Headphone amp with 0.000021% distortion

peranders · « **on:** 11 Oct 2004, 11:13 am »

I did one day discover the TPA6120 from Texas Instruments. I decided to let this IC have a go. The result can be seen below.

I have listened to the amp now and the result is that this may be the best headphone amp in the world. I don't think anybody can't say that this amp isn't in world class. The TPA6120 is an amp which can handle DC-100 MHz with ultra low distortion.

If you download the datasheet you can see the extremely good data this IC has.

peranders · « **Reply #1 on:** 20 Oct 2004, 07:37 am »

Anyone else who has tested this very good TPA6120?

Lost81 · « **Reply #2 on:** 20 Oct 2004, 07:44 am »

Very nice!

You designed and etched the PCB yourself?

-Lost81

peranders · « **Reply #3 on:** 20 Oct 2004, 08:27 am »

..... yes....

..... not etched, no. A bit hard to plating the holes at home. The pcb's are made professionally but the design is made at home with my Macintosh.

Lost81 · « **Reply #4 on:** 20 Oct 2004, 10:11 am »

Quote from: peranders

..... the design is made at home with my Macintosh.

Awesome!

Nice to meet another Mac user!

Cheers,
-Lost81

elnero · « **Reply #5 on:** 21 Oct 2004, 06:43 pm »

So is this an amp that's going to be available for sale or is it strictly a personal project?

ooheadsoo · « **Reply #6 on:** 21 Oct 2004, 08:05 pm »

peranders is a regular diy poster on head-fi

I was going to post a link pointing here but I checked and it's already been going around the block, at least about the chip, not peranders's new pcb.

I think he's selling the pcb for about $48 or something.

CIAudio · « **Reply #7 on:** 21 Oct 2004, 08:54 pm »

Our new VHP-1 uses the TPA6120, you can see preliminary info at http://www.ciaudio.com

peranders · « **Reply #8 on:** 22 Oct 2004, 07:33 am »

Quote from: elnero

So is this an amp that's going to be available for sale or is it strictly a personal project?

The project is "semi-personal" because it's too frightning for most DIY'ers altough it's not particular hard to solder by hand. The price is right and this is because I have made only a few pcb's.

peranders · « **Reply #9 on:** 22 Oct 2004, 07:53 am »

Quote from: CIAudio

Our new VHP-1 uses the TPA6120, you can see preliminary info at http://www.ciaudio.com

I have checked your amp. Looks real cute.

Is the TPA6120 the best headphone amp IC? What do you think? I don't know about that but it is real good, that's for sure.

peranders · « **Reply #10 on:** 29 Oct 2004, 09:13 am »

I have made some distortion measurements with the free RMAA software and my rather poor soundcard Canta Cruz Turtle Beach.

0.0058% at 39 ohms load and 0.0065% with Sennheiser HD545. Otherwise pretty good results. I will upload the results later.

peranders · « **Reply #11 on:** 29 Oct 2004, 07:53 pm »

Test results can be found here
http://www.sjostromaudio.com/hifi_files/qrv/qrv07_testreport/QRV07_HD545.htm

peranders · « **Reply #12 on:** 15 Nov 2004, 10:57 am »

No comments?

F_Dieckmann · « **Reply #13 on:** 24 Nov 2004, 10:21 am »

Very nice PCB as usual for all your efforts. As far as your specifications go.......

"Distortion: 0,00008 %, OPA134
0,00003 %, OPA627
0,00003 %, AD8610"

From: http://focus.ti.com/docs/prod/folders/print/tpa6120a2.html
* 80 mW into 600 ohms From a ±12-V Supply at 0.00014% THD + N

"Output impedance: 3 milliohms Damping factor: 16000 with 50 ohms load"

Since your headphone amp has an actual output impedence of 10 or 20 ohms depending on jumper setting, you actual damping factor is 5 or 2.5 (20 ohm option) into 50 ohms. I wonder if this is an issue with some low impedance headphones like the ecellent Grados which have an impedance of around 40 ohms as I recall.

Whoops!

peranders · « **Reply #14 on:** 24 Nov 2004, 11:42 am »

Hello, Fred, nice to hear from you! You are very missed at an another forum.

You are right about damping factor. I have copied text and numbers and some may be adjust a bit. The main thing Fred, TPA6120 is excellent as headphone amp!

Dan Banquer · « **Reply #15 on:** 26 Nov 2004, 01:26 am »

From what I have seen on headphone impedance curves is that they are very resistive, and generally very flat, with not much in variation. I have a prototype line stage/headphone amp that I have been experimenting with and it appears there is little difference between a 50 ohm output impedance and a 10 ohm output impedance. Although I am keeping the 10 ohm output impedance.
From what little experimentation I have done and some of the stuff I have read on the Headwize web site, I have come to the conclusion that incredibly low output impedance for a dynamic headphone amp does not appear to be crucially important.
What I did find when I did a lot of field testing on the unit was that dynmaic headphones, despite their rated frequency response, appeared to be anything but flat., and when faced with higher density music (other than a quartet or quintet) they tended to cloud up. (IM distortion?)
I guess it still is very tough to have one small cone try to do so much.
d.b.

peranders · « **Reply #16 on:** 26 Nov 2004, 06:44 am »

Dan, check my test results above. My headphone create a small hump in the bass.

Dan Banquer · « **Reply #17 on:** 26 Nov 2004, 02:12 pm »

That's certainly true; If I am reading the scale right that appears to be about 0.25 db up from the resistor curve. It almost looks like the headphone manufacturer used the impedance as a bit of a bass lift. What do you think?
d.b.

peranders · « **Reply #18 on:** 3 Jan 2005, 10:19 am »

I think this hump is not made deliberately. It has to the with the impedance of the headphone.

peranders · « **Reply #19 on:** 19 Apr 2006, 08:26 am »

Just checked the website of CI Audio, which configuration do you use, inverting or non-inverting? Any offset trim?

It's always nice to have a photo of the inside