RS28AS vs. RS28A, interesting finding

[Paul Carmody]

Not long ago, someone asked me if the RS28AS (that is, the shielded one) was compatible with the Core project. I wasn't sure, because it seems to me that whenever someone asks about swapping in the shielded for unshielded, people seem to say that the two are different, and crossover adjustments would have to be made. Well, this guy was nice enough to lend me a pair to test head-to-head, and here are the results, which I personally find surprising.

First, take a look at the shielded and unshielded tweeters' frequency response. I measured both of these in the Core cabinet, one after the other. I made sure not to move the mic at all. The two are overlaid. The unshielded is in black, the shielded is in red:
http://techtalk.parts-express.com/attachment.php?attachmentid=16180&stc=1&d=1306814570

Seems awful darn close to me. So let's look at the impedance. Same as before: unshielded in black, shielded in red.
http://techtalk.parts-express.com/attachment.php?attachmentid=16181&stc=1&d=1306814544

Finally, let's look at how the actual speaker works using either tweeter. I don't know about you, but this looks like a drop-in replacement to me.
http://techtalk.parts-express.com/attachment.php?attachmentid=16182&stc=1&d=1306814924

And just to make sure I wans't lent some fluke shielded tweeters, I tested them against one another, and they measured very close.
http://techtalk.parts-express.com/attachment.php?attachmentid=16183&stc=1&d=1306815027
http://techtalk.parts-express.com/attachment.php?attachmentid=16184&stc=1&d=1306815033

So, make of that what you will, but if you ask me I'd say that means that the unshielded and shielded tweeter are interchangeable. As I said, I found this surprising, because it seems to me as though this is not the commonly-held belief around the internets.

[Applesauce]

As I said, I found this surprising, because it seems to me as though this is not the commonly-held belief around the internets.

I was also solidly under the impression that they were at least somewhat - if not very - different drivers.

Thanks for this information. Is it possible this is a generational thing? i.e., an early run differed from a later run?

[Jeff B.]

Not long ago, someone asked me if the RS28AS (that is, the shielded one) was compatible with the Core project. I wasn't sure, because it seems to me that whenever someone asks about swapping in the shielded for unshielded, people seem to say that the two are different, and crossover adjustments would have to be made. Well, this guy was nice enough to lend me a pair to test head-to-head, and here are the results, which I personally find surprising.

First, take a look at the shielded and unshielded tweeters' frequency response. I measured both of these in the Core cabinet, one after the other. I made sure not to move the mic at all. The two are overlaid. The unshielded is in black, the shielded is in red:
http://techtalk.parts-express.com/attachment.php?attachmentid=16180&stc=1&d=1306814570

Seems awful darn close to me. So let's look at the impedance. Same as before: unshielded in black, shielded in red.
http://techtalk.parts-express.com/attachment.php?attachmentid=16181&stc=1&d=1306814544

Finally, let's look at how the actual speaker works using either tweeter. I don't know about you, but this looks like a drop-in replacement to me.
http://techtalk.parts-express.com/attachment.php?attachmentid=16182&stc=1&d=1306814924

And just to make sure I wans't lent some fluke shielded tweeters, I tested them against one another, and they measured very close.
http://techtalk.parts-express.com/attachment.php?attachmentid=16183&stc=1&d=1306815027
http://techtalk.parts-express.com/attachment.php?attachmentid=16184&stc=1&d=1306815033

So, make of that what you will, but if you ask me I'd say that means that the unshielded and shielded tweeter are interchangeable. As I said, I found this surprising, because it seems to me as though this is not the commonly-held belief around the internets.

Thanks Paul. That is very helpful. I have used both tweeters, but not at the same time and in different boxes, so I had no reference to tell how close they were. The plulished curves have a different Q with quite a bit of difference at the knee, but your plots don't show that to be an issue. This means that some designs may have some new life.

Jeff B.

[turn2]

Just out of curiosity, would it be a big bother to overlay the RS28F-04 onto those graphs?

[aarond]

Just out of curiosity, would it be a big bother to overlay the RS28F-04 onto those graphs?

Yeah, I thinking the same thing. The response and impedence graphs would be very interesting.

[arlis_1957@yahoo.com]

thanks paul. i didn't know that the a would go that low.

[Wolf]

thanks paul. i didn't know that the a would go that low.

Just because it measures that low, does not mean it cleanly does so. Lowest I've seen it is the 1.2kHz in Mark K's project with the RS225.

Later,
Wolf

[critofur]

Paul, good find - a little surprising considering how different the spec sheets looked, back when I looked at them long long ago...

Jeff, when you quote a message you don't have to include all the photos, thread is a lot neater if you don't. The only reason to do so would be if you replying lots of pages later and wanted to bump the pics up to the end...


Just out of curiosity, would it be a big bother to overlay the RS28F-04 onto those graphs?

The A and F (and now we know AS) are definately drop-in swappable at the crossover frequency range, with the difference being the F having a slight rise at the top end (which most people can't hear?).