Woofer Tester II - How accurate is it?

[richard c.]

Now that several of us have the "Woofer Tester II", I'm wondering what opinions people now have. Is it too good to be true or worthless?

I tested my Altec 416-8b's, my 2226 J's, and Altec model 14 woofers, and some model 14 recones. The results seemed to agree with the published specs.

Has anyone designed with the test results? Do the WT II results agree with any other accurate way to test a LF driver?

Richard C.

[Mr. Widget]

I think it is supposed to be quite good.

I'd be curious as to how closely it might agree with CLIO... Giskard, Zilch, Rob?


Widget

[remusr]

My WT2 has produced test numbers very close to JBL's published specs on new & NOS drivers, other than the VAS which is about 2/3 of JBL's. Similar results with JBL recones with sensitivity variably lower on Alnico's as could be expected. I am using the older software that came with it, not the updated available on the web.

[richard c.]

remusr - Are you talking about the Alnico magnet being partially demagnitezed? You don't see this kind of sensitivity difference with a ceramic / ferrite magnet? It's good to know that most readings correlate. Many people have complained about VAS readings - and that is a major feature for me. I didn't really want to build several VAS test boxes!

I was just wondering if anyone has seen any real bogus readings that they can't account for. I would like to hear from someone who has tested an old woofer with a very hard / dried out foam or rubber surround. The cloth accordion surrounds on my Altec 416-8B's seem to be OK as the Fs is still ~ 25 Hz on all 4 of them.

I did have trouble with the nickels bouncing around (buzzing) on one that I tested. I don't know if it affected the test.

I will check out the updated software...

Richard C.

[Zilch]

I'd be curious as to how closely it might agree with CLIO... Giskard, Zilch, Rob?

I have only compared them by eye, and deemed them "equivalent" on that basis.

With respect to accuracy, there do not appear to be absolutes in these determinations. It's a heuristic science, the validity being measured by the end result.

For example, CLIO manual goes into some detail about optimum drive levels for the determination. WT2 allows variable drive, as well.

That's a level of detail I don't yet require. If I were importing the measurement data into box/filter design software, I'd probably be looking more closely at the variables, tho....

I would like to hear from someone who has tested an old woofer with a very hard / dried out foam or rubber surround. The cloth accordion surrounds on my Altec 416-8B's seem to be OK as the Fs is still ~ 25 Hz on all 4 of them.

I did have trouble with the nickels bouncing around (buzzing) on one that I tested. I don't know if it affected the test.

The difference in Fs between hardened Lansaloy and new foam is easily seen, as is "overshooting" with brake fluid.

Method matters in all of this; I can't say my results ever match factory specs, except in relative terms. That doesn't diminish the utility, however, for my current level of work.

I believe we learned early on that the nickels have to be secured. I've thus abandoned the delta-mass appoach in favor of calibrating abandoned enclosures for different-size drivers as test boxes. Vas is, well, "difficult."

[remusr]

Richard - my ferrite 2231,2235&2245 woofers have not shown any loss in sensitivity based on the numbers produced with WT2. My Alnico 136A & 2231A's have dropped from JBL's 93 dB 1W1m to as low as 89 dB. Most are 90.5-92. I understand this is consistent with JBL info that Alnico's magnet will discharge over time with high heat/load, esp with typical rock band or PA usage. I have used the delta-mass method for VAS with taped-together groups of 5 US nickels.

[Joe Alesi]

Method matters in all of this; I can't say my results ever match factory specs, except in relative terms. That doesn't diminish the utility, however, for my current level of work.

I believe we learned early on that the nickels have to be secured. I've thus abandoned the delta-mass appoach in favor of calibrating abandoned enclosures for different-size drivers as test boxes. Vas is, well, "difficult."

Hello Zilch,

I hope all is well for you. As I was reading I had and idea of how to secure weight to the cone "securely", yet "without" damage to the cone- well maybe.

Suggestion:

Using a small Rare Earth magnet- I have some that are 1/4 inch cubes (see I even speak the US lingo!) and some small steel weights like washers (or even other magnets).

Place the magnet on one side of the cone and the washers/ steel disk (or other magnet) on the other side, with some suitable surface protection on the washers and magnets (say electrical tape) to avoid cone damage. The magnets hold the steel weights to the cone.

Provisos:

1. An even loading on the cone would be desirable, so perhaps 3 weights and magnets should be used, evenly spaced around the cone.

2. Keep the weights/magnets as far from the speaker magnet and voice coil gap as possible to avoid strong attraction. and something smashing into something it shouldn't

3. Putting too heavy a weight too far out on the cone may be undesirable from a cone flexure perspective.

Food for thought. All care, but no responsibility

Best
JA

[richard c.]

remusr - Have you ever had weak Alnico magnets recharged, and did it bring the sensitivity back to spec? This kinda changes the "old" perception that Alnico is always better, and ferrite magnets are crap! Mind made up, don't confuse with facts...

Zilch - Heuristic science...whew! At least it takes most of the subjectivism out of the equation... I guess I should glue together a couple of "standard" volume enclosures for testing purposes. I know I could use any known volume (and allow for the difference), but it's also hard to plug holes and measure the precise volume of an old enclosure.

Joe - I like the idea of magnets and washers. Now I will have to buy an accurate gram scale. The postage scale I now have isn't good enough for this.

I don't see how WT2 can accurately test sensitivity. I would also like to see it up against CLIO - something with a calibrated mic and the correct drive (1W/1m or other standard).

Richard C.

[Zilch]

I know I could use any known volume (and allow for the difference), but it's also hard to plug holes and measure the precise volume of an old enclosure.

That's where the heuristics come into play. It's whatever volume yields the "correct" result.

OR, a plastic bag, water, and a BIG scale....

[62.43 lbs/cuft]

[remusr]

Richard - I haven't recharged Alnico woofers - the only facility I can find in Western Canada (Vancouver) is apparently out of service for renovations until August at least and they want $200 w/ return S&H to do one.
Sensitivity - its all in the math; the software controls and moniters electrical motion of the voice coil.

[richard c.]

Zilch - Yes, they have a fairly concise description of the word heuristics as it applies to various areas of interest (not so much on DIY speaker building???).

How about measuring the volume inside a speaker cone by pouring water into a plastic bag placed into the upturned speaker? I saw that suggested once! He did mention that the weight of the water might damage the speaker...like an 18" worth $500. or so! I might be inclined to try "puffed" rice, instead.

remusr - That price sounds way too high, but I only have "Great plains Audio" for Altec drivers as my reference.

I guess if you use current to move the cone and then quickly sample the back EMF OR voltage created by the "generator effect" of moving a coil of wire thru a magnetic field. With calibration, I guess this would work.

Now, I am going to "search" LH forums for WT2 and WT II. I see there are references to woofer tester sprinkled thru out many of the threads for the past 2 years, here. I need to get a download of the latest WT2 software version, too.

Richard L.

[WTPRO]

Hi Richard

You are absolutely correct that when measuring drivers is concerned, set up and drive levels are very important. Neither of these are however a WT2 'problem' since it is entirely up to you how you do your mechanical setup and Vas test (3 Vas test methods are supported). If you follow this link you will see how the WT2 and WT-Pro are able to extract information over a very wide range of drive levels. The USB powered WT2 can only drive 3mA, but this is also easily enough to get valid 'TS parameters'. If you really need more drive level, then I would suggest the WT-Pro where you can test right up to the Xmax rating of the driver.

http://www.woofertester.com/cms_non_linearity.htm

Something to note here. In the first diagram, look closely at the scale where the WT2 BL vs Idrive curve transitions to the WT-Pro. The jump is on the order of 0.02 Newtons/Ampere on a scale of 4.50 N/A. Thats better than 1%, and we do this from <1uW drive to many watts of drive. I should also explain that in this case 'Idrive' is working into Zmax, so the voltage and therefor effective power is quite high. You can also select any two parameters along the right hand side and plot them against each other.

There are a fair number of other new topics in the FAQ, and more are on the way. If you have questions, please feel free to contact me directly at [email protected].

Hope this helps,
Keith Larson
Smith & Larson Audio

[Zilch]

Zilch - Yes, they have a fairly concise description of the word heuristics as it applies to various areas of interest (not so much on DIY speaker building???).

Well, I note it says "Fast and frugal" methodology there; we know what THAT equates to around here, and who its prime practioners are.

[Heh, heh.... ]

[richard c.]

I will study the info from your link, and use it the next time I play with one of my drivers. I will turn down the drive level as 94 dB (Altec model 14) is the lowest sensitivity speaker I have tried, and the nickles bounce even at these very low measuring levels.

I have a lot of reading to do...

Richard C.

[WTPRO]

Hi Richard

The jiggling nickel problem arises when you have high driver BL causing the cone to accelerate downward faster than gravity can keep the mass firmly attached. Another problem that comes up is when Mms is very high and you end up needing to use gobs of nickels. In both cases stacks of nickels or clay work better. Just be careful using clay. Many clay masses people use leave spots or tear up paper cones.

You may also find that you can use a smaller test mass. In the not too distant past, nobody had precision gear that could measure voltage and phase 5 digits of accuracy or see changes in resonance to fractions of a hertz. This meant that the original 'delta mass' test needed enough test mass that the resonance shift became significant, but not so much that the suspension non linearity's came into play. If you plot the change in Fs with test mass you should find that a smaller than 'optimum' mass can be used. We simply adopted that value because the voting public demanded it. BTW, this is not a problem if the cone is mounted sideways, but then you cant just set the driver on a firm surface or use a simple nickel.

The 'bunjiggy' setup is also packed with some interesting information, but again this is a set up compromise because the suspension direction of motion is up/down and will sag. Never the less, I tend to like it because it is incredibly simple and I can test drivers to very high levels. It also helps that the suspension sag in Q/Fs is also occurring in the Vas test, and therefor tends to cancel (results are generally good).

Never the less, suppose the driver was held in a jig/clamp/balance mechanism that could hold the driver in a side ways orientation. In this case the suspension is initially at rest (no sag). The only problem now is the change in Fs caused by the double moving mass phenomena. I have thought on a number of occasions that we could sell a few test jigs like this, but always come to the conclusion it would be more trouble than the added dollar return warrants. Here are some links to follow.

http://www.woofertester.com/bunjiggy.htm
http://www.woofertester.com/cms_deltamass.htm

Since this is the forum were JBL is spoken loud and often, I should also comment about drivers that have very high Zmax or inductance. The WT is limited by its USB supply voltage, so it is possible to clip the current drive output. If you watch carefully a 'CLIPPING' indicator will begin flashing, but so far it has been up to the operator to notice this. Eventually this will be written into our test methods (bomb the test in progress).

Finally, if anyone is in the Boston area, I would be very happy to give a demonstration.

Hope this helps
Best regards,
Keith Larson (WT-Pro)
Smith & Larson Audio