YAP power amp revisited, now at v2.1

Status
Not open for further replies.
MLED81

Hello Syn08

D102 is a MLED81 which is an infrared led is a red led a suitable replacement for this part or is there more too it.


Can I ask what the measured THD of this amp is at 100W @20Khz into 8R, the level of the individulal spectrum of components would be nice too.

Regards
AR
 
Re: MLED81

PHEONIX said:
Can I ask what the measured THD of this amp is at 100W @20Khz into 8R, the level of the individulal spectrum of components would be nice too.

Don't know for sure yet, but it's certainly under 1ppm (0.0001%). Currently I'm feeding the amp from two switching mode lab supplies (DLM 60-10). Their switching frequency is around 60KHz, so I can't get accurate spectra due to the switching artifacts. Anyway, the second harmonic is clearly dominant, around 8dB over the third, and spectral components are monotonically decreasing with the order.

You got the answer about D102/D117. D103 is a whatever red LED, it's role is only to (on board) signal the error (overcurrent latching) condition.
 
Slew rate

I was finally able, using my digital scope, to capture the (small signal) slew rate of the YAP 2.1 amp. The amp is non slewing, so it is now obvious that the input RF filter and the output coil is defining the rise/fall times.

I removed the imput cap, the output coil and zobel network and applied a 1MHz signal from a PM5193 50MHz pulse generator, with rise time around 1nS. The 50 ohm impedance output level was about 40mVpp, set to give a 500mVpp output (1MHz is slightly beyond the -3dB point, so the gain was lower than 28dB). A 50ohm resistor was placed at the YAP input to match the cable impedance.

The results are shown in the picture below. Rise and fall times are symmetrical and about 10nS, yelding a small signal slew rate (10% - 90%) of about 400V/uS. Doubling the signal input (not shown) approximately doubles the rise/fall times.
 

Attachments

  • yap_2_1_sr.gif
    yap_2_1_sr.gif
    20.8 KB · Views: 651
Re: J102 protection

PHEONIX said:
Does the connector J102 connect to a relay which disconnects the amplifier output from the speaker in fault mode. What relay do you use.

By the way how much power do you does this amp develop into 8R.

J102 goes to an "analog computer" that monitors all the fault conditions (overload (J102) output offset (using TA7317) temperature (including fans control), power supply faults) and manages the power sequencing, etc... It's a pretty complicated beast, the starting point is "Auxilliary circuitry" in the PGP amp (see my web site about), the YAP version is currently still in development.

The output power depends on the power supply; using two 800W toroids and +/-80W rails, the amp delivers continuous 2x300W RMS into 8ohm. This power levels will for sure require fan support at some time, though.

You don't need to do any changes in the schematic for rail voltages down to +/-50V.
 
Re: Slew rate

syn08 said:
I was finally able, using my digital scope, to capture the (small signal) slew rate of the YAP 2.1 amp. The amp is non slewing, so it is now obvious that the input RF filter and the output coil is defining the rise/fall times.

I removed the imput cap, the output coil and zobel network and applied a 1MHz signal from a PM5193 50MHz pulse generator, with rise time around 1nS. The 50 ohm impedance output level was about 40mVpp, set to give a 500mVpp output (1MHz is slightly beyond the -3dB point, so the gain was lower than 28dB). A 50ohm resistor was placed at the YAP input to match the cable impedance.

The results are shown in the picture below. Rise and fall times are symmetrical and about 10nS, yelding a small signal slew rate (10% - 90%) of about 400V/uS. Doubling the signal input (not shown) approximately doubles the rise/fall times.


Very Nice!

Best,
Bob
 
AndrewT said:
I would like to hear your comparison of measurement results to listening results.
I hope it turns out to be worth all the effort and design skills you have thrown at it.
I'll be patient.

It's on the todo list. I'm though not expecting any major subjective differences to (e.g.) the PGP amp, except for solving the 4ohm issues.

What I am really curious is to find out if there are any subjective differences between a MOSFET and a bipolar amp, both in the same objective performance range (YAP vs. VSOP).
 
Status
Not open for further replies.