DIYHiFi.org

For the sake of audio
It is currently Thu Oct 30, 2014 2:23 pm

All times are UTC




Post new topic Reply to topic  [ 17 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Sun Sep 28, 2008 5:08 am 
Offline

Joined: Mon Apr 23, 2007 11:07 pm
Posts: 575
Location: So. California
Can anyone share their OPA627 tweaking experience with respect to:

- "Optimum" input (PSU pin) voltage
- PS pin bypassing (values for caps)
- Snubber values

Thx,
-hm


Top
 Profile  
 
PostPosted: Mon Sep 29, 2008 9:48 am 
Offline
User avatar

Joined: Tue Dec 14, 2004 12:26 pm
Posts: 4927
Location: Lisbon, Portugal
Have you searched?

viewtopic.php?p=22981#p22981

Read Mikeg's post too.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein


Top
 Profile  
 
PostPosted: Mon Sep 29, 2008 10:53 am 
Offline

Joined: Mon Apr 23, 2007 11:07 pm
Posts: 575
Location: So. California
To my chagrin, I didn't search. "The other place" usually forces one to do so before posting so, in the few rare instances where such a feature is worth having this is one such case.

Anyway, your comments re: OPA627 were on the money, and worth repeating:

Also, the OPA627 really sounds better at +/-18V, with bigger decoupling caps (100uF) as CLOSE as possible to the pins, and biased to class-A with a resistor from output to the negative rail at between 5 to 10ma.
Remove the small 100nF bypass caps and use good, low ESR electrolytic caps.


Top
 Profile  
 
PostPosted: Mon Sep 29, 2008 1:30 pm 
Offline

Joined: Fri Nov 04, 2005 4:10 pm
Posts: 577
Location: uk
hollow_man wrote:
Can anyone share their OPA627 tweaking experience with respect to:

- "Optimum" input (PSU pin) voltage
- PS pin bypassing (values for caps)
- Snubber values

Thx,
-hm


Why bother with the 627? It is expensive, it's an old design , it does'nt like capacitative load, and there are better sounding opamps. Of you want to stick to 'classics' the AD797 sounds much better.

I just remove them as they sound hifi and 1D in many applications; very 'hifi' and not musical. People say that 'power' transfer design of the 627 is critical; I for one have not heard such a good implementation. :banghead: :banghead: :banghead:


Top
 Profile  
 
PostPosted: Mon Sep 29, 2008 1:55 pm 
Offline

Joined: Mon Apr 23, 2007 11:07 pm
Posts: 575
Location: So. California
fmak wrote:
Why bother with the 627? It is expensive, it's an old design , it does'nt like capacitative load, and there are better sounding opamps. Of you want to stick to 'classics' the AD797 sounds much better.

You may be right.

I've had them in my parts kit for about two years now. They were part of a components order for a headphone amp I began building then. I was fairly "young and dumb" to DIY back then; I would not make the same choices again.

The OPA627 is, as Mikeg noted, a fave for headphone-amp fans. I still consider myself a noob, so I can't venture a scientifically-credible opinion on the topic. That said, and from limited experience, I do like the AD op-amps better: AD8610/20, AD843, AD797 and the cheap-but-good AD825 ... gimme dynamics, bass slam, rhythm/timing/pace any day over "warm" tube sound!


Top
 Profile  
 
 Post subject: AD797
PostPosted: Mon Sep 29, 2008 10:58 pm 
Offline

Joined: Mon Apr 23, 2007 11:07 pm
Posts: 575
Location: So. California
fmak wrote:
hollow_man wrote:
Can anyone share their OPA627 tweaking experience with respect to:
- "Optimum" input (PSU pin) voltage
- PS pin bypassing (values for caps)
- Snubber values
Of you want to stick to 'classics' the AD797 sounds much better.

fmak/anyone: WRT to the orig. query, any tweaking advice on the AD797s specifically?


Top
 Profile  
 
PostPosted: Tue Sep 30, 2008 1:51 pm 
Offline

Joined: Wed Jun 29, 2005 6:34 pm
Posts: 1061
Location: East Coast USA
Have you tried an AD845?

You didn't say what your application is.

For low level signals like phono applications, a part like an AD797 is much better.

For line level applications with relatively low closed loop gain, something like an AD845 is better suited.

For a DAC I/V application, most opamps don't do very well.

If this is for an inverting amplifier like in a mixer, there's other choices.

If this is for a servo amp application, the LT1122 and AD711 work pretty well.

Most of these parts work much better with external buffer amplifiers driving the load and the feedback network. There's all sorts of reasons for this, including thermal modulation of the opamp chip when it has to drive a real load.

For opamp based phono amps and line amps, I've had my own best results with an AD797 followed by an LT1010 biased to 40 mA. For line amps, an AD845 followed by an LT1010 biased to 40 mA. The LT1010 definitely needs a real sink at that current. No current sink or source was used at the output of the opamp - it always sounded worse for me. Bypass caps really matter as does the quality of the voltage regulation. So does layout. Trying to keep the impedance presented to both the inverting and non-inverting inputs to the opamp is also important, but pretty hard to do, especially in a line amp that is connected to the wiper of a pot. A constant impedance attenuator works better here.

You can do better with discrete approaches and some open loop solutions. But for opamp based circuits these are really pretty good.

There's a ton of information on all of this on the Analog Devices web site, Walt Jung's web site, and the Linear Tech web site.


Top
 Profile  
 
PostPosted: Tue Sep 30, 2008 3:19 pm 
Offline
User avatar

Joined: Mon Dec 13, 2004 3:29 pm
Posts: 4180
Location: Tralfamadore
I know that you want specific recommendations for the OPA627, but.................

(I have never used it, so I can not supply that specific advice.)

But........CG's suggestions are very good. I am not a fan of the '797, although it works good for low-z, low-noise applications.

Working good and sounding good are usually not the same. But I assume that you have figured that part out by now.

'711 for servo......yep, done that. '845 for other stuff......yep, done that. JFET input, folded cascode. Good CMRR. At least from memory.

OP amps for I/V.........well if ya gotta, ya gotta. No idea which one to use. Not all are optimum. Even if you have a boatload, it may be in your best interest to consider others.

OK, enough o/t rantings on my part. We return you to our normally scheduled programme.

Jocko


Top
 Profile  
 
 Post subject: Using opamps
PostPosted: Tue Sep 30, 2008 4:53 pm 
Offline

Joined: Mon Apr 23, 2007 11:07 pm
Posts: 575
Location: So. California
CG wrote:
Have you tried an AD845?
No -- have not tried AD845, but I have tired its dual-channel version: AD843, in an outboard DAC:
viewtopic.php?p=28103#p28103
Quote:
You didn't say what your application is.
The "preamp" stage of a headphone amp. Specifically, the Tangentsoft PPA2. It's output is discrete :)
The PPA2's main features are here.
Schematics are here.
Tangent's comments on varoius opamps is here.

(Moderator: can you pls correct the spelling error in the title of this thread: bypassing not "bypsssing"! It'll make it easier for search engines to zero in.Thx)


Top
 Profile  
 
PostPosted: Tue Sep 30, 2008 5:37 pm 
Offline
User avatar

Joined: Mon Dec 13, 2004 3:29 pm
Posts: 4180
Location: Tralfamadore
Shows how close I am paying attention.......

Anyway, fixed!

Jocko


Top
 Profile  
 
 Post subject: Re: Using opamps
PostPosted: Tue Sep 30, 2008 8:55 pm 
Offline
User avatar

Joined: Tue Dec 14, 2004 12:26 pm
Posts: 4927
Location: Lisbon, Portugal
hollow_man wrote:
Schematics are here.


With that schematic, it is recommended that you use a jfet input opamp.
And in that context, R1 doesn't smell good to me, at such a high value. It only adds noise. A jfet input opamp doesn't need that "protection" - the AD815 with a volume pot behind it, does.

I don't like the sound of the AD8610, but that's IMO.
You may try on that circuit the OPA637, which sounds significantly better than the OPA627.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein


Top
 Profile  
 
 Post subject: Re: Using opamps
PostPosted: Tue Sep 30, 2008 10:37 pm 
Offline

Joined: Mon Apr 23, 2007 11:07 pm
Posts: 575
Location: So. California
carlosfm wrote:
hollow_man wrote:
Schematics are here.

With that schematic, it is recommended that you use a jfet input opamp.
And in that context, R1 doesn't smell good to me, at such a high value. It only adds noise. A jfet input opamp doesn't need that "protection" - the AD815 with a volume pot behind it, does.

I don't like the sound of the AD8610, but that's IMO.
You may try on that circuit the OPA637, which sounds significantly better than the OPA627.

Good info!

Tangent recommends the 637 here, but only for L and R, and not the ground channel (which apparently needs unity gain).
As far as R1, do the 627/637 need a value that high (not sure why that odd value -- 4.32K -- was chosen other than some sort of result from software calc)? If not, what do you recommend. Or how about AD825. A ballpark starting value would be nice.
Thx!


Top
 Profile  
 
PostPosted: Wed Oct 01, 2008 12:14 am 
Offline

Joined: Wed Jun 29, 2005 6:34 pm
Posts: 1061
Location: East Coast USA
An AD843 is a single device in a package - you must be thinking of some other part.

Some of these opamps give what appears to be greater sense of "high fidelity" because they have over-shoots or other artifacts that are associated with certain circuits and feedback conditions. Some people love it - some don't. I fall into the latter group.

This is the same kind of hi-fi sound you often get with the crappy power supplies found in most preamp and amplifier designs. Despite what you may have read elsewhere, the power supply is very much a part of audio signal path in most circuits. Feedback may reduce some of the effects of bad power supplies, but not all.

The Tangentsoft design is for low voltages and a simulated "ground" so that you can use small batteries for portable use. I don't think the AD797 is the right choice here at all for that application. The AD845 also prefers higher voltages because of the fabrication technology used. Some of the newer generation parts based on newer fab approaches may be better for this job. They're more optimized for low voltage and low current use. That may not be what might sounds best, but life is full of compromises.

That reminds me...

Why do all these headphone amplifiers have so much *gain*? Even with 300 Ohm headphones (like the Sennheisers), a volt RMS will drill a direct hole through your ear into your brain. At even head-banger listener levels, there's plenty of headroom with a unity gain amplifier following just about any source that people might have these days. An iPod produces about a volt RMS at its output when turned up to "11."

I'd think, and actually have thought, that what's really needed is a JFET buffer (pick your own favorite) following the volume control (pick your own favorite) into a well done unity gain output stage. The "diamond" style buffer in the Tangentsoft design is actually a good starting point for the output section, although there are better choices for the devices. You can get stupidly low distortion out of one of these into even Grado impedances, with the already low harmonics pretty much monotonic in nature. If you want to go further, you could use a suitably low gain front end that generates in phase and out of phase signals, each feeding separate diamond buffers. There's at least one manufacturer who does this with all of their power amplifiers, getting great results - all open loop. Then you'd drive each earphone with a push-pull signal, totally getting rid of that "ground", at least for the output section. It would require re-wiring your headphones, but life is full of compromises.


Top
 Profile  
 
PostPosted: Wed Oct 01, 2008 3:08 am 
Offline

Joined: Thu Dec 30, 2004 10:50 pm
Posts: 905
Location: Minnesota
Walt did that circuit how many years ago now. I believe it was his to begin with. I think.

It's been done in a bunch of different configurations.

And if you're going to do this via a single battery don't forget the input C.


Top
 Profile  
 
PostPosted: Wed Oct 01, 2008 8:00 am 
Offline

Joined: Mon Apr 23, 2007 11:07 pm
Posts: 575
Location: So. California
carlosfm wrote:
http://www.diyhifi.org/forums/viewtopic.php?p=22981#p22981
Read Mikeg's post too.

MikeG, in that post, noted: I would also add a .1 ohm / 3.3n snubber on each supply pin to common.
I've done this before with a DAC I/V. But will this tweak work in the Tangent PPA headphone amp with "virtual" ground (i.e., "common")? Which ground do I use: IG or OG?
Again, schematics for the PPA are here.

BTW: I do not use batt. to power this amp. Some info on my pwr source is here.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 17 posts ]  Go to page 1, 2  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group