frequency cut-off, than to lengthen the path of the rear wave. This is what makes the concept original.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Hello Lars,
No harmful differences to my eyes.
It is always a bit difficult to acertain which difference of pathlength is the most realistic to be introduced in an Hornresp model.
Better to use what seems logical and the more logical accords to the drawing I gave.
http://www.diyaudio.com/forums/atta...06-jean-michel-lecleach-horns-pathlengths.gif
Then, as always, there could be some discrepancy between the simulation and the measurement and, as ever, you'll need to perform some optimization...
Best regards from Paris, France
Jean-Michel Le Cléac'h
No harmful differences to my eyes.
It is always a bit difficult to acertain which difference of pathlength is the most realistic to be introduced in an Hornresp model.
Better to use what seems logical and the more logical accords to the drawing I gave.
http://www.diyaudio.com/forums/atta...06-jean-michel-lecleach-horns-pathlengths.gif
Then, as always, there could be some discrepancy between the simulation and the measurement and, as ever, you'll need to perform some optimization...
Best regards from Paris, France
Jean-Michel Le Cléac'h
Hello again Jean-Michel,
The simmed -3dB of 400Hz or 800Hz is a big difference. If the 800Hz is OK, then the wavelength giving 400Hz -3dB must be considered as harmful. But will this happen IRL?
I am a little confused as I didn´t think the rear wavelength should affect the higher frequencies this much. The low end doesn´t seem to be affected much.
The simmed -3dB of 400Hz or 800Hz is a big difference. If the 800Hz is OK, then the wavelength giving 400Hz -3dB must be considered as harmful. But will this happen IRL?
I am a little confused as I didn´t think the rear wavelength should affect the higher frequencies this much. The low end doesn´t seem to be affected much.
Hello Lars,
You are right, what I expalined is related to the low frequency. IN the medium frequency (over 150Hz or 200Hz) due to the increased directivity of the rear wave, there will be no interference between the rera and front waves.
But as it is strongly related to the geometry of the system, this is the kind of problem that Hornrresp cannot solve (for the moment).
Above 200Hz better to consider that the front wave only arrives to the listener and so better to not use the "combined" module.
(Note: for a horn bass-reflex VOT "style" having a vent on the front of the box the recommandation will not apply.)
Best regards from Paris, France
Jean-Michel Le Cléac'h
P.S. : I leave Paris for a geological trip and I'll be back only on Tuesday June 14th.
You are right, what I expalined is related to the low frequency. IN the medium frequency (over 150Hz or 200Hz) due to the increased directivity of the rear wave, there will be no interference between the rera and front waves.
But as it is strongly related to the geometry of the system, this is the kind of problem that Hornrresp cannot solve (for the moment).
Above 200Hz better to consider that the front wave only arrives to the listener and so better to not use the "combined" module.
(Note: for a horn bass-reflex VOT "style" having a vent on the front of the box the recommandation will not apply.)
Best regards from Paris, France
Jean-Michel Le Cléac'h
P.S. : I leave Paris for a geological trip and I'll be back only on Tuesday June 14th.
Path Length Difference in Hornresp
Hi Jean-Michel,
Just to clarify - the way that Hornresp defines path length difference, d1 in your path length drawing (attachment 1) should show the distance from the horn mouth to the nominal listening position, not from the horn driver diaphragm to the listener - see attachment 2 for further details.
To illustrate the difference using your definition of d1 as the basis for the comparision:
For d1 = 150 cm and d2 = 200 cm your drawing would suggest that the path length difference value required for Hornresp is d2 - d1 = 50 cm.
Assuming that the horn axial length (L12) = 50 cm, then the path length difference value actually required for Hornresp is d2 - (d1- L12) = 100 cm.
Hope this makes sense.
Kind regards,
David
Hi Jean-Michel,
Just to clarify - the way that Hornresp defines path length difference, d1 in your path length drawing (attachment 1) should show the distance from the horn mouth to the nominal listening position, not from the horn driver diaphragm to the listener - see attachment 2 for further details.
To illustrate the difference using your definition of d1 as the basis for the comparision:
For d1 = 150 cm and d2 = 200 cm your drawing would suggest that the path length difference value required for Hornresp is d2 - d1 = 50 cm.
Assuming that the horn axial length (L12) = 50 cm, then the path length difference value actually required for Hornresp is d2 - (d1- L12) = 100 cm.
Hope this makes sense
.Kind regards,
David
Attachments
I always was under the assumption that the path length difference could be used for things like an open baffle or H or U frame type baffle to calculated the roll of due to the path length. Now it seems there is a further use. Thank you David!
And Jean-Michel has been a constant source of great tips as well. Thanks to both of you.
And Jean-Michel has been a constant source of great tips as well. Thanks to both of you.
Le Cléac'h Horn - New Profile
Hi Everyone,
I recently had a close look at the Le Cléac'h horn flare - I was particularly interested in checking the accuracy of the constructed profile against the actual hyperbolic-exponential expansion rate upon which it is based.
For the purposes of my investigations I considered a Le Cléac'h horn having the following parameters:
Throat diameter = 2.54 cm
Cutoff frequency = 320 hertz
T = 0.8
I first compared Jean-Michel’s original and improved (2007) profiles against the actual hyperbolic-exponential expansion rate, with the wavefront area equivalent radius being plotted against axial length. See Lec_Original.png and Lec_Improved.png attached.
I then realised that it should be possible to construct an even more accurate Le Cléac'h profile. See the results in Lec_New.png attached. The Hyp and Lec curve values are identical to 11 decimal places.
The three horn profiles are shown together in Lec_Compare.png attached. It is interesting that the new profile makes the Le Cléac'h horn even slightly more compact.
Hornresp currently simulates Le Cléac'h horns using Jean-Michel’s "improved profile". Because the difference in performance between the improved profile and new profile is likely to be insignificant, the Hornresp model will not be changed. An option will however be added in the next release to allow raw data for the new profile to be exported, so that constructors can build the perfect axisymmetric Le Cléac'h horn, if so desired.
Jean-Michel has been given details of the method used to construct the new profile.
Kind regards,
David
Hi Everyone,
I recently had a close look at the Le Cléac'h horn flare - I was particularly interested in checking the accuracy of the constructed profile against the actual hyperbolic-exponential expansion rate upon which it is based.
For the purposes of my investigations I considered a Le Cléac'h horn having the following parameters:
Throat diameter = 2.54 cm
Cutoff frequency = 320 hertz
T = 0.8
I first compared Jean-Michel’s original and improved (2007) profiles against the actual hyperbolic-exponential expansion rate, with the wavefront area equivalent radius being plotted against axial length. See Lec_Original.png and Lec_Improved.png attached.
I then realised that it should be possible to construct an even more accurate Le Cléac'h profile. See the results in Lec_New.png attached. The Hyp and Lec curve values are identical to 11 decimal places.
The three horn profiles are shown together in Lec_Compare.png attached. It is interesting that the new profile makes the Le Cléac'h horn even slightly more compact.
Hornresp currently simulates Le Cléac'h horns using Jean-Michel’s "improved profile". Because the difference in performance between the improved profile and new profile is likely to be insignificant, the Hornresp model will not be changed. An option will however be added in the next release to allow raw data for the new profile to be exported, so that constructors can build the perfect axisymmetric Le Cléac'h horn, if so desired.
Jean-Michel has been given details of the method used to construct the new profile.
Kind regards,
David
