New Scan-Speak D3004/660000 Distortion Tests

[Joe Rasmussen]

You may have noted that the D3004/660000 Tweeter (D30 for short) is still listed as NEW - it has not yet been referred by the "Revelator" moniker, although they say its research is based on the R2904 (R29 for short) Revelator.

www.tymphany.com/datasheet/printview.php?id=55

Being a consultant I have been loaned a pair of D30 and told that it is only one of two pairs here in Australia. They look like this:



The most amazing look is the rear:



The distortion tests, Sample A and Sample B, measure slightly oddly compared to any tweeter tests I've done before:

Sample A:



Sample B:



There is more than reasonable consistency between samples, high order distortion 3-5 are very well suppressed, but 2nd harmonic is rather prominent and curiously flat over much of the working range of the tweeter. Maybe this is on purpose and designed to give this tweeter a certain 'sound'?

Now compare this with my current fave tweeter the NEW Peerless HDS 810921:



High order is as good as the D30, but 2nd harmonic is much better suppressed. I would be interested in hearing others interpretations of these measurements.

The Peerless HDS deserves to be adopted by the DIY fraternity. I don't know what the cost of the D30 will be but I suspect double or maybe more than the HDS.

I have not yet heard the D30 but have been using the HDS in my Elsinore Project speakers for about a month now and they continue to delight and even amaze me.

Test conditions: Mic 125mm from flush mount, gated sinusoidal, reference level 100dB. Please note 100dB is @ 125mm which is equivalent to 82dB @ 1 Metre. The reference level (black line) should not be taken as the frequency response of the drive unit as 125mm is chosen to be near optimum to measure distortion and NOT frequency response. Also, 100dB at the mic is well within its overload margin. After all, we don't want to measure the mic's distortion, right?

Joe R.

[AKN]

Hi Joe Rasmussen,

Interresting, thanks for sharing your results.
Tweeters in question has also been tested by Zaph among many other tweeters.
I have just ordered the Peerless HDS, can hardly wait for them to arrive.

[tktran303]

Hi Joe,

Thanks for the measurments.

I'm not sure if it's been engineered to give relatively high H2, but if those measurements are accurate and anything to go by (I see no reason why not), the Aircirc should probably deliver "musical tube-like treble"

On the other hand, both John Krutke and Mark K's measurements did not show this phenomenon. All HD components were very low, including H2.

Mark K's findings-
http://www.markk.claub.net/Testing/S...eerlessHDS.htm

John Krutke's findings-
http://www.zaphaudio.com/tweetermishmash/compare.html

Btw do you know the production date of that pair of 6600 tweeters?

There was a brief hiatus in production, and there has been no North American stock for over 6 months. I'm wondering if you tested an early (06) sample or recent 07 sample, and even if there are even any differences.

[ShinOBIWAN]

Originally posted by tktran303

Btw do you know the production date of that pair of 6600 tweeters?

There was a brief hiatus in production, and there has been no North American stock for over 6 months. I'm wondering if you tested an early (06) sample or recent 07 sample, and even if there are even any differences.

Yep, Mark and John tested these in the middle of last year. I presume the samples they had were amongst the first. Although not bad results that Joe has posted they aren't inline with John's or Marks so I have to assume that the measurements shown here are of an inferior production model, not so controversial when you consider they went AWOL for 6 month due to what I've seen widely claimed as 'production issues'.

I don't have any reason to doubt either Joe's, Mark's or John's testing methodology and since the HDS, also included in the measurements shown in this thread, was used to draw a comparison that further reinforces the idea of latest 6600 not being the equal of the first batches released mid last year.

I was quite taken by the results that John and Mark posted and shortly afterwards I tried for a couple of months to source a pair with no success. I'd like to see John revisit his tests with a later production sample just to be sure but I'd say this looks to be merely a good dome tweeter now rather than the class leader it showed last year.

It bodes well for the HDS which is cheap and looks to be the new performance king in dome related circles.

I'll close by saying there's a lot of speculation on my part so I wouldn't pay too much attention, I'm just speaking out loud.

[Tenson]

My interpretation of the measurements would be that it doesn't really matter. You are not going to use either tweeter below about 1.5KHz and that would be with at least 4th order roll-off. Looking only from 1.5KHz up, they are about equal. The HDS is better from 2KHz down.

So yes, the HDS looks to be a brilliant tweeter, especially for the cost. I would use it if I was looking for a no-compromise dome tweeter.

Of course something like the RAAL Shinobiwan just got is even better again

Still I think a lot depends where your Xover is. IMO if you are crossing at more like 5KHz or up, there are many tweeters that have top-notch performance on par with the HDS and ScanSpeak for a lot less, like the Seas metal domes.

[ShinOBIWAN]

Originally posted by Tenson
Of course something like the RAAL Shinobiwan just got is even better again

Your going to get me into trouble

That's a definite IMO situation. Which pretty much secures immunity from those inclined otherwise.

[Tenson]

Yeah, but I happen to agree with you

[Joe Rasmussen]

Originally posted by tktran303
There was a brief hiatus in production, and there has been no North American stock for over 6 months. I'm wondering if you tested an early (06) sample or recent 07 sample, and even if there are even any differences.

Without revealing confidences, I was told in January there was a delay and they arrived in February as I understand it. The delay had something to do with the 'finishing' - whatever that means? So these are actual NEW production samples. Whether that explains the difference between earlier (and pre-production it seems) and these current units, that is a definite possibility.

I am very confident of the measurement. When I saw it first time I was struck by what I saw, so the second sample was useful. I repeated the measurement on both samples and also other tweeters, such as HDS, as 'controls' - and all was consistent.

What I wonder is this: Is it deliberate? If so, what will they sound like? And will future samples be alike? It may be months, but I am likely to get my hands on other samples, those questions will find an answer.

In the meantime I hope many here will get to try the Peerless HDS. I look forward to their comments.

Joe R.
.

[SY]

Clearly there's an asymmetry between push and pull. And it's pretty consistent with frequency above 200 Hz. Suspension? It would be interesting to see the detailed geometry, including the magnetic system.

edit: How does the 2nd HD vary with level? That could pin down the mechanism.

[Joe Rasmussen]

Originally posted by SY
Clearly there's an asymmetry between push and pull. And it's pretty consistent with frequency above 200 Hz. Suspension? It would be interesting to see the detailed geometry, including the magnetic system.

edit: How does the 2nd HD vary with level? That could pin down the mechanism.

Hi Stuart, OK will try different (lower?) level asap and report back. I have to watch out overloading the mic and also balance against ambient noise. With existing test conditions it is consistent to at least minus -80dB with relation to reference level 0dB = 100dBSPL seen by the mic.

BTW, the day after the sun came out and I enjoyed a whole day in Amsterdam, roaming the canals, taking pics with my trusty Nikon. The sun disappeared the following day - but I was on a train back to Copenhagen. So I got lucky.

Joe R.



................................