[headfone]

Earlier, I posted this topic in the IEM sub-forum which has rec'd no replies thus far. I think DIY is more appropriate as the PPA2 -- or at least, my construction of it -- is the worst offender.

General question to those of you using PPA with IEMs:

If you turn the volume all the way down, do you hear hiss? If so, how much?

Thx!

 

[rds]

Yeah, I have that 'hiss' with my e3s and PPAv2.
It is actually a very pleasant white noise. Once I found that I had been sitting there listening to it for almost 20 minutes lol.
It is independent of volume and I think it is some issue with poor insulation in the shure cable. I found that if I twist around the section where the cable splits I can stop the hiss completely and the playback sounds totally normal.
You can try twisting the cable, but I've found it's very hard to get it in the right position. I think basically what happens is you end up separating the area with the poor dielectric and thus kill the hiss.
Oh and it is an issue with 3 channel buffered amps. If you have another one like the M3 it would be interesting for you to try it. You should end up with the same hiss.

 

[funch]

That condition is common with the PPAv2 if the amp is using the
default 2R2 output resistors. I experienced it with my first PPAv2,
so I switched to 3R3's and the hiss went away.

 

[rogerlike]

Quote:

Originally Posted by funch /img/forum/go_quote.gif I experienced it with my first PPAv2, so I switched to 3R3's and the hiss went away.

Yea, no hiss with my PPAv2

 

[headfone]

Quote:

Originally Posted by rds /img/forum/go_quote.gif It is independent of volume and I think it is some issue with poor insulation in the shure cable. I found that if I twist around the section where the cable splits I can stop the hiss completely and the playback sounds totally normal. You can try twisting the cable, but I've found it's very hard to get it in the right position. I think basically what happens is you end up separating the area with the poor dielectric and thus kill the hiss. Oh and it is an issue with 3 channel buffered amps. If you have another one like the M3 it would be interesting for you to try it. You should end up with the same hiss.

Yup: independent of volume.
I have Shure’s, too. Couldn’t get the cable-twist trick to work.
I have a Xin Super Macro amp, which has 4-channels. No hiss. Ditto on Go-Vibe (dunno if this is 3-channel and/or buffered).
A heavily-modified Meier Corda II has moderate hiss; a modded HeadRoom Little has almost none.
Quote:

Originally Posted by funch That condition is common with the PPAv2 if the amp is using the default 2R2 output resistors. I experienced it with my first PPAv2, so I switched to 3R3's and the hiss went away.

Great tip! Was SQ affected by the R change?

I experience hiss on both my SE530s and EC2s. I have Senn IE-8s on order. It’ll be interesting to compare the two IEM brands on the hiss issue. So, I’ll hold off on the resistor mod ‘till the Senns arrive.

 

[amb]

The hiss is not a characteristic of 3-channel active ground amps. It is an indication that the amp's gain is too high for IEMs, which are usually extremely efficient. The default gain of the Pimeta, PPA and M³ is 11 (R4 = 10K, R3 = 1K). It is appropriate for high impedance headphones and a weak source. For IEMs I recommend dropping the gain down to around 2. Changing R3 to 10K will do the trick. It is arguable whether voltage amplification is needed for IEMs at all. Since they are usually low impedance (less than 32 ohms), just a unity-gain buffer should do, such as a JISBOS.

 

[headfone]

Quote:

Originally Posted by amb /img/forum/go_quote.gif The hiss is not a characteristic of 3-channel active ground amps. It is an indication that the amp's gain is too high for IEMs, which are usually extremely efficient. The default gain of the Pimeta, PPA and M³ is 11 (R4 = 10K, R3 = 1K). It is appropriate for high impedance headphones and a weak source. For IEMs I recommend dropping the gain down to around 2. Changing R3 to 10K will do the trick. It is arguable whether voltage amplification is needed for IEMs at all. Since they are usually low impedance (less than 32 ohms), just a unity-gain buffer should do, such as a JISBOS.

Hmm...I would like PPA2 to work for a wide variety of cans: IEMs, Grado, Senn, AKG, beyer, etc.
For models like HD-650, I find that the volume control needs to be well past 12-o-clock for moderate-level listening.
Perhaps a two-way switch that selects between the above-noted gain Rs will be a viable option.

 

[NelsonVandal]

If you don't hear any hiss with IEM's, you need to remove some ear wax. All (op)amps are hissing, some more and some less.

 

[nikongod]

Quote:

Originally Posted by headfone /img/forum/go_quote.gif For models like HD-650, I find that the volume control needs to be well past 12-o-clock for moderate-level listening.

What gain did you use? voltage gain: 3, 6, 9.75, 12, etc?
What opamps?

There is nothing wrong with using all of the attenuator's range. The last (maximum) 1/3 of most attenuators is better matched than the first 2/3 anyways. If the cans play at a winning SPL at maximum spin you are OK.
Quote:

Perhaps a two-way switch that selects between the above-noted gain Rs will be a viable option.

You may prefer the results of just switching a smallish (25-50ohms) resistor between the outputs of the amp and the headphone.

You could try getting better opamps too. 3:1 says you would jsut hear the source hissing after that... sorry about your wallet.
Quote:

Originally Posted by NelsonVandal /img/forum/go_quote.gif If you don't hear any hiss with IEM's, you need to remove some ear wax. All (op)amps are hissing, some more and some less.

I sort of agree with this philosophically, but not everything hisses audibly. certainly at an SNR of -120db (vs actual listening level, not maximum level!) you would call it quiet.

 

[headfone]

Quote:

Originally Posted by nikongod /img/forum/go_quote.gif What gain did you use? voltage gain: 3, 6, 9.75, 12, etc? What opamps?

Gain=11 (don't recall if this is PPA2 "default"?)
Opamps: 637s for L/R, and 627 for "ground".
Quote:

There is nothing wrong with using all of the attenuator's range. The last (maximum) 1/3 of most attenuators is better matched than the first 2/3 anyways. If the cans play at a winning SPL at maximum spin you are OK.

Agreed. IIRC, most volume sections note 0db at max setting and negative infinity at min. Not sure what the most linear range is tho'.
Quote:

You may prefer the results of just switching a smallish (25-50ohms) resistor between the outputs of the amp and the headphone.

How does this affect SQ, other than eliminate hiss, of course?
UPDATE: I tried this w/47R (three, in series with all three channels, at the output jack). Hiss was gone, but SQ was significantly reduced.
Don't have any other R values on hand to experiment further.
Quote:

You could try getting better opamps too. 3:1 says you would jsut hear the source hissing after that... sorry about your wallet.

See above for my opamps. I have some AD8610 "in stock", too (saving these for the day I get tired of the BB sound!).
Relatively speaking, I think opamps have far less to do with hiss than gain (which I didn't compensate for during construction; but as many of posters in this thread have correctly deduced to be the culprit) Quote:

I sort of agree with this philosophically, but not everything hisses audibly. certainly at an SNR of -120db (vs actual listening level, not maximum level!) you would call it quiet.

Going off-topic here … but to a large degree, the significance of an opamp is often over-rated. Swapping opamps is a quick-n-dirty “fix” to issues that have better -- and often less-$ -- science-backed solutions. IME, even "Jellybean" opamps can be made to sound (and measure) very good using proper bypassing, decoupling and snubbing techniques (example 1; example 2, example 3). Of course, these same techniques work on the "better" opamps, too.

 

[headfone]

I didn't see Tangent's Electronic Calculator 'till now.
Gain, Noise, Output DC Offset ... very handy!

 

[headfone]

Quote:

Originally Posted by amb /img/forum/go_quote.gif The hiss is not a characteristic of 3-channel active ground amps. It is an indication that the amp's gain is too high for IEMs, which are usually extremely efficient. The default gain of the Pimeta, PPA and M³ is 11 (R4 = 10K, R3 = 1K). It is appropriate for high impedance headphones and a weak source. For IEMs I recommend dropping the gain down to around 2. Changing R3 to 10K will do the trick. It is arguable whether voltage amplification is needed for IEMs at all. Since they are usually low impedance (less than 32 ohms), just a unity-gain buffer should do, such as a JISBOS.

On his PPA2 page, Tangent notes that..."The only value you're likely to need to change is R4, to adjust the gain. You could instead adjust R3, but this would upset the impedance balance at the inputs of the op-amp, increasing distortion."

So, given Tangent's comments above and that

(R4/R3 + 1) = gain

would lower values for both R4 and R3, say 5K, be better?

 

[NelsonVandal]

.

 

[NelsonVandal]

Quote:

Originally Posted by headfone /img/forum/go_quote.gif Gain=11 (don't recall if this is PPA2 "default"?) Opamps: 637s for L/R, and 627 for "ground".

Gain 11 is crazy for use with IEM's.
Quote:

Originally Posted by headfone /img/forum/go_quote.gif Relatively speaking, I think opamps have far less to do with hiss than gain (which I didn't compensate for during construction; but as many of posters in this thread have correctly deduced to be the culprit)Going off-topic here … but to a large degree, the significance of an opamp is often over-rated.

You're dead wrong. Nothing in an amp such as the PPAv2 colors the sound as much as the opamps - power supply, "audio grade" resistors or capacitors etc.

 

[tangent]

Quote:

Originally Posted by headfone /img/forum/go_quote.gif would lower values for both R4 and R3, say 5K, be better?

The PPA docs are correct: you want to change only R4, not R3. Make R4 1K, not R3 10K.

If you've added a bass boost adjustment pot, you might also want to put something in R7, if you find that the bass boost adjustment is now twitchy, with too little rotation giving too much effect. Something in the low K-ohms range will moderate the pot's effect on bass boost.