Explendid amplifier designed by Michael Bittner, our MikeB

anatech said:
Hi Jan,
Nope, it's not filtered at 22 KHz. The photo was at 10 KHz, but 1 KHz looked the same. I'll have to let you know the -3 dB point. My manual says that it has a bandwidth from 20 Hz to 200 KHz for level measurements.

-Chris


Right. You know, I always chuckle if I see those opamp data sheets showing a THD that rises slowly until 10kHz or so and then bend downwards. Clear sign that they switched in the 22kHz low pass on the Audio Precision. ...

Jan Didden
 
The measured THD was 0.017%. The Leader LAG-120B I'm using measures 0.032%

Hi Chris, if i read this right your generator has a higher distorsion figure from his own than the amplifier result, so a nulling effect maybe here?
What is the right value for the bias pot?
Loek
 
Hi loek,
I think that the generator output is picking up more noise that the amplifier test due to the relative impedances. But you are correct, the measured oscillator performance is much worse than the amplifier. There is every possibility that the amplifier's performance is much better than measured in this setup. I am betting that is the case and it is also why I posted the measured distortion from the generator. I think the other members would agree with me on this.

I am using a 500 R bias control, the other components are unchanged in this circuit.

-Chris
 
Hi Chris,

Your welcome!

Why not use a CD player as the sig source - may not be as good as the best sound cards at 24 bits, but still well under 0.005 percent typically, for a decent player. And, having proper isolation, much better than a sound card with a noisy PC spraying all that EMI around...

Cheers!
 
Hi Clem,
I don't feel like burning a CD for this. A good suggestion though.

Hi Mike,
No risk, no fun ?
But ........ I like my meter. :bawling:

I have the pair pounding away warming up. DC offset on the other was -1.7 mV. They don't have any glaring problems and sound much better than many commercial products that come across my bench.

This will be interesting.

-Chris
 
I tried the 22ma bias on the MJL0281/0302, comparing 110ma against 22ma...
To be honest, i didn't hear a difference :cannotbe:
I guess, as ilimzn pointed out, these new devices are built different, using ring emitters or multi emitters and what ever technique. For the MJL21196/5 the 110ma bias is valid, but obviously no longer for these new devices.

Chris, i have 2 minor differences to yours, i have c14 = 22pf and 10ohms as REs to Q1/2.
Have you already replaced the ceramics with the cute micas and the input cap with the Wima 4.7uF ?

Mike
 
Hi Mike,
I'm using 10 pF for C14.

All the new cute caps are in unit #2 The old nasty ceramics are in unit #1. I want to measure any differences if any, then listen. I burned it in for a day.

Intial impressions: Really good quality amps. Dynamic and are in no way wimps in the bass region. If they can get even better, then wow!

To be honest, i didn't hear a difference
Then save on the heating bill. Even sticking the speakers in parallel didn't appear to strain these amplifiers at the lower bias setting. That's a good sign. On my large heatsinks there is a positive temperature coefficient. We'll see what it does on smaller heatsinks later on. I'm sold on them for their intended use.

-Chris
 
Groov, simply click on my www-button, there you get most of the important information needed for symasym.

Borys, the rmaa measurings are strange, it might be the measuring setup itself, like the right channel was not correctly connected. (grounding ?)
Also, the left channel seems to get induced power supply artifacts ? (HF-garbage, but might be the soundcard)
Just searching for perfection... The 2nd harmonic might have been cancelled out, can you show loopback measuring of your soundcard ?

Carlos, you might have been confused, the picture is from Borys, not Jan. But i would also be interested in his opinion about symasym.

Chris, it's very logical that symasym plays fine on 4ohm speakers, my DIY-speakers are big (40L) vented speakers with 4ohm. (2 way, 2 woofers + 1 tweeter, no series resistor to the tweeter,90db)

Jan, Symasym is also an attempt from me to prove that feedback is not evil. Symasym is definitely a high feedback amp, using 57db of feedback.

Mike
 
MikeB
In RMAA test i have connected only one channel becouse i have only one dummy load (8,2R) the second one wasn't connected at all. I'll repeat the meas with stereo dummy loads and everythink will be clear.
With FJL high tones improve much against 2pair MJL. Could it be result of much lower capacitance (faster) of output devices ??


Thanks all for your help!!