DIY Headphone Amplifier

[PJPro]

Cannot believe it! Components are here. Arrived this morning. 1.5 days from ordering to delivery from Texas. Fantastic service!

Been rooting around in the box of goodies....."Oohh, oohh, what are those big orange ones?". Errr, the ones that look too big to get in the case I've ordered and too big for the protoboard. Damn!

I'm going on about the input capacitors. Tangent recommends a certain size but, elsewhere on his site, recommends going for something with more capacitance....so I did (same model as the smaller one though). And they're huge.

I'll have to modify the component layout to make it all fit! But it will go in the case. Just.

I'll post some piccies shortly.

 

[PJPro]

OK. So here are the parts from Mouser. Note the big orange drops at the top of the picture.

And here is my trial fitting of one of the Vishay 0.47uF input capacitors to the protoboard and case. It's tight, but it will fit!

 

[PJPro]

OK. Fitted the power supply circuit (buffered virtual ground with noise reduction).

Here's the schematic.

It's actually proving to be a little tricky. Fiddly rather than difficult. But it does work so far. There is an even split of the voltage across positive and negative circuits to the virtual ground (within 600ths of a volt).

Need to press on with the amplification circuits, case modifications and the panel components.

 

[PJPro]

The size of the case has caused a compromise to the circuitry of the power supply. I had intended to go for the larger 1000 æF 35V electrolytic capacitor. However, this was just too big to fit the case. So, I had to go with the 470 æF version.

The size of this capacitor is partially responsible for how bassy the amp sounds. I believe it has something to do with the level of stored "energy" in the circuit which can be quickly made available to the op amp for additional punch.

 

[Olli1324]

Would I be correct in my belief that changing that capacitor requires no extra modification? Just a rip out the old one, solder in the new one jobby?

 

[Anonymous]

In case if you need help, make sure to check out www.headwize.com if you haven't done so, I believe the original designer of cMoy is an active member or a Mod over there

Also, there are many great sounding DIY amps on www.amb.org

 

[PJPro]

Olli1324:Would I be correct in my belief that changing that capacitor requires no extra modification? Just a rip out the old one, solder in the new one jobby?
Yeap. That'd be right. Having thought about it, I probably could have bunged in the 1000uF capacitor...it would have just had to be on it's side.

 

[PJPro]

Cankin:In case if you need help, make sure to check out www.headwize.com if you haven't done so, I believe the original designer of cMoy is an active member or a Mod over there

Also, there are many great sounding DIY amps on www.amb.org


Yes, that's where I originally found out about the Tangent site.

 

[PJPro]

The amplifier circuitry is now in place.

In the middle of the board you can see the DIP8 gold plated socket to accept the Texas Instruments (Burr Brown) Op Amp. The two big orange things are the input capacitors (Vishay). These are causing further problems, have heavily influenced the layout of the components on the board (I have had to deviate from Tangents recommended layout) and have made the addition of resistors and will make subsequent tweaking of resistors difficult.

The right input in is, err, on the left and the left is on the right. The right channel is carried by the white hookup wires, while the left is carried by the red.

These inputs are fed via the input capacitors, throught vishay input resistors (which are linked to the virtual ground) into the op amp. The corresponding outputs go through vishay resistors to the output components (mini-jack).

The power supply is used to power the op amp. You can probably see the blue (positive) and black (negative) hookup wires coming from the power suppy circuitry to the top right and bottom left connectors on the op amp socket.

Things haven't improved on the fiddly front. I'm not entirely convinced by my soldering iron tip. I think my hook up wires are a tiny bit too big and the tin/lead/silver solder (62%/36%/2%) needs a little more heat to make it melt than a simple tin/lead mix.

Have to say that I'm finding my eyesight a little bit of an issue. The size of the components is soooo small that, working in relatively poor light, it's a little difficult to focus. Thanksfully, the helping hands (from Maplin £9.99) came with a large magnifying glass.

Despite all of that, I am enjoying the project and things are coming along nicely. The next thing to do is sort out the panel components (drilling the panel to accept the controls), add the LED power light and resistor to the circuit and then wire it all together.

While the performance of this amp is going to be limited by its very design, I have spared no expense on components and, where optional parts have been listed, I've gone for the highest quality available (at least as far as I know).

Roll on the testing. I can't wait!

 

[Anonymous]

what a fun project, thanks for the commentary, from scanning this it seems you've got all of these including soldering iron etc for about 80 -100 quid, would that be right ?

cheers Gavinÿ

 

[John Duncan]

zzgavin:Cambridge Audio DACmagic when I finish writing the book
?

 

[PJPro]

zzgavin:
what a fun project, thanks for the commentary, from scanning this it seems you've got all of these including soldering iron etc for about 80 -100 quid, would that be right ?

cheers Gavin

Maybe. I'm going to work out all the costs so that others can make a judgement on whether it's a worthwhile actvity. To be honest, it's a bit of fun that results in something useful (hopefully).

One unexpected consequence of building this amp and all the initial research is that my wife has definately warmed to the idea of just buying one and thinks the £500 for a Stello, all things considered, is quite reasonable. So, this might be hobby which can be used to justify kit purchases later.

"Darling. I'm thinking about building a new amplifier. But don't worry it's hardly going to cost me anything. It'll just need a bit of my time. I don't think it'll take much more than a week to research, a further week to identify and order all the parts and perhaps 2 weeks of evenings and weekends to assemble it. And I'd need to spend, ooooh, a thousand pounds to buy something similar off the shelf. It's the mark up for transportation and the retailer that really costs you know......"

 

[Anonymous]

JohnDuncan:zzgavin:Cambridge Audio DACmagic when I finish writing the book?

ÿ

An encouragement to myself mainly, I'm writing a book for O'Reilly the technical book publishers on social web applications.ÿ

When I finish drafting the book I get my next advance and with that some more hifi toys, the changing dollar pound exchange rate brings unexpected benefits.ÿ

 

[Anonymous]

Thanks PJ, a costing would be helpful. Post book I'll want some other projects and electronics has appeal

 

[PJPro]

The panel components are now hooked up to the rest of the circuitry.

So, what we have is two mini jacks: the one on the left is the input, the one on the right the output. Between these is the on/off switch come volume pot. This volume pot is made by Alps and I've seen it being used on a number of (what I would regard as) quality amps. The Stello uses an Alps pot for instance.

The twisted wire at the top of the amp is the LED/Resistor. This fits into the hole in the top right hand corner of the panel.

So, the power comes in from the battery, with the positive switched though the pot. This then feeds the power supply circuitry, LED and ultimately the op amp.

The audio stream comes in the left mini jack, passes through the pot and onto the input capacitor. It is processed by the op amp and gets sent out of the right mini jack.

The green earth wires connect the panel components to the virtual earth on the protoboard.

I made a bit of a mess of the LED. The raw component comes with long exposed metal wire connectors; one for positive and one for negative. To the live connector I have to solder a resistor and then wrapped it in insulation tape. I then used hookup wire to connect the resistor and negative connection of the LED to the protoboard. I then wrapped more insulation tape round the lot to prevent shorts. It does the job but can't help feel that some sort of shrinkable sleeve would have been better.

Souldering up the pot was a bit of a nightmare. I almost gave up. Trouble is that if you leave your soldering iron on the component for too long you end up knackering the resistors inside....and you can even melt the pot itself. After a bit of swearing, walking up and down, huffing and puffing I managed to sort out the technique.

To hook up the pot, the first thing I had to do was make sure I was doing things in the right order. Obvious and I did this from the off. But I might as well say it. I did the middle row first....and within this row the connectors nearest the opamp first.

The next trick was to only strip a tiny bit of insultation of the wire (prevents shorting out the wire next to it). I then dropped a small blob of solder onto the exposed tip of the wire. I then heated the connection on the pot with the iron and placed the pre-soldered wire onto the iron/connector. The heating and fixing of the wire takes three or so seconds.

All that remains to do is to fit the amp into it's case, remove the battery connector (which is used for testing only) and solder the circuitry up to the proper battery connectors within the case.

I'll admit that I've already tested the amp at this point. I used some crappy active PC speakers and my Beresford as the source. Risky but I didn't have anything else.

What can I say? Deep joy?

I'll apologise for the quality of the pitures in this thread right now. For some reason my usually well behaved compact cannot focus on the protoboard.

 

[PJPro]

Just to add a bit of clarity to my descriptions, please see the image below provided by TangentSoft.

It's not a precise match to what I have built as my amp uses the buffered power supply rather than the simple supply shown in the image above.

 

[chebby]

You say it is going to be battery powered.

Would there be any advantage (in future) to making a seperate power supply box one day? Something with a little 40va toroidal transformer maybe?

My USB DAC has the standard wall-wart but also the option of an upgraded PS for £105. (I guess something similar would probably cost you about £15 to make yourself)

The Trichord Dino PS uses a little 40vA toroidal transformer for instance.

Hey, I am no expert. Just enjoying what you are doing vicariously. My knowledge of electronics ended with A' levels 30 years ago

Just wish I had the confidence and knowledge to do something similar.

What about a DAC next? Wolfson WM8740 24 bit/192khz DAC chips only cost £3.49 each if you are buying just one!

 

[Anonymous]

zzgavin:
An encouragement to myself mainly, I'm writing a book for O'Reilly the technical book publishers on social web applications.ÿ

I hope we aren't your guinea pigs?ÿ

 

[PJPro]

The amp is now finished!

Amp with cover off

Amp with cover on

Grados SA80s, Beresford DAC and DIY Amplifier

The session with the oscilloscope is set up for 12:00 tomorrow. I've made some test leads to allow an audio signal to be passed into the amp and the oscilloscope to receive the signal coming out. See below.

 

[Messiah]

So how does it sound?????

 

[Olli1324]

It certainly looks smart. Nice job. I can't wait to hear your report on sound quality.

 

[PJPro]

chebby:You say it is going to be battery powered.

Would there be any advantage (in future) to making a seperate power supply box one day? Something with a little 40va toroidal transformer maybe?

My USB DAC has the standard wall-wart but also the option of an upgraded PS for £105. (I guess something similar would probably cost you about £15 to make yourself)

The Trichord Dino PS uses a little 40vA toroidal transformer for instance.

Well, if I understand correctly, a battery offers the lowest noise levels. That's why that approach has been adopted by Grado and the RA1. Mains is more convenient. Having said that, I intend to have a go at producing a mains supply designed by Tangent. He does say that these are not really for beginners due to the large levels of charge held in the capacitors.....so need a little more experience before I take the plunge.

chebby:Hey, I am no expert. Just enjoying what you are doing vicariously. My knowledge of electronics ended with A' levels 30 years ago

Just wish I had the confidence and knowledge to do something similar.

Actually, this project has been fairly easy once I had assembled all the parts.

chebby:
What about a DAC next? Wolfson WM8740 24 bit/192khz DAC chips only cost £3.49 each if you are buying just one!
Yes, that would be my plan. But there doesn't appear to be quite the same number of DIY DACs on the web.....and I don't think I'm knowledgable enough to come up with something myself.

 

[Anonymous]

Octopo:zzgavin:An encouragement to myself mainly, I'm writing a book for O'Reilly the technical book publishers on social web applications.ÿ

I hope we aren't your guinea pigs?ÿ

ÿ

only in as much as the internet is my research space. Actually this is a nice community and it gets a brief positive mention in my book in terms of a magazine editorial teams interaction with their community. Mostly I'm hear to learn about hifi and related issues, which I'm learning a lot from you all.

 

[PJPro]

I've had one or two issues lately.

Issue 1: Too busy at work to make the time for testing. So, oscilloscope testing delayed.

Issue 2: Hmmmm. See this link.

 

[PJPro]

I won't go into the details of the issue right now, but it appears that I may have a simple fix.

See the link identified above if you want the details now.

I am hoping to be able to apply the fix tonight.