TS parameters. who's correct???

[dallaire]

Well, did some interestiing stuff today(I think anyway) I tested my four inch peerless mids for TS parameters, and what I found left me kind of wondering who's right, who's wrong? Just for kicks I left the drivers in garage overnight (Phoenix), about 50deg. So I then measured Fs and it was way high as expected, so temp does matter a lot. After three hours of break in time on bench, this time in house, after acclimating to house temp, about 75deg., the Fs was down 20hz! This was all to be expected. What was not expected was when placeing driver on top of two inch by two inch sqaure metal post, four foot high, I measured about .985 volts at resonance, which I thought was accurate. So I then figured I would try hanging the speaker as some say to do. Nothing around speaker within four feet or so. What did surprise me was the voltage at "the same resonance" was down to about .965, these aren't exact voltage values, don't have papers with me right now, but are close enough for explination. So then which is correct to use????? Hanging reading, or top of metal post reading??? Seems I have heard people say DO NOT HANG THE SPEAKER as this will not work. Than I have read that some say this is the only way So I was wondering, what does the manufacturer do? or is thier method totally different?

Anybody?
dallaire

 

[ocool_15]

I would measure the drivers at regular room tempurature as thats where I would expect you to use them. I don't quite understand using the metal pole but I expect that to affect your results. You should measure the drivers with the cone facing horizontally. If you have the cone pointed upwards your fs will be lower. For smaller driver I usually will suspend the driver by the speaker wire, you can usually wrap it around the driver in a way to reduce stress on the terminals while still pointing the cone horizontally.

 

[Conrad Hoffman]

There's a good saying in the test world, "test it the way you're going to use it." IMO, that means cone facing the way it will be in the speaker, room temperature, and realistic amplitudes. And don't expect things like speakers to give useful three-place results.

 

[dallaire]

Yea, I think I begining to think the hanging method is the way to go, like you have said. The reason I was using the two by two pole was to elevate the speaker from boundry problems. I wonder if the fact that the magnet sticks to the pole quite strongly, if this affects something. The pole being metal and all.

Anyway, I think tomm. I will redo measurements using the hanging method, and go from there.
Thanks again,
dallaire

 

[hermanv]

I think the peak voltage variation is due to the Q of the mounting setup. i.e. an enclosure or box resonant at Fs will have he highest voltage. I think the absolute voltage is unimportant unless you are trying to measure driver Q.

Most European manufacturers use an "infinite" baffle. A baffle made from a 1/2 sheet of plywood (4ft by 4 ft) or MDF would give reasonably accurate answers down to about 100 Hz (or is it 1/4 wave. in which case the baffle works to about 50 Hz?).

 

[dallaire]

Yes, I was tring to measure Q's and everthing needed to design box for mids. I woke up this morning and began testing my mock up baffle about ten by twenty-seven. I tested closed back mid box that was made using parameters from testing. What I found was removing the back off the box not only yielded a lower response but was "much" smooter, open back mids are the way for me! I am very suprised!, thanks for the suggustion. now I need to adjust baffle dim. so it will support down to 300-350hz, right now it will only support about 400hz max. as I didn't want to have to contend with baffle step effecting mids. I am using the program called "EDGE" that was a free download, it is surprisingly accurate, when testing estimates created by program.

Anyway this is where I am at, thanks again for the tips!
dallaire

 

[hermanv]

There are a number of ways to dampen woofer resonance peaks, each with it's advantages and limitations.

* A conjugate network is a series RLC network resonant at the woofer peak, wired across the voice coil. By designing in the correct Q a very flat impedance response results. Conjugate networks offer an additional advantage, the crossover network will function more ideally because the load resistance is more constant.

* A port tuned to woofer resonance also tames the peak and generally results in the highest acoustic output at the lowest frequency the woofer will support. Port Q is tricky and probably results in overhang or ringing if the woofer Q and port Q don't match correctly. Also ported designs have steep roll off slopes below port resonance. Either of these reasons could be why some people dislike ported designs.

* Another alternative is to design an enclosure too large or too small so any woofer resonance is not made worse by the cabinet. Either under or over size requires a conjugate to work best. The oversize enclosure offers the slowest acoustic roll off and sounds best to many people. The problem is that for a decent low frequency cut-off the box can get quite big. The under size enclosure will cause the cut off frequency to be higher than the woofer is capable of supporting.

I know very little about transmission line woofer design so I can't comment about how it relates to resonance peaks.

For one of the best guides around go to:
http://ldsg.snippets.org/