[tangent]

Quote:

Originally Posted by vixr /img/forum/go_quote.gif its awesome...I highly recommend it.

The TANC-F is officially on the schedule now. I'd press the button on it now if it weren't for the recent YJPS and PIMETA v2 board runs, plus a need for a PPA v2 board run.

 

[wuerzig]

Quote:

Originally Posted by tangent /img/forum/go_quote.gif scope the output.

I think this might be due to the case not being grounded yet. Also, I remember this looked different in analog mode (didnt take a pic sorry).


Quote:

Originally Posted by tangent /img/forum/go_quote.gif try running square waves through the amp.

Input vs. output. Is that good enough?

The power-on "thump", left and right. Its interesting that the channels differ, but then I only scoped that once.

I also measured the frequency where the signal drops 3dB to be at about 2 Mhz (darn my english vocabulary is missing a word here, whats that called, the upper frequency threshold?), but with much degredation due to what you see in the second image.

Sorry for the crappy pics, I took them with my cellphone.

 

[tangent]

I'd bet on that first one being low-level oscillation rather than induced noise. Uncased PPAs are generally pretty quiet. Are you using the schematic value for R11, or a lower one you saw recommended above in this thread?

Your other oscillograms look fine. The square waves aren't ringing badly, and the turn-on thump voltage is acceptably low.

 

[rds]

What load were you driving when you scoped the amp?

Even with R11 at 100 ohms I get some hiss in my Shure se530s (IEMs), so I'm thinking there is some oscillation going on there. I'll try to scope it this week and see what's happening.

 

[tangent]

With IEMs, I would use a gain of 2, no more. That rules out OPA637s and other uncompensated chips, but it sounds like a good trade for noise floor to me.

 

[wuerzig]

Quote:

Originally Posted by tangent /img/forum/go_quote.gif I'd bet on that first one being low-level oscillation

I didnt expect the amp oscillating at 100Mhz. I will investigate further.

Quote:

Originally Posted by tangent /img/forum/go_quote.gif Are you using the schematic value for R11?

Yes.

Quote:

Originally Posted by tangent /img/forum/go_quote.gif Your other oscillograms look fine. The square waves aren't ringing badly, and the turn-on thump voltage is acceptably low.

I thought so, thanks for your input

Quote:

Originally Posted by rds What load were you driving when you scoped the amp?

Some random sennheiser headset, I think the PC150.

 

[tangent]

Lower R11 to 100 or 220 ohms, and let's see if that helps. If it does, that might be enough to push me to change the schematic.

 

[wuerzig]

In my effort to identify the signal from the first oscillogram, I connected the probe to nothing but a short piece of wire (~3cm). The same thing happens, so my guess is that this is noise after all.

I wont have access to the scope for a couple of weeks, so trying to lower R11 will have to wait until I get my hands on it again.

 

[rds]

I did some measurements of my PPA v2 (with r11 = 100 ohms, MJE243/MJE253, OPA827) driving 40 ohm Ultrasone Pro 900s. I had to keep the voltage low so as to not blow the drivers. Unfortunately the equipment I used is not the best, so there is some noise but I thought the measurements look very good.

520Hz square wave

3.2kHz square wave

12.1kHz sqaure wave

I couldn't find any high frequency oscillation

 

[tangent]

It'll look even better if you turn on the 20 MHz bandwidth limit to get rid of some of that HF noise. I take it from the use of photos instead of screen grabs that you don't have the TDS2MEM or similar?

 

[rds]

Ah, I didn't realize that it had a bandwidth limit ...and it's right on the screen too

I know it can be connected to a computer ...by a serial cable iirc. But I didn't have a cable with me, and even if I did I'd probably be too lazy.
I realized I should have taken a shot in the MHz range too. I'll probably bring the PPA in again with my YJPS (once it's done) and take some more measurements.

 

[FallenAngel]

Interesting thing I found out today with my PPAv2 - AD744 (comp output) doesn't like class-A bias in the amp (anything above 2mA and it starts to output DC like crazy) and AD829 (comp output) REALLY doesn't like class-a bias and goes haywire with even 0.5mA.

Oh well, live and learn; now 3x AD744 compensation pin output at 1.5mA class-A sounds damn good and without any DC offset.

YJPS is one damn great PSU by the way, very impressed with it compared to how I remember STEPS and Simga11 with the PPAv2.

 

[NelsonVandal]

Quote:

Originally Posted by FallenAngel /img/forum/go_quote.gif Interesting thing I found out today with my PPAv2 - AD744 (comp output) doesn't like class-A bias in the amp (anything above 2mA and it starts to output DC like crazy) and AD829 (comp output) REALLY doesn't like class-a bias and goes haywire with even 0.5mA. Oh well, live and learn; now 3x AD744 compensation pin output at 1.5mA class-A sounds damn good and without any DC offset.

When used "comp pin out" (output buffer is bypassed and the signal is taken from the voltage amp stage), both AD744 and AD829 are single end class A and external biasing isn't beneficial in any way.

 

[MisterX]

Yep, check the the simplified schematic in each of the respective datasheets and you should see why you had a problem when you tried to apply "class-A bias"

 

[FallenAngel]

Thanks guys, now I have to find a way to bypass it.

Probably just going to cut R10.