Need help resolving odd sound from 2" mid-dome - new speaker build

[Efalegalo]

Hi All,

I recently gifted away a pair for Wayne 4Pi Speaker Cabinets and in return, the person gifted me a vintage set of ITT LPKM 130/50 2" mid-domes.

Look through my driver inventory, I realized I had a pair of Dayton Audio RS225S-8 (vintage shielded version) and Peerless DX20BF00-04 tweeters.

With the parts on hand, I proceeded to build a ATC lookalike mini-monitors, with offset dome mid and offset tweeter.

As I take measurements of the ITT (using sweeps via ARTA) - I hear some sort of resonance / odd tune to the sweep between 100hz -700hz, and thereafter the sweep sounds more normal. It sounds tube-y - if that makes any sense. Perhaps I can record in on my phone and upload to Youtube is that helps. One would have to listen for it using headphones.

The mid's resonance frequency is approximately 250, but despite crossing the mid at 800hz - I still hear it. I've even tried including a parallel LCR circuit to suppress the resonance, but I still hear it.

Can fellow forum members help me troubleshoot the issue? Let me know what measurements I can provide (e.g., distortion, near-field with and without LCR, CSD, etc.).

[5th element]

My recommendation is this.

First of all use a signal generator, in ARTA, to find a specific frequency you think sounds particularly bad, or plagued by this issue.

Then load up the spectrum analyser in ARTA and set the signal generator to this frequency. Then look at the spectrum that is generated and observe the harmonics that are produced. Alter the drive level, as required, to make it sound the way you want and see what the spectrum looks like. If the driver is producing anything that isn't a harmonic then this will display it. Sometimes when things vibrate in a speakers mechanical components they do so at frequencies not typically/harmonically related. Or at least not at frequencies that are typically picked up by standard harmonic sweeps.

If the spectrum looks clean then the driver is clean, there's not much else to say

[Efalegalo]

Additional Measurements

Sincere thanks for posting, 5th. I've never tried this feature in ARTA, but an attempt at doing what you suggested is included below.

600hz - 1000hz - click image to enlarge. See anything odd?

[5th element]

Sorry for the late reply!

What sticks out to me here is that the levels of distortion are rather high. You're getting very high levels of odd order distortion, 3rd and 5th are much higher than they should be. This is at odds with your distortion measurement above from STEPs, which shows pretty low levels of distortion. Also the signal level is very low. ARTA is recording a level at lower than -70dB.

I think you've got a hardware problem here in the measurement chain. You could simply be measuring the left channel when the signal is coming in on the right channel and what you're looking at is the channel separation or something.

[Efalegalo]

Coming back to this. New measurements

Can someone please explain to me why distortion would be high at lower volume (e.g., 0.4v here) vs at higher volume (e.g. 2.8v or 5.6v)?


See measurement. Also - under the assumption the distortion measurements were performed correctly, could someone suggest the lowest recommended crossover point?

Also - 5th Element - I think the issue was that the STEP measurements were taken at a higher volume than ARTA Spectrum Measurements. I've included re-do at 1000Hz and 600Hz @2.8v

[Bear]

Can someone please explain to me why distortion would be high at lower volume (e.g., 0.4v here) vs at higher volume (e.g. 2.8v or 5.6v)?

The first hypothesis is that the drive level isn't sufficient to get above your noise floor in your measurement equipment. After that, my mind would wander to a suspension/mechanical issue. But others with more experience will have more informed answers.

[5th element]

Yeah usually the problem is a low signal to noise ratio when it comes to making distortion measurements at low drive levels. Most of the time you need to place the microphone close to the diaphragm otherwise the noise swamps things.

That said your FFTs, from ARTA, show plenty of SnR though.

It's possible that what you're hearing is some kind of linearity issues with the suspensions. And by this I mean the suspensions are being responsible for increasing the distortion at lower listening levels. If it isn't signal to noise issues then it has to be something. I've seen some drivers, with awful motors, that have very poor distortion performance at low levels. But these have terrible performance at low levels and it only gets worse as the drive level increases.

[Alaric]

This probably isn't pertinent, but I used to have a pair of decent speakers (A/D/S L1530) with soft dome mids and tweeters. One one song, during one musical sequence (repeated through the song) I would get a a rattle that was generated about two feet from the midrange dome. If I put my ear closer it disappeared, but from two feet away and further it was a very noticeable rattling sound. That same album on any other stereo wouldn't do it, just with soft dome midrange drivers. The guys at the shop I bought the gear from finally concluded it was just a lucky coincidence that there was some weird harmonic that was the driver's weak point and one of my favorite songs happened to use that sound a lot. I accepted that at the time (I was 22 and knew almost nothing about anything).

Although, now that I think about it, 40 some odd years later, it may have just been Martin Barre's fingers strumming across the steel strings and I didn't know anyone else with a rig with that level of resolution. I never took the speakers back to have them checked (They were 5 feet tall and I had a broken leg).
Well, I contributed nothing with that ramble. The thread title kicked in a memory from damn near a half century ago. My apologies, guys.