diyAudio (
-   Headphone Systems (
-   -   The Objective2 (O2) Headphone Amp DIY Project (

RocketScientist 1st December 2011 08:20 AM

@qusp, this is getting OT, but the trick with transformers is the frequency range you want. They can work fairly well but only over a limited bandwidth. The audio transformers in the cheaper ground isolators have serious trouble below 100 hz and start to get weird above 10 khz. The $100 ones from Jensen are much better but still have their limits. RF transformers are useless at audio frequencies.

There are USB isolators but the ones that work with USB 2.0 cost more than some of the Picoscopes. Most only work as 12 Mbps not 480 Mbps. All the USB scopes I know of with isolated USB only run at 12 Mbps and have really slow screen updates--which is especially bad for say trying to follow a music signal to observe clipping, or whatever. There's more about scope trade offs, etc. in my (rather dated in terms of my blog) Testing Methods article.

RocketScientist 1st December 2011 08:43 AM

REGULATOR FAILURES - I've been told there's some grumbling on Head-Fi about regulator failures with the O2? Has anyone here had a problem? 78XX and 79XX regulators are normally about as fragile as anvils.

I get lots of O2 related emails and private messages every day and not a single one has mentioned regulator failures. So it's a bit odd the fuss is ONLY at Head-Fi where a significant gang has tried to do everything they can to make the O2 look bad to protect various interests there.

The regulators in the O2 are thermally protected. Even if someone manages to get them too hot, they should just temporarily shut down. The only obvious way to fry them would be to use more than a 20 VAC wall transformer and exceed their maximum input rating. Obviously that's hard to do accidentally. And the impedance of the wall transformer and 2000 uF of primary capacitance forms a natural RC filter against line transients.

The only other thing I can think of is if someone used the wrong diodes (like the Schottky parts) for D3 and/or D4 and they shorted and are pumping AC into the regulators. But I would think the caps would ooze goo all over if that were the case.

I'm blocked from head-fi, so if anyone here cares to perhaps direct those concerned to post here or contact me via the blog, that would be great. Something seems very odd.

greenalien 1st December 2011 09:45 AM

@RocketScientist - you seem to have hit on the one downside of the project - you release a design for a whole bunch of people with varying skills to build, and when some of them end up with something that doesn't work, they blame you, the design, the components - anything but their own incompetence!
I don't think you could have made the build process any easier, this is the best-documented project that I can ever recall seeing, but some of the queries posted on this forum would suggest that not everyone has read your build instructions...


78XX and 79XX regulators are normally about as fragile as anvils.
What happens if they have been mounted too high so that one or both of the metal tabs contact the case?

BuildMeSomething 1st December 2011 10:42 AM

RocketScientist, the person who vocally reported faulty voltage regulators on head-fi has been directed here on numerous occasions by various contributers to that thread. TBH, the first diy O2 thread was locked and the second is now a bit of a joke (though quite entertaining to read).

It'd be quite surprising if the person who reported the faults posted here. His overall view of the O2 and yourself has been quite negative, to put it politely.


qusp 1st December 2011 10:51 AM

I posted in there regarding the regs simply because there does seem to be a number of similar failures of the same part with the same symptoms, that to me deserves a closer look, thats all. i dont think anyone is questioning the thoroughness of the documentation and preparation for this project, I know i'm not and certainly any reasonable person wouldnt; but you just dont know until real numbers of people out in the real world build a kit and do remember thats what it is, not a design that will be built under consistent conditions

as mentioned there is a large number of people with different skills, different access to tools, different access to parts (some of which may even end up being sold fakes) and if there is a number of similar problems i would say although its probably not the designers fault, it is a fault. nothing that cant be recovered even if it does turn out to be linked to the design. with this many kits out there, i would be surprised if there wasnt a problem.

could it be linked to improper handling of the parts before they were packed?

also i think its unwise to use words like grumbling

Jokener 1st December 2011 11:00 AM

I need to attend to building my own O2.
People that bought their kits from me have already finished theirs and here I am, with nothing to show...
I think I'll go out and buy a new soldering station first, though.
A Weller station from the 80s is good and well, but I think a shiny new Ersa i-Con 1 is in order...
(Or would you guys recommend strongly against it or strongly FOR something else?)

I just don't think that the "problem" would be this limited if it were caused by the design.
Of course the design has flaws. There is no such thing as a perfect design.
You can even argue about a simple piece of wire and if its perfect, so a design as complex as this... you get the drift.
But the "imperfections" are documented in the blog, the board layout and all the other information that is available.
But lets wait and see what that turns into.

qusp 1st December 2011 11:19 AM


Originally Posted by Jokener (
But lets wait and see what that turns into.


possibly unwise to claim victory after less than one month in service of the official kits/pcbs; my guess is RS designed this for much more than that.

and more specifically at you: to attempt to simply shift blame the builders by doing things like putting the word 'problem' in inverted commas, will not go well, especially since it was promoted by many as a design my cat could successfully finish because of the documentation and support.

give it a closer look first and that appears to be what RS is keen to do, which is all i'm saying. there areat least a couple others that have reported the same issue, hopefully one will come here so it can be gone over. good chance it really will turn out to be user error, but it shouldnt be assumed

Atilla 1st December 2011 11:37 AM

Jokener I assembled this kit with an Ersa 'gun', where the only temperature control is the trigger, so I doubt you can have problems with anything :)

Anyway, I'd like to report that I haven't managed to blow this up yet, which is super-odd, given my track record so far.

I didn't even manage to short the S2 to the via, even though I forgot about the potential issue. It's been plug&play so far, absolute breeze to assemble and extremely enjoyable to listen to.

I went trough a pair of crappy earbuds, by clickng on high gain, volume to the max and leaving it there until they melted. Just to see how that'll go.

So far so good, now I'm running its batteries to depletion to test the shutdown. It takes me quite a while already.

jtktam 1st December 2011 01:16 PM

I don't know how anybody can blow anything up doing this amp

it's like 1 on a scale of 1 to 5

if you are not careful and put stuff in the wrong place, or don't check for shorts or fitment issues, then don't blame the board

I built over 10 now and every one works 100%

no rectifier issues!


greenalien 1st December 2011 01:43 PM

@Jokener - if you're going to use 60/40 solder then any decent small soldering iron will be fine - however, if you're going the lead-free route, a temperature-controlled iron is pretty much essential.
My recommendation would be to stick to 60/40, it melts at a much lower temperature, and thus is less likely to cause component damage during assembly.

All times are GMT. The time now is 07:27 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Copyright 1999-2016 diyAudio

Content Relevant URLs by vBSEO 3.3.2