Thiele Small Parameters for GW-8004?

[Ludo]

Has anyone measured the Thiele Small parameters for the Goldwood GW-8004 coaxial driver? It looks like an interesting driver for a low cost, 2-way, sealed box system.

[Ludo]

Re: Thiele Small Parameters for GW-8004?

So, out of curiosity and despite my prior dissapointments with bargain-basement drivers, I plunked down my hard earned cash for a couple of GW-8004s.

The drivers came nicely packaged and the stickers on the boxes read: black steel frame, rubber surround, poly-mica cone, 20 oz magnet, 1.4" aluminum voice coil, 190W power handling, 30Hz to 20kHz frequency response, 88dB/W/m sensitivity, 8 ohm impedance, and 2" aluminum/mylar tweeter. From the photo, you can see that the tweeter is directly wired across the woofer without any x-over.

Measured TS parameters:

Driver 1 (2)
FS = 68 (68) Hz
Qm = 5.2 (5.4)
Qe = 2.4 (2.7)
Re = 7.2 (7.1) ohms
Le = 0.37 (0.35) mH

Listening impressions: sorry but straight out of the box these things sound like c#%p!

Next steps:

1) Looks like a sealed box is the only way to go with these, but first I'll need to measure Vas. Maybe start with 1/2 cu-ft box?

2) Cut the tweeter leads from the woofer cone and measure the impedance and frequency response of both drivers independently.

3) Hook them up to a 2-way, active x-over to see if there is any hope in spending time on a passive x-over design.

4) Measure the x-max and run some box sims to figure out the "optimum" enclosure.

Can a decent sound be extacted from these puppies? What will the x-over look like? How many components will be needed? Will the bass extension and quality be reasonable. Will the box be humongous? Am I wasting my time?

Louis

[Paul Carmody]

Re: Thiele Small Parameters for GW-8004?

Can a decent sound be extacted from these puppies? What will the x-over look like? How many components will be needed? Will the bass extension and quality be reasonable. Will the box be humongous? Am I wasting my time?

One more question...
Could you repeat the design on another pair of the same drivers, or would the specs be grossly off?

[Ludo]

Re: Thiele Small Parameters for GW-8004?

One more question...
Could you repeat the design on another pair of the same drivers, or would the specs be grossly off?

Not sure? Do the TS specs on these types of Goldwood drivers vary much from lot to lot? How about the frequency response?

My limited experience with Goldwood piezo tweeters and a couple of Pyramid titanium horn tweeters was that there was noticeable frequency reponse difference unit-to-unit even comming from the same box.

Finally figured out how to convert the WT3 screen shot to .jpg so here are the impedance and phase plots.

Louis

[Chris Roemer]

Re: Thiele Small Parameters for GW-8004?

Not sure? Do the TS specs on these types of Goldwood drivers vary much from lot to lot? How about the frequency response?

My limited experience with Goldwood piezo tweeters and a couple of Pyramid titanium horn tweeters was that there was noticeable frequency reponse difference unit-to-unit even comming from the same box.

Finally figured out how to convert the WT3 screen shot to .jpg so here are the impedance and phase plots.

Louis

With an Fs near 70 and a Q around 1.6, they should have about a +5dB around 70Hz with an F3 between 50 and 60. You can see that they're crossed at 3k. You bought them as coaxials, not sure why you'd cut them apart now. Obviously they'll be weak in any real bass, but just get a sub.

Where did you buy these anyhow, I don't see them in PE's catalog.

Chris

[Ludo]

Re: Thiele Small Parameters for GW-8004?

With an Fs near 70 and a Q around 1.6, they should have about a +5dB around 70Hz with an F3 between 50 and 60. You can see that they're crossed at 3k. You bought them as coaxials, not sure why you'd cut them apart now. Obviously they'll be weak in any real bass, but just get a sub.

Where did you buy these anyhow, I don't see them in PE's catalog.

Chris

Chris,

I buught them at PE: http://www.parts-express.com/pe/show...number=290-378

I want to cut them apart because, as far as I can tell, the mylar tweeter is just wired in parallel with the woofer without any x-over; I should probably ohm them out to make sure. I suspect this may be one of the reasons they sound so horrible, either because the tweeter is getting driven hard below its operating frequency range or the woofer has some breakup modes at higher frequencies. As they are now, these drivers may be ok for a paging system but the are unlistenable for music.

Louis

[Chris Roemer]

Re: Thiele Small Parameters for GW-8004?

Chris,

I buught them at PE: http://www.parts-express.com/pe/show...number=290-378

I want to cut them apart because, as far as I can tell, the mylar tweeter is just wired in parallel with the woofer without any x-over; I should probably ohm them out to make sure. I suspect this may be one of the reasons they sound so horrible, either because the tweeter is getting driven hard below its operating frequency range or the woofer has some breakup modes at higher frequencies. As they are now, these drivers may be ok for a paging system but the are unlistenable for music.

Louis

Well, by the looks of your WT3 impedance plot, they ARE crossed at 3k. That "peak" in impedance looks just like any other tm pair with a crossover net with an Fc of 3k. I guess MAYBE just parallelling a tweeter and woofer COULD look like that. (Let's take a look)

NOPE. Unless they're 16ohm drivers, simply parallelling an 8n woof & tweet would drop your ohms down to around 4 near 200 cps or so. I think they're crossed, you just can't see it.

Chris

[Ludo]

Re: Thiele Small Parameters for GW-8004?

Well, by the looks of your WT3 impedance plot, they ARE crossed at 3k. That "peak" in impedance looks just like any other tm pair with a crossover net with an Fc of 3k. I guess MAYBE just parallelling a tweeter and woofer COULD look like that. (Let's take a look)

NOPE. Unless they're 16ohm drivers, simply parallelling an 8n woof & tweet would drop your ohms down to around 4 near 200 cps or so. I think they're crossed, you just can't see it.

Chris

Chris,

You're absolutely correct. There is a small cap hidden behind the faceplate of the tweeter for the high pass. I'll take some FR measurements, hopefully later this weekend (need to work on my daughter's car first), and post them.

Louis

[Ludo]

Re: Thiele Small Parameters for GW-8004?

Well this is looking a lot more promissing than I first thought. After 2 days worth of burn-in at 30Hz and cutting the tweeter leads on one of the units at the junction with the woofer cone, the Fs dropped from 68 to 47Hz and the Qts dropped from 1.6 to 1.2!

I also made some quick frequency response measurements with the driver still not in a box but just a bunch of towels stuffed against the back side of the basket. The frequency response of the woofer is very smooth and rolls off gradually above 3kHz. The tweeter response drops off below 3.5kHz and is flat out to 20kHz; not too shabby.

Then I hooked everything up to a 2-way, 4th-order, active x-over set to 3.5kHz and spent 30 minutes listening. The bass, as expected, was very subdued without a box but there was definately something nice there worth pursuing. The midrange was very clear and without any signs of graininess. The trebble was also full and precise. My overall impression is that the original sound of this driver was severely compromised by the simple, first-order high pass used on the mylar tweeter.

The other good news is that the tweeter assembly can be easily removed with a single screw allowing access to both tweeter leads for a complete x-over upgrade; possibly 3d-order? Not sure if there is any benefit in bandlimiting the signal to the woofer yet, but that is something to try down the road.

I guess the next step is to try out a passive x-over for the tweeter but I really should get the driver into a box first. :o

Louis

[Taterworks]

Re: Thiele Small Parameters for GW-8004?

Even if you don't attempt to lowpass the woofer, you might still want to consider the addition of a series notch filter to control cone breakup effects. You'd need a measured response curve to do this properly, however.

I admire your willingness to experiment.

[Ludo]

Re: Thiele Small Parameters for GW-8004?

Even if you don't attempt to lowpass the woofer, you might still want to consider the addition of a series notch filter to control cone breakup effects. You'd need a measured response curve to do this properly, however.

Yup, I took a quick frequency response measurement but didn't save it. It looked real smooth except for some low level stuff between 6 and 8kHz.

I admire your willingness to experiment.

Thanks, I am willing to make a few mistakes and hopefully learn something in the process. It's making mistakes on cabinets that really scares me!

By the way here is the tweeter impedance curve without the cap.

Louis

[Ludo]

Re: Thiele Small Parameters for GW-8004?

Finally got the Goldwood coaxial in a box and made some additional measurements. This is with the tweeter leads cut from the woofer cone which had the benefit of reducing the Fs and Qt as I mentioned in a previous post.

So here are the specs for one modified unit:

Fs = 47 Hz
Qe = 1.7
Qm = 4.8
Re = 7.1 ohm
Le = 0.61 mH
Vas = 35 l
Sd = 214 cm^2
X-max = 2.75 mm

I've attached the frequency responses for the woofer, with and without box stuffing, and for the tweeter with a 22uF capacitor. The system response is with the stock 3.3uF capacitor for the tweeter. The roverall esponse is not very flat but at the same time it is not terrible. The cabinet stuffing definately helps tame some of the bass boominess and this could be tuned down a bit by doubling the volume of the 0.44cu-ft box I was using for the testing.

The sound of this coaxial is pretty much dominated by the forwardness of the upper midrange and harshness from the tweeter, but although most would not consider it high fidelity, it is relatively easy to listen to; ok I only tried it for 15 minutes. I think the midrange and trebble shortcomings can be cured based on my experimentation with an active x-over but I'm not sure how to tame the bass lump?

Louis