Mark K's Speaker Pages

...when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the state of Science...Lord Kelvin


Dayton RS28A

 

UPDATE 12/08

The Dayton RS28a has been retested since the ferrofluid has been removed. Follow this link to the new data for the nonferrofluid.

 

The data below pertains only to the older ferrofluid versions!

Usher 9950 and the new Dayton prototype. The Seas 27TDC and millenium will be used for comparison.

Frequency response

 

Impedance

NOTE-this impedance curve is suspect. The subsequent two production tweeters I received measure like the PE site, with a much lower Fs and, when I loaned jonmarsh my prototype, he got a curve like the PE site with that protype. Go figure. I suspect I had misaligned the VC or the felt was touching or, well who knows. I must have removed the VC ten times with the protype. Why do I post it? Well, that's what I originally got. I would go with the impedance curves that are on the PE site. If I have the time and energy, I will add an updated curve.

 

Off axis performance.

Usher prototype

 

Usher 9950

 

Waterfall

Dayton prototype

 

Usher 9950

 

Linear distortion

Comparison with millenium.

 

 

 

 

 

 

Nonlinear distortion

At each frequency, the Dayton prototype will be first, followed by the Usher 9950.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Some more interesting nonelinear distortion graphs

 

A comparison of distortion between the Dayton prototype and the Seas 27TDC at 1k

 

Now, a comparison of distortion between the Dayton prototype and the Seas 27TDC at 7k

I didn't reverse the graphs at 7k, because, well, there's no reason to. The 27TDC has modest, but extra distortion across the board.

 

Now, a comparison between the Dayton prototype and the Seas Millenium

First, at 1k.

 

 

Addendum--

If you look at my second set of Usher 9950 nonlinear graphs, they are slightly worse than the original one I tested. Even so, using my original 9950 data, which is more favorable to the 9950, the Dayton prototype represents an improvement. See below.

 

Conclusion

Another WOW! The Dayton prototype, which appears to be an Usher product, based on the 9950, but modified, appears to be an improvement over the original 9950, especially with respect to nonlinear distortion. From the above graphs, the Dayton/Usher metal dome has better nonlinear distortion across the board, compared to both the Seas 27TDC and millenium. You can compare these distortion spectra with the others on my site and it's clear that this Dayton is a winner. Whether or not it's the "best" depends on your weighting of the criteria, but it is certainly on the "A" list.  It should easily cross at 1.4k or below with a 4rth order or steeper crossover and is an excellent match for the rest of the RS series. I had misgivings about grills and diffusors, based on my recent testing of the LPG26T, but the Dayton prototype showed no significant problems with this. The FR graph/transient distortion is from 2k to 11k. It falters a bit around 15k, but not really any worse than the 27TDC or Millenium.

UPDATE 04/13/05. When I received the production pair I ran the tests again. One unit tested close to the prototype, though not quite as good. The other almost did, but it had notably more nonlinear distortion at 1 and 1.3k. All the units were fine at 1.5k and above. I'm not sure if this unit is an abberation or if there is some production variability. I will test the two I have coming. That will be a total of five samples. For now I would say, if you want to use these below 1.5k, consider testing them or sending them to me to be tested. This may be like the cpu overclocking thing. Some cpu's will oc more than others. PE rates the tweeter at 1.6k, which is certainly achievable with all the units. You may have to cherry pick if you want 1.2k.

UPDATE 05/04/05. I received a second prototype pair, which tested very well, in the same league as the original prototype. On the flip side, Zaph sent me a pair which performed horribly. I discussed this with Darren at PE, and he is aware of the unit to unit variability. He has implemented some in house QC checks to weed out any problem tweeters. Another pair is on it's way. I will update this page when I get them. If you really want to see the graphs, look here.

UPDATE 01/09/06. I received another pair for testing. Results here.

UPDATE 02/13/06. Yet 3 more pairs. Results here.