I am building a 3-way using the following drivers:
D2904/710003 Tweeter
12M/4631G Mid
18W/4531G Woofer
The initial cross over frequencies are 3500 / 350 using first order. I thought I would start at that point, listen, measure and tweek. My baffles are designed for an odd order.
This will ultimatley be bi and/or tri amped SE. Until I have enough amps built it will have to be passive.
My problem is that I get a harsh buzz/hash at times in the tweeter, almost as if it is blown. I am using a 10uF cap. after switching tweeters out it remains present. When I lower the cap to 8uF it is much improved but I can still hear it at times.
I think that this is a problem with hitting the resonance since I am using a 1st order cross.
The question is how do I make a series notch filter for this speaker not having a number for Qes and Qms. fs =520Hz Re = 3.0Ohm.
I am in the process of building my measuring equipment from my pc using Sound Easy, but not done with it yet. For the first go around to get these up and working I would like to get it in the ballpark if possible by using published values.
D2904/710003 Tweeter
12M/4631G Mid
18W/4531G Woofer
The initial cross over frequencies are 3500 / 350 using first order. I thought I would start at that point, listen, measure and tweek. My baffles are designed for an odd order.
This will ultimatley be bi and/or tri amped SE. Until I have enough amps built it will have to be passive.
My problem is that I get a harsh buzz/hash at times in the tweeter, almost as if it is blown. I am using a 10uF cap. after switching tweeters out it remains present. When I lower the cap to 8uF it is much improved but I can still hear it at times.
I think that this is a problem with hitting the resonance since I am using a 1st order cross.
The question is how do I make a series notch filter for this speaker not having a number for Qes and Qms. fs =520Hz Re = 3.0Ohm.
I am in the process of building my measuring equipment from my pc using Sound Easy, but not done with it yet. For the first go around to get these up and working I would like to get it in the ballpark if possible by using published values.
Yes and no
Tweeter ressonance will always be a problem in any xo
But it doesnt cause what you hear
Tho, the ressonance does mess with the filter slopes
10uf is way too big, and probably 8uf too
Especially with "6db"
You need "12db" to protect tweeter
If you use "6db" on mid driver also, the noise you hear could come from mid driver
Use a paralel zobel(cap/res) after series inductor
btw, if you have a series resistor on tweeter it also changes the size of series cap
Dont trust a theoretical/calculated xover, you have to trust your ears
Your tweeter is resonanting. With first order, the most I can go without a Series Notch is with a 4uF. With 8uF, it gets irritating. Sometimes even painful. This is with a Vifa D25AG35 that has a rear chamber.
With a Series Notch, you may be able to get away with 8uF. You can still get a fairly good network with only Fs and Re. Having Qes and Qms is more accurate though.
According to the D2904 datasheet, Fs=750Hz, Re=3.5 ohms.
A basic LCR is 40uF + 1mH + 5 ohms. Use a cheap E-cap for the 40uF. No point spending on one of those polypropylene. Values don't have to be exact. 4.7 ohms will work. Series two 100uF e-caps for 50uF. What you want is to see how effective the LCR is with 8uF.
Regards
Mike
With a Series Notch, you may be able to get away with 8uF. You can still get a fairly good network with only Fs and Re. Having Qes and Qms is more accurate though.
According to the D2904 datasheet, Fs=750Hz, Re=3.5 ohms.
A basic LCR is 40uF + 1mH + 5 ohms. Use a cheap E-cap for the 40uF. No point spending on one of those polypropylene. Values don't have to be exact. 4.7 ohms will work. Series two 100uF e-caps for 50uF. What you want is to see how effective the LCR is with 8uF.
Regards
Mike
Just checked my datasheet for the D2904/710003 and it show 520Hz and 3.0. I hope to get SoundEasy working over the holidays and I'll have a real number.
Years ago I use to have Thiels and absolutely loved them. Per Jim Thiel, I asked him in person, he used first order crossovers, however, he stressed that it took ALOT to make them work. So that is were I am starting. The baffle is sloped back approx 15 degrees to better align the acoustic centers. It is an approximate start
It is DIY learning time at its best.
I will say that I had to replace a few tweeter on the Theils. Usually after a party.
Who knows what I'll end up with. I am going to try the LCR first and then move from there. I'll post back with how it works.
The 8uF vs. the 10uF made a HUGE difference. In all but the extreme vocals the rasp is gone. Still irritating however as was stated above.
Years ago I use to have Thiels and absolutely loved them. Per Jim Thiel, I asked him in person, he used first order crossovers, however, he stressed that it took ALOT to make them work. So that is were I am starting. The baffle is sloped back approx 15 degrees to better align the acoustic centers. It is an approximate start
It is DIY learning time at its best.
I will say that I had to replace a few tweeter on the Theils. Usually after a party.
Who knows what I'll end up with. I am going to try the LCR first and then move from there. I'll post back with how it works.
The 8uF vs. the 10uF made a HUGE difference. In all but the extreme vocals the rasp is gone. Still irritating however as was stated above.
Photo and update
Just a photo of the project. No finish or inlays yet. Infact bass doesn't even have a port or lining yet.
The LCR took out the rasp. Even went back to the 10uf for a minute and it was still gone. Now I am really motivated to get the measurement system up.
The system will be driven by 300b and/or kt88 amp(s). I do have my alephs for back-up however.
Just a photo of the project. No finish or inlays yet. Infact bass doesn't even have a port or lining yet.
The LCR took out the rasp. Even went back to the 10uf for a minute and it was still gone. Now I am really motivated to get the measurement system up.
The system will be driven by 300b and/or kt88 amp(s). I do have my alephs for back-up however.
Attachments
Thanks R-Carpener,
I use to be in your neck of the woods, now I am on the other end of the state near Jamestown, NY.
A while back I had Sound Easy working. I got side tracked on other projects and never set the new computer up to run it. I just need to put some grunt work into it to get back to measuring.
I never did much with tweeters though, other than a simple impedance and SPL. Can't remember if Sound Easy can calculate the T/S without changing mass.
I do wish there were other DIY people in the area.
I use to be in your neck of the woods, now I am on the other end of the state near Jamestown, NY.
A while back I had Sound Easy working. I got side tracked on other projects and never set the new computer up to run it. I just need to put some grunt work into it to get back to measuring.
I never did much with tweeters though, other than a simple impedance and SPL. Can't remember if Sound Easy can calculate the T/S without changing mass.
I do wish there were other DIY people in the area.
Sorry, I was half asleep when I wrote that. I meant you can determine it from the curve in the spec sheet. If you notice, the impedance scale is a "power of two" kind of scale. Probably the easiest way to extract this impedance info is to use SPLTrace, rather than trying to interpolate off the graph. SPLTrace will spit out a impedance file and you can possibly figure out the T/S parameters from the formulas using that information.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.