| carlosfm |
| quote: | Originally posted by dip16dac
It shouild be connected to a zero ohm source to have no DC output voltage. The 4.02 K ohm resistors are only there to limit the DC offset when the source is disconnected or turned off. |
Of course.
The evaluation board doesn't have an input buffer.
If you connect a pot there you will have DC.:att'n:
The source may not have a variable output, you need a pot.
I also don't want to use a buffer because I want to evaluate the chip by itself, to see how it sounds alone. |
|
|
| fit |
| quote: | Originally posted by peranders
Good job and the soldering doesn't look so bad. :up:
I think have made the right decision to include a buffer because of the large input bias currents of the TPA6120. With this bufer you can connect any source without having problems with offset voltage.
What supplyvoltage do you use? How hot does the IC get? |
soldering would look better if I use liquid flux - I havent any yet...
supply voltage +/- 15V is taken from main PSU on may EP90ES AC3 decoder - in fact this is a headphone amp for this device - LMs produce +/- 11,66V (120Ohm/1kOhm)
chip is quite warm (buffers are hotter) - but so far I didt put much load on it - still testing (ear & osciloscope) - I can say - bass sounds "deeper" than from my STR930 amp... |
|
|
| peranders |
I use plain 60/40 tin 0.7 mm. Works perfect for me.
 |
|
|
| Upupa Epops |
| Pretty nice toy, Per. I like it :cool: . |
|
|
| carlmart |
| quote: | Originally posted by peranders
I use plain 60/40 tin 0.7 mm. Works perfect for me.
|
You soldered all that by hand?...
Carlos |
|
|
| peranders |
Indeed, not very hard if you know the tricks and it didn't take so much time other.
Soldering iron
Tweezers. This pair of tweezers is just fantastic, perfect balance :nod:
I have two soldering irons when I solder all 0805 parts.
Some day I will descibe my tricks in more detail and with pictures also. |
|
|
| carlmart |
OK. Let's see:
1) The solder pen I have: Hako 936. What temperature and tip?
2) The tweezers I have are good but might not be adequate for the job.
3) What is that template? What material?
On one article I read it said that using super-glue to fix the part before soldering it was a good trick. Do you think so?
How to fix the part and the heat are my main worries.
Apparently you need to use solder with some silver on it.
The next thing is how to do my prototype pcbs for SMD, which is not easy.
Carlos |
|
|
| peranders |
To put it short:
1 Put tin on both pads.
2 Place the part with a pair of tweezers
3 Take two irons and warm for a second or two. The part sucks into palce. Boom finished.
Version two with only one iron:
1 Put tin on only one pad
2 Place the part and warm the tinned pad and adjust the placement with the tweezer.
3 Solder the other pad.
You don't have to glue the parts. |
|
|
| fit |
| quote: | Originally posted by Upupa Epops
I have another question ( for good judging of sound is one day too little ;) ) : Have you any control of this amp by high speed scope ? I am asking about stability, 'cos in dataseeet is many warnings about this :xeye: . |
so far no problems with stability (geez I have 14 caps in decoupling ) - I can see 50Hz however when the input floats in the air :D |
|
|
| carlosfm |
...it's gettin' better.
It's been :hot: in'-in for some days.
Today, arriving home, I've been listening and comparing with my pre.
My pre seams more open on the midband and treble, more detailed, smooth but "open".
The TPA was smooth but a little shut-in, lacking some harmonics and directness.
It was better than some days ago, but not convincing.
As I don't trust my test pot, I picked another (better) one I have here for some time and didn't use yet.
It's a log. 10k carbon pot, all metal (including metal shaft), smooth and silent operator and good looking indeed.
I bought some of these some 6 months ago but haven't tried yet.
They were only available in 10k...
I made a passive "pre" in a very small box, with single RCA inputs and outputs.
I've just connected it half an hour ago.
And...:eek:
The deam thing oppened up, much more detail, much more "alive".
It doesn't give me sleep now.:D
:scratch:
Gonna wait some days more, no hurries.
I'm going to be busy the next days, so let it :hot:.
This sound now reminds me the best pre I've heard in my life, an unforgivable experience: the Jeff Rowland Coherence.
I still have to try the AD815, I'm just curious. The Jeff Rowland uses this one. |
|
|
| ericallan |
| quote: | Originally posted by carlosfm
Of course.
The evaluation board doesn't have an input buffer.
If you connect a pot there you will have DC.:att'n:
The source may not have a variable output, you need a pot.
I also don't want to use a buffer because I want to evaluate the chip by itself, to see how it sounds alone. |
How did you solve the DC problem - did the 4.02k resistors make a difference, or is there another way without resorting to a buffer?
Also, to use as a pre-amp, is the value of the feedback resistor the only change needed? |
|
|
| carlosfm |
| quote: | Originally posted by ericallan
How did you solve the DC problem - did the 4.02k resistors make a difference, or is there another way without resorting to a buffer?
Also, to use as a pre-amp, is the value of the feedback resistor the only change needed? |
I posted this some pages before, check it out.
I did this on veroboard and I'm not using the standard values of the evaluation board.
Sorry, I'm in a hurry now, have to go.
Later, |
|
|
| carlosfm |
The AD815 is already singin' on my main system.:cheerful:
I could only get my hands on the SMD version.:bawling:
This thing needs to be properly heatsinked, it gets :hot: .
No DC-offset on the input (~0.2mv).:up:
Around 33mv DC on the output.:xeye:
Starts with around 16mv, and goes slowly increasing to stabilize at around 33mv.
I won't use input caps, here, but maby I will use output caps.:dodgy:
Each current-feedback chip is a different animal...
Gonna let it :hot:-in. |
|
|
| Upupa Epops |
| Which one is singing better - TI or AD ? :xeye: |
|
|
| carlosfm |
Take it easy, I'm reading some threads...:D
It's late now, tomorrow when I arrive home I'll take a listen to the AD815.
But real comparisons only in a week or so, I have much to do now.
I'm making some "improvements" in the house, I have no kitchen right now, this seams like camping!:clown:
Tomorrow night gotta do some wall painting... |
|
|
| carlosfm |
This thing seams really special, but some issues still to solve.
To match impedances and lower input/output DC I used 220R from +IN to ground.
But this attenuates the signal too much.:bawling:
I removed the resistors and I have ~1V at the output.:xeye:
With a 10k pot connected, I have no DC at low volumes, but it increases with the volume.
Of course, to lower DC to "normal" levels I really need 220R (after the input caps).
I need the gain I have (2.8x), and I can't change the feedback resistor to a much higher value...
I tested signal trafos at the output and :up: 0mv DC.
What's more, these little things cost me pennies and are very transparent, much better than most caps.
I only have two, tomorrow morning gotta go buy some more, if they still have them.:bawling:
The plan is to make something a la Jeff Rowland Coherence: input and output trafos.
So... out with the input caps, in with the trafos.
Without an input buffer, this seams the most rational approach.
Wait some days, my wife wants the kitchen ready before the pre.:D
Well... so do I... |
|
|
| analog_sa |
| quote: | | I tested signal trafos at the output and 0mv DC. |
Most input transformers will get quite upset with dc on the primaries. What type are you using? Do you intend to use the same at output? They might not sound so transparent there.
In any case this is an unexpected development, well done! |
|
|
| carlosfm |
| quote: | Originally posted by analog_sa
What type are you using? Do you intend to use the same at output? They might not sound so transparent there. |
Do you recommend using them only in the input?
I may use caps at the output.
At the input I have around 15mv DC with a 10k pot at max volume. At minimum I have 0.
What type?
Now someone's gonna laugh...
I made tests with some "ground loop isolators" and found some that I can't detect any difference in sound when they are inserted on a circuit.
Curiously, this type I'm talking about is made for car audio and costs... 5.5 Euros.:D
I bought another one today.
I tested some that cost more than double the price and didn't like.
This thing has no brand, it just says "High quality ground loop isolation", and it's a cylinder. Female RCA inputs, male outputs.
I scavenge it and take the trafos. |
|
|
| analog_sa |
| quote: | | Do you recommend using them only in the input? |
You'll have to measure the dc with the secondaries in circuit. It will probably be very low. In any case if you don't notice appreciable reduction in bass the cores are not saturated. Is the pot connected directly to the primaries? This may be less than optimal as at low volume settings the primaries will be seeing higher driving impedance and as these transformers likely have lowish primary inductance there will be less bass and higher distortion at low volumes. Ideally you should drive the primary directly from an opamp output but this effectively means an extra stage.
How big/heavy are the cores? |
|
|
| peranders |
| I have ordered prototype pcb's...... for my QRV-07 amp . |
|
|
| Upupa Epops |
| Per, on site of 07 you have photo of 04, attention :) . |
|
|
| peranders |
| I know:cool: . I just duplicated the whole QRV04 section. 50% of the pcb look as the QRV04 but for now consider the pictures as place holders until the QRV07 is alive and well. |
|
|
| carlmart |
As long as a similar to Walt's is built and others compared to it...
Walt Jung is very picky with replacing parts on his designs. And I think he deserves some respect there.
When this design was published in Electronic Design he was very adamant on builders using those exact parts.
Carlos |
|
|
| Nicke |
| quote: | Originally posted by carlmart
As long as a similar to Walt's is built and others compared to it...
Walt Jung is very picky with replacing parts on his designs. And I think he deserves some respect there.
Carlos |
Iīll agree,what I meant was that it would(probably)be a better solution.
| quote: | Originally posted by carlmart
When this design was published in Electronic Design he was very adamant on builders using those exact parts.
Carlos |
Well,I use the AD744/811 as a preamp(exactly like Walt designed it and it works pretty good
:) |
|
|
| carlmart |
| quote: | Originally posted by Nicke
Well,I use the AD744/811 as a preamp(exactly like Walt designed it and it works pretty good
|
Yes, that was his first implementation of the combo circuit.
In fact I must say the circuit I was talking about was a preamp that was similar to the one you mentioned, but used an AD812 in the second stage. It was published in ED in December 1998.
http://www.elecdesign.com/Articles/...?ArticleID=7206
I am curious to know if somebody built it and what were the findings.
Carlos |
|
|
| peranders |
| quote: | Originally posted by Nicke
Iīll agree,what I meant was that it would(probably)be a better solution.
|
Nicke, I think you have misunderstood. TPA6120 is a standalone unit with extremely low distortion. You don't need any feedback to the AD8610.
Walts article was about something else. |
|
|
| Nicke |
| quote: | Originally posted by peranders
Nicke, I think you have misunderstood. TPA6120 is a standalone unit with extremely low distortion. You don't need any feedback to the AD8610.
Walts article was about something else. |
What I meant was that it could be better to use something similar to Walt Jungīs circuit(with AD8610 & TPA6120) instead of using a AD8610 followed by a TPA6120,each with itīs own feedback loop.
Didnīt you intend to use a AD8610 before the TPA6120?
Why dont you try both ways... |
|
|
| carlmart |
| quote: | Originally posted by Nicke
What I meant was that it could be better to use something similar to Walt Jungīs circuit(with AD8610 & TPA6120) instead of using a AD8610 followed by a TPA6120,each with itīs own feedback loop.
Didnīt you intend to use a AD8610 before the TPA6120?
Why dont you try both ways... |
I agree with Nicke that perhaps you should try both ways and see which sounds better or has lesser problems.
Even if by now your project is probably too advanced for further tests.
Carlos |
|
|
| peranders |
What do you think? I can't wait to get the boards soldered!
 |
|
|
| Upupa Epops |
| Precision and quite logical arranged PCB. Guys, go on, your designs can't be better ! :cool: |
|
|
| peranders |
Thanks, Pavel.
It will be interesting to see how difficult it will be to solder the PowerPad. I'll guess I'll have the answer real soon now. |
|
|
| fit |
| if holes are metallised use of hot-air gun solve the problem of PowerPad :) if not - I think first you have to "flood" holes under TP with the tin |
|
|
| peranders |
| I have soldered most parts now, only a few are left, which I don't have. This pcb is looking real cute. The TPA6120 is left and I'll wait untill everything else is tested. |
|
|
| TNT |
P-A!
"Made with a Macintosh" ;-)
What CAD prog do you use?
BR/ |
|
|
| peranders |
| quote: | Originally posted by TNT
What CAD prog do you use?
|
RUN Electronic Design System, a german software. |
|
|
| peranders |
The PowerPad IC was rather easy to solder :nod:
1 Get a thin layer of fresh tin on the PowerPad
2 Solder all pins
3 Burn the PowerPad with 450 degrees C! Works just fine!
What do you think? Isn't the pcb cute?
 |
|
|
| carlmart |
| quote: | Originally posted by peranders
3 Burn the PowerPad with 450 degrees C! Works just fine!
|
How do you do that burn? On a an oven or with the solder pen from below?
Congratulations on the job!
Carlos |
|
|
| peranders |
| I just turned up my soldering iron to max temp. :hot: |
|
|
| digi01 |
P-A,
Congratulations on the job!
what is your PCB board size?have you consider offer the pre-solder(SMD parts) PCB?
enjoys
digi:cool: |
|
|
| peranders |
Thanks!
The size is 76 x 137 mm and no to your last question. The interest is zero I'm afraid. |
|
|
| digi01 |
| quote: | Originally posted by peranders
Thanks!
The size is 76 x 137 mm and no to your last question. The interest is zero I'm afraid. |
keep the fate man.this is really a good design.
IMHO,maybe those smd parts is troublesome to novice,right? |
|
|
| peranders |
| Today I have listened to the amp. I like it to be short. :nod: |
|
|
| peranders |
| I have measured the offset voltage and I got 2 and 11 mV with gain of 2 and 1 kohms resistors. In other words, this TPA6120 needs a buffer. The AD8610's have approx. 60 uV! |
|
|
| carlosfm |
| quote: | Originally posted by peranders
I have measured the offset voltage and I got 2 and 11 mV with gain of 2 and 1 kohms resistors. In other words, this TPA6120 needs a buffer. |
That's exactly what I have found with my tests, P-A.
Like I said, TI's demo board is no good... at all.:D |
|
|
| peranders |
| What do you say, those of you that have listened to the TPA6120, best ever, fair of what? |
|
|
| carlosfm |
I say it's good, but I haven't spent so much time with it.
My goal was not to use an input buffer, although I know it needs it.
The thing is, the AD815 sounded better to me (even with output caps) and I put much more efford in developing a good compromise without an input buffer.
Don't take my word for granted, maby if I've done the same thing to the TPA6120 it would sound as good/better?
I say to everyone: try it.
In my implementation it was a little too "smooth" sounding, but I'm sure that it has to do with the caps.
P-A, if you don't use caps in the signal path, I'm sure you will have different results, of course.
As you have low DC, you don't need them.
As an headphone amp, this chip will put "slam" on your head, you will be surrounded by dynamics.:D
So, P-A, you have made a very nice board.
Have you listened to it?
Is it better than the op-amp+BUF634 solution? |
|
|
| peranders |
I have listened to the amp now a couple of hours and it may be the best I ever have made but the tricky part is to switch amps fast which isn't easy. The sound is excellent but it's hard to say how much different the sound is from my QRV-04 with AD8610 + BUF634. This combination is also extremely good. The major difference between the two amps is that QRV-07 consumes a bit more power. It seems that the groundplane cools rather much. The IC is warm not more.
If someone wants to try out a really exciting amp, TPA6120 is the one.
Bigger picture of the TPA6120 here
Bigger picture of the backside of the pcb here |
|
|
| peranders |
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. |
|
|
| sek |
Hi Per-Anders,
unfortunately the measurement images don't get displayed. Forcing to display them produces an error message claiming the images 'contain errors'.
But the numbers look good. I've seen such results from RMAA on headphone amps (on tangentsoft's site), where the lower output impedance of the headphone amp is capable of driving the soundcard input at lower distortion than the soundcard output alone could. That's probably why we see less distortion with a 39Ohm resistor than with no DUT at all. ;)
I think you're in the need for a better sound card, now. ;)
Sebastian. |
|
|
| peranders |
Even old crappy Netscape 4.08 Macintosh from 1998 can show this report proberly.
Check you browser that you can show png pictures proberly.
I think we can say that my amp isn't substantially worse than the soundcoard.:) |
|
|
|
