With any quality loudspeaker the change in the driver after running should be very little. As you have shown, there will not be enough change to affect the box design in any way. The days of needing to break in a driver are 20 years and more ago, or with very inexpensive drivers.
Mark
Thank You Mark, i keep that in Mind 
BTW i got an Massage from UPS some Devices are on the way from Saelig! 
Let see what i get next Day from the ordered Devices (Order QA472, QA451B, QA461 (462) still open)
I guess the first Both
Have a nice Day
Robert
Found an nice Device for the Analyzer Enclousure Build.
Have this Attenuator lying around with 10KOhm in Stereo and 24 Stepps.
Need to measure it but i guess it has -3,2dB per Stepp.
So i could use it also Differential up too -73dB Log
Every stepp only one resistor in serie, one Parallel.
Probably this one: but my is already more than 15 Years old and now time to use it
https://www.don-audio.com/Elma-A4-Audio-Rotary-Switch-Series-Type-SMT-10k-Thin-Film-Audio-Resistors-4-Wafers
But no thinfilm Resistors and was only around 160€ with normal 1% Welwyn Resistors.
from THEL Audioworld.
Measurements all Stepps Loopback Chanel L Out to In Single Ended:
And here some Noise Measurings @ -20dB, -40dB and -60dB:
One newbe question:
Why the Noise Numbers didn’t show as negative Numbers ? Like THD.
Means it:
Signal is @ -20dB, Noise is 81,87dB away from it? and not @ -81,87dB right?
I guess for better Noise Measurings would be good to add after the Attenuator an descend Amplifier to get the Signal back to 0dBV right? -20dB Atten. then +20dB gain Amp and so on right?
Also no intefearence with Input Impedance i guess.
I can do it next days, got already the Cosmos APU.
It has an decsend Amplifier in it +34dB and +60dB
But i ordered Cabeladapters, i hope i get them next day from Amazon
And Pssst QA472 and QA451B is incomming @ Monday AM.
Than i could do some Tests with my Class D Hypex Amps NC122MP and NC250MP.
Robert
Hi. “SNR” Tile is not the noise but the signal-to-noise ratio. If you want to display the noise, you have to select the “N-D dBV” tile, and you will see that if you select it, the noise will be expressed by a negative dB number.
Thx, i will try it next day.
It means Noise - Distortion right?
Robert
Yes. But it is actually the measurement of only the noise in the analyzed signal, while the “N+D” tile measures the noise component plus the distortion component of the analyzed signal.
Hello,
Replaced the Mesurings for N-D and looks much better
But at the End it gives the same SNR Numbers:
-105db N-D +20dB Gain =85dB SNR
-114dB N-D +40dB Gain = 74dB SNR
-118dB N-D +60dB Gain = 58dB SNR
Right ?
The 50Hz Spike is a bit worse, may be Shielded Wire would Help.
This 50 Hz Spike happen every time i didn’t measure Differential!
I guess it cals Comman Mode Error or so right?
Tested my Cosmos APU and if i measure it Single Ended i get this Spike @ 50 Hz and if i use Differential input i didn’t see it.
Like Here:
Question to the Experts!:
Will an Isolation transformer (230V) Help For an DUT? For the Common Mode Problem?
Something like this may be:
In combination with Mains Filtering or so?
Any Sugestions or the only way is to measure Differential?
Robert
Hi Frunse. Of course they must be the same numbers! SNR= 20log(S/N) which by the properties of logarithms (using the expressions of the various terms in dB) becomes: SNR = S - N. From which:
N= S - SNR, where N corresponds to N-D.
Hi, at this link you will find useful information on minimizing induced noise in measurements made with the QA40x.
An isolation transformer will hardly solve these problems, since there are many different ways to introduce these disturbances (radiated mode and conducted mode, capacitive coupling and inductive coupling, etc.). In any case, the one in the figure is not an isolation transformer, but a VARIAC that does anything but isolate.
Hi yes wrong Picture i mean this LOL:
So probably good to use an Battery (Akkumulator) for INUC, USB Hub and QA Devices if its Possible.
DUT like an 1000Watt Amp won’t be possible by sure but for DAC and Smal signal Analogue Circuit Measuring wil be.
Robert
Some more work on the Enclousure.
for
QA403
QA472
QA451B
QA462 (If released)
QA455 (If released)
On the Left you see the Digital Out Interface
On Bottom of the QA403 i will integrade my little Monacor Amp with 40Watt In and Outs
Robert
Test this Day my QA451B with 2 Amps and got good Results and also i guess found an Bug, not sure.
First THD+N for MONACOR AKB-60 (45Watt amp not Moded)
Second DENON POA 800 (my oldest Amp )
So now about the Bug (guessing)
If i start the Automated Mesurement PWRTHDvsPower i need to type a value @ this field:
Input Level Range Adder (dB)
But if i want to use the Full range let say at 42dB and try to type 42 there
he dont start the Measuring and give me advice to not type more than 30dB there ??? Why???
May be i need to use an newer Software Version for QA403
Robert
Hi. That field is not related to the input attenuator setting value, but to the value to be added so that the input attenuator (which in this test is handled automatically since “Autoset Input Range” is selected) maintains the distance of that value in dB from the signal produced by the amplifier on the load.
Oh man, now i understand
So if i want to have on Input and headroom add of +12db.
He will automaticaly use the Input Value +12dBV away from Signal right?
With External Load Resistors:
Looks better now
Also with the Internal Load of QA451B:
The AKB-60 has an TDA7294 in, so an well known Device.
But the Resistors get really fast Hot to 60° for this Measuring and switched off automaticaly!
You can see it on the red Trace the last Measurpoint the Load was already OFF.
Anyway i will move it in an other Enclousure may be i can add some Thermal Pads and a nice Aluminium Heatsink from both Side to get some more seconds of running Time
The whole Enclousure is from Alluminium so may be it is enough to conect the topp and bottom of the PCB (with Thermal Pads) to the Enclousure
How about this Heatsink Design
With QA462 on Bottom (Not yet released) may be also cooling the Power Amp there
Robert
After building me an XLR to BNC Adapter i was able to make an Compare on the Monacore Amp between Single Ended and XLR Input with an little Suprise, but may be i am doing something wrong!?
XLR seems to be 10dB in THD+N Worse than Single Ended but i was sugesting it will be better ???
One Question: i get 6dB More Output Level on XLR right?
I setup the Input to -12dB and the Output to -6dB but not sure if i doing right.
I guess +6dB would be better but this don’t changed the THD+N value in General.
Also getting an Errormassage after the Measuring ands and the Value adds to the Graph’s.
But cant see where the Overload happen ??? I used the External Resistor Load!
Robert
Hi. The fact that you are using external resistors has nothing to do with the overload that is occurring on the amplifier. You are probably using an input signal to the amplifier (DUT) that is overloading it. If I may take the liberty of giving you some advice: read the QA403 manual that will help you a lot. You can find it in “Help” “Open User’s Manual PDF”
What i am not understand is the Attenuator is at 30dBV and PK VRMS was max @ 14,61 Vrms.
@ 30dBV you can Measure till 31Vrms on QA403 Imputs so how this could be an Overload?
QA403 Output has 500 mvRMS so is this Error related to QA403 Inputs or to the Amp Inputs?
I will Check the Manual but i guess this Erro is measument dependent so i need to look @ QA Wiki for this kind of Measuring.
@ same Power (1Watt 8,35Ohm) the Differential looks also worse on the lower Side off 1kHz Signal (i guess an Design error or because the XLR Conector are directly next to Mains Conector, that i would never build ilke this )
 on the PCB
would help a lot. I will Change this also next.
Robert
Ok Guys like i told now the last Test with this little Amp
Removed the AC from PCB and Connected directly to an Lab Powersupply.
Frequenzy Response: (No little Peak @ 100Hz anymore!)
THD and THD+N Differential now so much better:
before:
So now i go for this Linear Regulated Powersupply for this Amp:
https://www.audiophonics.fr/en/linear-regulated-psu/lv1084-20v-5a-dual-controlled-linear-power-supply-module-p-8426.html
2x LT1083 up to 7,5A Regulator
Any Suggestion what i could do better here??? Adding Mains Filter Probably?
For the INUC and USB Hub and all QA USB devices i will use this PSU for Mains and also an big switch with 2 4mm Connector for 12V Battery Supply for in Car Measuring.
https://www.audiophonics.fr/en/diy-kits-and-boards/audiophonics-psu-s4-hp-linear-power-supply-module-dc-micrel29752-p-10435.html
Robert
Hallo Mat could you provide me the Color code for youre Front Plates?
My brother in law could do the Powder Coating and an friend of my could Digital Print it like you, or Laser engrave it.
like here: (Laser Engraved)
The Digital Print dont realy works good on Plain Aluminium Plates so Powder Painting comes Handy And why not Red
It could look like this
Back Side: only some USB and one Network Connector (XLR kind of Connectors) needed.
On the Left side: Battery 12V input, and Switch between Mains 12V and Battery 12V Powering.
On the Right: AMP secondary Output, Parallel to the Front.
On the Middle the Mains Connector with Mainsfiltering.
Some more Impressions:
Can’t wait to start Building it
But the Powersupplys will send to me first next week, if all goes the right way.
Robert
Hallo Mat could you provide me the Color code for youre Front Plates?
It’s Cardinal T241-RD129
