Measuring speaker stuff starting with Thiele-Small parameters

Hello,

I am packing up and moving.

There are too many speakers, some all original, others recones with OEM parts and others are Frankenstein speakers made with voice coils from one model cones from another model voice coil and frames from another. I have one set of 15 inch Frankenstein JBL’s that are near 2235’s maybe.

What I want to do in this thread is measure speaker stuff starting with Thiele Small parameters of the 15 inch Frankenstein JBL’s.

This is just for fun. I have the time and space plus new to me Audio Precision APx1701 speaker measuring tools.

I am going to start by attaching the Frankenstein drivers to a fiberglass ladder to measure the Thiele Small parameters. I am thinking of using the added weight method as opposed to the known volume box to do the measurements.

Anyone with thoughts to add?

Thanks DT
 
Hey Chris,

Double A’ batteries sound useful and handy. The instructions I am looking at say to use plasticine, you know that molding clay stuff. I am a bit concerned that the oil in the plasticine will stain a paper cone, I will stick some to a white paper plate as a test to see if it stains or not.

I have the impression that the added weight method could be more accurate than attempting to guess the volume of a box and a driver.

Thanks DT
 
The best I've found are little magnets(5-10grams) like those that you find in closet doors.
Put one on each side on the cone so they attract each other,
also put a little piece of thick paper or cloth between the magnets and the cone to prevent scratches.
For larger woofers I put a pair of magnets on each side.

Easy to install and to remove without damaging the cone.
 
Last edited:
I can see where the strength of neodymium magnets could be an issue. But I think this can be addressed with one or a combination of these approaches:

  1. Use a small magnet - 1/4" dia. or smaller
  2. Use a magnet holder - to increase the size/mass and make it easier to grasp the magnets for removal
  3. Use layers of tape or a self-adhesive plastic/rubber foot - to place over the magnet to protect the cone surface and increase separation to weaken the force
 
Audiomatica Clio 4.51 standard small signal TSP evaluation has got 1Vrms of output voltage in a measurement setup that includes driver only in the circuit, the so called internal mode of measurement, and the output impedance of clio card is 100R.


You can make trial measurements with lesser voltages and investigate if there are significant differences. I get consistent results by having 15R of a resistor in the circuit and a driver, powered by a headphone out of a laptop sound chip at Vout=0.5V rms. In that case, the output impedance of a sound chip is 68R.
 
Hello JohnPM and Lojzek,

There is a range; 0.1 - 1.0 volts.

I will start low and work my way up.

The AP hardware has a test amplifier with a fixed gain of 20 and has the recommendation for small signal of not exceeding 1 volt output. Kind of interesting that the instructions are so fuzzy on the topic.

Thanks DT
 
Never attach a magnet or any ferrous material to the cone. If you do this near the dust cap, so it doesn't excite any rocking modes, it is going to affect the magnetic forces in the gap. If you do it near the surround, there will be rocking modes and it will change the parameters.

If you use a 50/50 mix of blue tack and standard hardware store blue stick for hanging things on the walls, the mixture will not be sticky enough to damage any cones.

If you want the most accurate results for the perturbation derived parameters, add the known weight to the cone and take a nearfield frequency response before and after. In the passband, subtract one response measurement from the other. Given the known added mass, this makes it very easy to determine the drivers moving mass perfectly. You then use this to calculate everything else.

To determine an optimum drive level, take multiple impedance curves starting at the lowest level and increasing to the maximum level. Look at the frequency of Fs. The best accuracy will happen when you use the drive level where the Fs measurement is the lowest.
 
Another useful fact is that a US nickel is 5 grams.

I like the non-residue blue tack, and I have a mass scale to accurately measure within 0.1 gram. Harbor freight had an inexpensive mass scale on sale recently (around $10). Don't know how accurate it is, but you could check against a quantity of nickels or AA batteries, etc.
 
Last edited:
Hello,

An old friend told me about the Zen of Motorcycle Maintenance; the motorcycle needs to know that you are not afraid to break it, only then will the motorcycle cooperate.

I am afraid of breaking my APx 1701 transducer interface adaptor.

This afternoon I spent an hour with customer service in Beaverton. I am all set now for monitoring voltage and current across the sense resistor and monitoring voltage across the driver under test.

All I need now is to put together a XLR test lead to connect the analyzer to the DUT.

Thanks DT
 
If you are going to measure low impedances with the AP, you should just build a current source. It is extremely accurate, the wiring is super simple and there is no chance of damaging the input circuitry.

Jack,

The amplifiers in the apx1701 are pretty much a voltage source. The specified output impedance is 0.13R including the 0.1R sense resistor.

Please tell me more about what you are thinking about with a current source and low impedance measurement!

I am cautious about the voltage; one day a long ago I stuck a probe into a tube circuit and shorted B+ volts into my the SYS2522. Duke fixed it for me.

Thank You DT