Power amp measurement system

[phofman]

I do not know, it probably very much depends on the SSR model used. -120dB at 100 ohm is 0.1mOhm. Perhaps grounding the circuit input (behind the input resistor) instead of inserting the relay into the path?

 

[LKA]

I have good experience with SSR speaker protection. The simulation is attached.

 

[LKA]

the voltage divider and its protection circut

 

[LKA]

the voltage divider prototype enclosure, inside is still empty, pcbs are on the way

 

[Cobalt1]

the voltage divider prototype enclosure, inside is still empty, pcbs are on the way

it's my voltage divider with MCU and LiPo battery
step 1dB

 

[LKA]

I finished and measured the 10k passive notch filter. Depth is 60dB, pretty good.

 

[jan.didden]

Looks good! I'm waiting for my PCBs to be delivered.

Jan

 

[LKA]

Looks good! I'm waiting for my PCBs to be delivered.

Jan

yea ... next step is a post-notch 30-40dB amp (same box as for notch, two 9V battery powered)

 

[29285]

Why not use 2x 3.7V Li-Batteries - last much longer than 9V blocks. And you do not need so much voltage swing to feed a soundcard.

 

[Mark Tillotson]

If you use rechargables its not such an issue.


A single 9V with virtual ground would be adequate too.

 

[LKA]

I tried to compensate the notch1k filter in ARTA. The compensation file was generated by STEPS.

green trace - soundcard loopback
black trace - loopback with notch + compensation (*.mic file)
blue trace - green black difference

H1 diff -0.8dB
H2 diff +1.37dB
H3 diff -0.04dB

 

[LKA]

I modified my dummy load a bit.
- zener diodes removed
- "raw" monitor port has permanent ground connection now
- RC compensation added

 

[LKA]

I'm going to try this circuit as post-notch preamp. (modified, originally front-end for my HEC amp)

 

[LKA]

preamp quick test

the chain : Viktor1K -> Notch1k -> PreampK4v3 -> XFi

Viktor1K output: +4.4dBv
Notch attenuation: H1 -55dB, H2 -9dB, H3 -5dB
Preamp : +40dB
X-Fi input: -10dBv

After correction:
H1 = -10+55 = +45dB
H2 = -114+9 = -105dB (at least)
H3 = -117+5 = -112dB (at least)

I hope I am correct ... THD < -145dB

 

[trobbins]

I find that disposable aluminium baking trays are a simple way to supress mains hum on the workbench - one to hold everything, and one to act as cover.
https://www.amazon.co.uk/10pcs-Disposable-Aluminium-Baking-Trays/dp/B00JF0TWS2

 

[LKA]

I find that disposable aluminium baking trays are a simple way to supress mains hum on the workbench - one to hold everything, and one to act as cover.
https://www.amazon.co.uk/10pcs-Disposable-Aluminium-Baking-Trays/dp/B00JF0TWS2

cool

I'll use steel enclosure (162x70x217 mm) for the final preamp. Inside : R-core transformer, super reg (jung) and two K4v3 modules (20dB and 40dB amplification).

 

[trobbins]

Yeh, some form of total test setup Faraday cage (with all equipment inside, including batteries and interconnecting cables), with only USB cable exit, can often supress local stray hum ingress.

I do the same with my EMU0404 when testing an amp on the bench - the EMU sits in an aluminium tray, with exit hole for 100:1 probe to go to bench DUT, and exit hole for line output lead, and another for USB. Battery powering for EMU (and if needed also for USB isolator) sits in tray as well. A ground clip link from line input BNC cover to tray also seems to help sometimes, as does inverting a tray over the top to make a total cage.

 

[phofman]

...if needed also for USB isolator...

Which USB isolator do you use? EMU 0404 is high speed (480Mbps). Thanks.

 

[trobbins]

I have used a ADuM4160 based pcb that was on ebay years ago - it has a 'full speed' data rate setting, so I can get 48kHz sampling rate from the EMU if needed. I typically used the USB isolator for cheap USB soundcards I was checking out, as I was keeping the EMU for reference measurements back then. I haven't had to use it for the EMU now I think about it - and use a battery power interface for the EMU when checking for hum loop parasitics.

 

[phofman]

I see, full speed only. Thanks.