GM said:
Thanks, GM, I'll double check my T/S measurements, .. and try some known drivers too .
Displacement measurement
At long last I got my act together and measured the displacement of the PD1550 TH to compare with Hornresp and AkAbak (outline details of the design are earlier in the thread).
I showed earlier that the model did not predict the response exactly, but was within 1dB everywhere that counts.
As the excursion is quite a strong function of frequency (with nulls below and above), it seemed best to seek out the maximum around 30Hz and compare with the corresponding maximum in the model, rather than just compare at a particular frequency.
The measured peak was 2.4mm +/- 0.3mm peak with 8V rms in at about 32Hz. The measurement was done using a flat steel probe stuck to the magnet such that it could be slide back and forth to just touch the cone while playing (it was very easy to hear when it did touch). Its position and the gap to the rest position of the cone were measured several ways. The most convincing and stable result being given. The voltage was measured only after the excursion, so that I was not biased towards a particular answer.
The hornresp model gives 2.1 mm peak with 8V rms, peaking at just under 30Hz. The AkAbak model (slightly different geometry, not exported from Hornresp) gives 2.1mm peak at just over 30Hz.
In conclusion I was not able to distinguish any significant discrepancy between the model and measured excursion for the PD1550 TH near the in-band maximum.
The previously claimed discrepancy does not appear to apply in all cases, of course I have tested only one design.
I do not plan to investigate further.
Ken
At long last I got my act together and measured the displacement of the PD1550 TH to compare with Hornresp and AkAbak (outline details of the design are earlier in the thread).
I showed earlier that the model did not predict the response exactly, but was within 1dB everywhere that counts.
As the excursion is quite a strong function of frequency (with nulls below and above), it seemed best to seek out the maximum around 30Hz and compare with the corresponding maximum in the model, rather than just compare at a particular frequency.
The measured peak was 2.4mm +/- 0.3mm peak with 8V rms in at about 32Hz. The measurement was done using a flat steel probe stuck to the magnet such that it could be slide back and forth to just touch the cone while playing (it was very easy to hear when it did touch). Its position and the gap to the rest position of the cone were measured several ways. The most convincing and stable result being given. The voltage was measured only after the excursion, so that I was not biased towards a particular answer.
The hornresp model gives 2.1 mm peak with 8V rms, peaking at just under 30Hz. The AkAbak model (slightly different geometry, not exported from Hornresp) gives 2.1mm peak at just over 30Hz.
In conclusion I was not able to distinguish any significant discrepancy between the model and measured excursion for the PD1550 TH near the in-band maximum.
The previously claimed discrepancy does not appear to apply in all cases, of course I have tested only one design.
I do not plan to investigate further.
Ken
from reading past posts, it seems that i will need a subsonic filter to protect the eminence from bottoming out.
i cannot find many, im guessing i should use 1 at 20hz but this set at 10
http://store.xmfanstore.com/linedriver.html
will it suffice?
also is it possible to use my pc as a subsonic filter, so the card would cut off unwanted output below 20hz
i cannot find many, im guessing i should use 1 at 20hz but this set at 10
http://store.xmfanstore.com/linedriver.html
will it suffice?
also is it possible to use my pc as a subsonic filter, so the card would cut off unwanted output below 20hz
I much prefer the sound and realism when the speakers are driven by a wideband amplifier. <2Hz to >50kHzNaudio said:
But I listen at volumes that are unlikely to overstroke my 15inch and 18inch drivers.
Just use and LR24. Rod Elliot's site has schematics you can tweak the values in to get a 20Hz HPF and even PCB's.
Just doing some pondering on this quite Friday morning. I’ve been building horns of all sorts for over 15 years and cannot recall there ever being a more significant advancement in high efficiency speaker design than the tapped horn. Really, the last major advancement in my mind is the invention of the compression driver back in 1926 by Albert Thuras and Edward Wente of Bell Telephone Laboratories. For high efficiency in the low frequency range, the tapped horn’s size and SPL output just can’t be beat. I have ditched my very large downward firing subwoofer horns (combined 340 liter volume) for two tapped horns that take up all of about 80 liters total for the pair. I can’t tell you how much happier the wife is now. I definitely scored some points when I removed my old subwoofer horns and put the tapped horns in.
A couple observations I’ve made between a classical front loaded horn subwoofer verses the tapped horn. 1.) Even through both are types of horns, the tapped horns actually have less group delay. I actually have better sound ingratiation with the mid-bass horns. Time alignment is a real option now! The old front loaded horns sound like they were ¼ to ½ a beat behind the tapped horns. Insert the slow vs. fast bass debate here. 2.) The tapped horns are 6dB less sensitive than the front loaded horns. However, the tapped horns surpass them in sensitivity below 38Hz. Due to the drop in mid-band sensitivity; the tapped horns need 4X the power as the front loaded horns. However, I have 350W on hand, so this is a non-factor because I never needed all that power anyway and still won’t. 3.) It is more fun to watching people’s reaction to the amount of bass coming from this rather small and tall box. Most people thought the old front loaded horns were a poor trade off for size verses performance. I really didn’t care what people thought, but I am much happier with the tapped horns. It still blows my mind how well they work and how small they are. In my opinion the tapped horn has been a win-win for everyone.
Best Rgs, JLH
A couple observations I’ve made between a classical front loaded horn subwoofer verses the tapped horn. 1.) Even through both are types of horns, the tapped horns actually have less group delay. I actually have better sound ingratiation with the mid-bass horns. Time alignment is a real option now! The old front loaded horns sound like they were ¼ to ½ a beat behind the tapped horns. Insert the slow vs. fast bass debate here. 2.) The tapped horns are 6dB less sensitive than the front loaded horns. However, the tapped horns surpass them in sensitivity below 38Hz. Due to the drop in mid-band sensitivity; the tapped horns need 4X the power as the front loaded horns. However, I have 350W on hand, so this is a non-factor because I never needed all that power anyway and still won’t. 3.) It is more fun to watching people’s reaction to the amount of bass coming from this rather small and tall box. Most people thought the old front loaded horns were a poor trade off for size verses performance. I really didn’t care what people thought, but I am much happier with the tapped horns. It still blows my mind how well they work and how small they are. In my opinion the tapped horn has been a win-win for everyone.
Best Rgs, JLH
Hmm, I'm going with Jensen's 1952 Transflex tapped TL, the TH's conceptual predecessor, though IIRC Tom said he wasn't aware of it at the time of his creation. The whole speaker system was dubbed the 'Reproducer of the Future' and how prophetic it turned out to be as only its BL woofer horn would gain popularity as the Imperial: http://img505.imageshack.us/img505/9528/transflex18tl.jpg
FWIW I built a proof-of-concept decades ago and it was pretty impressive with sine wave input and what little truly LF source material I had even though I didn't have the right woofer, but with no practical way to blend it to ~70 Hz FL mid-bass horns that I was aware of at the time I re-used the wood and the LS-15 has been languishing in the closet ever since.
GM
FWIW I built a proof-of-concept decades ago and it was pretty impressive with sine wave input and what little truly LF source material I had even though I didn't have the right woofer, but with no practical way to blend it to ~70 Hz FL mid-bass horns that I was aware of at the time I re-used the wood and the LS-15 has been languishing in the closet ever since.
GM
GM said:
I was well aware of the Jensen Transflex. I think Freddy pulled it up about 5 years ago. It didn't spark much interest because no one knew what to do with it. I kind of discount the Transflex because not even Jensen understood how it worked. They were unable to get reliable and reproducible results. That's why it died in the 1950's. It wasn't until Tom Re-figured / Re-discovered it, and showed how to get reliable results, it then became a viable option in loudspeaker technology. Thanks must also go to David McBean for the tapped horn portion of Hornresp. I would give full credit to Tom in this case. It’s fine and dandy if you stumble across something interesting like Jensen did with the Transflex. However, it is He who figures it out and is able to apply it that gets the pay off.
The tapped horn is a very big deal to me. It obsoletes all large subwoofer horns. I'm so glad mine are now gone. It also makes Bruce's Seismic horn sub obsolete. We can now use a 10" $50 car subwoofer in an enclosure 1/4 the size and smoke what the Seismic horn sub does. The tapped horn is also much easier to DIY because there is no complex folding needed. You can go from spending almost $1000 and weeks on a traditional sub horn, or a few $100 at the most and a day or two for a tapped horn. That is very significant in my book.
Hi JLH - Please post details on your 80L TH. I love my 300L TH but there are times (and places) where it is just too big.
Here is an example of a $40 car subwoofer used in a tapped horn that can out perform the Seismic horn sub.
And here is another example for a $50 car subwoofer that can do it too!
Enough said!
And here is another example for a $50 car subwoofer that can do it too!
Enough said!
LR24 = Linkwitz - Riley filter with a 24dB/octave slope.Naudio said:
http://sound.westhost.com/project09.htm
Don Snyder said:
Here is my tapped horn that replaced my old front loaded subwoofer horns. These are about 42 liters each. So, I traded two 170 liter horns for two 42 liter horns. A good trade in anyone's book. Basically I just needed them to cover from 30Hz to 90Hz. With room gain, they are 3dB down at 31Hz. This is actually better than my old front loaded horns. Also note that I'm using custom 16 ohm versions of the B&C Speakers 8PE21 woofers.
JLH said:
Hi JLH,
To give credit where credit is due, the Jensen Transflex design was actually analysed in detail by Peter Tappan way back in 1958. His work is documented in the January 1959 JAES article - Analysis of a Low-Frequency Loudspeaker System".
The introduction to his paper reads:
"The purpose of this work has been to provide a theoretical analysis of the performance of the Jensen Transflex loudspeaker enclosure system, to devise from this analysis a method for selecting suitable enclosure dimensions for a given speaker and amplifier, and to verify the analysis by experiment".
The tapped horn model used in Hornresp is an extension of Peter Tappan's work.
Kind regards,
David
JLH
Would you repost those images separately, for the old and feable eyed among us.
From what I can see the curves look pretty good.
Paul
Would you repost those images separately, for the old and feable eyed among us.
From what I can see the curves look pretty good.
Paul
aceinc said:
Here is a bigger screen shot of the tapped horns. This is for the $40 Coustic car subwoofer.
Here is the one for the $50 MTX car subwoofer.
Have fun.
David McBean said:
Thanks for that information David. I assumed (bad monkey, very bad....) that since nobody did anything with it, they didn't understand it. If any of those guys are still alive today, they must be kicking theirselves for not running with the TH idea.
Rgs, JLH
David have you seen....
this patent by Arnold Klayman US # 5,177,329? Can this style of cabinet be designed with Hornresponse? Thanks.
this patent by Arnold Klayman US # 5,177,329? Can this style of cabinet be designed with Hornresponse? Thanks.