Some speaker driver measurements...

Well the schema in previous post explicitly places Rsense between dut and ground, whereas most schemas i know of place Rsense between amp and dut.
I use a Rsense of effectively 0.135 ohm between amp and dut, and works fine. Value determined in Arta Limp by measuring a resistor of known value -6.8 ohm- as dut , and adjust the value entered in setup so in Limp the measured dut value is a match.
And yes balanced input in audio interface. ( RME Babyface Pro FS) I went for this low value because of the fact that Purifi also measures with an Rsense of 0.1 ohm, so my ts and imp measurements correlate with theirs. It also makes voicecoil current harmonic distortions possible.
 
  • Like
Reactions: markbakk
I think the benefit of placing the sense resistor between DUT and ground, would be to keep voltage at the soundcard input low. It may avoid the need to use a voltage divider at the soundcard input to step the voltage down to safe levels for the soundcard. With 0.1 ohm at the ground side, you could send 20V to the driver, and only see around 0.5V at the soundcard for a 4 ohm speaker.

I will have to think about this a bit and run some tests, if it works well it may simplify my test set up a fair bit.
 
There is a possibility the basket could produce resonances in the 350 to 400Hz range. That was found in some of the Purifi test's that there was a slight increase in distortion depending on how the driver was mounted.
An ENTIRE basket significantly resonating? 🤔

I find that hard to believe, that's a lot of mass and stiffness.

Is also the frequency range where spider first resonance mode occurs usually.
Something like this, or being more specific, could be a mismatch in acoustic mechanical impedance between cone and surround.
Or anything in the same kind of idea.

Those acoustic mechanical resonances can be very tricky.
With just a very unfortunate combination and set of parameters, you could have some kind of resonance on your hand.
Which could also explain why the way of mounting had an influence on it.
(especially in relation to the compliance)
 
  • Like
Reactions: IamJF
The tattle tale is a rise in all harmonics at the frequency of interest.

from a DIY point of view, the easiest way is to decouple the woofer from the baffle via use of grommets or other viscoelastic material and hand tighten the screws just so. that
Stop when the material is partial compressed.

However the effectiveness of this depends on the assembler- overnight-tighten the screws and you’re back to square one.

In an assembly line this is too fiddly and needs an extra QC step.

Mounting the driver by the magnet is an alternative and is very effective, and less fiddly.

The Radiant Acoustics Clarity 6.2 is done this way, IIRC,

The 3rd alternative that driver manufacturers use is a viscoelastic material between the motor and the basket.
 
I see your point. This “resonance” is ever so subtle, and likely below audibility.

On the grand scheme of things level of importance is low.

Sometimes we just want to know what we are measuring

Eg.
Before-
1739312914807.png

Problem with driver?

After:
1739312957261.png



Difference?
Viscoelastic material between frame and cabinet, and between screws and frame.

Answer: “problem” with mounting
 
From my past experience dry or wet mounting we called it then.
Simple example: vibrationsensor in M16 Bolt shape, mounted dry on large lathe to measure bearing noise was regardless of torque peaky . Mounting with some matter (some fat) in between the peaky behaviour was gone. We attributed it to the not 100% contact, in other words pockets of air in case of dry mounting. This allowed some movements and/or reflections between sensor and heavy lathe at particular frequencies
When still playing vinyl records, the whole mounting of the player was "wet" and diff was very audible.