PPA v2 construction discussion

[funch]

OK, so I cleaned the flux from my board, and installed two 6.04K resistors in R4 = gain of 6.7. Hooked it all back up and tried it again with my 32 ohm HD497's. This time no hiss at idle, but as soon as I added volume, the hiss was back. It still plays music, but with the hiss and crackling. I've read post from a few others who experienced hiss/noise with low impedence cans, but not with high impedence, so I tried my HD 600's, and sure enough, it works great with the 600's.

I'm so relieved that it works. I was begining to doubt myself on this one!

So how does one get this amp to work with lower impedence 'phones? Or is it strictly an amp for high impedence 'phones?

*Edit*
I just hooked up the bass boost (pot and 0.1uF cap) and even with it at full turn, there's no problem. Happy, happy, joy, joy.

 

[tangent]

funch, villekille, there are posts earlier on in this thread with advice from ppl about dealing with instability. The AD843 is particularly unhappy in this amp in the default configuration. Low impedance loads are also risky, and combining them, as in villekille's case, is almost a guarantee of instability.

Just off the top of my head, some of the advice had to do with:

- smaller buffer input resistors
- different values for output resistors
- different choices for output transistors

 

[kklee]

Quote:

Originally Posted by tangent /img/forum/go_quote.gif - different values for output resistors

That tweaked my memory. With my PPA problem, not only did I adjust C6G, I upped the output resistors to 4R7 as well.

 

[funch]

I'm wondering about my choice for C4. I used 100uF Muze KZ caps. Is this OK, or should I have used the recommended 220uF? I went with the smaller ones because I wanted a better cap, and the 220 wouldn't physically fit.

 

[funch]

Well, I just finished replacing the carbon 2R2 output resistors with some metal film 4R7 ones. Now, even with the volume turned all the way up and no input signal, there's no hiss or noise whatsoever. Quiet as a mouse, and, even with very little time on it, it sounds great.

 

[villekille]

I also changed the output resistors to 4R7 and the hiss is gone

Can you check how many mA is the current at idle? Mine takes about 100 mA and the buffer bias is set to 15mA (72mV with 4,7 ohm output r), no class A biasing. Using opa134x2 + opa227. How can it be that high?

 

[luvdunhill]

Quote:

Originally Posted by villekille /img/forum/go_quote.gif I also changed the output resistors to 4R7 and the hiss is gone Can you check how many mA is the current at idle? Mine takes about 100 mA and the buffer bias is set to 15mA (72mV with 4,7 ohm output r), no class A biasing. Using opa134x2 + opa227. How can it be that high?

I had a hard time getting that particular combination working ;( I got noise at about 25% volume. Now that I have an oscilliscope and some testing tools, I'm going to try again, but FYI it didn't work out of the chute for me.

 

[funch]

Quote:

Originally Posted by villekille /img/forum/go_quote.gif I also changed the output resistors to 4R7 and the hiss is gone Can you check how many mA is the current at idle? Mine takes about 100 mA and the buffer bias is set to 15mA (72mV with 4,7 ohm output r), no class A biasing. Using opa134x2 + opa227. How can it be that high?

I just realized that this question was probably for me.

In my previous post, I had measured my idle current at around 110mA. According to Tangent, it was probably high. But the amp is working fine, so I'm not going to worry about it. At least not until it goes boom, if it ever does. Nothing gets too hot, so I'm just going to enjoy the music.

Speaking of which, I just changed the output resistors from 4R7 down to 3R3, and still no hiss, or noise with my 32 ohm HD497's, so that's where they're going to stay. I did this because in comparison with my M^3 through HD600's, the highs in the PPA were a little constricted with the larger resistors. Maybe some of that added distortion that Tangent talks about in his doc's. No such problem with the 3R3's. I suppose if someone wanted a smoother, less detailed sound, then they might prefer the 4R7's in the output. As for me, I'm

.

 

[DanP]

I am new the Head-Hi and I am looking to build a PPA v2 amp with the battery board. I have been able to locate all of the parts except the MC33340P charge monitoring chip. This chip is out of stock every where. Does anyone know of a source for this part or a replacement part.

Thanks

 

[JDAPJ]

Hi!

I would like to know if anyone of u experts here could please explain this for me:

In the instructions to PPA it says that Q4 should at least have a Idss of 30 mA.
However, when I run it in a spiceprog and set the recommended Icq of 20 mA
it seems to be enough with 5-10 mA to get the 40mV drop over 2.2 ohms.


And to the next q:

What is the real benefit of using currentmirrors instead of the simpler approach with a transistor+led+resistor ?

Hope someone have the time to read this...

Thanx!

/PJ

 

[amb]

Quote:

Originally Posted by JDAPJ /img/forum/go_quote.gif In the instructions to PPA it says that Q4 should at least have a Idss of 30 mA. However, when I run it in a spiceprog and set the recommended Icq of 20 mA it seems to be enough with 5-10 mA to get the 40mV drop over 2.2 ohms.

It's because Q4 has a resistor and trimpot to set its current, and is not run "wide-open" at its Idss current. In order for the trimpot to have a useful adjustable range to suit device variations, Q4 should have an Idss much higher than its actual intended operating current.

Quote:

What is the real benefit of using currentmirrors instead of the simpler approach with a transistor+led+resistor ?

It allows a single trimpot to set both the top and bottom currents. Using the transistor/LED/resistor scheme, you'd have to independently adjust the two CCSes, and it would be difficult to make them track each other.

 

[JDAPJ]

Now that was just what I needed to know, thanx a lot amb!

Had my thoughts in this direction and you confirmed it, again thanks!

/PJ

 

[FallenAngel]

Hey guys,

I was debugging my PPAv2 tonight and couldn't get very far as I'm seeing the same thing over and over again.

The amp sounds great but it's oscillating

Gain of 11 (freaking loud!)
OPA637 on L/R = 7 x 7mA = 14mA
OPA627 on G = 7 mA
Buffers = 3 x 25mA = 75mA
No Class-A
No LED
No Bass Boost (S2 jumpered, no second POT)

Total = 91mA
Reading = 140mA!

Where am I losing 50mA?

I also tried adjusting the buffer bias to 30mA each and I didn't get a total of 120mA, I still got 50mA over at about 165mA! I do think the amp sounds better this way.

I used most of the recommended parts.
- 2N8087/2N8088
- MJE243/MJE253
- 2.2Ohm output resistors (Yes, I know it would have been smarter to order a set of 3.3Ohm, but I haven't but will be considering it soon)
- C6G is 33pF NP0/C0G ceramic.

I read a little before that having the amp in a solid case is required and I got the same results when I was testing uncased as when I put it into the recommended Hammond.

I think the solder job was ok, shouldn't have any cold joints, perhaps the opamp socket soldering is a little worse than it could since it's harder to have the pins in physical contact before soldering. This will be looked into and double checked as soon as I get a chance.

1) What is the recommended way to connect the case to the board?
2) What would you guys recommend doing next to track down the problem?

Thanks in advance.

 

[steinchen]

Quote:

Originally Posted by FallenAngel /img/forum/go_quote.gif - C6G is 33pF NP0/C0G ceramic.

try 100pF and the oscillations should be gone

 

[FallenAngel]

Thanks Steinchen, I'll be replacing those as soon as I get my hands on a pair of 100pF NP0/C0G ceramics. I've got a few X7R but I was reading these aren't good enough in this position so I'll be ordering in a day or two (combining projects, eh).

Yes, I said pair because I have 2 PPAv2 amps, both with similar parts and opamps except the second has BD139/BD40 output transistors and it is showing the EXACT same oscillation

. Just my luck I guess.