If that tweeter is anything like the other 26s, don‘t worry too much. Mainly K2 harmonic distortion. About the waveguide: crossovers actually become almost easy with them. And it works a lot of the time. But I can’t predict the behavior of this tweeter in one.
Last edited by a moderator:
With most tweeters of this size distortion rises under 2kHz. As your midwoofer can play 3kHz without any problem and has a very brave behaviour even higher, why induce distortion from the tweeter?
https://doc.soundimports.nl/pdf/brands/SB Acoustics/SB17MFC35-8/pdf_SB Acoustics_SB17MFC35-8_1.pdf
See, it is not the 1.700Hz that may create a problem, but any x-over has a broad region where it still plays quite loud together with the other chassis.
If you have a look at the tweeter response, it ramps up a bit, under 2.000 Hz, as it has a Qts of 1.12
https://doc.soundimports.nl/pdf/bra...-C000-4/SB26STC-C000-4-TO-BE-DISCONTINUED.pdf
The SB tweeter is an excelent chassis with a very acceptable price point. I use SB tweeter too, by the way.
There is nothing to gain from using it too deep with your specific woofer.
Maybe have a look at typical measurements of 1" dome tweeter, you can see how distortions usually rise at some point around 2kHz.
https://doc.soundimports.nl/pdf/brands/SB Acoustics/SB17MFC35-8/pdf_SB Acoustics_SB17MFC35-8_1.pdf
See, it is not the 1.700Hz that may create a problem, but any x-over has a broad region where it still plays quite loud together with the other chassis.
If you have a look at the tweeter response, it ramps up a bit, under 2.000 Hz, as it has a Qts of 1.12
https://doc.soundimports.nl/pdf/bra...-C000-4/SB26STC-C000-4-TO-BE-DISCONTINUED.pdf
The SB tweeter is an excelent chassis with a very acceptable price point. I use SB tweeter too, by the way.
There is nothing to gain from using it too deep with your specific woofer.
Maybe have a look at typical measurements of 1" dome tweeter, you can see how distortions usually rise at some point around 2kHz.
@Turbowatch2 the SB17 midwoofer starts to breakup sooner than what SB Acoustics says.
Here it is the measurement made by my Italian shop. As you can see...
Here it is the measurement made by my Italian shop. As you can see...
You couldn’t get teh tweeter closer to the midWoofer. Given the ideal (pretty difficult) goal of attaning a quarter-wavelength at the XO every little bit helps. Looks like it vould be at least a cm closer.
dave
dave
You should measure the response of the tweeter before you finish everything. It strongly depends on position. If you get something unpleasant and have to try a new panel, move the tweeter closer to the woofer as Dave wrote.
That is great. Please measure what you got now. Only for the tweeter, the woofer doesn't matter that much right now. No need to mount it on the box, just measure as far away from anything standing around. Compare to the curve from the data sheet. The difference is a result of your baffle size and position of tweeter. You may try to chamfer the sides for a second try.
Try moving tweeter and woofer as close as possible in the next baffle try.
There should be a program you can use to simulate this instead of wasting too much wood.
Try moving tweeter and woofer as close as possible in the next baffle try.
There should be a program you can use to simulate this instead of wasting too much wood.
VituixCad
Last edited:
i know VituixCAD i've already made a simulation using measurements given from my local shop. I don't like that hump at 4.5k tho...
Last edited:
You need a measurement that has definition of the baffle it was made with. Then you simulate the position on the baffle, so you can see what response you get. You only simed for some unknown measurements. Why do you trust your speaker shop more than the manufacturer? Do they have an anechonic measuring room? I don't think so.
Might lead you nowhere. This tweeter is able to give a very smooth response, placed right on a baffle. Simulations are a simple garbage in = garbage out machines.
Might lead you nowhere. This tweeter is able to give a very smooth response, placed right on a baffle. Simulations are a simple garbage in = garbage out machines.
Developing a quite modest 2-way like this doesn't require anechoic rooms, although that one in Haan makes me jealous again and again 😉. What you do need is some experience and knowledge, a computer with either ARTA, REW, HolmImpulse or other app, some small parts and wires, a power amp, a microphone and an usable sound card. But we know that I hope.
Ciro uses measurements from some speaker shop for simulation, instead of the manufacturers data sheet. I wanted to point out why the measurement of a professional speaker factory are different from those done by a dealers "in house".
So if you put data not done by specification into a simulation, your results may be far off.
So if you put data not done by specification into a simulation, your results may be far off.
That may be true, but whereas I respect SB, their data sheets aren't well documented (where has the time gone that Vifa incorporated high res CSD plots in their data sheets...). And likely a lot of enthousiasts could produce results like SB does, DAAS isn't exactly the most elaborate test bench, adequate certainly, but no more than that.
You are right, DAAS is quite old, but what else do you want to know from a data sheet? We as DIYS community only have access to these fantastic chassis because there is a commercial market for them. DIYS is only a tiny part and for many (arrogrant) manufacturers more a drag than a profit.
This is nothing like a car or motorbike, something you buy and use, documented in any data possible.
This is just a sole component and if you use it, you do your own measurements in situ anyway (or not). So the most important thing about a data sheet is to respect the norm for measuring, which gives you a basic idea of what you get. DIN baffle, level and impedance.
Even as it may impress the customers of a small loudspeaker shop, telling people the manufacturers data are wrong and presenting your own backyard made measurements is no real help IMO.
Sure, Klippel with any possible thing done would be nice, but to be honest, I don't miss that. I read some DIYS magazines and pick my distortion figures from there, as this is what I want to know. Also comparing Klang&Ton or Hobby HIFI responses with manufacturers show how honest they are.
By the way, in the past a lot of large sales companies just wrote the response curves by hand it seems, just as our Chinese friends do today. Lie to the world and smile...
This is nothing like a car or motorbike, something you buy and use, documented in any data possible.
This is just a sole component and if you use it, you do your own measurements in situ anyway (or not). So the most important thing about a data sheet is to respect the norm for measuring, which gives you a basic idea of what you get. DIN baffle, level and impedance.
Even as it may impress the customers of a small loudspeaker shop, telling people the manufacturers data are wrong and presenting your own backyard made measurements is no real help IMO.
Sure, Klippel with any possible thing done would be nice, but to be honest, I don't miss that. I read some DIYS magazines and pick my distortion figures from there, as this is what I want to know. Also comparing Klang&Ton or Hobby HIFI responses with manufacturers show how honest they are.
By the way, in the past a lot of large sales companies just wrote the response curves by hand it seems, just as our Chinese friends do today. Lie to the world and smile...
FWIW, this is about a different, but very similar driver (SB15MFC-08) so maybe not too OT.
Measurements shown: by the factory, from Zaph Audio site, from Dibirama, and the actual measurement in a 9L closed box with 22x35 cm front baffle.
Note the baffle step, which is not present on any of the other (wide baffle) measurements.
Measurements shown: by the factory, from Zaph Audio site, from Dibirama, and the actual measurement in a 9L closed box with 22x35 cm front baffle.
Note the baffle step, which is not present on any of the other (wide baffle) measurements.