Two samples of used speakers measure differnetly???

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
How different is acceptable. I bought some used Focal TC120TDx tweeters and 6W 4252 drivers, and they measure a lot differently from each other. I can include the graphs of the impedance to show you, as thats all I have measured really at the moment. This is using the Woofer Tester 3 from dayton. Addittionaly, while one tweeter measures .2 mh of inductance, the other measures none??? That doesn't make sense does it? WIth the midbass drivers, one of the cones seems to be of slightly lower mass I think. One of the woofers has an FS of 50.4 hz free air and the vas comes out at like .66 cubic feet. The other driver has an fs of 49hz and a vas of .72 cubic feet. Neither of those numbers are even close to the manufacturers specifications for the driver. The Focal spec sheet has a vas of .9 cubic feet and an FS of like 42hz.

As for the tweeters, those are the drivers that I think are more concerning. The first tweeter file I will attach has specs which are close to the manufacturers specs. The impedance plot looks the same, but the FS is still around 150hz higher than speced. However, the other driver has a completely different looking impedance plot, doesn't even look like it should, and its fs is almost 1000hz higher, at over 2000hz, instead of 1080hz (Spec is 860hz or something like that). Is this make sense, or is there something possibly wrong with these drivers?

I did try using the different impedance plots (zma files) in speaker workshop to see what kind of difference it would make to the response, and it was very minor, but that isn't taking into account differences in the response of the drivers (Which I still need to measure).
 

Attachments

  • tc120tdx acomp.jpg
    tc120tdx acomp.jpg
    19.3 KB · Views: 367
hmm Ok I decided to recalibrate the WT3 and make sure it was warmed up plenty, then use Nickles and Bluetack for the added mass to remeasure VAS. My scale isn't too good, and measures 3 nickles, a little tape, and a small dab of blutack as 15.4 grams. The cone measures 5.2 inches from the middle of the surround, but only 5 inches or so of actual measurable cone area. I thought I read once that you should measure to the middle of the surround though. Is this right? That measurement has a pretty sizable effect on the vas reading. Anyway, going with 5.2 and 15.4 grams, I get a vas reading of .86 cubic feet for one and .94 cubic feet for the other. The FS reading is, again, different, with 48hz for one and 52hz for the other. Should I be concerned about these differences, or are they pretty minor overall?
 
After I took those measurements I decided to retest the tweeters to get the T/S parameters and compare again. I again had pretty wildly different numbers. I then retested the woofers, just moments after doing it before, and suddenly the numbers became drastically different. Maybe I should average all my measurements and go with those!!!! In all seriousness, when I remeasured them the woofer with the lower FS dropped to close to spec, around 42hz, and the vas was around .96 cubic feet. Then I measured the other, FS was around 51hz and the vas was around .72 cubic feet. That is over 20 percent different. While Focal is known to be a high quality speaker designer, and the drivers appear to be well made, I actually know nothing about how good they really are.
 
Neither speaker has ever been repaired, so if one is broken, that is beyond me. I don't believe either was ever damaged. The results aren't perfectly repeatable, but they are close enough. I mean, if I keep hitting the button over and over again, I get identical results. If I unhook the WT3, reconnect it, and try again, then I get different results. I didn't realize the speaker would act as a microphone, nor that it needed to be clamped in anyway. They give no information in the directions other than, hook it up and push the measurement buttons. I could try a better free-air mounting situation I suppose.

As for using "unreliable" clips, well yes I am, I'm using the alligator clips that came on the WT3. I would like to say that I can't imagine that Dayton would design this thing with inappropriate leads, but I too thought that was a strange choice. They are steel and magnetic, and I thought that steel leads within a magnetic field would greatly throw off inductance measurements, no? Any recommendations though for fixing this, I mean, its a premade device, so while I could solder on new ends, I don't have a tons of options. I had thought about getting the higher current copper clips and using them, would those be an option?

If the tweeter is not measuring any inductance, would that imply its blown? Thats the only thing I can think of, I mean, lets say the coil overheated and was shorting itself, it would massively reduce the inductance., no?
 
diyAudio Member RIP
Joined 2008
pjpoes said:
How different is acceptable.......WIth the midbass drivers, one of the cones seems to be of slightly lower mass I think. One of the woofers has an FS of 50.4 hz free air and the vas comes out at like .66 cubic feet. The other driver has an fs of 49hz and a vas of .72 cubic feet. Neither of those numbers are even close to the manufacturers specifications for the driver. The Focal spec sheet has a vas of .9 cubic feet and an FS of like 42hz.



If you have not done alot of testing I would consider those numbers to be actually pretty close. Cms is the biggest variable to T/S testing and its a pain in the ****. Cms (compliance or softness or hardness of the suspension) changes both with room temperature and how much the driver was just excersized (hysteresis in the suspension)

Anytime you have a lower Fs with the same driver you will have a higher Vas and its a result of a higher Cms (suspension went softer)

The suspension can go softer the hotter it is in the room and the harder you work the driver just before the test.
 
Are you supposed to work the driver before testing? I just took them off the shelf, plugged them in, and tested.

No I have very little experience testing drivers, this is my first foray into such things. I hadn't been willing to invest the money into the equipment, but now feel that, as my crossover designing knowledge, skills, and understanding improve, the more important such tools are. I haven't even constructed a proper testing baffle for measurements, gotten a good mic stand, er rather, any mic stand, or setup an area to test in with even the remotest anechoic properties. I'm thinking of buying the Joseph D'Appalitto book on testing loudspeakers, so that I can begin to better understand these things. If there are some online resources I can read in the meantime, I am all ears.

If we ignore the mdbass for a bit, what about the tweeter. I guess I wasn't completely clear when I wrote this, I knew the midbass drivers were close, probably close enough. It's the tweeters that concern me most. Those graphs are actually of the tweeters, and I can't imagine that over a 1000hz difference in FS is considered acceptably close. I think something is worng, but don't really know yet.
 
One of the biggest issues is the delta mass measurement method. It looks to me that your mass is moving. If you don't have a nice clean sharp impedance peak, you can't get accurate parameters. The delta compliance method is much more accurate and repeatable, but it can work only if the mass and driver are both perfectly stable. We went through a lot of work to get to be able to make repeatable T/S measurements. It is quite difficult to get a nice clean impedance peak, especially as the motor strength of the driver goes up. The following post I did on our forum talks a little about what we went through.

http://www.aespeakers.com/phpbb2/viewtopic.php?f=4&t=927

I need to get some good pics of the "shark cage" as it is called where we clamp down the driver to measure.

John
 
John, Thank you.
I would love a nice pic of your testing rig, I'm getting currious what I should be using.

I was on your website just last night looked at your page on T/S parameters. I found it interesting, and also note that you appear to use multiple different testing pieces. How is the accuracy of the WT3 compared with your more expensive gear?
 
pjpoes said:
How different is acceptable. I bought some used Focal TC120TDx tweeters and 6W 4252 drivers, and they measure a lot differently from each other. .


I had a pair of speakers I bought for a disco but used one with a valve amp for guitar practice. One day the valve amp died so I just used a semiconductor amp instead after that.

Its wasnt until I came to do my first disco that I realised one speaker was considerably louder than the other.
I measured the impedance of the speakers and one had become about half of its specced impedance.

I then realised it was the speaker that had been connected to the valve amp that was low impedance. Clearly the valve amp dying also fried the voice coil on the speaker.

I ripped the speaker apart and the voice coil stunk and was partly fried.
 
pjpoes said:
Neither speaker has ever been repaired, so if one is broken, that is beyond me. I don't believe either was ever damaged. ...

How do you know all that. Did they belong to a member of your family?

About the test box a lot of people use it but it's not the professional way. Sure is close for a fast reading. I've never used it, I'm going to work with one system from Germany as soon as I am set.
 
I of course do not know for absolute certainty that the drivers were never serviced, but Focal insists they do not repair or rebuild any of the w-cone midbass drivers. I've also had a heck of a time trying to get replacement tweeter diaphragms. They will not sell me any, Orca claims you can't get them, and I've never had any other speaker shop tell me they can. I think the only possibility of repair would be to a midbass driver, and the surrounds are definitely the stock ones (I have many other focal midbass drivers, and they are identical), and I see no signs of repair (such as glue or unevenness). Also, the person I purchased them from assures me he never repaired them. I'm inclined to believe that these midbass drivers have not been repaired, nor has the tweeter.

What makes you say that the test box isn't the professional manner. I just left the Purdue Acoustics research center, where one of the groups is Harman International, with their supplied acoustic testing equipment. I befriended one of the graduate students on the project, who helped me verify my measurements, and according to him, they have been using either the test box or added mass method, depending on driver size. I don't even know of any other ways to accurately calculate V(a)s, and I believe the test box method is considered the best. All of the test boxes were labeled either Harman International, Harman, or JBL, so it seems that one of the largest speaker manufacturer/producers uses the test box method.

By the way, is the system you are using from Germany the ATB system? If its the ATB Pc Pro, I have that as well, the WT3 is more accurate, at least according to both Meniscus Audio and the Manufacturer-Kirchner. If its the ATB Precision, then you are quite lucky, thats a 5600 dollar piece of test gear, but again, they have those at Purdue, and we used both the WT3 and ATB Precision, with identical t/s results. Also, to reiterate my point, the ATB precision, ATB PC Pro, and Linear X system all use the same three standard methods for measuring Vas. They use a test box, added mass, or given SPL (The least accurate of the three).
 
Well pjpoes, is good to know.
Is good to gather all that information, it is the ATB PC Pro from Kirchner, as I read from you on your other thread. I have been pressing for a forum in english in my personal mail to him. I remember that Svante G. refer to it, as one of those boxes comparing to his measurements (simulated software data?) differently. Sometimes is difficult to arrive to this kind of understanding as you make it, until I have more data/information to confirm it, I am grateful and of course I believe in what you are saying. Great information anyway.
 
Well the ATB PC Pro is still a great piece, and I wouldn't feel badly about its measuring. They are perfectly accurate for what we are doing. I have two people trying to convince me to spend the money on the Linear X LEAP system and I/V box for measurements. I mean, I could always just head over to one of the electrical engineering labs and do these the old fashioned way and get results accurate to within the limits of the HP gear, but I dont think its needed. Anyway, I saw another posting talking about the superiority of the Linear X method, and I'm sure its better, but even with the limited measuring abilities of the WT3 or ATB device, I still find my speakers measure close to as predicted. To that end, i notice others on this forum and other DIY sites who end up with final speakers that measure close to the simlutations, so I think these impedance curves are good enough. For crossover design, I really think the resonant peak in the impedance and its levels around that are what need to be most accurate. As long as it gets that right, the d.c. resistance right, I'm not sure the slight differences in rising impedance with frequency is going to throw our results off by all that much.

However, again back to my point, the ATB PC Pro will use one of the three standard methods, so what "more sophisticated" methods were you planning on using? I understand that one the engineers at Linear X strongly dislikes the std delta mass method because of the assumption that compliance is constant. I'm sure the Linear X system would allow designs that are that much better, but I can't afford it right now. I use the added mass method mostly, as its a lot easier to do. I have been using bluetack or modeling clay mostly so it doesn't move around, but it tends to fall off when the driver is in the vertical position people suggest. I'm still working on making a test rig, as some have suggested, along with finding a way to add mass that won't fall off. I mean, I don't want to damage the speaker cone with something overly sticky, the clay makes me nervous enough.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.