Hand-matching single op amps?

[dano12]

I saw an advertising blurb relating to an overdrive that talke about using the benefits of using two single opamps instead of a double and "hand-matching" the opamps.

Obviously, matching transistors is pretty important, but has anyone ever heard of this?

[R.G.]

I love it when the innate urge to advertise ANYTHING that will give an advantage overpowers the need to make sense.

Opamps DO vary. They vary in "edge conditions" from part number to part number. They vary from unit to unit within the same part number, like all electronics stuff.

However:

THE WHOLE POINT OF AN OPAMP IS THAT IT IS INTENDED TO BE USED IN A WAY WHERE THE EXTERNAL PARTS MAKE ALL THE DIFFERENCE AND THE UNIQUENESS OF THE OPAMP VANISHES BEHIND THE FEEDBACK AND HIGH GAIN.

Then there's that monolithic thing. Dual opamps are made by "printing" two independent opamps side by side on the same piece of silicon. They are NOT made by building singles and tossing two of the singles into the epoxy body. Parts made on the same slice of silicon experience the same diffusion, etching, temperature treatments, etc. and are as identical as semiconductor devices can be made. While there do exist a few parts like the LM394 supermatch dual transistor where monolithic devices are further laser trimmed to get rid of even the residual differences in monolithic devices, it requires an insane amount of equipment to do this.

It is possible - barely - that someone could have access to the hugely expensive equipment needed to even measure the differences of a couple of single opamps and match them better than a monolithic pair. It is more likely that someone has deluded themselves into believing they can do this with a DMM, or worse yet, by ear. The odds of this really happening is so small that I would wager money on it not being true at a repeatable factual level, and I'm not much of a gambler. It's even more likely that someone just HAD to have a gimmick to advertise, and since carbon comp/handwired/point to point/silver solder/ and even Icelandic virgins have been overused, they settled on hand matched opamps.

My response to the advert is "WALOHCS! Prove it."

[Sir H C]

Hard to better match than if it is two well laid out op-amps on the same die, and the same package.  Some quads do use two dual die (but from the same wafer just a different dicing).  Agree fully with RG.  This is why those dual transistors or transistor arrays match better than singles.

[MartyMart]

Sounds like "Walohcs" to me too.
About to hand build some 990 opamps ( discrete ) for pro audio use, now they should sound nice !!
MM.

[slacker]

I don't know about the hand matching part, but depending on the circuit you could get some differences in tone or whatever by using 2 different types of single instead of a dual.
For example a rail to rail opamp might sound "better" in one of the stages and a bog standard 741 might sound "better" in another.
Probably just marketing fluff though.

[R.G.]

Yep, deliberately mismatching different types can make a difference.

Of course, that's the reverse of what the advertising blather said.   

[slacker]

Depends on your or their interpretation of the word matching. It could mean pairing 2 identical singles or it could mean pairing 2 different but complimentary singles.

[DougH]

I can see a new Gear Page thread now: "Hyuk Hyuk, Hey Goober, I just got one o' them thar new-fangled hand-matched tube screamers! I got the purest handcrafted tone possible now!"

AFAIC, this is just another variation on the whole "handcrafted" aesthetic shtick that has crept into so much boutique music electronics hyperbole over the last few years. This is designed to give you the impression that a dedicated 'artisan' (with a slight paunch, flannel shirt, bushy beard, and round wire frame spectacles) is hunched over his workbench poring over every molecular detail of his beloved, er, umm, 'creation', in order to provide you the "purest tone" humanly possible.

It's crap, pure & simple.

[The Tone God]

Depends on your or their interpretation of the word matching. It could mean pairing 2 identical singles or it could mean pairing 2 different but complimentary singles.

If it was just the term "matching" I would agree but the OP is quoting the term "hand matching" which gives you the impression that someone is matching the opamps like transistors in some effects so I would say they are trying to be misleading. If they are "hand matching" for real then they are wasting everyone's time for the most part.

If the seller was on a forum I would ask the general batch of whys and hows this needs to be done to see how they justify this operation. IMHO its just another angle of hype.

Andrew

[dano12]

It's crap, pure & simple.

Along with the hand-matching of opamps, there are other things advertised that just sound like snake oil:
- Quad Eutectic solder
- Every solder joint treated with Cardas activated rosin flux
- Copper enclosure for shielding against radiated EMI and ESD
- Cryogenically treated wire

[slacker]

If it was just the term "matching" I would agree but the OP is quoting the term "hand matching" which gives you the impression that someone is matching the opamps like transistors in some effects so I would say they are trying to be misleading.

Fair point, I was just trying to give them the benefit of the doubt, but having seen the rest of the spiel I think it's fair to say it's probably BS.

[R.G.]

Along with the hand-matching of opamps, there are other things advertised that just sound like snake oil:
- Quad Eutectic solder
- Every solder joint treated with Cardas activated rosin flux
- Copper enclosure for shielding against radiated EMI and ESD
- Cryogenically treated wire

Wow. It takes real nerve and a certain cynicism to claim things like that.

"Quad eutectic" is a term with no meaning. Metal alloys are either eutectic - meaning going directly from solid to liquid with no slushy/plastic middle ground - or they are not. Or perhaps "Quad" is a brand name. 

However, using a brand name - like "Cardas" - means literally nothing at all in a commodity market, like that for activated rosin flux.

Copper is OK for EMI shielding. However, you can do the same with aluminum and most other metals as well; copper offers no unique advantages, and unless there is somewhere to bleed the charge away to, merely shielding ESD is of no consequence. Conductive plastic would ESD shield as well as copper, maybe better since the high resistivity would lessen any sudden discharges.

And cryogenically treated wire. Wow. Where to start?
What would you say if I came to you and told you I could tell how cold this copper wire got at one time a few years ago by just listening to music conducted through it? That is the implied claim.

There are simply so many variants on the label "DANGER! Do Not Touch! 50,000,000 Ohms!"

[Arno van der Heijden]

This stuff makes me go laugh and sad at the same time....

Which manufacturer are we talking about here?

(BTW That quad eutectic solders seems to exist for real: http://www.soniccraft.com/products/connections/solder/cardas.htm)

[R.G.]

(BTW That quad eutectic solders seems to exist for real: http://www.soniccraft.com/products/connections/solder/cardas.htm)

So they do! I went to the Cardas web site for some background. It appears that "Quad Eutectic" means "an alloy of four metals (tin, lead, silver and copper) which is the correct mix to be eutectic". So that really should read "Four-metal eutectic solder", but I suspect that does not sound as good as "quad eutectic". George Cardas claims to have invented four-metal eutectic solder; that claim seems to be at least embellishment on the face of it, as metallurgists have been beating their heads on solder composition for at least eighty years.

And while we're on it, I looked up the original paper that soniccraft used as a paraphrase source. It's the usual mishmash of halftruths and non sequiteurs that haunt the audio field. I guess it's worth it to some people to sell solder either (a) is a stock industrial product that's little-known, (b) people mix up for you to your spec, or (c) that you cook up yourself in a pot out back, which is entirely possible - for $50 a pound.

Nota bene: I did not dig any deeper into this. It smells so strongly of audio tweeko blather that there might even be some grain of truth in there, hidden from me and I just didn't have the stomach to look. This note is to sooth any True Believers who read this and are outraged enough to dig out a truth I missed.

Sigh. Cardas probably believes his hype. Here's a excerpt from one of Cardas' papers on cable breakin, for your education:

A note of caution. Moving a cable will, to some degree, traumatize it. The amount of disturbance is relative to the materials used, the cable's design and the amount of disturbance. Keeping a very low level signal in the cable at all times helps. At a show, where time is short, you never turn the system off. I also believe the use of degaussing sweeps, such as on the Cardas Frequency Sweep and Burn-In Record (side 1, cut 2a) helps.

I bet you didn't know you were traumatizing your cables by moving them, did you? 

[PerroGrande]

I've had several cables in therapy for years...  The cable that connects my laptop screen to the base is positively a mental case.

The choice of solder used isn't driven by "sound" (I'm not sure what solder sounds like, except when the connections are not "properly done") -- it is driven by the intersection of cost and suitability for the task at hand. 

Kester (a real solder company) notes:

The 2% silver is required when soldering to silver or silver plated components/leads. The small percent of silver in the solder prevents the silver on the leads from migrating into the solder resulting in a weak or brittle solder connection.

and

The Sn60Pb40 has a plastic range and puts down a slightly thicker coating of solder. Sn60 is often preferred for lead tinning and other solder coating applications. Sn63Pb37 is eutectic and as such has no plastic range. Generally it flows better than the Sn60 and is the preferred alloy for wave soldering and surface mount applications.

Apart from the necessity to include silver in the alloy when the target material is silver/silver-plated (which is legit), the rest strikes me as snake oil...  $44/pound snake oil that I would be willing to bet could NOT produce a distinguishable or measurable difference in audio performance when compared to a "properly done" connection made with an appropriate traditional solder.

I'm not going to have time to spend on this, however, as I'm going to be off in my basement hand-matching op-amps using my crappy oscilloscope, ancient voltmeter, and 10% carbon resistors.  And while I'm doing that, I'm going to start packaging up my new product: "The Big Dog's Tincture of Nitrogen-70" -- which, when added to a room, will improve the transmission of vital longitudinal waves (when compared to a vacuum)...  Only $89.95/cm3...  God, I'm in the wrong business!

[R.G.]

God, I'm in the wrong business!

You and me both. When I read that stuff I first laugh at the sheer silliness, then the nerve of someone to be willing to expose their least-attractive parts in public, and then I cry that there are enough people who actually believe it that they not only buy, they DEFEND this stuff, sometimes to extreme limits.

P.T. Barnum was, as always, right.