Orange drops or Mallory 150s in 5E3 build?

[g1]

I'll say it again, there's no human on earth that can tell the difference between different style caps of the same value in the same circuit under the same conditions in a double blind test.

That is your opinion and you are fully entitled to it. It may even be true but good luck proving that one. What was your sample size again?

tubes... to say one " brand" sounds good is a statement bourne of ignorance and inexperience.

And if someone who has repeatedly experienced one brand sounding better than others (due to better QC, lack of microphonics or noise or whatever), that experience is not valid? Borne of ignorance?
I'm not one for the golden ears mojo stuff, but your statements are completely over the top.
I would have thought the same about resistors, until I read RG's excellent study. Then I realized different compositions may react to certain dynamic factors in different ways. Maybe not enough for most people to notice, but possibly enough for some people to notice. To deny even such possibility would be just plain foolish.

[guitician]

That must be the machine all those lousy caps are made on. I'll bet the thing damages the metalized layer of the film in the process of winding the cap.
Anything to made a buck I guess...

[Chuck H]

I have a JL Hood book here where he briefly discusses his short career as a consultant for a company that made plastic film for packing, packaging, kitchen wrap AND capacitors. At the time (as it is now) polystyrene and polypropylene were the film caps of choice for higher quality, but still affordable audio products. The polypropylene film came in several grades with the highest grade being intended for capacitors. All of the capacitor companies used a lower grade to save money on the premise that none of their customers ever heard any difference between the highest grade and cheaper grade film.

He does also note, though, that he thinks it's reasonable to assume there is some good reason that listener preferences tended toward the polypropylene product in general. And then he goes into all sorts of yammer regarding dipole charging and alignment for different dielectric materials and notes that polyester is a poor performer. He also states outright that just because there isn't a hugely significant test difference in cap performance that listener preferences must exist for a reason and shouldn't be dismissed.

That isn't to say that I'm all in this camp. But definitely a little. And I also think that any difference isn't better/worse as JL Hood tried to categorize it. WRT guitar amps, and their purpose as sound "makers" rather than strictly amplifiers, it's just a matter of personal preference. I think it makes good sense to try and match a cap type/construction when cloning for this reason. But I wouldn't get too hung up for a new design. And I don't. In other words, it would seem silly to me for someone to plainly state "I prefer the sound of polyester capacitors." because any influence the cap type does have is interacting with many other, and more influential aspects of design.

[cyclone]

Groove tubes and other resellers at the end of the tube production era had no stock to select FROM but the leftovers from outside edges of the tolerance bellcurve. All the good ones, the middle 5-10% went to military and large manufacturers still using them in huge quantities. Thats a simple fact of supply and demand and simply how manufacturing works, then and now. It's not slander to say that GT bought the fringe tubes and graded them, thats why GT made up the grading system, to turn that into a selling tool. Good for them, they had the best tubes around for many years. Used them a lot back then, output tubes were hard to find in matched sets, which was generally more important.


In one of those Popular Electronics from the 50's from the link I posted, there's an article about dud tubes and old tubes commonly being relabled, resold or turned in for warrantee claims. It was a multimillion dollar industry and a big concern to the major manufacturers of tubes. So, some of those old tubes you buy on Ebay may be relabled duds from the 50's!! So there's no telling whats on any tube just by the label.

New tubes mfrs simply have no quality control, they sell everything. The vendors and end users have to find the middle 20% and the bell curve is much flatter, meaning many fewer tubes are close to bogey of a given run.

And on coupling caps, my opinions are not opinions, they are the results of carefully conducted double blind tests with musicians over the last 35 years of doing this stuff. Double blind tests under controlled conditions. It is just science at the most basic level. It doesn't cost anything but less time than I spent writing this post. I know science scares superstitious people, thats ok.

Post less test more.

[Chuck H]

I somehow dismissed that you were speaking about a specific era for GT. And I actually agree that many of the NOS tubes relabeled were indeed A LITTLE off spec. If they weren't functional they would have been destroyed. GT still tested them for some level of relative goodness and gave performance descriptions with their relabeled model number. I know that happened because even GT eluded to all this at that time. Though I can't remember where I read it. So, sorry for that. You're mostly right then since that statement applied to a specific era and circumstances. I also think it's possible that GT developed their grade number system from testing those tubes. In the end though it was just used as a marketing ploy. They already had the numbers measured for each tube, why not use them? Good call IMO.

Just to be clear on it, the modern game has changed and is quite different. As you note, many modern tubes are out of spec. and sold like that from the manufacturer without excuses. The new game from relabelers is just to try and scrable up enough close to spec. tubes to be turned at a profit and cull the grossly bad ones. This use to be the case with non relabeling venders. But now, IME, those vendors just flip product leaving it to the consumer to make a claim whenever a grossly bad tube is received. Blessedly there are still a couple of vendors that will do additional testing and culling for an additional price. This actually makes sense for players and small time builders because it allows one to buy what they need for a given project If you only have one or two a year the added expense is small and the service saves a lot of time and trouble on our end.

Regarding your carefully conducted experiments... Good for you. My results are different from yours. This is where you say "Good for you." right back at me. Then we both move on with those things our relationship can support. 99% of the time, when people openly disagree it isn't because they want to learn from the ensuing discussion. They want to change the other parties information. Well, if both disagree-ees are doing this it's plain to see the futility. So let's have the good sense to keep relationships limited to what they are capable of providing. Doing otherwise only alienates people from each other further and then nothing happens. And there's nothing to be learned or gained from nothing.

[Pedro Vecino]

Groove tubes and other resellers at the end of the tube production era had no stock to select FROM but the leftovers from outside edges of the tolerance bellcurve

I've heard speak bad of Groove tubes on other occasions. Nothing new.
I went Groove Tubes client from the moment it was introduced in Europe in the mid-80s approximately in Frankfurt. Except some purely circumstantial problem all models in its first decade had perfect specifications without any problem. From the first to the last. 6550 GE, 6L6GC Sylvania, EL34 China, EL34 Tesla, 12AX7 GE, 5751 GE, 12AT7 Philips, 12AU7 RCA, 12AY7 and 6072 GE, ECC83 Tungsram, ECC83 Ei, ECC83 Tesla (transition to first JJ more exactly), 6V6 Russia and 6V6 RCA (I can forget some model). Wait: also 6Pi3c labeled as 6V6HQ (with a much large current consumtion than a 6V6)
At that time in Europe there were many thousands of Tungsram, RFT and others stored (I bought several hundred almost two decades later in sealed boxes of 100 paying his price) and Ei in Yugoslavia was working. Same with Russian 6V6: industrial quantities.
It could be that at that time in USA the supply of GE 5751 and probably some other model was limited by military requirements due to low production but that in relation to the global number of models and other more serious problems meant nothing.

[g1]

Post less test more.

I don't know if that was meant for me or for Chuck or for both of us, but it's terribly rude and condescending, and way below the usual high standard of your posts, which I always enjoy reading. So I'll assume you were kidding around and that I'm being overly sensitive. In the interest of finding common ground I'll remind you that the two of us have the distinction of both being told by SGM that he had repaired more amps than either of us had ever seen (let alone repaired) .

And on coupling caps, my opinions are not opinions, they are the results of carefully conducted double blind tests with musicians over the last 35 years of doing this stuff. Double blind tests under controlled conditions. It is just science at the most basic level.

My data over a similar time frame agrees with yours. I'm just not ready to come to the same conclusion. I agree with your hypothesis, but it is not yet proven fact. That is basic science. Between us, if we've checked 7000 people, that would work out to around .0001% of the people on earth. Still some room for doubt. I can hear a lot of things guys I play with can't. There has to be a lot of people with way better hearing than I have. I don't believe most people could hear any difference, I sure can't. But there's no way I'd ever put up a million bucks for anyone who could pass such a double blind test. You sound like you would .
Sorry about all the nitpicking, I just think the claim that no one on earth could possibly pass such a test requires a belief of religious proportions. I may agree with it being probable, but I don't have enough faith to consider the opposite completely impossible.

[cyclone]

There's a big difference between "I swapped out coupling caps in the same amp, and they sounded different" than conducting blind tests and reporting results. my results are that in literally hundreds of double blind tests on coupling cap types that I have conducted, playing at any volume, with audiophool amps and audiophools, nobody can tell the difference Me included, and believe me I "wanted" to hear the difference. ceramic disk, cowboy leatherette (LOL) Bacon, herculon, corinthian leather


I'm not trying to insult anyone, I'm honestly sorry if it was taken that way. Everyone should be testing every part they put in an amp. People should conduct real tests and report real results on these forums not taste tests on electronic parts that simply don't fit a given application. Orange Drop. What does that mean? Some of them have leads like fence wire. Which series and model of capacitor?

[cyclone]

I can hear decoupling caps though

[Chuck H]

I can hear decoupling caps though

I get that. Decoupling caps have a greater variance in their flaws WRT circuit function. Small interactions between stages occur in these flawed areas causing small positive and negative feedback loops. At least this is how I interpret it when I examine a typical circuit.

I've been choosy about using higher performance filter/decoupling caps for that reason. What I haven't examined is whether or not some of the "failings" caused by these flaws may actually have tonal benefits?!? That is, perhaps certain coveted classics sound like they do partly because the decoupling was less than great in key areas. I think dai h. did some experiments with adding small resistance to the ground end of the decoupling caps some years ago and found that there was indeed an audible effect. Better or worse would be subjective and probably design dependent. One example would be many new Fender combos that have the HV rail decoupling in phase stages. Oscillations occur when the caps stop doing a complete job. No room there for experiment. But on most amps the decouplers are arranged on out of phase pairs. There's a lot of things that affect tone that I haven't taken the time to examine closely.

[dai h.]

hey chuck,

my memory of that (couple of experiments I tried) was this (in a "guinea pig 50W Marshall rigged up as typical split cathode circuit w/very slight parts value alterations but pretty much the stock circuit) :

1) 0.5(0.3?)mH inductor in series = no audible effect (as far as I could tell)

2) very small R in series = no audible effect (I'd guess the value was too small (I found the pic of the experiment and the R was a very tiny 0.27 ohms). If memory serves in some old thread R.G. recommended MUCH higher values to simulate old caps.

3) 3uF film(two 1.5uF metal polyester) in place of (16uF? alu electro) = seemed audible (slightly smoother/nicer sounding). I would guess some might consider it an improvement or at least useful.

(I also "Y-ed" the filters for the Bright and Normal channels (i.e. separate decoupling R and C). In hindsight, I should have adjusted the dropping Rs to make sure I obtained the exact original B+ at those nodes (but I didn't feel this was a huge difference).)

As far as cap types, I seem to have a similar position to chuck's. That is, there have seemed to be audible differences, but generally not to the degree they are crucial "make or break" in significance. I would recommend trying different values over say, trying every 22nF you can get your hands on. Possibly, you could group them from cleaner to dirtier (less distortion to more distortion) types (say clean stuff would be polypropylene, polystyrene, class 1 ceramic, slightly more distortion as metal polyester, polyester, dirtier as non-temp. compensating ceramics), and my understanding is the audible-ness depends on circuit position (and maybe voicing--as in if the response has more of the freqs human ears are sensitive to then any possible differences are probably more audible?).

Also, it might be good to get a sense of perspective since something like speakers will be far more significant (as mentioned in the past by Bruce C. and probably others), and very subtle differences might not matter if recorded and the high end is rolled off by the engineer (or ear fatigue sets in--and maybe aging--my understanding is that you lose response and sensitivity as you grow older though as with lots of things seemingly genetics and environment probably has some influence).

[Pedro Vecino]

The effect of the capacitors is not only perceivable through the ears. In addition and more importantly, through the physical contact with the instrument.
A sound is a sound for a viewer, but for a guitarist is also a texture that modulates. Different types of capacitors (as tube models) are involved in the composition of the texture together with the design and its instrument. Is that particular texture and its association with the needs of the musician which allows a wide, suitable channel for his expression.
If do not have this in mind (or can not interpret it for whatever reason -it may be many-) experiment with capacitors is simply a waste of time.
I have never filed an amplifier with capacitors replaced by another model waiting the consecration of the musical spring. But when I modified a design, choosing a capacitor value and a specific composition, yes I have considered it and the final result is not the same selecting different compositions/models.
However the high pass filter capacitors, treble capacitor and volume bypass capacitor in a Marshall not sound the same being ceramic, silver mica or polystyrene. 2200pF silver Mica capacitor associated to vintage and modern modes on a three-channel Rectifier does not sound like one of polystyrene of the same value. And of course, a Tweed Deluxe with Sprague 715p does not sound like one with polyester foil film capacitors. The list is endless.
Differences depend on what other elements permit. If the Tweed Deluxe sounds ugly as a devil, capacitor changes is not going to turn into a jewel. Same with Marshall and everyone else. I've never known an amplifier in which a simple action turns it into something wonderful.

[Dave H]

My data over a similar time frame agrees with yours. I'm just not ready to come to the same conclusion. I agree with your hypothesis, but it is not yet proven fact.

So, in 35 years of double blind testing no one has detected a difference but you still believe there is one. That sounds like religion to me

[g1]

No, I don't really believe there is any significant difference to be heard. But my data (and maybe cyclone's too) is skewed in that I don't really deal with the kind of people who do believe that.
So since I can't 100% disprove it, I don't make 100% absolute statements.
Another thing, all the arguments for either side demand only that the caps measure exactly equal in micro-farads. Do you think ESR could make a difference? How about ceramic discs that are generally accepted to be more microphonic, can anyone actually hear that?

[Jazz P Bass]

Excellent 'Myth Busters' type article.

Atlantic Quality Design, Inc., The Truth about Tone Capacitors