MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

My take on SMT and DIY
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author My take on SMT and DIY
metasonix
I really don't think the situation is as dire as Muff claimed it is. Yes, the industry IN GENERAL is going SMT with a vengeance--especially for the packages on new integrated circuits and other semiconductors.

However: there is still a booming market for low-production electronics, such as "boutique" guitar amps, hand-made effect pedals, industrial equipment, etc. So it's very unlikely that all through-hole or "leaded" components (meaning, having wires to connect them with) will disappear completely. They are getting more costly, but the ongoing demand is too great. So it will not be a problem to get resistors and caps for building tube amps, for example. I get a lot of parts from Mouser, and they still have leaded passive components available for decent prices.

Example: The carbon-composition resistor was discontinued about 10 years ago. It was the older way to make a (small) resistor, dating from around 1920. Carbon-comp resistors were THE standard component in the electronics industry, until carbon-film resistors took over in the 1970s, right around the time that tubes disappeared from consumer electronics. (Carbon film is more stable than carbon-comp, but has problems at radio frequencies above 150 MHz. It is easier to mass-produce, I hear.) Carbon comp looked dead, as its special properties enjoyed very little demand--until Victoria Amplifiers got together with some other boutique guitar amp builders and a few makers of specialized RF electronics, and paid Ohmite and KOA to restart low production of carbon-comp resistors. They're comparatively very costly (about 20c each for 1/2 watt, versus <5c apiece for carbon-film) but supplies are no longer a problem for low production. Even Mouser is carrying them now.

and btw: resistors and capacitors don't have wire leads because it makes them easier to use in assembling PC boards. Printed circuits are an afterthought. When PC boards (finally) started to catch on in the 1950s, all the components available were meant for point-to-point construction--where the resistors and caps were hung between terminals in a chassis. So the biggest PC board users, mass producers of radios and TV sets, simply used available resistors and caps. They managed to get the component makers to change some things--such as offering the parts preloaded onto reels, for automated insertion into the holes of PC boards, and-or with pre-bent and trimmed leads. The basic design of the components remained the same, until SMT came along in the '80s.

SMT is a great idea for low-cost portable electronics, and it has made some things possible that weren't, even 15 years ago. It's STILL difficult to breadboard with, and nothing the manufacturers of components have devised to date is helping that situation. To hand-assemble with SMT parts really requires use of a stereo microscope and fine tweezers (and much patience). Breadboarding of a prototype is often still done with leaded parts--if possible.

We are having no problem getting through-hole components, except for one IC chip that is used in the tube tester we build for Vacuum Tube Valley. In my experience, costs for electronic components of all kinds are going up across the board anyway.

(I can STILL buy all the vacuum tubes I want. Collectors are still offering me thousands of free TV tubes, for little or nothing. Sad that most of them are of no immediate use to me. One problem I am having is that the supply of new-old stock 7-pin miniature tube sockets is starting to dry up. Any NEW tube sockets are either cheap Chinese ones of poor quality, or horribly expensive and only available in 10,000 piece orders.)

ROHS is a whole different area of insanity and fear. It's still perfectly legal to use lead solder in commercial electronics -- IF you can take one of the special exemptions. We do, Analogue Systems does, and a few other synth manufacturers do. Given their low production compared to cellphones or PCs, ROHS is not a major problem for analog synth manufacturers--yet.
Kwote
great post.
felix
yes, very enlightening, particularly the part about RoHS special exemptions.
pizzamon
Just wanted to say that I found this an interesting read as well, thanks.
Randaleem
metasonix wrote:
ROHS is a whole different area of insanity and fear. It's still perfectly legal to use lead solder in commercial electronics -- IF you can take one of the special exemptions. We do, Analogue Systems does, and a few other synth manufacturers do. Given their low production compared to cellphones or PCs, ROHS is not a major problem for analog synth manufacturers--yet.

Hi Eric,

Great post! Do expound on the special exemptions. I've not been able to perceive one which applies. Or rather, i've not experienced how the EU is applying the existing ones. Any guidance on or offlist would be great!

Kind regards, Randal
flight
Excellent post sir!
I'd like to chime in with my own observations:
First off, I agree completely. Leaded components aren't going anywhere any time soon. The only devices that are likely to disappear are ICs that are still being manufactured with "archaic" (read: older than 2000) technology - which is what happened to the ICL8038, NE566 and the CA3080 I believe.

I try to use as many SMD (Surface Mount Devices) in my designs as possible, but there are many components that have not yet made the transition to SMT (Surface Mount Technology). One major example is the 2.2μF metal-film capacitor on the Plague Bearer. I have yet to find SMT capacitors that are larger than 1μF. Another problem the choice of dielectric - with SMT metal film you are largely limited to polyphenylene sulfide (PPS), PET (polyethylene terephtalate, A.K.A Polyester or Mylar), and PEN (polyethylene naphtalate). No polypropylene or polystyrene.

The rest are mainly mechanical items such as headers and pots. There are a few SMT headers, but 99% are just standard headers with the pins bent 90° to lay flat on the PCB, which just makes for a bigger footprint and zero protection against lateral stresses aside from the solder joint itself.
BananaPlug
Quote:
Example: The carbon-composition resistor was discontinued about 10 years ago. It was the older way to make a (small) resistor, dating from around 1920. Carbon-comp resistors were THE standard component in the electronics industry, until carbon-film resistors took over in the 1970s, right around the time that tubes disappeared from consumer electronics. (Carbon film is more stable than carbon-comp, but has problems at radio frequencies above 150 MHz. It is easier to mass-produce, I hear.) Carbon comp looked dead, as its special properties enjoyed very little demand--until Victoria Amplifiers got together with some other boutique guitar amp builders and a few makers of specialized RF electronics, and paid Ohmite and KOA to restart low production of carbon-comp resistors. They're comparatively very costly (about 20c each for 1/2 watt, versus <5c apiece for carbon-film) but supplies are no longer a problem for low production. Even Mouser is carrying them now.


How does carbon comp benefit a boutique guitar amp builder?
Randaleem
BananaPlug wrote:
How does carbon comp benefit a boutique guitar amp builder?

Hi BP!

Carbon comps handle overloading better and mmore gracefully than carbon film and metal film.

As a sound analogy, Think moog modular VCO vs. Cwejman. The second is tight and precise, the first is wonderfully loose.

Comp is essentially conductive mud, so you have variation which isn't present in the thin films, as the electrons have a greater number of paths from one end to the other. A bit of chaos effect perhaps?

Anyways, overdriving the resistor(s) in a boutique pedal can result in some nice warm grunge that metal film simply doesn't give.

A search on "advantages of carbon comp resistors" may prove fruitful.

They also look cool on the PCB! The smooth, even, shiny and true regular right cylinder, and the color bands are clear and crisp, compared to the oft-barely readable bands we get on our CF's and MF's. Like a vintage phenolic knob vs. the shiny plastic clones we're seeing now!<G>

Kind regards, Randal
Ken MacBeth
...hey! Great post Eric- good explanation too. I came to similar conclusions myself. My stuff is and always will be styled using the 'old tech' as long as that technology exists. There is no manufacturing advantage to using older carbon comp parts/ old capcitor types and styles and through hole parts- the advantage gained is the sound. There IS a difference! My Dual Oscillator and 'backend' filter rely on these parts for the sound- a warm thick and rich sound. I bough a ton of SMD's a while ago and built an equivalent of the M3X type oscillator- this is a close hybrid of the Mk II and Mk III mini oscs- the result was a decent bottom end- a good high end but the non tangible middle area seemed to have gone. They could be rock solid as expected- and I did notice a trace of something that I avoid- 'soft syncing'- this is different to the soft syncing people love- in this case- it's the slight locking of two oscs together just as they hit unison- then it sounds like one osc....thats tricky to deal with.
Back to the old parts- and old style layout. You like the old sound- you need old style parts and old style pcb layout. With SMT- the layout is a completely different disipline- it has different priorities. Old style we all hope to maximise the use of one layer- preferably the bottom layer- and if we can't connect- then we go on to the top layer. My average in design is normally 85% bottom layer, 15% top layer- and that's where most of my power tracks go...

...etc!

Ken
Ken MacBeth
...this is the stuff that I'm using in whats going out there- a tad different to the tiny surface mount parts!
Kent
Randaleem wrote:
BananaPlug wrote:
How does carbon comp benefit a boutique guitar amp builder?

Hi BP!

Carbon comps handle overloading better and mmore gracefully than carbon film and metal film.

As a sound analogy, Think moog modular VCO vs. Cwejman. The second is tight and precise, the first is wonderfully loose.

Comp is essentially conductive mud, so you have variation which isn't present in the thin films, as the electrons have a greater number of paths from one end to the other. A bit of chaos effect perhaps?

Anyways, overdriving the resistor(s) in a boutique pedal can result in some nice warm grunge that metal film simply doesn't give.

A search on "advantages of carbon comp resistors" may prove fruitful.

They also look cool on the PCB! The smooth, even, shiny and true regular right cylinder, and the color bands are clear and crisp, compared to the oft-barely readable bands we get on our CF's and MF's. Like a vintage phenolic knob vs. the shiny plastic clones we're seeing now!<G>

Kind regards, Randal


The above may be valid, but I'd like to mention three things:

1) In high voltage circuits, there may well be benefits to using Carbon Comps in select areas of the circuit such as when certain overload characteristics are desired. Just peppering a circuit with a component type and 'wishing' or 'believing' that it is making the sound of said collection of circuits subjectively 'better' seems like the kind of "Valve Voodoo" that is rife in the 'Gullible Boutique Guitarist Market' these days. It's a bit like adding motor oil to every part of the car and thinking that the whole car will run smoother.

This also leads one to confirmation bias and simply filtering out the information that one doesn't wish to hear & accepting what confirms one's held position.
Specialist Speaker Cable Manufacturers exploit Confirmation Bias to the hilt. A/B/Y tests show otherwise.

2) I've yet to encounter a 9VDC powered distortion box than can actually push a resistor into distortion. Is there one?

3) The last I checked, and it has been quite some time, Carbon Comps have much looser tolerances than any of the other types of comps mentioned. I think that this has a lot to do with the perceived sound of the entire system in which they are used. A 'System' is built upon of a collection of sub-circuits, transducers, transformers and other items which heavily, heavily, heavily influence the performance & tone of said system. Overall system design has a much greater influence upon system performance than an individual type of part. All of those old amps have really loose tolerance values in resistors and capacitors and, age and wear included, I think that this has quite a lot to do with their sound.

Just for the record, Randaleem; I'm not attacking you personally. I'm simply expressing my thoughts on this mystical & subjective subject for the benefit of the reader. This isn't really a rebuttal of you; it's more like "other things to consider". Dancing Star
Randaleem
Kent wrote:
The above may be valid, but I'd like to mention three things:

Just for the record, Randaleem; I'm not attacking you personally. I'm simply expressing my thoughts on this mystical & subjective subject for the benefit of the reader. This isn't really a rebuttal of you; it's more like "other things to consider". Dancing Star

Hi Kent,

No worries. I respect your opinions greatly, and look forward to your posts! I don't feel attacked in any way.

I'm the first to agree about the valve voodoo, which I refer to as the "400 dollar wood knob". You probably saw that one? And I'm on the same side of the speaker wire argument as you it seems; mine use lamp zipcord. No silver and full of oxygen!<G>
So getting to the specific issue at hand; Carbon comp resistors are different in these ways: A)They're noisier. Part of this is due to the conductive mud, i.e., their literal composition. Because of this they also are not linear in their resistive behavior. Do both of these facts affect the sound. Surely they can. Hard to argue otherwise. Is it a good thing? Much harder question, that! As you say, It depends upon the entire circuit and must be judged on a case-by case basis. People are not all going to agree. And it won't require golden ears!<G> b) Their tolerance is looser. See above paragraph for the result of this, along with your comments. IOW, I agree.
C)They can and do handle transients differently(better?) than modern resistors. Partially this is because their inductance is lower, the comp mud thing against the thin film which has generally been laser trimmed in a helical fashion. so in answer to your 1) above, I think we basically agree. I don't believe I suggested simply peppering them about wishfully? But there are places where they can be a better choice than a newer resistor.
As for your 2) above, I don't know. I'm FAR from an expert on pedals. But I DO know that design with intention, making use of the added noise and other "qualities" of comp resistors, properly applied and understood, DOES make a difference in the sound. Simply pretending that a resistor is a textbook current limiter leaves far too much real of the world out of the component, and therefore the design. Even basic college electronic texts will get into this kind of difference.
On 3), again I can't find anything I said which would constitute disagreement. I agree with what you've written.

Summed up, here's what I believe. 1) the type of components and all other factors DO matter. To a greater and lesser degree, to be sure, but there is an effect. Choices have consequences. Intentional design matters more. With the understanding that "intention" can be pro-active or re-active. For the same reason we have two types of loops in programming. The key to intentional design is the feedback loop, whether the test is before or after. During intentional design, both Wishing and believing are tested in this feedback.

Nice post. Kind regards, Randal
BananaPlug
Fascinating stuff. Thanks everybody. Reminds me of debates about capacitors. One person had some unconventional ideas about decoupling capacitors but I don't remember what his point was.
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group