How to measure CLASS D unbalanced amplifier THD ?

Ives

New Member
Thread Starter
Joined
May 26, 2020
Messages
2
Hi,

Question for all you smart guys :-)

I can measure normal CLASS A or A/B amps THD but what do I best put in the measuring return line for CLASS D amplifiers to do a balanced conversion ? I cannot measure between GND and AMP OUT.
I could use a small audio transformer but I'm not a big fan of that (although I assume the calibration will straighten the curve) . Normally I use a Behringer UCA222 USB-interface.

Any suggestions are greatly appreciated !

Thx
Ives - Belgium
 

Antonio Di Motta

Active Member
Joined
Feb 8, 2018
Messages
179
Location
Italy
Hi, the best thing to do is to use an operational amplifier as a differential amplifier. Here is a schematic of principle.

On the internet you can find formulas to calculate the resistance values based on the gain you need. In the case of unitary gain, you can set all the resistances to the same value, say 10K.

32039
 

Ives

New Member
Thread Starter
Joined
May 26, 2020
Messages
2
Hi, the best thing to do is to use an operational amplifier as a differential amplifier. Here is a schematic of principle.

On the internet you can find formulas to calculate the resistance values based on the gain you need. In the case of unitary gain, you can set all the resistances to the same value, say 10K.

View attachment 32039
Perfect thank you for the suggestion and feedback. Much appreciated.
 

Antonio Di Motta

Active Member
Joined
Feb 8, 2018
Messages
179
Location
Italy
I guess you want to evaluate the harmonic distortion at the maximum power of the amplifier and since we are talking about class D or the configuration of two class B amplifiers the output level is very high, certainly> 100W on 8 ohms and therefore the configuration that you I proposed not good because it is necessary to attenuate the signal by at least 30db. In this case the value of the resistances must be R1 = 33 * R2
(R3 = R1 R4 = R2).
Try to see if this scheme works for you, it is a part of my analyzer, it has very high input impedance 1M ohm and a decoupling capacity of 0.1uF as an integer you can also use the cheaper NE5532 I think the distortion introduced should however, be very low. Consider that with this configuration you can measure the levels up to about 100W equal to 28Veff on 8 ohms.
Regards
Antonio
 

Attachments

Top Bottom