[aos]

I promised this document long time ago and I have been working on it for a long time. It's not yet finished but there's probably enough info for people who were interested and if I continue tinkering with it, I'd never be really ready to put it up.

Be warned, I love to write and this is a very long document, and also the document was written more for my sake than for others. It has too much self-reflection. So if you feel like you'd puke on that kind of content, you can skip to the end and see the schematics. But there is quite a bit of generally useful info in the document (I hope).

http://members.shaw.ca/ifkovic/artic...nverter v2.htm

 

[zoboomofo]

Thanks for sharing your success with us! It LOOKS like it was a lot of hard work.

 

[aeberbach]

aos, great job. Thanks for taking the time to do this.

What software did you use to create your boards, and will you be offering any PDBs for sale?

 

[aos]

Actually, Chu Moy suggested that I simplify the thing a bit in order to make it possible for an intermediate DIY'er to finish it (as it stands now, you need to have advanced skills).

I think I'll be able to design a simpler thing around the same core. The only drawback is need for two 9V batteries (or at least one) and one say 7.2V lithium battery (you can use another 9V but it won't last more than 3-4 hours), and forget about built in charger.

On the positive side, the sound is going to be quite a bit better (based on my experimentation) plus you'll get headphone amplifier in the box that can be used on its own with separate line-in. Parts count will be reduced as will the size and the number of tiny components.

I'm not going to reinvent the wheel, the amplifier will be the acclaimed EL2001 / AD8610 multiloop (the same topology is used in many variants around here thanks to Apheared's evangelization and ppl behind the scenes).

In the meantime, the design is here for people who have the means and the need. The price for parts for the current design sits at below $300 Canadian but it can be significantly reduced by choosing different capacitors and the new design should have less parts...

 

[Serow]

So, did you imagine that it would take that much time and effort to build it when you started out? Great job - I bet there were a few times when you finished wading through a stack of datasheets, none of which would work, and just wanted to throw the damn thing in the garbage!

 

[aos]

No kidding, it was a LOT of work and there WERE a lot of frustrations with datasheets. I did nothing but datasheets for weeks, at home and at work (it was slow then, it isn't now

). I am almost certain there was that one night when I dreamed about finding out a chip that meets my specifications...

The only way I was able to do it is that I didn't do much else when at home, and I did a lot of stuff at work too. One Thursday I was up till 4AM to finish PCB layout so that I can use copier at work on Friday to print it on trasparencies which would leave me the weekend for making it - stupid since I could've just came for 10 minutes on Saturday, I live within walking distance anyway. But sometimes you just can't stop doing it.

In retrospective I didn't need nearly that much time, lots of it was spent on needless research and rehashing same things over and over.

 

[zoboomofo]

Hey aos, you know how the Stax earbud uses one of those whatchamacallits that takes the voltage from an AA then jacks it up high for the headphone? Well have you considered using that to make the batteries a) less space consuming and b) cheaper to change? I think cmoy had an article on these dc-dc converters. Anyway, I'd be interested in your thoughts on these things and how feasible they are to use.

 

[tangent]

HeadRoom just redesigned their Airhead amps to get rid of the DC-DC converter. Now why do you think they'd do something like that?

Things that make you go "hmmmm..."

 

[Joobu]

While a DC-DC converter (switcher) would allow the use of less batteries, batteries would run out sooner since the same power would have to be extracted from less batteries. I believe the folks at Headroom use Datel DC-DC converters in their portable products so it's not an impossibility.

The biggest barrier, perceived or otherwise in the use of switchers is the interference they create. Most switchers for audio applications operate beyond 20 kHz to alleviate the audible impact. Switchers are heavily used in commercial products and definitely here to stay. Now whether they will ever be accepted by audiophiles is a different question.

 

[zoboomofo]

okey-dokey thanks guys.

 

[Joobu]

It was probably cheaper to not include switchers in the Headroom low-end products. The mid-level amps (portable) still appear to have DC-DC converters.

I feel switchers have a bad reputation due to their misuse. They are more difficult to use so I suppose it isn't surprising.

 

[aos]

For consumer electronics 2AA batteries are kind of standard. They're cheap, small and have good capacity.

9V batteries are on the other hand not that widespread. They're more expensive and have pretty low capacity. Therefore they're relegated to more specialized uses where power drain is small and/or device is going to be on only occasionally.

It's very likely that prices of batteries reflect the sales volume. Consumers prefer 1.5V batteries by a wide margin.

I think AA batteries would last longer even with DC-DC conversion which has efficiency of about 80% if I remember well.

Anyhow, in my case, I have 6AA batteries. Their prices aren't particularly high, box has space allocated for them and there is a built-in charger, so I think it's still quite practical. In fact, I worked hard to get to that point and I'm quite happy with results on that side.

Modified version that I might start building this weekend, on the other hand, uses additional 2 9V batteries (in addition to 6AA's or one Lithium-ion 7.2V battery) and is anything but practical. It should sound much better though.