Resistor type, quality, location. ???

[Gainzilla]

Hi all,

Apologies in advance as I know this has been beat to death. Please bear with me. I'm planning a new build and I'm trying to decide what resistors to buy. The amp is going to be a two channel affair (fender clean, Marshall-esque 4 stage lead, built in a 100w donor chassis).

I suppose if I'm looking for "mojo" I would use NOS carbon comp resistors for the fender stage. What about the crunchy channel? I was thinking about carbon or metal film for most locations and 1watt Carbon comp on all but the first stage plate resistors (PI included). Does that seem reasonable?

Next, does brand or dollar really matter? I'd love to spring for Takman or Kiwame, or even the Mil-spec dales I've used before, but there will likely be upwards of 70 resistors in this thing. That's a lot of money for "mojo." Is it fine to just use mojo resistors in the signal path and something like Xicon elsewhere? What is your workhorse resistor? What do you use if it's for you, and not for profit?

Also, is there any real reason why resistors from Tayda should be avoided? 1% carbon films are like 10 for a nickle or something crazy like that. I've heard a lot that overall circuit design (and resistor type) has more influence on the amp than esoteric resistors, but where does quality factor in? Reliability? Quality control/consistency? Tone?

I do believe that good quality components make a huge difference, but it's also true that there are diminishing returns. We aren't building hi-Fi, we're building rock and roll machines. They're supposed to be colored or have personality/attitude. If love to hear your thoughts. Should I spring for high-end, roll the dice on Tayda Carbon films, or just play it safe with a Xicon or similar?

Thanks in advance! Any advice helps. I found the Takmans on sale through December for .65/piece. I might spring for a few select values if it's worth it.

Cheers!

[xtian]

Totally biased, personal opinion:

CC resistors look period appropriate, but add nothing to the sound until they are old, rotten and slightly malfunctioning. They do add noise, however.

CF and MF resistors from reputable mfgrs are very good, reliable, and reasonably accurate. And cheap!

Pricy resistors may buy you more accuracy, but I think that's not worthwhile in a guitar amp. They do, however LOOK awesome.

I like these "pillbug" Vishay/Dale, 1w MF resistors:

http://www.vishay.com/docs/31544/packthre.pdf


They're adorable!

[img:510:394]http://image1.shopserve.jp/transmitsoun ... aydale.jpg[/img]

[Gainzilla]

Yeah, I agree, purely from an aesthetic point of view it would be nice to have a sexy looking amp. Key word, nice. I'm not trying to build a show car that I put on display with the hood open. I'm trying to build a high performance muscle car for actually driving. Aesthetics play a role, but I would trade all the sexy in the world for a jaw dropping tone.

Where do you typically get those little bugs? I like how readable they look as well.

Cheers!

[R.G.]

I suppose if I'm looking for "mojo" I would use NOS carbon comp resistors for the fender stage. What about the crunchy channel? I was thinking about carbon or metal film for most locations and 1watt Carbon comp on all but the first stage plate resistors (PI included). Does that seem reasonable?

Unfortunately, no. Carbon Comp mojo ultimately is about two things; (1) excess noise and (2) voltage coefficient of resistance. I wrote a lot of this up back in 2002. Here's my cut on it:
http://www.geofex.com/article_folders/c ... oncomp.htm

Bottom line, there's only one, maybe two places a CC will produce a sweeter sound, and that's in the phase inverter or maybe just before it. Otherwise the signal levels are too small (in most amps; there are all kinds of odd amps around) to give the CC a chance to distort.

Next, does brand or dollar really matter? I'd love to spring for Takman or Kiwame, or even the Mil-spec dales I've used before, but there will likely be upwards of 70 resistors in this thing. That's a lot of money for "mojo."

You'll get all kinds of answers from people who really want to believe in magic, but as long as you're buying major-mass-manufacturer goods, not small-secret-sauce-maker items, buy on the specs - and go read the specs. Resistors have all kinds of limitations that people simply ignore - things like the voltage limitations as well as voltage coefficient, power rating, surface temperature *at* the power rating, self inductance, self resonance; it goes on. The makers often tell you all this stuff in the datasheet for free.

The major manufacturers **have** to get this right, because their customers that matter are big enough to hire enough lawyers to make errors costly. Not so with Hyper-La-De-Da Resistors From the Fifth Planet.

The reality is that resistors are made for big-volume electronics manufacturing, and the rest of us are parasites who get to hitchhike on the rolling feast. The big manufacturers turn out excellent goods for each other, and are quite blind to us.

Is it fine to just use mojo resistors in the signal path and something like Xicon elsewhere?

Mojo resistors are, almost by definition, a creation of advertising. They are created *to* advertise and sell for high prices, and also so that buyers can either have an inflated ego about having such finery inside their homebuilt cases, or so they can use the mojo references in their sales advertising in turn.

Bottom line, use mojo resistors where it makes you feel good. If it makes you feel good, it's probably worth the money as long as you believe the mojo. Sadly, if you ever question the mojo, the magic has a way of evaporating.

What is your workhorse resistor?

Metal film, metal oxide, and wirewound (which, curiously enough, have the lowest excess noise). If I'm trying for some original tube-amp fak... er, mojo I'll use some CCs in the plates of the PI.

What do you use if it's for you, and not for profit?

Why ever would those be different?

Also, is there any real reason why resistors from Tayda should be avoided? 1% carbon films are like 10 for a nickle or something crazy like that.

A better way to state that question is: what major manufacturer do Tayda's resistor come from? Or are they floor sweepings from some other enterprise?

I've heard a lot that overall circuit design (and resistor type) has more influence on the amp than esoteric resistors, but where does quality factor in? Reliability? Quality control/consistency? Tone?

You're at the start of a long climb in learning. You have to ask yourself this about all of these questions: can you write down clearly what you mean by the question, and how you would measure it?

I was required by my then-current employer to complete formal training in engineering for quality in design, manufacturing, and delivered products, on top of undergraduate and graduate degrees in EE as a prerequisite to getting hired at all.

The short answer on quality is this: quality is **conformance to requirements**. It's not high price, pure materials, well-thought-of in the trade press, mentions in trade magazines, "wow factor" to folks passing the booth at trade shows, whatever. If it meets the requirements, it's of high quality.

That then begs the question of "what are the requirements", doesn't it? The next short answer is that if you can't write down the requirements, you don't have a real good chance of getting quality. It's like the exchange between Alice and the Cheshire Cat:

The refrain "If you don't know where you're going, any road'll take you there" was essentially a paraphrase of an exchange between Alice and the Cheshire Cat in Chapter 6 of Lewis Carroll's Alice in Wonderland:

"Would you tell me, please, which way I ought to go from here?"
"That depends a good deal on where you want to get to," said the Cat.
"I don't much care where—" said Alice.
"Then it doesn't matter which way you go," said the Cat.
"—so long as I get SOMEWHERE," Alice added as an explanation.
"Oh, you're sure to do that," said the Cat, "if you only walk long enough."

So to expound on your list:
Reliability?
Reliability is a statistical item. For major manufactured goods, it can be calculated from the failure rates and methods in the datasheets from the manufacturers. Don't know the manufacturer?? Uh-oh...

Quality control/consistency?
Big deal here. I was part of the push for quality in the 80s and 90s when the world came up with ISO9000 as a way to measure it. You cannot assess quality control or consistency from anecdotes in an internet forum. Sadly, you're going to have to go learn the stuff and do the research to find it out. Even if 100% of everyone here had bad luck with Honest John's Reliable Resistors, that says *nothing* about their true failure rates without also specifying lots of other stuff about who, when, where, and how they were applied.

Tone?
Write down a clear, concise, and measurable definition of "tone" and I can tell you exactly what resistors - and caps, and tubes, and transformers, and on and on to use. Otherwise, picking things for "tone" is rather like picking out girls at a dance: you have your preconceptions about them, how they look at the moment, and their reputations as filtered through your "friends" as a guide. You may really think that redheads are the gas, but the "friends" are hung up on "tall".

I do believe that good quality components make a huge difference, but it's also true that there are diminishing returns.

Good, as far as it goes. However, consider Lord Kelvin's maxim:
“When you can measure what you are speaking about, and express it in numbers, you know something about it, when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely, in your thoughts advanced to the stage of science.”

Thanks in advance! Any advice helps. I found the Takmans on sale through December for .65/piece. I might spring for a few select values if it's worth it.

I realize that this is an heretical view in the amp world, but unless you know what you're buying, and what you get for your dollars, in a clear and measurable way, you're asking for what would make other people think your amps are cool. There may or may not be any relationship to what you wanted when you started.

[Phil_S]

I believe R. G. knows of what he speaks and listen to him if are able to do so. Unfortunately, if you are like me and lack a technical background in amp sorcery, more or less you will need to rely on what others say. On that score, I believe I have built some very decent sounding amps with metal film resistors throughout. Early on, I came to the conclusion that mojo, if it exists, is hidden from view and I don't know where to find it.

I will tell you that I have bought el cheapo resistors on eBait from some Pacific rim country and I got what I paid for. By this I mean really thin leads and that makes me not want to use them simply because they require rather gentle handling. That makes me question whether they will stand up the whatever punishment they will get at something north of 300V. The good news is, when I spend $2,50 for 100 el cheapo resistors and decide I don't like them, I have nothing to feel bad about and in the bin they go.

I'll give a BIG +1 on R. G.'s emphasis on reading the spec sheet. I'm not an engineer, but I can still understand a lot. There are a few important things to look for. At the top of my list is the voltage rating. Some resistors are only rated for 250V or less and that is, IMHO, way low for use in most places from the PI forward to the power section. The other thing I look for is hefty leads. We're not mass manufacturing with a wave soldering process. Get a product that can stand up to what we need to do.

Sometimes these two factors alone are enough to compel the purchase of 1W rated over 1/2W. Remember, there are plenty of spots in an amp where the math will tell you 1/2W is more than adequate, but that's not the right choice where you have low amp/high voltage present. It's not enough to just do the math. You have to look at the spec sheet and then at the application to make sure you didn't exceed the voltage spec.

The simple truth is that with a little patience you can buy decent resistors at Mouser and other big suppliers and benefit from the pricing they get. If you don't like those places, there are lots of smaller vendors that will sell to you, but, as R. G. says, do you know who actually made them? If you don't know, you can't read the spec sheet.

Good luck.

[arcturus]

Also, is there any real reason why resistors from Tayda should be avoided?

I recently finished a build with Tayda 1watt carbon films. Their values are pretty consistent.
On the down side, i found the outside package of the resistors to vary in format from value to value, and the most annoying feature is that the leads are very thin for what i expect to find in a 1watt resistor, even my locally sourced 1/4w resistors have thicker leads. This shouldn't be a problem for us, since we work with low currents.

[pdf64]

...Bottom line, there's only one, maybe two places a CC will produce a sweeter sound, and that's in the phase inverter or maybe just before it. Otherwise the signal levels are too small (in most amps; there are all kinds of odd amps around) to give the CC a chance to distort...

I'm trying to get my head around whether the CC LTP plate resistor induced distortion will be common mode or not? Probably not, as the signal wave should be in opposite polarity.
Also, irrespective of a global NFB loop to counteract it, will the local NFB in a LTP / cathodyne / floating paraphase act to linearise the signal?

My thinking is coming to the view that for a NMV type classic design, the 1st stage may be the best for a CC plate load, because

1/ the signal level there is likely to be sufficient for the distortion to happen, if only on the initial part of the signal envelope

2/ for any later stages, the next stage/s will likely enter grid clamping / cut off at rather lower signal levels than the CC distortion happens, thereby masking the CC effect.

The benefit of a 1st stage CC plate load being that the volume control can be turned down, thereby allowing the CC effect to be heard without excessive 'contamination' from later stages.
The key drawback being added CC noise, but the classic designs aren't super high gain, and their noise seemingly tolerable for decades.

For designs with a master volume of some sort, then it's probably best for the CC plate load to be immediately prior to that.

[martin manning]

I am yet to be convinced that there is any value in using anything other than good quality modern resistors that have the required specs. Quality 1% MF resistors are available in 0.5W size for less than $0.20 each in small quantities. Paying any more than that is just throwing money away. Of course, if someone is willingly throwing money away, and you can get a cut, I guess that's okay.

[M Fowler]

I couldn't agree more, Martin.

Mark

[cbass]

I only like resistors that you can't get anymore

[Structo]

[Phil_S]

I only like resistors that you can't get anymore

I can get you some of those.

[Jered]

"the rest of us are parasites who get to hitchhike on the rolling feast."

Classic!

[V2]

half watt MF resistors will be rated for 300V; most 1W MF resistors have the same limit. Use good-quality CFs (1W) if you need 500V.

[R.G.]

If your amp building has veered off into how the amp looks inside instead of how it works, you can stop reading right here.

If you are more interested in how to do what with what components you can get, this may be interesting.

Resistor power and voltage ratings are obviously per resistor. If you need a resistor that will stand 500V across it, but can only get 300V resistors, use two R/2 resistors in series. Yeah, yeah, that's so obvious as to be dull, but you'd be surprised how many people are stymied by not being able to think of it.

If you need a resistor that will withstand a power of P, but only have resistors rated at P/2, you can also parallel two resistors of twice the necessary resistance.

If you need a resistor of R ohms at P watts, you can make up the resistor by two R/2 resistors rated at P/2 watts; the powers ratings add. Or you can use two R*2 resistors in parallel at P/2 watts. The powers add again.

If you need a resistor R with voltage rating V, and you only have resistors with V/2 rating, you can use *four* resistors of value R in series/parallel and get an equivalent resistance R, voltage rating V and a power rating of four times the power rating of the sub-resistors.

This is kind of trivial stuff that may only be useful late at night or on a weekend when you have to cannibalize parts, but it can also make up parts that are simply not commercially available. Arrays of lower-rated parts can be surprisingly robust against any one part's failure, and can be extended almost infinitely.

They just look funny.