HIGHLY Preliminary Measurements (takato14)

so yeah I bought a measurement mic and threw together a measurement rig

I was literally holding the microphone and baffle assembly in hand while I did this because I havent made it down to the hardware store for a chunk of PVC yet so take these graphs with an absolutely enormous grain of salt

I had two baffles to use on these, one is a hard plate, the other is a soft plate, for simplicity's sake (and because I just got called into work) I did hard plate for all of these

I'm already having some issues with headphones that rely on a seal (4AAA) as I could not get bass from it without the hard plate and the measurements are looking very mountainous and improper, more on this later when I'm not rushing to get to work on time

Sansui SS-100 (modded):



Sennheiser HD700 (modded):



Koss Pro/4AAA (modded):



These measurements are NOT corrected for the mic calibration at the moment so again, grain of f'ing salt

brb work, more to come in like 10 hours

 

Maybe my creatology foam is too cheap or something, its certainly not name brand, and the "soft" plate is just the rear side of the hard plate covered in said foam LOL

I've still got a lot to mess around with before continuing (like actually being able to put the headphones ON the thing without holding it) but so far the results are far better than expected right off the bat, really excited for this if I'm being perfectly honest

 

A few updates:

Switched to REW, ARTA seems slightly dated and I was getting more consistent results with REW in general. Also I couldn't figure out how to get distortion measurements from it. *shrug*

Put together a pseudo-finalized rig for now using a cut up Pringles can instead of PVC and various different foams as internal damping materials, pictures tomorrow (I do intend to replace it with PVC ASAP and remeasure again)

Remeasured most of my stuff multiple times, heres the current plot for the Sansui including HD:



Most distortion content below 100Hz should probably be ignored due to poor noise floor in my measurement environment

Hopefully tyll will post his plots of the Sansui soon but these are largely consistent with his raw FR data save for the upper midrange, which seems to resonate centered around 2kHz (this resonance is more consistent with my impressons than my own plots, FWIW)

I took CSDs last night as well but the noise floor in my room was atrocious and I forgot to save the picture, ill grab that when I get home

 

I'd rather use ARTA but I couldn't figure out how to generate the distortion plot. Then again REW is also very impressively intuitive with how it saves files. I dunno, if someone could help me figure out THD+N I could just compare results from the two side by side and stick with whatever is more consistent.

 

One more thing: I also took measurements of various different damping configurations on the Sansui:

Here's what they measure like when configured as they were when I sent them to tyll:



And here's what I managed with some material swaps and rearrangement:



It's still difficult to tell how much of this is noise floor, unfortunately. I'm gonna try a few more things when I get home to see if I can get my room a bit quieter.

 

it's not just noise floor... if you cause micro vibrations, it'll effect bass distortion measurements very significantly.

So yeah... while you're measuring, hold your breath, and pray there won't be a vehicle driving by your house.

 

does anyone have any advice for how to mount this assembly anyways? I measured and cut a section of Pringles can about the distance between my two ears but the mic's USB cable juts out the back of the assembly by almost an inch, making it impossible to put the headphones on it. I'm still stuck with holding the damn thing in my lap and pressing the cup against the baffle, working against the force of the headband as it digs into the side of the housing. I'm not so much worried about the inconvenience as I am about the variability this might introduce to the measurements.

 

Congrats!

 

I think most of us with the UMIK-1 opted for a 90-degree connector that can acommodate a somewhat stretched headphone.