So what in solder "sounds" bad, and what can be done about it?

[Alan0354]

But does the sign of the dR/dV matter? Could it add to the second harmonic produced by the tube with one sign, and subtract with the other?

The CC might have opposite effect from the tube, but I venture to guess the tube will dominate the CC no matter which way. I don't know, just a guess. If you look at the plate current vs plate voltage with constant grid voltage, the slope of the straight portion is pretty steep. That means with a given grid voltage, dI/dV is pretty high and the gain drop as the plate voltage decrease which give the round bottom of the sine wave when plate voltage is at the lowest point. This is second harmonics right there.

[soundmasterg]

Yes, the late Bob Pease was one of my favourite electronics writers. His book on troubleshooting is packed full of information that has helped me no end, both at work and in hobby projects.

The irony is that he died by crashing his car after publishing a book on how to drive safely.

The other sad thing was that this happened on the way home from the funeral service of Jim Williams, my other favourite electronics writer, who died of a stroke the week before.

Which troubleshooting book are you talking about? I always like those.

Greg

[Steve Conner]

"Troubleshooting Analog Circuits", published by EDN.
Amazon.com: Troubleshooting Analog Circuits (EDN Series for Design Engineers) (9780750694995): Robert Pease: Books

In the cover photo it's hard to tell where his beard ends and the tangle of wires begins- a proper analog wizard!

Re the Art Of Electronics: The graph showing non-linearity of a carbon comp resistor is in Figure 6.53, page 373 of the second edition. I think RG Keen and Morgan Jones may also have measured it.



SAME LAB 18 MONTHS LATER

M.S.: Well, the experiments are finally done.

Me: Thank God. Did we discover anything?

M.S.: We proved that Seymour Duncan's silver wire pickups aren't worth the money. Even if I did have to buy 5 sets and cut them up for analysis.

S.D.: (In bath of cryo treated asses' milk) Doubled my sales of the things this year!

Me: I guess it was worth the grind, to help bust some audio B.S.

M.S.: Oh, you were in the control group, you've been playing the exact same guitar since Day 1.

Me: Cries

Ghost of Leo Fender: Fools! Real progress doesn't come from tinkering with minutiae that require blind tests to distinguish them. It's a step change that even Joe Public can recognize. My revolutionary Stratocaster still outsells all your boutique trinkets. And the irony is, even if you did come up with a game changer, everyone would still buy Strats.

All: Amen.

THE END

[Mike Sulzer]

Re the Art Of Electronics: The graph showing non-linearity of a carbon comp resistor is in Figure 6.53, page 373 of the second edition. I think RG Keen and Morgan Jones may also have measured it.

Thanks for the reference. I did not think of looking in the regulator section last nignht.

[Mike Sulzer]

The CC might have opposite effect from the tube, but I venture to guess the tube will dominate the CC no matter which way. I don't know, just a guess. If you look at the plate current vs plate voltage with constant grid voltage, the slope of the straight portion is pretty steep. That means with a given grid voltage, dI/dV is pretty high and the gain drop as the plate voltage decrease which give the round bottom of the sine wave when plate voltage is at the lowest point. This is second harmonics right there.

Certainly the tube dominates for the largest signal swings, but since the resistor effect is pretty gradual, it might have a small, but nonetheless greater, effect at intermediate levels. I do not know, and this gets pretty complicated.

By the way, I think it is the spacing between the the lines for constant grid voltage that results in the triode non-linearity. That is, you get less change in the voltage across the plate resistor for a given increment in grid voltage as the grid voltage becomes more negative, that is, lower plate current.

[David Schwab]

Ghost of Leo Fender: Fools! Real progress doesn't come from tinkering with minutiae that require blind tests to distinguish them. It's a step change that even Joe Public can recognize. My revolutionary Stratocaster still outsells all your boutique trinkets. And the irony is, even if you did come up with a game changer, everyone would still buy Strats.

And actually. Leo was never finished tinkering with pickups and guitars. So judging from that HUGE assortment of prototype pickups on the table, minutiae was not lost to Leo.

Amen to that!

[imaradiostar]

"Troubleshooting Analog Circuits", published by EDN.
Amazon.com: Troubleshooting Analog Circuits (EDN Series for Design Engineers) (9780750694995): Robert Pease: Books

In the cover photo it's hard to tell where his beard ends and the tangle of wires begins- a proper analog wizard!

Re the Art Of Electronics: The graph showing non-linearity of a carbon comp resistor is in Figure 6.53, page 373 of the second edition. I think RG Keen and Morgan Jones may also have measured it.



SAME LAB 18 MONTHS LATER

M.S.: Well, the experiments are finally done.

Me: Thank God. Did we discover anything?

M.S.: We proved that Seymour Duncan's silver wire pickups aren't worth the money. Even if I did have to buy 5 sets and cut them up for analysis.

S.D.: (In bath of cryo treated asses' milk) Doubled my sales of the things this year!

Me: I guess it was worth the grind, to help bust some audio B.S.

M.S.: Oh, you were in the control group, you've been playing the exact same guitar since Day 1.

Me: Cries

Ghost of Leo Fender: Fools! Real progress doesn't come from tinkering with minutiae that require blind tests to distinguish them. It's a step change that even Joe Public can recognize. My revolutionary Stratocaster still outsells all your boutique trinkets. And the irony is, even if you did come up with a game changer, everyone would still buy Strats.

All: Amen.

THE END

Amen.

I really need to buy a few of Bob's books. One of my coworkers keeps one at his desk and it always makes for an informative and entertaining read.

Steve, I think you speak to the truth of the matter. An audio engineer friend in Nashville has regularly told me that "there is no golden signal path." If you spend time in pro audio and studio circles you'll meet those people that say you NEED a vintage tube mic into a telefunken preamp into a fairchild limiter to get "the sound" but the truth (and many famous recordings) reveal that it is simply not the case.

Too many modern amps have too many knobs that don't do a whole lot. Who out there is really listening to these things?

My cousin (another engineer- yeah, I lived in Nashville for too long) said he'd take stepped controls with audible changes between positions over pots any day. Easy to return to a setting, easy to say better or worse at a given setting. This makes a lot of sense- focus on the changes you can hear and make good decisions.

Jamie

[Alan0354]

Certainly the tube dominates for the largest signal swings, but since the resistor effect is pretty gradual, it might have a small, but nonetheless greater, effect at intermediate levels. I do not know, and this gets pretty complicated.

By the way, I think it is the spacing between the the lines for constant grid voltage that results in the triode non-linearity. That is, you get less change in the voltage across the plate resistor for a given increment in grid voltage as the grid voltage becomes more negative, that is, lower plate current.

This can get complicate fast as you can look at it in different ways as jumping between constant grid curves, or on a given grid curve and look at the dI/dV slope. Without going crazily deep into the math, I venture to say they say the same thing. In your way, you are plotting the load line and look at the curve of the load line to explain this. In my case I just look at the transconductance change due to plate voltage. You can see the non linear effect.

[Jazz P Bass]

I think all you need is a "Feel Good Knob".

[Mike Sulzer]

This can get complicate fast as you can look at it in different ways as jumping between constant grid curves, or on a given grid curve and look at the dI/dV slope. Without going crazily deep into the math, I venture to say they say the same thing. In your way, you are plotting the load line and look at the curve of the load line to explain this. In my case I just look at the transconductance change due to plate voltage. You can see the non linear effect.

That makes sense; it has to be the same physics no matter which way you look at it.

[FunkyKikuchiyo]

I think all you need is a "Feel Good Knob".

Just don't twist around your knob in public... even if it does feel good.

[Mark Hammer]

Well you can always rely on the interwebs for facts. No rumours there.

A few comments.....

1) As several have alluded to, it's important to differentiate between solder and soldering technique. A 30W iron with a pristine clean tip may work great for 60/40 solder but if it needs some re-tinning and you're using 63/37 or lead-free, you may end up with more than your fair share of cold joints, and what people end up hearing is not the impact of the solder, but rather the impact of cold joints that just happen to use that solder.

2) I find, with annoying consistency, that people generalize across disparate contexts regularly, and presume that what is good advice and a best practice in context A must necessarily be a requirement in context B, because, after all.....it's electronics....and it's sound, right? So people will stumble onto something that makes perfect sense for hgh-end power amps used for reproducing broad-bandwidth multi-source music signals where all harmonic information needs to be phase-aligned in order for ear/brain to correctly "assign" harmonics to the right fundamentals and create coherence, and assume that it also applies to how to wire up an EB-O or a fuzzbox. Whats missing is a pervasive sense of X matters here, but not there.

Could solder composition matter? Sure. I have no doubt. Does it matter in the tiny quantities employed in pickups or their connections, particularly considering just how many additional stages it will pass through on the way to the speaker cone/coil that may use any of a variety of solder types from different companies? There, I do have my doubts.

[RedHouse]

...Whats missing is a pervasive sense of X matters here, but not there...

That will never catch on Mark, blows all the arguemental opportunity's away!

All kidding aside, yes I do agree, and it seems to be pre-requisite on all the forums these days, etched in stone somewhere on mount doom it says:

"Thou shalt not post without context malignment"

[guitician]

I once built a guitar amp from a late fifties console chassis. All I did was "tack solder" parts from my "pulled parts" bin until I "felt good" with what I was hearing. With all the types of caps I had the 1KV ceramics sound the best. And CC were the flavor of the day at the time, so I used them too. All the parts were being held by globs of 60/40 solder, but when I finally wrapped all the terminal strips with the leads and made the final connections, it was just as I had it sounding before, great, and with the most treble of any amp I have ever played, yet it still remained sweet sounding. It may be that random chance is one of the greatest audio tools there is.

[defaced]

A 30W iron with a pristine clean tip may work great for 60/40 solder but if it needs some re-tinning and you're using 63/37 or lead-free, you may end up with more than your fair share of cold joints, and what people end up hearing is not the impact of the solder, but rather the impact of cold joints that just happen to use that solder.

I respectfully disagree with part of this statement. Mixing 60/40 solder with 63/37 solder is not going to give any significant change in results. There's nothing interesting happening between 40 and 37 Wt% Sn in the Pb/Sn binary system. A mix of those two solders are going to behave more or less like 60/40 because Alpha Pb will solidify as a separate phase ahead of the eutectic composition just as it does with 60/40 solder.