[daggerlee]

Quote:

Originally Posted by Dexdexter /img/forum/go_quote.gif Thanks for the update, dl. I'm following your op-amp adventures with keen interest!

I really cannot wait to hear the AD chips in here...I am looking forward to their 'sparkle.'

Also I can confirm that my Etys sound great through the Heed, when paired with an impedance adapter. Currently using a 100 ohm one which sounds great, but a 75 ohm one (which turns the 4P into a 4S) also sounds great. But before, when I was driving the Ety 4Ps without an adapter, they did sound bad, because of volume mismatch at low volumes.

After reading that the Heed is able to drive K1000s...well now I am positively intrigued... I thought I was done with Head-fi forever once I upgraded my DAC...

- Well actually boomana helped clarify, apparently they drive them but not to their full potential. Looks like that's another $800 saved

 

[daggerlee]

Received my opamps today but not my adapter, which means I have only been able to roll in the AD823.

From brief observations, I find the AD823 to have more treble 'spark,' that is cymbal hits and such seem faster and sharper than the OPA627's. On the other hand, the OPA627's have a midrange - lower range presence and fullness that the AD823 lacks - this can make treble heavy records sound almost 'metallic' and can leave some drums sounding dull (I am thinking of the drums in Prince's when doves cry).

Of course I haven't really done any long term listening or comparisons (and I don't think I will have time

) Am eagerly awaiting my adapters so I can roll in the AD8620 and the LM4562.

Also, as a sidenote, I will NOT be going back to the NE5532 that was originally in the amp, not for sound reasons, but because with the NE5532 in there I measure DC offset at around 30 mV per channel, which while not harmful to headphones (I've heard anything under 50mV shouldn't be a problem) is still much higher than I'd like. With the OPA627's they measure at less than 0.5 mV per channel, same with the AD823.

 

[GASTAN]

I love my K1000 with Heed (beats W5000 & RS-1 on Ear 150+ HD, as well as on CanAmp
well, I have yet to try them with more powerfull amp, but using them with CanAmp makes me almost happy (I just hope for slightly more bass with big amp)

--G

 

[upstateguy]

Quote:

Originally Posted by daggerlee /img/forum/go_quote.gif Also, as a sidenote, I will NOT be going back to the NE5532 that was originally in the amp, not for sound reasons, but because with the NE5532 in there I measure DC offset at around 30 mV per channel, which while not harmful to headphones (I've heard anything under 50mV shouldn't be a problem) is still much higher than I'd like. With the OPA627's they measure at less than 0.5 mV per channel, same with the AD823.

Hi DL

Why do you think that is?

Is it possible that it is intentional, and part of a design feature we haven't uncovered yet?

Btw, did the sound-stage change with the 627s?

Regards

USG

 

[daggerlee]

Quote:

Originally Posted by upstateguy /img/forum/go_quote.gif Hi DL Why do you think that is? Is it possible that it is intentional, and part of a design feature we haven't uncovered yet? Btw, did the sound-stage change with the 627s? Regards USG

Well I'm not sure, but I don't think high DC offset is an intentional part of any design. I think it is just a result of the fact that the NE5532 is a fast, bipolar opamp, and not part of the Heed design itself. By the way, not all canamps may measure that high, but the two I measured did. I have heard of other canamps measuring a more reasonable 15 mV per channel with the NE5532.

I've just remeasured with the NE5532 in, and it's at around 22-24 mV per channel. Perhaps it was just variation among NE5532 opamps - the one I'm using isn't the one that originally shipped with the amp. Perhaps the batch that Heed used for my canamp and the other one I measured were just a bit higher than usual. Or perhaps my measuring method is flawed, but I doubt it...

 

[Lad27]

Quote:

Originally Posted by daggerlee /img/forum/go_quote.gif Well I'm not sure, but I don't think high DC offset is an intentional part of any design. I think it is just a result of the fact that the NE5532 is a fast, bipolar opamp, and not part of the Heed design itself. By the way, not all canamps may measure that high, but the two I measured did. I have heard of other canamps measuring a more reasonable 15 mV per channel with the NE5532. I've just remeasured with the NE5532 in, and it's at around 22-24 mV per channel. Perhaps it was just variation among NE5532 opamps - the one I'm using isn't the one that originally shipped with the amp. Perhaps the batch that Heed used for my canamp and the other one I measured were just a bit higher than usual. Or perhaps my measuring method is flawed, but I doubt it...

High DC offset (and for this part of circuit ANY DC voltage) is indeed not desirable. Normally designer will place a DC blocking capacitor on the signal line.

As with so many other things around here, there are two camps - one claiming that no capacitor is good (claiming degradation of sound quality) and they will happily live with some serious DC offsets - good example is a massive 'Mod your Zhaolu' thread.

The other camp will go for zero offset - that is good idea, providing it is implemented right - the cap should be at least polystyrene, preferably with extremely low ESR parameter. These caps are not all that common though, adding to the cost of amp, so more often than not just plain-jane ceramic or other cheap caps are there.

I would personally prefer good DC blocking caps (not an easy mod for uninitiated - it would certainly involve cutting off PCB track).
If that cannot be done I'd rather live with up 10-20mVdc offset than have cheap'n'nasty cap in the audio path.

I haven't opened mine yet, but I'll do so soon and make some measurement. Then I decide what next.

Maybe opamp rolling? I just got that feeling than CanAmp is what it is (soundwise) thanks to NE5532. These are really cheap compared to Burr Browns or Analog Devices opamps, but there are millions of 5532's all over audio (hence the low price?). They are really good design, very, very solid performers. So will see how big "Mod your CanAmp" ends up.

 

[daggerlee]

Lad27 I'd be curious to see what sort of measurements you are getting with the NE5532. I definitely think opamp rolling is the key, as the OPA627 and AD823 I have rolled in so far both demonstrate less than 3 mV DC offset per channel, so I definitely think the high DC offset readings I measured from the NE5532 are characteristic of the opamp itself and not the Heed design.

After a bit more listening, I think I prefer the OPA627 to the AD823 at least for now. This is due to my setup - the OPA627's lend a nice richness to the midrange and lower frequencies that fills out the K701's sound, whereas the AD823 were a bit too thin for my taste. As far as soundstage, I couldn't detect a noticeable difference between one and the other...

I received my adapters today so will later be rolling in AD8620 and LM4562, I am excited about both of them.

 

[Lad27]

Quote:

Originally Posted by daggerlee /img/forum/go_quote.gif Lad27 I'd be curious to see what sort of measurements you are getting with the NE5532. I definitely think opamp rolling is the key, as the OPA627 and AD823 I have rolled in so far both demonstrate less than 3 mV DC offset per channel, so I definitely think the high DC offset readings I measured from the NE5532 are characteristic of the opamp itself and not the Heed design. After a bit more listening, I think I prefer the OPA627 to the AD823 at least for now. This is due to my setup - the OPA627's lend a nice richness to the midrange and lower frequencies that fills out the K701's sound, whereas the AD823 were a bit too thin for my taste. As far as soundstage, I couldn't detect a noticeable difference between one and the other... I received my adapters today so will later be rolling in AD8620 and LM4562, I am excited about both of them.

DL,

3mV dc sounds good, that's negligible. Glad to hear that soundstage is intact - that is what drew me away from Grado's into Beyer team

.

As I said I haven't opened mine, if amp is going to go "belly up" it probably would within first month, so I'm giving it a grace period.

I've looked at internal guts of 5532 (being involved in IC design in my distant past) - I can't see any reason why the DC offset is there.
Unfortunately I don't have the CanAmp schematic and no time to go tracing every component just to see how circuit topology may contribute,
but again, it all points back to 5532 itself. I'm speculating that perhaps some manufacturing batches could be out of spec.

I'm really curious about AD8620 - it should (according to datasheet) perform really well.

Let us know DL.

 

[leener]

hello,
canamp has an output along with input, what is this output for?
i'm sorry may be this question is rather stupid but i'm NO good at all in all this. i mean can i connect let's say CDP + EQ through canamp to something else (for ex. another Amp2) so that i could use EQ with both canamp and Amp2 (which i can use with speakers) ?
well, any way - what is this output for?

CDP1 > EQ > CanAmp > CDP2

thnx in advance, leener

 

[Shpongle_Dongle]

Quote:

Originally Posted by Lad27 /img/forum/go_quote.gif The other camp will go for zero offset - that is good idea, providing it is implemented right - the cap should be at least polystyrene, preferably with extremely low ESR parameter. These caps are not all that common though, adding to the cost of amp, so more often than not just plain-jane ceramic or other cheap caps are there.

To prevent roll off you'd need at least a 220uF value in series with the output with nearer to 1000uF being preferable as a catchall for all 'phones from 32R to 600R a simple calculator where you enter the impedance as that of your headphones (ie: Grado would be 32 ohm in the resistor field) http://www.muzique.com/schem/filter.htm So film caps are pretty much no go as output caps due to the sheer size of them, a 220uF film cap would be the size of a house, non polar electrolytics are best in this application IMO.

Quote:

I would personally prefer good DC blocking caps (not an easy mod for uninitiated - it would certainly involve cutting off PCB track). If that cannot be done I'd rather live with up 10-20mVdc offset than have cheap'n'nasty cap in the audio path.

Actually very simple to impliment and not necessary to cut any track. The LH channel wire that connects from the board to the headphone socket simply needs to be cut with a 220uF non polar capacitor soldered in between (in series) same with the RH Wire.... these series capacitors will null any DC that is present at the output. You're pretty much guaranteed negligible offset with FET / diFET chips such as the AD 823, OPA627, OPA2132P, OPA2134P, AD8620 etc etc


Quote:

Maybe opamp rolling? I just got that feeling than CanAmp is what it is (soundwise) thanks to NE5532. These are really cheap compared to Burr Browns or Analog Devices opamps, but there are millions of 5532's all over audio (hence the low price?). They are really good design, very, very solid performers. So will see how big "Mod your CanAmp" ends up.

NE5532 are superb chips but not politically correct with audiophools who prefer to pay mega bucks for flavour of the month chips which probably don't sound anywhere near as good as the NE5532.

 

[Rav]

Quote:

Originally Posted by leener /img/forum/go_quote.gif hello, canamp has an output along with input, what is this output for? i'm sorry may be this question is rather stupid but i'm NO good at all in all this. i mean can i connect let's say CDP + EQ through canamp to something else (for ex. another Amp2) so that i could use EQ with both canamp and Amp2 (which i can use with speakers) ? well, any way - what is this output for? CDP1 > EQ > CanAmp > CDP2 thnx in advance, leener

It's a pass through, so you can connect another amp to the canamp's output. What comes out of the output is the same as what goes in to the input, the canamp doesn't amplify the signal or anything.

 

[Shpongle_Dongle]

Quote:

Originally Posted by Lad27 /img/forum/go_quote.gif I've looked at internal guts of 5532 (being involved in IC design in my distant past) - I can't see any reason why the DC offset is there. Unfortunately I don't have the CanAmp schematic and no time to go tracing every component just to see how circuit topology may contribute, but again, it all points back to 5532 itself. I'm speculating that perhaps some manufacturing batches could be out of spec.

Scroll down to Mark's "DC Issues" section http://www.mhennessy.f9.co.uk/preamp...e_research.htm More in-depth on the 5532 here: http://www.dself.dsl.pipex.com/ampins/webbop/5532.htm

 

[Dexdexter]

Quote:

Originally Posted by leener /img/forum/go_quote.gif hello, canamp has an output along with input, what is this output for? i'm sorry may be this question is rather stupid but i'm NO good at all in all this. i mean can i connect let's say CDP + EQ through canamp to something else (for ex. another Amp2) so that i could use EQ with both canamp and Amp2 (which i can use with speakers) ? well, any way - what is this output for? CDP1 > EQ > CanAmp > CDP2

First and foremost, it is a line-out pass-thru to facilitate connection to a tape-monitor circuit of an amplifier.

So, yes, you could do CDP > EQ > CanAmp > Amplifier,

or CDP > EQ > Amplifier > Tape-Out > CanAmp,

or CDP > Amplifier > Tape-Out > EQ > CanAmp,

but not CDP1 > EQ > CanAmp > CDP2

I'm sure it has other possibilities and perhaps someone might post about those.

 

[Lad27]

Quote:

Originally Posted by Shpongle_Dongle /img/forum/go_quote.gif Scroll down to Mark's "DC Issues" section http://www.mhennessy.f9.co.uk/preamp...e_research.htm More in-depth on the 5532 here: http://www.dself.dsl.pipex.com/ampins/webbop/5532.htm

Thanks for links. Very good sites. I've bookmarked them.

 

[daggerlee]

Have been rolling opamps - so far I have to say my favorite is the LM4562, it has the nice roundedness of the OPA627 sound, but sounds as 'fast' as the AD8620. Although truth be told, at least to these ears the differences are minimal, and I'd be happy with any of the three - OPA627, AD8620, or LM4562.