RS28AS or VIFA XT25TG30

[millers]

Which one is better? Are the vifa worth $100 a pair or not? Thanks, Scott

[Pete Schumacher ®]

> Which one is better? Are the vifa worth $100
> a pair or not? Thanks, Scott

You'd be hard pressed to find a much better tweeter than the RS28, regardless of price. I'd spend the money on the Dayton's. They also look great.

[scottsehlin]

IMHO, both are world-class tweeters. The Vifa is a little more tricky to work with, while the Dayton crosses low easily and is generally well behaved with very low distortion. The Vifa can be more delicate and extended in the extreme high frequencies if it happens to work in your application.

If we knew a little more about your application, we might be able to advise better.

[Paul Carmody]

Sorry to **** in, but as time goes on, the more I want to try using the XT-25 (despite its juvenile-innuendo-snicker-inducing looks). From what I have gathered, there is something tricky to getting it to filter correctly. However, I don't know what that is. What happens if it is not filtered correctly?

I do know that both you (Scott) and Wolf used it in a series XO. Since I don't find myself headed down that road for a few more projects--still trying to grapple parallel XO--what does a designer need to know to make it work in a parallel XO?

-Paul

> IMHO, both are world-class tweeters. The
> Vifa is a little more tricky to work with,
> while the Dayton crosses low easily and is
> generally well behaved with very low
> distortion. The Vifa can be more delicate
> and extended in the extreme high frequencies
> if it happens to work in your application.

> If we knew a little more about your
> application, we might be able to advise
> better.

[Pete Schumacher ®]

> Sorry to **** in, but as time goes on, the
> more I want to try using the XT-25 (despite
> its juvenile-innuendo-snicker-inducing
> looks). From what I have gathered, there is
> something tricky to getting it to filter
> correctly. However, I don't know what that
> is. What happens if it is not filtered
> correctly?

> I do know that both you (Scott) and Wolf
> used it in a series XO. Since I don't find
> myself headed down that road for a few more
> projects--still trying to grapple parallel
> XO--what does a designer need to know to
> make it work in a parallel XO?

> -Paul

Below that, 2nd harmonic distortion rises pretty rapidly. The impedance bump is low in frequency, but pronounced. You may need an impedance conjugate filter depending on your design slopes.

[dthomas]

> Sorry to **** in, but as time goes on, the
> more I want to try using the XT-25 (despite
> its juvenile-innuendo-snicker-inducing
> looks). From what I have gathered, there is
> something tricky to getting it to filter
> correctly. However, I don't know what that
> is. What happens if it is not filtered
> correctly?

> I do know that both you (Scott) and Wolf
> used it in a series XO. Since I don't find
> myself headed down that road for a few more
> projects--still trying to grapple parallel
> XO--what does a designer need to know to
> make it work in a parallel XO?

> -Paul

I like that tweeter but it wants to cross at about 2200 to 2500 Hz. If you tame the impedance rise at fs it is no harder to use than any other tweeter.

It is a nice sounding tweeter, very detailed and has great transient response for a fabric dome but still keeps that airy sound. Definitely a tweeter worth considering. If you mate it with a smooth midbass that can cross at 2500Hz easily it makes a great sounding tweeter.

Dave

[Paul Carmody]

"Impedance conjugate filter" = zobel? Or is it something else altogether?
What is it, and why does the tweeter need it?

> Below that, 2nd harmonic distortion rises
> pretty rapidly. The impedance bump is low in
> frequency, but pronounced. You may need an
> impedance conjugate filter depending on your
> design slopes.

[dthomas]

> "Impedance conjugate filter" =
> zobel? Or is it something else altogether?
> What is it, and why does the tweeter need
> it?

A zobel is used to flatten rising impedance as frequency goes up.

A conjugate network is used to flatten the rise of a drivers impedance at resonance. This typically only done to tweeters. In the case of a tweeter you can design a series LCR filter to go in parallel with the tweeter. The LCR network would be centered at the tweeter's resonant frequency. This will flatten or greatly lower the tweeters rise in impedance at fs. In the end it helps the tweeter to behave acoustically to the crossover network. If you leave the rise untreated it could cause the tweeter to not roll-off as expected and result in high distortion created by a higher than expected acoustic output from the tweeter at fs.

Dave

[scottsehlin]

> Sorry to **** in, but as time goes on, the
> more I want to try using the XT-25 (despite
> its juvenile-innuendo-snicker-inducing
> looks). From what I have gathered, there is
> something tricky to getting it to filter
> correctly. However, I don't know what that
> is. What happens if it is not filtered
> correctly?

> I do know that both you (Scott) and Wolf
> used it in a series XO. Since I don't find
> myself headed down that road for a few more
> projects--still trying to grapple parallel
> XO--what does a designer need to know to
> make it work in a parallel XO?

> -Paul

The other comments in this thread are generally on the money. You can use a parallel network with the XT-25, but you will either have to cross high or use a conjugate notch to deal with the sharp impedance rise around Fs. You can look at Zaph's ZD-5 project for an example of this approach.

For my "Hawkeye" project, I was using a really small box and was concerned with the number and size of crossover components I was stuffing into that box. Also, I just really couldn't get excited about the sound of the MG14-XT25 combo with a more typical LR4 2.5 kHz crossover target (I thought I might be able to get away without the notch at that point).

The Acoustic Reality Series Crossover approach helped me get the crossover down closer to 2 kHz without resorting to a notch filter - and it sounded a lot better than my other attempts. Had I been willing to use the notch filter, I might have stayed with a parallel approach.

[Wolf]

> "Impedance conjugate filter" =
> zobel? Or is it something else altogether?
> What is it, and why does the tweeter need
> it?

If you remember I had these last year at Richmond, and I flipped the switch on the xover in one exercise. This was in/out for the LCR or conjugate filter.

A lot of times, when you have a nasty impedance peak like this, the tweeter will be still outputting acknowledgable levels of a frequency you can hear that sounds "off" or like it doesn't belong, essentially low-end tweeter distortion. Other times it can create a suckout in the adjacent drivers' FR, as they cancel out when the Fs is not reduced in the tweeter. It's sorta like reducing a breakup for a metal-cone midbass driver, but from the tweeter, not the woofer.

I measured the parts with the WT2 to get the flattest impedance curve when inserted, and they are within an ohm from 20-20k, or a very flat Z curve. I would imagine the same LCR would apply- unless the impedance peak is in a different place.

FWIW, I really like both the RS28 and the XT25. Both are sweet when used in the correct ranges. My nod for ruggedness is the RS, due to the aluminum face, and top-end for the XT25.
Later,
Wolf

[Wolf]

> The Acoustic Reality Series Crossover
> approach helped me get the crossover down
> closer to 2 kHz without resorting to a notch
> filter - and it sounded a lot better than my
> other attempts. Had I been willing to use
> the notch filter, I might have stayed with a
> parallel approach.

So- you didn't use a notch at all? I'm surprised. The difference in mine was subtle, but still there, and I preferred it with the notch in the end. I believe the xover point was about 2250-2300Hz in my project. Would you care to post the schem again?
Later,
Wolf

[millers]

> If we knew a little more about your
> application, we might be able to advise
> better.

MTM with 2/ DC160S-4 rear channels 80 watts

[scottsehlin]

Provided Link: DIY Iowa 2007 Budget Class


> So- you didn't use a notch at all? I'm
> surprised. The difference in mine was
> subtle, but still there, and I preferred it
> with the notch in the end. I believe the
> xover point was about 2250-2300Hz in my
> project. Would you care to post the schem
> again?
> Later,
> Wolf

The schematic for the "Hawkeye" can be found on the page linked below. My thought process was a little more along the lines of "can I get away without using the notch" rather than an A-B with and without the notch. I've been considering revisiting that design with a currently in-production woofer, and if I do that I definitely should experiment with the notch.

One difference I see between our designs is that my second cap value is considerably lower, which will give a steeper low end rolloff to the XT25 and maybe reduce the need for notching the impedance peak.

[Wolf]

I had a 10uF cap on the tweeter, and it still made a difference. The notch and larger cap made the design golden, IMO.
Later,
Wolf