I have a question about how sensitive a back horn will really be in a real world application. I've always thought that if you have a back horn loaded driver that has a 96 db @ 1W/1M efficiency, you'd get that same efficiency in room with that sort of cabinet.
I was reading Art Dudley's online review of Ed Schilling's "The Horn" in Stereophile, the review of the older version that uses the Fostex FE108E Sigma driver (He currenly uses the FE126). The efficiency of this driver as per the Fostex spec sheet is 90 db @ 1W/1M, but when the speaker was actually measured, they got an efficiency rating of 94 db @ 1W/1M.
How is this so? I would think that with a 90 db efficient driver, you'd get a loudspeaker with the same efficiency. Can someone explain this to me? I don't rule out that maybe their measurements could be wrong, but say what you will about their editorial content and reviews, they should be able to make a fairly accurate measurement of this sort.
For those that are interested, here is the link to the measurements in the review:
http://stereophile.com/loudspeakerreviews/104horn/index4.html
I was reading Art Dudley's online review of Ed Schilling's "The Horn" in Stereophile, the review of the older version that uses the Fostex FE108E Sigma driver (He currenly uses the FE126). The efficiency of this driver as per the Fostex spec sheet is 90 db @ 1W/1M, but when the speaker was actually measured, they got an efficiency rating of 94 db @ 1W/1M.
How is this so? I would think that with a 90 db efficient driver, you'd get a loudspeaker with the same efficiency. Can someone explain this to me? I don't rule out that maybe their measurements could be wrong, but say what you will about their editorial content and reviews, they should be able to make a fairly accurate measurement of this sort.
For those that are interested, here is the link to the measurements in the review:
http://stereophile.com/loudspeakerreviews/104horn/index4.html
Sensitivity is measured on an IEC baffle. Baffle step diffraction typically takes up to 6dB off this. So it is not uncommon (in theory at least) to expect up to 6dB less and to knock the midrange off with the crossover.
Large baffles and back horns can attempt to reclaim the lost sensitivity. In the case of the back horn, you can get bass that is greater than the original sensitivity. But back horns don't help in the midrange. What good is loud bass if you don't have midrange to go with it?
If you only want bass, then all is good. If you design the horn to have just enough efficiency to give you the right level of bass (to rated sensitivity), that's OK too. You could also horn load the front, or you could let the horn be efficient and actively EQ the bass back down, scoring much more output from your amp.
Large baffles and back horns can attempt to reclaim the lost sensitivity. In the case of the back horn, you can get bass that is greater than the original sensitivity. But back horns don't help in the midrange. What good is loud bass if you don't have midrange to go with it?
If you only want bass, then all is good. If you design the horn to have just enough efficiency to give you the right level of bass (to rated sensitivity), that's OK too. You could also horn load the front, or you could let the horn be efficient and actively EQ the bass back down, scoring much more output from your amp.
This speaker has a small baffle, but it appears that the overall measured response is 94db at one watt input, while the driver itself specs at 90.
I understand that back loaded horns amplify the bass output of low Qts drivers, but how do you account for the increased efficiency from 90 to approximately 94db 1W/1M?
I understand that back loaded horns amplify the bass output of low Qts drivers, but how do you account for the increased efficiency from 90 to approximately 94db 1W/1M?
There appears to be a little peaking at the top end, perhaps some cone breakup. It wouldn't be hard to horn load the bass up to that level. I'm not certain if this is done here, but I feel it wouldn't be right to exploit that.
Rarely do I find that sensitivity lines up. How does a manufacturer come up with it?...I encourage a grain of salt.
Rarely do I find that sensitivity lines up. How does a manufacturer come up with it?...I encourage a grain of salt.
As I understand it the smallest baffle size for an IEC baffle is 1350 mm x 1650 mm (for an 8" driver - see http://www.rjbaudio.com/Audiofiles/FRDtools.html) or roughly 41" x 65". This is a (relatively) huge baffle so I can't agree with Indm that the differences in sensitivity related to the baffle step because you'd expect the opposite to what has been measured -- the driver in IEC baffle should be more sensitive. Unfortunately I can't explain the difference though.
Couple of additional points. Firstly -depends how the original driver sensitivity reading was made. Note, in the published plot of the 108ESigma for instance, the (on average) +4db rise above 2KHz - about 15KHz over the 90db nominal. And then it depends how the driver / cabinet sensitivity reading was made too.
You can boost a driver's sensitivity with a horn over it's passband, and above it's nominal sensitivity rating, though that's not a good idea generally as it will become bass-heavy. It all depends on how big the horn is. The bigger the horn is, in both length and mouth size, the more acoustic gain in the LF passband it will provide. The usual aim is to balance this out to best match the driver's typical sensitivity, but not always. People sometimes tune for a bit more gain to give a subjective boost to the LF, to add a little more weight.
As a general rule, go by the driver's sensitivity rating: if you get more, then fair enough.
You can boost a driver's sensitivity with a horn over it's passband, and above it's nominal sensitivity rating, though that's not a good idea generally as it will become bass-heavy. It all depends on how big the horn is. The bigger the horn is, in both length and mouth size, the more acoustic gain in the LF passband it will provide. The usual aim is to balance this out to best match the driver's typical sensitivity, but not always. People sometimes tune for a bit more gain to give a subjective boost to the LF, to add a little more weight.
As a general rule, go by the driver's sensitivity rating: if you get more, then fair enough.
BHD said:
I suppose I should have posted this on the Horn Shoppe forum, but since I don't own a pair of his speakers I didn't want to waste his time. I was just hoping that someone here could come up with an answer.
Thanks to all who took the time to try, though...
It's been my experience that Ed is always willing to invest his time in the education of a "future convert". It's at the very least, always an entertaining conversation, as long as you can stick to audio -just don't start talking politics.
Scottmoose said:Or things which make a loud bang, and vomit forth a small metallic projective with more than common force I gather.
there's nothing common about anything closely related to Ed
and a single projectile might be survivable
and come to think of it, be careful even when discussing audio
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.