Cavalli EHHA Embedded Hybrid Headphone Amp - Page 120

Need to add one more thing that I observed the transistor besides Q15 i.e. Q16 is cool. On the opposite side both the transistors are almost equally hot.

^ Isn't Q15 the one which lies in corner in your new case, seems poorly ventilated to me.

Not a ventilation problem I think. Checked the AMB forum, possibly its brother in parallel (Q16) is bad as its heatsink is damn cool while Q15 is carrying its load and overheating. Will replace it as suggested in the AMB forum if I find one in the market otherwise will need your help to source it!

It explains why the power rails are delivering enough power and the amplifier is running fine. Rest of the MOSFETs on the S22 as well as the BJTs on the amp boards are equally hot (can touch them and count to 10).

I auditioned an EHHA RevA made by local Head-fier BK8569er and loved how it sounded.  I would describe it as sounding fast, clean, linear, airy.  In comparing it to my SOHA1, I kept coming back to comparing the EHHA to the stax headphone sound while the SOHA had the more dynamic sound.  Similar to Stax, my only complaint with this amp was it seemed to lack dynamic punch or a sense of weight on certain phones.  But that's a minor quibble and not necessarily a bad thing.  Other than dynamics I thought it clearly bettered the SOHA1 in every way and made the SOHA sound colored. 

So I bought one on the forums made by a HF'er from a glassjar kit.  I'm not here to name names, I know he's a good builder and the amp seems to be well done.  But it doesn't sound nearly as good as the other EHHA I heard (and borrowed for a week).  In fact, as far as I can tell my SOHA1 beats this EHHAreva in every way including speed and detail, things the other EHHA was clearly superior in before.  BK8569er has been kind enough to check out the amp for me and do some tests for obvious problems and hasn't found anything glaring, and he confirmed it looked like a good build at least on the surface.  He also lent me some tubes which didn't change much.  I also tried switching from BJT's to Mosfets. 

Maybe I just shouldn't be buying DIY amps if I know nothing about DIYing.  But I like owning things that have so little marketing buzz behind them and instead have lots of love!

So I'm wondering if anyone here would be willing to have a look at it.  If you own an EHHA it might be better so you can compare but not necessary.  I'm not listening to it at all and not sure what else to do. 

Quote:

Originally Posted by rhythmdevils 

I auditioned an EHHA RevA made by local Head-fier BK8569er and loved how it sounded.  I would describe it as sounding fast, clean, linear, airy.  In comparing it to my SOHA1, I kept coming back to comparing the EHHA to the stax headphone sound while the SOHA had the more dynamic sound.  Similar to Stax, my only complaint with this amp was it seemed to lack dynamic punch or a sense of weight on certain phones.  But that's a minor quibble and not necessarily a bad thing.  Other than dynamics I thought it clearly bettered the SOHA1 in every way and made the SOHA sound colored. 

 

So I bought one on the forums made by a HF'er from a glassjar kit.  I'm not here to name names, I know he's a good builder and the amp seems to be well done.  But it doesn't sound nearly as good as the other EHHA I heard (and borrowed for a week).  In fact, as far as I can tell my SOHA1 beats this EHHAreva in every way including speed and detail, things the other EHHA was clearly superior in before.  BK8569er has been kind enough to check out the amp for me and do some tests for obvious problems and hasn't found anything glaring, and he confirmed it looked like a good build at least on the surface.  He also lent me some tubes which didn't change much.  I also tried switching from BJT's to Mosfets. 

 

Maybe I just shouldn't be buying DIY amps if I know nothing about DIYing.  But I like owning things that have so little marketing buzz behind them and instead have lots of love!

 

So I'm wondering if anyone here would be willing to have a look at it.  If you own an EHHA it might be better so you can compare but not necessary.  I'm not listening to it at all and not sure what else to do. 

i would look for measured differences in the the bias points, including the tube, VAS, and quiescent current on the output devices.  do both amps sport the same pot and output devices?  how did you roll the output devices - are they socketed and was the bias reset?  pics could be helpful.  

forgot to mention, the amount of NFB can influence the sound as well.  what are the open and closed loop gains for each amp?  

I'll add a bit of color since I apparrently own the "magic" RevA and also messed with rythmdevils unit at his place quite a bit.

With his original BJT setup I verified correct R17/22/23 and quiescent current.  Heatsink temps were normal.  I swapped in my GE/Amperex and later my Telefunken tubes for long term testing.

The output devices were socketed for easy changes, so we swapped in MOSFETS, changed to the appropriate R17/22/23, and adjusted the quiescent current per the setup instructions.  Everything measured and behaved normally.

R13/14 is 75k, but R11 is 1.5k as opposed to the BOM value of 1.2k.  jdkjake informs me that this will have the consequece of reducing the gain from ~9 to ~7.5 and increase NFB ~1.7dB.

The build itself is very well done with lots of attention to detail and solid workmanship.  The grounding scheme is executed differently than mine, and this build employs an Alps pot and dual Avel-Lindberg transformers.  FWIW, my own build closely follows the MOSFET BOM with individually sourced components and uses a TKD pot and a shielded SumR transformer.

I didn't do any critical listening, but to my ears, with his gear, rythmdevils unit functioned correctly and without major fault.  It made music.  Rythmdevils probably has some very senstive ears, but he reports that none of the changes we made (rolling tubes, BJT --> MOSFET, etc.) had any appreciable influence on the SQ.  I find that a little surprising - as if something is being masked.  As I recall the only source was a gamma-2 via optical.  I think at this point he wants to turn it over to an expert to work some magic on it - any takers?

BK

Quote:

Originally Posted by BK_856er 

I'll add a bit of color since I apparrently own the "magic" RevA and also messed with rythmdevils unit at his place quite a bit.

 

With his original BJT setup I verified correct R17/22/23 and quiescent current.  Heatsink temps were normal.  I swapped in my GE/Amperex and later my Telefunken tubes for long term testing.

 

The output devices were socketed for easy changes, so we swapped in MOSFETS, changed to the appropriate R17/22/23, and adjusted the quiescent current per the setup instructions.  Everything measured and behaved normally.

 

R13/14 is 75k, but R11 is 1.5k as opposed to the BOM value of 1.2k.  jdkjake informs me that this will have the consequece of reducing the gain from ~9 to ~7.5 and increase NFB ~1.7dB.

 

The build itself is very well done with lots of attention to detail and solid workmanship.  The grounding scheme is executed differently than mine, and this build employs an Alps pot and dual Avel-Lindberg transformers.  FWIW, my own build closely follows the MOSFET BOM with individually sourced components and uses a TKD pot and a shielded SumR transformer.

 

I didn't do any critical listening, but to my ears, with his gear, rythmdevils unit functioned correctly and without major fault.  It made music.  Rythmdevils probably has some very senstive ears, but he reports that none of the changes we made (rolling tubes, BJT --> MOSFET, etc.) had any appreciable influence on the SQ.  I find that a little surprising - as if something is being masked.  As I recall the only source was a gamma-2 via optical.  I think at this point he wants to turn it over to an expert to work some magic on it - any takers?

 

BK

 

all things being equal, as long as all bias points are reasonably matched, including 1.5mA through each triode, both amps should sound identical.  the fact that RD's amp is from a GJ kit and your's hand selected shouldn't matter.  while more expensive, as long as there's no noise, i don't believe the SumR will improve the sonics.  

i've done comparisons between the TKD and Alps - as in 2 identical amps side by side with the only difference in one sports an Alps and the other the TKD.  i've done this A/B comparison with 2 different amp designs as well.  they do sound subtly different from one another and can influence the final sound.  each have trade offs and i don't have a preference for either.

So, this being DIY, no two realizations will be *exactly* the same. Potentially different parts from potentially different manufacturers from potentially different batches with potentially different tolerances, well, you get the idea.

That said, I would drop the MOSFETs back in and bring R11 down to 1.2k  --OR-- bring R13/14 down closer to 56k. The R11 mod is easier.

During my experiments, the NFB setting was very important to the overall sound presentation. Changes of 2db were fairly pronounced while 1db changes were noticeable upon critical listening. Perhaps you are more sensitive to these nuances than others.

If that does not yield the desired results, then you may need to compare each and every part used between the two builds to see what might be the issue. Perhaps an off spec'ed part or two got inserted. Perhaps a part or two got swapped. Impossible to say without a deep, carefully comparison between the two instances. Tedious no doubt, but, certainly an exercise in intellectual curiosity that might yield the sound that you seek. Or not. Hard to say.

As for pots, I used both alps and TKD. The TKD seemed to be the better choice, but, I may have been biased by the seductive smoothness of it's operation. The TKD is one smooooth pot! Much love here for the TKD.

Thanks for the responses!  And a big shout out to BK for meeting up at least a few times and looking at the amp (and hanging out of course :-)  I'm constantly surprised and amazed by Head-fiers. 

To clarify, the amp doesn't sound bad necessarily.  It just doesn't sound very good.  It is pretty flat and lacks depth, air, detail, speed, etc.  It honestly sounds like a portable amp and I can't imagine anyone preferring it over the SOHA1 I have here.  So I can't see it being a simple matter of preference.  But who knows.  I should compare it to my portable PPAS and see which one wins.  I honestly think the PPAS might sound better just in the way of dynamics, sound stage, imaging, etc.

I don't have the ability to look over the amp like that due to some fairly serious health issues I'm facing right now and lack of knowledge about it.  So the 2 options I see is to sell it to someone for whatever it's worth or try to have it fixed.  I'm open to offers for either one.  I'm sure with the right tweak this would suddenly be an awesome EHHA.  It's got the Mosfet and BJT's socketed, and I have mosfets in there now. 

Anyone want to fix or buy?  I hate to see it just sitting here.  It's also kind of a big investment not to be using it.  Kind of sad, would be great with my orthos. 

Quote:

Originally Posted by BK_856er 

The build itself is very well done with lots of attention to detail and solid workmanship.  The grounding scheme is executed differently than mine, and this build employs an Alps pot and dual Avel-Lindberg transformers.  FWIW, my own build closely follows the MOSFET BOM with individually sourced components and uses a TKD pot and a shielded SumR transformer.
 

I don't know the EHHA amp very well at all, but you might look at this...

What are the differences between the EHHA and BETA22 in terms of sound?

      Quote:

Originally Posted by FOXY 

What are the differences between the EHHA and BETA22 in terms of sound?

Depends...

I prefer the HD800 with the EHHA and Beta22 with the LCD-2.

Overall I prefer the LCD-2 with the Beta-22 by a very wide margin. Possibly my very early serial numbered HD800’s were just too bright for me and the EHHA tamed them.

Does anyone here know of a replacement for the 2SC2705?