New Owner! FR question

[nickmo]

So far I love my Sierras! I do have one question, though. Why is it that the frequency response chart that ships with each Sierra doesn't match the frequency response listed on the Sierra-1 product specs page?
For example my center channel is ±3 from 100Hz, ±4 from 80Hz, ±6.5 from 40Hz according to the graph that came with the speaker.Am I reading the graph wrong? Or do I not understand what "44Hz - 22kHz ± 3dB" means? I think it means that between 44Hz and 22kHz there is no more than a six dB difference.
Here is the graph:
http://gallery.me.com/swensonn/100197/P1040890/web.jpg

Thanks,
Nick

[curtis]

The individual measurements that come with the speakers do not show the bass response because there is not an easy/accurate way measure below 250hz when they are assembled.

See that second to last line on the page you posted?

[nickmo]

Ha! I knew it was me being anxious!:D I could hear great bass while I was listening to them, but I couldn't the graph out of my head. Thanks, Curtis!

nick

[davef]

So far I love my Sierras! I do have one question, though. Why is it that the frequency response chart that ships with each Sierra doesn't match the frequency response listed on the Sierra-1 product specs page?
For example my center channel is ±3 from 100Hz, ±4 from 80Hz, ±6.5 from 40Hz according to the graph that came with the speaker.Am I reading the graph wrong? Or do I not understand what "44Hz - 22kHz ± 3dB" means? I think it means that between 44Hz and 22kHz there is no more than a six dB difference.

An excellent and extremely valid question...

If you notice on the frequency response graphs included with your speakers, it states directly on the graphs that the measurements are only accurate down to approximately 250Hz. Measurements below this frequency do not accurately reflect performance (not even close).

Taking full range anechoic measurements on a loudspeaker production line is a practical impossibility. The speaker must either be measured in an anechoic chamber or measured by ground plane technique, thus requiring the speaker to be placed outside, in a large parking lot with no reflections (any obstacles) in at least a 30 foot radius around the speaker.

We could also use a close mic technique which would require us to measure the output of the woofer and the port and then sum the two using a weighted average technique, then "splice" this data back into the overall measurement. This would require us to move our production line measurement microphone multiple times for each speaker tested and position it with extreme accuracy each time. Also not logistically possible for a production line -- and having dropped a few microphones in my day at about $1000 a capsule to replace, not something I want to risk many times each day :o

A 20Hz wavelength is almost 60 feet long... As you can see, it is impossible for us to include full range anechoic measurements for every Sierra-1 speaker we ship (although we would certainly like to) Few loudspeaker companies these days even bother to measure every loudspeaker they ship…

Enjoy!!

[azanon]

See that second to last line on the page you posted?

Yeah.

So why not remove the disclaimer, and make the graph stop at 250hz since the information below that frequency has no usefulness other than being misleading for those who miss the disclaimer (such as the OP)? Being the computer age and all, surely those graphs are modifiable with whatever software you have to account for the limited measuring techniques for the assembly line.

Also, are the Sierra's the only ones that come with graphs? This is the only Ascend speaker I haven't bought yet, and I've never received a graph.

[nickmo]

Thanks for the reply, Mr. Fabrikant. I totally missed the disclaimer. Your speakers sound wonderfull. I originally purchased them for home theater, but I keep digging out old music just to see how it sounds on the Sierras:). The top entries in my Netflix queue have all been replaced by various live concert recordings on Blu Ray:). I can't get over how AMAZING music sounds through the Sierras.

Thanks again for your customer service, hard work and brilliant speaker design!
Nick

[scape]

So Ascend, or maybe just you with those expensive mics haha, measure for the whole spectrum the final production unit before it is put into mass manufacturing-- and then each speaker, or batch, that is manufactured gets a simple mic test from the FR you described above? just curious how you get the measurments for the website. this probably also explains why the subwoofer measurements are not here yet, it's probably a real pain in the... haha

[davef]

So Ascend, or maybe just you with those expensive mics haha, measure for the whole spectrum the final production unit before it is put into mass manufacturing-- and then each speaker, or batch, that is manufactured gets a simple mic test from the FR you described above?

Sorry -- I don't follow you on this :confused:


just curious how you get the measurments for the website. this probably also explains why the subwoofer measurements are not here yet, it's probably a real pain in the... haha

Full range measurements for our loudspeakers (as posted on our website) are measured by one of the various methods I posted above


Taking full range anechoic measurements on a loudspeaker production line is a practical impossibility. The speaker must either be measured in an anechoic chamber or measured by ground plane technique, thus requiring the speaker to be placed outside, in a large parking lot with no reflections (any obstacles) in at least a 30 foot radius around the speaker.

We could also use a close mic technique which would require us to measure the output of the woofer and the port and then sum the two using a weighted average technique, then "splice" this data back into the overall measurement.

The Sierra-1 low-end response was measured by ground-plane technique and then spliced into the upper range frequency response measurement - thus resulting in what is known as a quasi-anechoic full-range frequency response measurement.

[GirgleMirt]

Originally Posted by scape
So Ascend, or maybe just you with those expensive mics haha, measure for the whole spectrum the final production unit before it is put into mass manufacturing-- and then each speaker, or batch, that is manufactured gets a simple mic test from the FR you described above?
Sorry -- I don't follow you on this :confused:


Yeah what I think he meant was that during the design/conception phase, full measurements are taken, but once the design is concluded and actual manufacturing starts, the only measurements are the ones you provide with the Sierras.

In that case I'd say that's correct. You don't really need to take full measurements for each sample because you just want to check if the tweeter, woofer, crossover, etc are performing normally, and if something isn't, you'll definitely see it in the FR posted. It's really a QA thing: Quality assurance. If the woofer or crossover are for some reason messed up, you'll see it in the measurements, you really don't need the full measurements just to make sure every speaker is functioning as it should. :)

Ex: Don't remember exactly what was the crossover point, but it's somewhere at like 1800hz, so what you see on the right of ~1800hz is the tweeter, and on the left is the woofer, so if one is not working right, or the crossover is messed up, you'll see it on the graph. The port output is not measured because it's constant given woofer/enclosure, and if we know the woofer is working as it should, the bass from the port will be constant.

[scape]

Yeah what I think he meant was

yes, you put it much more clearly-- long days at work are melting my brain, haha

[davef]

Yeah what I think he meant was that during the design/conception phase, full measurements are taken, but once the design is concluded and actual manufacturing starts, the only measurements are the ones you provide with the Sierras..

Got it - and the above statement is 100% correct.



In that case I'd say that's correct. You don't really need to take full measurements for each sample because you just want to check if the tweeter, woofer, crossover, etc are performing normally, and if something isn't, you'll definitely see it in the FR posted. It's really a QA thing: Quality assurance. If the woofer or crossover are for some reason messed up, you'll see it in the measurements, you really don't need the full measurements just to make sure every speaker is functioning as it should.

Very well said and entirely true. Although any potential problems in the low frequency response of the speaker are typically revealed in the impedance measurement we take but including a printout of this measurement with the speakers would probably cause even more confusion.


The port output is not measured because it's constant given woofer/enclosure, and if we know the woofer is working as it should, the bass from the port will be constant.

The port tube is a constant and if the woofer is operating correctly, the port tube will indeed be operating properly. Additional QA steps are taken such as rub and buzz testing on top of sine wave sweeps -- these are meant to reveal mechanical problems such as air-leaks or voice coil rubs which are difficult to detect by measurement in a production line setting.

[davef]

Yeah.

So why not remove the disclaimer, and make the graph stop at 250hz since the information below that frequency has no usefulness other than being misleading for those who miss the disclaimer (such as the OP)? Being the computer age and all, surely those graphs are modifiable with whatever software you have to account for the limited measuring techniques for the assembly line.

We could remove that section of the graph but I am not sure if this will help since the graphs would then look nothing like a typical frequency response measurement and might cause even more confusion.

One option I have considered, and am still considering, is to simply splice in the average low frequency response of the Sierra-1 so that the response is then full range. Problem with this is that doing so is somewhat complex, but not beyond our abilities, but doing so also presents an ethical dilemma for me as part of the response is not the actual measurement of the speaker being tested. I know that nobody would ever question it but it just doesn't feel right to me :(



Also, are the Sierra's the only ones that come with graphs? This is the only Ascend speaker I haven't bought yet, and I've never received a graph.

Yes, these are the only speakers that we do include printouts of the actual measurements. However, I have included printouts of our other loudspeakers when specifically requested. The printouts do add additional processing time for each pair of speakers and this does add additional costs so it is not a cost-effective option for all of our products.


Thanks for the reply, Mr. Fabrikant. I totally missed the disclaimer. Your speakers sound wonderfull. I originally purchased them for home theater, but I keep digging out old music just to see how it sounds on the Sierras:). The top entries in my Netflix queue have all been replaced by various live concert recordings on Blu Ray:). I can't get over how AMAZING music sounds through the Sierras.

Thanks again for your customer service, hard work and brilliant speaker design!
Nick

Thanks Nick -- I am very pleased that you are enjoying the speakers and the fact that you are now listening to more music is the biggest compliment we could ever receive. Thank you!! :D