I started doing some measuring for my next build. I am going to do a budget pair on SB drivers.
I ran a quick impedance test on the tweeter. How strange. WAY off spec. So, burned it in for a day with music. Let it cool and well, the Fs did drop 40 Hz. Better, but that odd bump, and the spec is still way out of published.
A bit of fooling around, and I found the issue. It is the thin plastic back cup. If I held the driver tight against a corner, the glitch goes away and the specs drop to very close to published. Next to see how I can fix it.
I ran a quick impedance test on the tweeter. How strange. WAY off spec. So, burned it in for a day with music. Let it cool and well, the Fs did drop 40 Hz. Better, but that odd bump, and the spec is still way out of published.
A bit of fooling around, and I found the issue. It is the thin plastic back cup. If I held the driver tight against a corner, the glitch goes away and the specs drop to very close to published. Next to see how I can fix it.
Attachments
First test, three layers of Dynamat on back of housing. Helped, but not as much as holding it. ( 630 Hz) I also noticed the flange is pretty flimsy and resonates at about the Fs. Bad news. It is worse in a baffle than it is sitting on the desk. (660 Hz) Next test, clean off the Dynamat and epoxy on a piece of MDF.
Why am I worried about problems at 600 Hz? Because I intend to put in a notch filter as I am trying a pretty low crossover.
Why am I worried about problems at 600 Hz? Because I intend to put in a notch filter as I am trying a pretty low crossover.
Epoxied a piece of 1/2 inch MDF to the back making the thin plastic rigid. Raised the resonance to 660. Mounted in a baffle, 676.
Now, what to think about this.
#1 the rear chamber is not well designed. If it as flexible enough to cause these differences, it would be susceptible to modulation from the woofer.
#2 To prevent #1, should be in a sealed sub-chamber.
#3 Well, I can't build a speaker with me holding the driver with pressure on it. Seems like the best result is with three layers of Dynamat. I an guessing the rear of the chamber is acting like a passive radiator, and the Dynamat changed it's mass, so it changed its tuning. What I can't explain is why so much better result with me holding it. I would have thought baffle mounting would increase the load on it and lower the resonance, but the opposite seems to be true.
#4 You need to measure the impedance in the box where you will use it. If I used the out-of-box, or worse, OEM spec, to build a notch filter, it would be way the heck off.
Dissi, Well I don't have that review. I only have one driver. Their numbers are equally far off. ( Spec being 600). As I have concluded the hump is a function of the design, so I can rule out a defect or damage. Could it be that somehow my suspension has gotten that much stiffer so it's resonance is in a range where the chamber back is revelant?
Well, I guess I can go set it up in the shop and play music for a week and see if it limbers up. I will say, I have not had a tweeter this far off before. Woofers yes, not a tweeter. As they say, stay tuned.
Now, what to think about this.
#1 the rear chamber is not well designed. If it as flexible enough to cause these differences, it would be susceptible to modulation from the woofer.
#2 To prevent #1, should be in a sealed sub-chamber.
#3 Well, I can't build a speaker with me holding the driver with pressure on it. Seems like the best result is with three layers of Dynamat. I an guessing the rear of the chamber is acting like a passive radiator, and the Dynamat changed it's mass, so it changed its tuning. What I can't explain is why so much better result with me holding it. I would have thought baffle mounting would increase the load on it and lower the resonance, but the opposite seems to be true.
#4 You need to measure the impedance in the box where you will use it. If I used the out-of-box, or worse, OEM spec, to build a notch filter, it would be way the heck off.
Dissi, Well I don't have that review. I only have one driver. Their numbers are equally far off. ( Spec being 600). As I have concluded the hump is a function of the design, so I can rule out a defect or damage. Could it be that somehow my suspension has gotten that much stiffer so it's resonance is in a range where the chamber back is revelant?
Well, I guess I can go set it up in the shop and play music for a week and see if it limbers up. I will say, I have not had a tweeter this far off before. Woofers yes, not a tweeter. As they say, stay tuned.
youre talking about the rear and not the mounting surface right? if so maybe some 2 pack epoxy putty or P40 fiberglass would sort things out. Lead flashing?
EDIT. Your last post just beat me. So the whole rear chamber is flexing that much! Perhaps its removable, and a suitable replacement could be made?
EDIT. Your last post just beat me. So the whole rear chamber is flexing that much! Perhaps its removable, and a suitable replacement could be made?
Last edited:
I am surprised that driver from a company such as SB shows something as pronounced as this. Now if I bought a pair and fs was 600Hz instead of the 500Hz spec I wouldn't lose any sleep over it, but the fact that the chamber is affecting the response in such a way is worrying. I do wonder if your driver has a defect, plastic chambers tend to be very solid.
Edit - having looked at the tweeter some more, the only thing I can think is that the chamber is slightly loose in places. Zaph's measurements have fs on 500Hz too, maybe it's a quality control or shipping issue?
Edit - having looked at the tweeter some more, the only thing I can think is that the chamber is slightly loose in places. Zaph's measurements have fs on 500Hz too, maybe it's a quality control or shipping issue?
Last edited:
Seems well attached. I'll know when I pop it open. It is about two years old but has been in a safe environment. I sure would hope they don't age and stiffen if not used. I'll set it up for a week long burn in tomorrow before I rip the chamber off. It is not a expensive driver, but I don't want to trash it for no reason. If it does change much, then there is further proof why you have to break in before designing the crossover.
Well you can't fault SB on the build quality, it's there in spades. It'll be interesting to see what the results are like with the chamber removed. It is possible that the best results might come from leaving the chamber off and mounting the tweeters in a small sealed chamber within the cabinet.
Odd behavior. I made a series of chambers, 2" pvc and block of MDF. Used hot glue to attach them for a good seal. Very rigid, so not acting like a passive radiator like the thin plastic one was. 1/8 inch, 1/4 inch etc.
A small chamber gave as one would predict a higher Fs, Higher Qts, and higher Xmaz.
What is strange, is as the chamber increases, the impedance looks more like a ported chamber. The larger the chamber, the further apart the humps and the more even. I was expecting the Q to be dropping and Fs dropping to some minimum, not a second peak.
Using cotton batting for fill, firmly stuffed, I got a Zmax of 7.8 Ohms, Qts of .54 and an Fs of 593. Just a little shelf where the second peak is, around 850. I also get a bit of a "glitch" at about 530 which I always got except when I was holding it in my hands, pushing on the back. I suspect this has something to do with the distance between the "nib" in the back and the support tripod for the center of the dome.
I am not sure what this is telling me.
A small chamber gave as one would predict a higher Fs, Higher Qts, and higher Xmaz.
What is strange, is as the chamber increases, the impedance looks more like a ported chamber. The larger the chamber, the further apart the humps and the more even. I was expecting the Q to be dropping and Fs dropping to some minimum, not a second peak.
Using cotton batting for fill, firmly stuffed, I got a Zmax of 7.8 Ohms, Qts of .54 and an Fs of 593. Just a little shelf where the second peak is, around 850. I also get a bit of a "glitch" at about 530 which I always got except when I was holding it in my hands, pushing on the back. I suspect this has something to do with the distance between the "nib" in the back and the support tripod for the center of the dome.
I am not sure what this is telling me.
Tweeters often have foam or felt plugs in the pole vent or situated on top of the pole piece itself, these can sometimes be very dense such that the air flow is impeded. I would imagine that if you placed a chamber after the plug you could get some sort of dual loading going on that could give you twin peaks or so.
Well in the ring radiator design you've got the voice coil and on either side there's a radiating area. On one side you've got the large roll surround and on the other you've got the dimple. Maybe these two are creating the effect. It is interesting of course to see that the dual peaks go away in the original measurements simply by you bracing the back of the chamber with your finger. This kind of points away from a dual chamber type thing, but something has to account for it. Perhaps the chamber cup has a resonance at ~500Hz, it seems to small for that to be the case though.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.