Klh 17

secondslc · Nov 28, 2008

Zilch said:
These KLH-17s may have had good bass when they were new, but they do not now, they did not require re-doping, which only made them worse, based upon my measurements of two pair now, and we need to rethink some common assumptions here.

You're my hero Zilch, proving the masses wrong with measured results. Kloss was not stupid when he made these speakers, he was just limited by technology and price. Later products from him got better.

Yeah, i re-doped my surrounds and convinced myself that it sounded better, but at least we got someone trying to find better drivers and make a true upgrade.
:thmbsp:

Zilch · Nov 28, 2008

secondslc said:
You're my hero Zilch, proving the masses wrong with measured results.

There's certainly also merit to the "Forget the measurements, trust your ears; it's the MUSIC, stupid!" approach, but that just never worked very well for me.

The successes of Indignia and EconoWave stand in testament to the validity of the numbers here. Many of those members who have built them have also discovered new pleasures in this alternative perspective on the hobby.

That's the bottom line, as I see it -- having fun and learning stuff, and anybody can do it. In the end, it IS all about the music.... :thmbsp:

Paratima · Dec 2, 2008

The numbers am what they am

KLH17 woofer response...

Well, I tried the canned approaches, but they were inconvenient, to say the least. One set of tones didn't have the selection I wanted. The other set was OK, but if I lost my place, my only option was to start over. This does NOT make for happy camping! :thumbsdn:

So finally, I fired up Audacity and cut my own speaker tester tone CD, where each tone is on its own track, and each track is self-identifying. That is to say, my voice, announcing the current selection of hertzes. Works! :thmbsp:

Image uploads are temporarily out of order on the forum, so I had to dink around with it considerably to get it to display off another site. Guess I could link to my xls file offsite, but that's more work. If anyone wants the actual numbers, they can ask. May not get them, but they can ask. The numbers am whatever they am.

Paratima · Dec 2, 2008

Test setup:

Zilch · Dec 2, 2008

Without looking back at my results, I'm reading that as -6 dB just under 60 Hz, and -10 dB at 50 Hz. Do you agree?

There's also ~2-3 dB of West Coast "boom" in the region of 80 - 200 Hz, apparently.

Excellent work! Who'd have thought this could be done with a R/S SPL meter?

Try it on some other closed-box speakers, if you have any to compare. :yes:

Note: I have tamed the Goldwood rising response, and am now working on the HF comp and attenuation for Selenium on the PE waveguide. These are sounding mighty nice, now, I must say.... :thmbsp:

Paratima · Dec 2, 2008

Zilch said:
Without looking back at my results, I'm reading that as -6 dB just under 60 Hz, and -10 dB at 50 Hz. Do you agree?

Yup, exactly.

There's also ~2-3 dB of West Coast "boom" in the region of 80 - 200 Hz, apparently.

Excellent work! Who'd have thought this could be done with a R/S SPL meter? :thmbsp:

I'm gettin' sucked in, ain't I? :headscrat Or am I already there?

Try it on some other closed-box speakers, if you have any to compare.... :yes:

Stand by.

Zilch · Dec 2, 2008

First time I did this sort of thing was with a pal on another forum half-way around the world. We were both stunned by producing the same results with the same conclusions in a similar investigation.

It's no BS. Stuff is what it is, and that is ascertainable. Jack, Skywave, and I can reproduce each others' results so well it's difficult to tell who's is which.

[Except for the NYC accent in Skywave's....

]

jackgiff · Dec 2, 2008

What did you do to tame the Goldwood's rising response? How much cabinet volume, and what is the tuning frequency?

I built speakers for my nephew using the GW-10PC, and thought they were pretty nice for the price.

Zilch · Dec 2, 2008

Lowpass is 3.0 mH/6.8 or 7.5 uF, not sure which 'til the HF is done.

KLH-17 calcs to 0.892 cuft, gross.

I'm tuned to 33.5 Hz with 2 x 1.5" ports, 8.5" long.

F3 = 57 Hz, F6 = 43 Hz, F10 = 31.5 Hz, theoretically.

Tuning didn't measure that, though, last time I did it.

I'll verify all that before finalizing.... :yes:

RayW · Dec 2, 2008

Paratima said:
I'm gettin' sucked in, ain't I? :headscrat Or am I already there?

Doesn't that SLM have an audio out? If so, then download Speaker Workshop (http://www.speakerworkshop.com/) and you should be able to do MLS testing instead of the manual version. Then you'll really be on your way. Come to the dark side...we have cookies.

Ray

Zilch · Dec 2, 2008

RayW said:
Come to the dark side...we have cookies.

:lmao:

Paratima · Dec 2, 2008

SpeakerWorkshop downloaded, Master.

Zilch · Dec 3, 2008

Leveling the rising response of the Goldwood GW-10PC/8 in this alignment lowered its sensitivity to ~85 dB, and the Selenium D220Ti required ~24 dB total HF comp and attenuation using the E'Wave crossover at 1.88 kHz to balance with it.

Before:

After:

Full-range Sinusoidal:

Low efficiency, but yes, bass, fully 20 Hz lower than the stock KLH-17, plus E'Wave 20 kHz HF.... :thmbsp:

Schematic and Sim:

Paratima · Dec 3, 2008

Rather than bringing that high-efficiency tweeter to its knees, here is yet another idea:

http://www.parts-express.com/pe/showdetl.cfm?Partnumber=275-040

...start with an admirably flat and noted for low distortion yet lower efficiency tweeter, or, from the Indignia...

http://www.parts-express.com/pe/showdetl.cfm?Partnumber=275-070

Just thinking, mind you. :scratch2: Not that I have any really clear idea about such things...

Zilch · Dec 3, 2008

Yes, BUT:

1) Compression driver dynamics,
2) Controlled directivity,
3) XO below 2 kHz and actually perform there, and
4) Ultra-low distortion.

If these were biamped, would there be issues with running the HF on flea power and no passive attenuation, just HF comp?

Another option would be higher efficiency woofers and a sub, or multiples, but the objective here is to get real bass from these small boxes themselves. KLH promoted these as having good bass AND efficiency. Not by modern standards, apparently.... :no:

Paratima · Dec 3, 2008

Oh. :tresbon:

So, you gonna show us some pix of the fancy port tubes you rigged up?:scratch2: I seem to recall something about 2 x 1.5 inches by 8.5 inches and angled down.

And how do they SOUND?

secondslc · Dec 3, 2008

Zilch said:
Yes, BUT:

1) Compression driver dynamics,
2) Controlled directivity,
3) XO below 2 kHz and actually perform there, and
4) Ultra-low distortion.

If these were biamped, would there be issues with running the HF on flea power and no passive attenuation, just HF comp?

Another option would be higher efficiency woofers and a sub, or multiples, but the objective here is to get real bass from these small boxes themselves. KLH promoted these as having good bass AND efficiency. Not by modern standards, apparently....

So, are we porting these suckers? Or will you give us a ported and non porte d option?

Zilch · Dec 3, 2008

secondslc said:
So, are we porting these suckers? Or will you give us a ported and non-ported option?

I have them ported, but I want to verify the tuning vs. sims, and make sure the other Goldwood performs the same, i.e., that the one I've designed with is not anomalous in some respect. I should probably swap Seleniums, as well.

I'll try to do that later tonight. Right now, I'm listening with them cranked to see how hot R2 gets.... :yes:

Paratima said:
And how do they SOUND?

You wouldn't want me to deprive you of the thrill in the discovery, now, would you? :nono:

THAT, and whether you like it or not, are for you to determine.... :yes:

Zilch · Dec 4, 2008

Two Systems:

The two Goldwood woofers are not perfectly matched, but are within ~1 dB of each other in sensitivity.

#1 Selenium D220Ti is 9 months old, #2 is brand new. Close enough. Both systems are +/- 1.5 dB from 500 Hz - 20 kHz. :thmbsp:

Thumbnails below should answer questions about the port/duct configuration, but BB6P says I'm tuned too low at 30 Hz, and that I should reduce the length by half and remove about half of the fiberglass I have in there. I'll try that, so let's not lock into what I presently have for venting and damping just yet.

I have not yet resolved the issue of why they do not measure nearfield anything like how BB6P says they should behave closed-box. It's as if the drivers do not match the database parameters. I'll have to measure these drivers' T/S parameters myself to see if they match up.

I'ts weird. I don't measure or hear any difference, ported vs. closed, as if the driver is totally dominating this alignment at low frequencies. No, the elbows should not matter one whit.... :dunno:

RayW · Dec 4, 2008

Zilch said:
3) XO below 2 kHz and actually perform there, and

I've used a couple of domes around 1500Hz. They won't part your hair like the EWG but they did an admirable job.

Zilch said:
I have not yet resolved the issue of why they do not measure nearfield anything like how BB6P says they should behave closed-box. It's as if the drivers do not match the database parameters. I'll have to measure these drivers' T/S parameters myself to see if they match up.

Where did you get the numbers? PE? Goldwood spec sheet? I've seen some stuff on mfg spec sheets that was so far off it looked like a different driver.

I'ts weird. I don't measure or hear any difference, ported vs. closed, as if the driver is totally dominating this alignment at low frequencies. No, the elbows should not matter one whit....

That is weird. I would think you'd have to have a monstrously high Qt for the driver to not show an effect from a box that size.

A note on elbows (not for you, Z, for folks who might not know): The kind of elbow Zilch is using will work fine. You need to keep the cross sectional area of the port constant. The small elbows that have a sharp corner on the inside can cause anomolies (turbulence) at that point. Use ones like Z.

Ray

Klh 17

mmmmmarantz

Curve Junkie

E'Wavin'

E'Wavin'

Curve Junkie

E'Wavin'

Curve Junkie

Well-Known Member

Curve Junkie

Parrothead with a badge

Curve Junkie

E'Wavin'

Attachments

Curve Junkie

Attachments

E'Wavin'

Curve Junkie

E'Wavin'

mmmmmarantz

Curve Junkie

Curve Junkie

Attachments

Parrothead with a badge

Similar threads