New, Cheap Dynalo...

[Pars]

The caps he used had a Dv/Dt of 1200 or so IIRC. I figured out which ones they were at one time, but could never find them... probably have to go direct which he can do easily, but us mere mortals can't

[Beefy]

the way experiments typically work is as follows: you start with a hypothesis. your hypothesis is that the caps make a difference. you design an experiment.......

Yep to all of that. The BIG one to do at the design stage is to do experiments where you also try and disprove your hypothesis. Confirmation bias is a bitch, but do the negative tests in the first place and people can't challenge you.

My suggested test would be to swap the caps between the two units. See if the 'preferred' listening follows the caps or remains with the rest of the amp.

[luvdunhill]

I'm increasingly finding that less capacitance sounds better than more. For the currents needed for a low power headphone amp, try to keep it around 100uF. I'd be curious to hear what you think.

[Beefy]

I'm increasingly finding that less capacitance sounds better than more.

You fail at audiophile

[Maniac]

I'm increasingly finding that less capacitance sounds better than more. For the currents needed for a low power headphone amp, try to keep it around 100uF. I'd be curious to hear what you think.

Same capacitor design with different capacitance can have different response to the same frequency input, and also like many have said before, cheaper cap with large capacitance is not as good as better (often more expensive) caps at lower capacitance.

[Maniac]

I seem to recall reading this on the DynaHi thread on hf. I was skeptical, but I do know that Dr. Gilmore used some crazy fast Illinois Caps in his original Dynalo supply psu (last caps in the PSU). Just curious how big of FMs you were using? 470uf I would guess... I thought I would try 100uf in the one I am nearing completion on, but might try some film in it and see if I hear anything.

I think I used 470uF FMs, those were physically pretty small ones. IMHO for analog stuff, stay away from caps with ultra-high high freq ripple current and almost no low freq ripple current, get ones that are somewhat more even.

If you plan to try a few different caps on the amp board itself, I'd suggest soldering it more or less like SMD parts, instead of sticking it through the mounting position and solder. Less wear and tear on the PCB when you switch to different caps. And when you decided on one, then do solder them via the through hole.

[Maniac]

Yep to all of that. The BIG one to do at the design stage is to do experiments where you also try and disprove your hypothesis. Confirmation bias is a bitch, but do the negative tests in the first place and people can't challenge you.

My suggested test would be to swap the caps between the two units. See if the 'preferred' listening follows the caps or remains with the rest of the amp.

I'm not about to design a product or an experiment by committee, as the result is only used for internal use, I was only sharing some info with people here and again I'm not trying to write my PHD thesis here or publish a paper on AES or some scientific journal. The full detail of the test result is only used for development, and I doubt any manufacturer would really post all details on their tuning process for everyone to critique, comment, and change the whole deal with a ton of comment from the web.

Even if all those comments are good and valid, and with lots of manufacturer doing that, you'd likely end up with more than one manufacturer sounding close to the same, and the entire development cycle dragged on and on and on. That just doesn't work for any business, that only works if you got a lot of leisure time doing leisure pursuit.

As for comparing the caps' difference, your suggestion will actually make the test invalid, as constant resoldering and desoldering the cap will likely damage the cap from some point on. The better method is to install another set of the same cap on the second unit, and that is when the tester reported that the difference is getting hard to tell.

Yes, some may say that there can be differences between the same part, but finding out if the same batch of parts can have that much of a difference is also useful. Which we didn't find any meaningful difference between caps from the same batch.

[Maniac]

if only there were some kind of socket and pin system that could be used...

that would be another control invalidation, regardless of your anecdotal experience.

most of them already do. there are only so many ways to design and build a good amplifier. i personally doubt that many could give a reliable preference between two reasonably designed and built amps of reasonably similar capabilities in a proper double blind. the differences, really, are all in our heads. experts can't even reliably tell whether somebody is playing a Strad or a dime store fiddle when they are merely separated from the player by a cloth curtain. and double blinds in wine testing are laughable, with experts not able to tell that they are tasting the same bottle of wine over and over. i think there are lessons to be learned here.

Socket and pins??? Are you kidding me? This is not some large filter caps that are placed far and away from the main circuit, this is decoupling cap, the contact performance is crucial here. Degrading it with a socket and pin just makes the test pointless.

Design by committee is one way to fail, like Apple Newton.

[Maniac]

one would imagine that, if the expensive caps do indeed perform better, then there would still be a increase in performance when both types of caps are equally handicapped. it's a valid control.

you apparently know very little about the Apple Newton. first, it wasn't designed by a committee any more than any other highly technical and complicated technology is (in fact, it was John Sculley's baby the way the iPod and the original Macintosh were Steve Jobs' babies), and it was hardly a failure, ushering in many new technologies and showing that a PDA could be a practical product, even in an imperfect early version. where do you get this stuff?

Clearly you do not know most of those sockets available on the market will become so loose in about 20 inserts and pulls that it is no longer useful for anything other than recycling material. If you don't understand the failure mode of the parts in the middle, and how they work, then I really would suggest that you check those out and perhaps study how those could affect the test result before offering them as a solution. Since it is clearly not a solution at all but another bigger can of worms.

As for Apple Newton, it is still a failure, it failed. It ushered in a lot of technology, but was not able to remain viable. And faded away in history with nothing more than some technical firsts to its name.

[Maniac]

i concede the point. that's a lesson, too, but for you.

faded away? hardly. the Apple we know now is a direct descendant of the Newton. the clean design aesthetic of many current Apple products was ushered in with the Newton, and the Newton gave Apple early portable industrial design experience (you of course know Jonathon Ive designed the Newton, i'm sure) and helped allow the company to avoid missteps when it reentered the portable market, the market that has taken Apple from being a relatively minor computer company to a dominating juggernaut. the human/machine interface of the Newton was also a proving ground for software and ergonomic techniques and ideas, many of which Apple has used to build devices that shine most brightly due to their elegant interactivity. if that is fading away then your inductive reasoning is solid and your conversational skill pithy. now, if you'll pardon me, there is a dust mote over in the corner that seems to present more intellectual elasticity than just about anything you've posted in the last several thread pages, so i'm going to divert my attention.

Well, don't let me hold you back.

[Maniac]

you know, you're right. i think you do need a brief time out period to decide whether your presence beings anything to this community. thinks for bringing it up.

So that's your excuse for editing other people's profile about double blind, which I never claim I did.

[Beefy]

I'm not about to design a product or an experiment by committee, as the result is only used for internal use, I was only sharing some info with people here and again I'm not trying to write my PHD thesis here or publish a paper on AES or some scientific journal. The full detail of the test result is only used for development, and I doubt any manufacturer would really post all details on their tuning process for everyone to critique, comment, and change the whole deal with a ton of comment from the web.

Sure, but you're spending what I assume is big dollars on custom caps for what may or may not be flawed data. And if you weren't really happy with the idea of these whizzbang caps, you wouldn't have posted a picture of them in this thread. Unless you do the tests, you are lying to us and to yourself.

As for comparing the caps' difference, your suggestion will actually make the test invalid, as constant resoldering and desoldering the cap will likely damage the cap from some point on. The better method is to install another set of the same cap on the second unit, and that is when the tester reported that the difference is getting hard to tell.

Same shit, different shovel. You wouldn't need to swap it back and forward several times, but just once. And you would have to make 100% sure that the listener never knew what you had done.

[n_maher]

So that's your excuse for editing other people's profile about double blind, which I never claim I did.

Let me boil it down for you to its simplest form.

You claim you performed a test. Your "test" was not a valid test and therefore it's results are nearly meaningless. Just because you performed a random series of events does not make it a valid test. If you would like to claim that you, through empirical data gathering, discerned that most users like a certain setup better than another that's fine. It's just not a conclusive result.

[LarsMarkelson]

Hey GPH and other HA-006+ owners,

HA-006++ Special Edition pre-order just went up. It's supposed to have a number of upgrades... stepped attenuator this time for volume control for one... and a bunch of other stuff that may or may not improve the sound. Anyway, I'll be able to compare to my HA-006+ now when I get it, I pre-ordered one myself just now. The coolest thing is that you can choose your own gain (1x / 3x / 11x) this time. I chose 1x since I plan to use it with JH-13's in the future. Oh and I got free shipping. Maniac's been posting in the other forum about it under the review for Sheer Audio HA-006+ (DynaLo) ? thread.

Here's the link to pre-order, http://www.acoustic-fun.com/catalog/product_info.php?products_id=83

[GPH]

Hey GPH and other HA-006+ owners,

HA-006++ Special Edition pre-order just went up. It's supposed to have a number of upgrades... stepped attenuator this time for volume control for one... and a bunch of other stuff that may or may not improve the sound. Anyway, I'll be able to compare to my HA-006+ now when I get it, I pre-ordered one myself just now. The coolest thing is that you can choose your own gain (1x / 3x / 11x) this time. I chose 1x since I plan to use it with JH-13's in the future. Oh and I got free shipping. Maniac's been posting in the other forum about it under the review for Sheer Audio HA-006+ (DynaLo) ? thread.

Here's the link to pre-order, Acoustic-Fun

Cool, I'll be interested to read your thoughts. I'm planning to change the gain myself soon on my unit, 11 is way too high for the JH13. Stepped attenuator, I'm not a big fan unless the steps are really close. How many steps is he using?

[Salt Peanuts]

Stepped attenuator, I'm not a big fan unless the steps are really close. How many steps is he using?

21 steps, according to the website.

[swt61]

That's nine more than AA.

[LarsMarkelson]

Hey GPH and anyone else using HA-006+ with JH-13,

Even with the volume knob at 0 I am getting a definite hiss in the right ear with my JH-13. It becomes noticeable on the left side too when I move the knob above 9 o'clock, but is a good deal less than the right.

Are you guys experiencing the same?

[GPH]

No hiss at listening volumes for me, but I pick a very small hum with the JH13.

EDIT: I have to add that I'm not sure if the hum is due to the HA-006+ or something else in the chain.

Edited April 30, 2010 by GPH

[Pars]

I would check DC offset on the amp, particularly if I was going to be plugging expensive IEMs into one of these.

[LarsMarkelson]

I would check DC offset on the amp, particularly if I was going to be plugging expensive IEMs into one of these.

How do I do this? I have the back of the amp set on 110V and am using the amp in the US... is this what you're talking about?

The hiss is really bothersome. I have to turn the music up a good amount to mask it, and even then I feel like it is negatively affecting SQ.

I'm pretty much sure it is the amp as well. I plugged directly into Pico DAC line out and no hiss at all, and also listened straight from my Sansa Clip+ and no hiss either.

[Pars]

DC offset is measured at the output with the pot turned all the way down. With a DMM (digital multimeter) set to DC volts, you measure from L (tip) to ground (sleeve) and R (ring) to ground (sleeve). These should ideally be <1 millivolt (0.001 Vdc), but anything under 5-10mV should be fine. Easiest way to do it is to use a blank 1/4" phone plug and do your measurements across the contacts; avoids having to open the unit.

Excessive DC offset could trash your IEMs. I'm not pointing fingers at the manufacturer or anything, but since this is a DC coupled amp, there are no coupling caps to protect you from this.

One other thing you could try is unplug the amp from your source, turn the volume all the way down, and then listen. If you still hear the noise, then it is the amp. If not, blame your source.

[GPH]

I'm pretty sure there's something wrong with your unit Lars, I'd get it checked. I just measured DC offset on mine and I get 2mV if you want to use that as a reference. Also, I only hear hiss at very high volumes, far outside my comfortable listening range and the only time I can hear a small hum is with the JH13, otherwise it's completely silent.

[Shahrose]

Has anyone compared this amp to the GS-1 or Gilmore Lite? If this is only a little better than the Glite, then I'll pass, but if it turns out to be as good as a GS-1 then it would be a terrific bargain.

BTW I'm referring to the upgraded version: HA-006++ selling for $450. Pics here: http://www.head-fi.org/forums/6573925-post84.html

[LarsMarkelson]

Has anyone compared this amp to the GS-1 or Gilmore Lite? If this is only a little better than the Glite, then I'll pass, but if it turns out to be as good as a GS-1 then it would be a terrific bargain.

BTW I'm referring to the upgraded version: HA-006++ selling for $450. Pics here: Head-Fi: Covering Headphones, Earphones and Portable Audio - View Single Post - review for Sheer Audio HA-006+ (DynaLo) ?

This actually hasn't been released released yet. I have one on pre-order right now, should be shipping soon - Maniac said he just got all the parts in.