Military grade components? Also, perfboard

Started by mvotre, September 18, 2014, 04:02:18 PM

Previous topic - Next topic

mvotre

hi there.

First, sorry for some english errors (not a native speaker).

I had a account here since 2007, but just now I'm returning to DIYing. Already made a distortion and a TS a looong time ago.

Now, I'm studying all again, first to build some small amps (found a never known love with my Smokey amp those days), and them some pedals for myself. But I got this questions :

1 - what about those so called military grade/premium stuff for the components? I found that the mil. generally means components with stupid tolerances to high/low climates and stuff like that. Some high quality components seens (from what I saw) to have some smaller tolerances (like 1% instead of 3% of variation in the values - made numbers here). SO, the question is: for some home made pedals, wich I will just use at home, high grade components make a difference in sound? Since the values in guitar effects are small, those variations will affect tone? Tried to find some tests/info on Google, but couldn't find nothing. With so, all components are better, or just some? I'm not building to try to save money or something like that, so if high quality components really have some audible differences, I will first build with cheap local found pieces, and them biuld a better version, with the expensive stuff. If it just sounds virtually the same, cheaper parts = more projects  :icon_cool:

As I said, I tried to google that, and even searched in this forum, but couldn't find anything. If someone have some links, please share!

2 - I will build a Ruby amp and other similar small amps first, and will buy some perfboard. It can look fairly nice (as some examples here show - http://mavweb.mnsu.edu/quistd/Classes/EE-336/Perfboard.html). For personal projects, can i just leave it like that, or I would get some performance issues, and should make a pcb after the initial testing phase? Again, not a problem to rebuild it, if the end result can be better.

Thats all. Hope those questions were not asked every day, and my next post will hopefully be to show my first working amp (high hopes for that - my first distortion took a hell of a long time to get it right). Sorry for some bad phrases, and for the long rambling  ::)

R.G.

Quote from: mvotre on September 18, 2014, 04:02:18 PM
1 - what about those so called military grade/premium stuff for the components? I found that the mil. generally means components with stupid tolerances to high/low climates and stuff like that. Some high quality components seens (from what I saw) to have some smaller tolerances (like 1% instead of 3% of variation in the values - made numbers here).
Ignore that. Just use normal commercial components. Even most militaries are using normal commercial stuff, but testing it first.
Quote
SO, the question is: for some home made pedals, wich I will just use at home, high grade components make a difference in sound?
Maybe. Maybe not. Probably not. There may be some cases where a super component makes a different. There are certainly places where you could use, for example, a 1K, 10K, or 100K resistor without changing anything at all. It is generally a waste of money to use highly-touted "super"components, with a few very special exceptions. Those are mostly pre-tested germanium transistors. Super capacitors and solid-platinum resistors are primarily a way to separate you from your money.
QuoteSince the values in guitar effects are small, those variations will affect tone?
Maybe. Capacitor tolerances do affect tone when the capacitors are used for frequency filtering. They almost do not matter at all for power supply filters as long as they are big enough. Same for resistors. Where a resistor sets a critical gain, or a critical frequency response, yes, tolerance matters. In other places, you just need some resistor, and the value is almost indifferent.

Unfortunately, the real answer is that you have to know what the circuit does and how it does it - really, really understanding the operation of the circuit - before you can say that something matters or not. For instance, would a car engine be better, more powerful, or last longer if it's made out of platinum instead of aluminum or cast iron? The only way to know is to know that temperatures, pressures, and stresses are inside the engine, and also what temperatures, stresses and thermal characteristics are for the metals. Otherwise, you can't know ahead of time, and you can waste a LOT of money trying to find out.

QuoteTried to find some tests/info on Google, but couldn't find nothing. With so, all components are better, or just some? I'm not building to try to save money or something like that, so if high quality components really have some audible differences, I will first build with cheap local found pieces, and them biuld a better version, with the expensive stuff. If it just sounds virtually the same, cheaper parts = more projects 
Very few people really understand the circuits' operation, so there is little information out there, as you found out. You have the right plan - try with parts that are easy and cheap to get, then experiment a little with fancy stuff if you want to. You may find the cheap, easy stuff is just fine for music pedals.

Quote
2 - I will build a Ruby amp and other similar small amps first, and will buy some perfboard. It can look fairly nice (as some examples here show - For personal projects, can i just leave it like that, or I would get some performance issues, and should make a pcb after the initial testing phase? Again, not a problem to rebuild it, if the end result can be better.
For personal projects, leave it there. There are some special cases where PCB is really needed for performance, but mostly pedals are not that way.  The real, super, incredible advantage of PCBs is that if you use similar components, every one of many comes out the same. Wiring and other variations do not happen, so they are consistent. IF you're building only one, and don't want to buy a pre-existing pcb, use perf.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

mvotre

wow man, coudn't ask for a better answer. A huge thanks!  :icon_biggrin:

saturday it's shopping time. But first I will read lots about those small circuits (I really want to know - at least a little) what the heck I'm doing, instead of just following the schematics. Off course, this ir for the long run. At first, need some builds ready.

amptramp

For a while, there were three temperature ranges for integrated circuits, commercial (0 to +70 °C), industrial (-25 to +85 °C) and military (-55 to +125 °C).These devices all came off the same production lines - the only difference was the testing.  When MIL-STD-883 came out governing military integrated circuit processing, the military process flow was quite a bit more complicated than commercial.  The requirements for computer CPU's and memories eventually forced manufacturers to step up their processing in order to get adequate yields for integrated circuits with large numbers of internal components.   Eventually, it got to the point with some manufacturers that the only difference between commercial and military was the final burn-in of 160 hours and in some cases traceability data.  It was a bit more complicated than the MIL-PRF-19500 requirements for discrete semiconductors.  Note that commercial devices will still survive in the military temperature range - they just don't have established reliability or specifications beyond the temperature range they are tested at.

There are some applications for military resistors and capacitors, but these usually have commercial equivalents that are just as good for our use.  As far as differences in sound are concerned, the military equivalents of any given technology such as metal-film resistors or mylar film capacitors offer no advantage over the commercial versions.

The MIL-SPEC era was a golden era for electronics design, but the 1992 Perry initiative originated by Admiral Perry that eventually led to the military abandoning most MIL specs and standards and ended the high costs and dubious benefits of the enormous collection of specs and standards.

PRR

> what about those so called military grade/premium stuff

Back in 1927, a home radio had about 8 resistors, had to sell CHEAP, the resistors were the worst junk which would mostly live-out a 30-day warranty. This was most of the resistor market. Yes, sometimes those "30-day" resistors lasted 80 years, but you could never be sure.

Military radios sometimes used such junk, but since military systems are larger, have more parts, more parts to fail, have to work in all weather, and a failed radio could allow deaths (radio was key in one early WWII battle around a hill),  the military learned to demand much better quality parts.

This rose to a Big Problem with mainframe computers. When you have millions of parts, ALL of which HAVE to work to get the correct answers every time, you can't be fooling with unreliable parts. For some years there were Computer Grade parts.

Fast-forward to the 1980s. A $100 cassette deck or VCR, sold by the millions, might have hundreds of resistors. If they are junk, a lot of new decks won't work when bought, which gives a maker a bad reputation. Also the large quantities of parts being made justifies a lot more effort getting clean materials and reliable processes.

As R.G. says, eventually the quality of "ordinary consumer parts" got to be as good as or even better than "military" parts made the old ways. The military has to extra-test, because that is what they have always done, they have the test-setups, and because they DO have to work in cold and heat and thin-air that you never want to meet.

The "military" stuff won't for-sure sound better. (Might possibly sound worse if they compromise dielectric stability for deep-freeze survival, but even that is unlikely.) 

Use parts suitable for what you are building. Ordinary everyday shirt-sleeve environments call for parts like you find in old VCRs (or new BluRay players, except SMT parts are awkward for DIY). If you are building a Moon-Rocket that you will ride in, where if a controller failes you will die, you will want parts made for extreme conditions and do a lot of testing.

Buy from reputable vendors. The ordinary parts today are VERY good, and in stompbox work will do the right thing for your whole lifetime.

Avoid mystery vendors (eBay). Some are good, and some are sweeping-up factory-rejects and selling them to fools.
  • SUPPORTER

Thecomedian

#5
Perfboard can be as professional as you want it to be. There's nothing wrong with it. Good soldering practice and good layout can make a perfboard look as good as a PCB and it's cheaper, for limited numbers of the product. Making PCB costs more for limited numbers, even if DIY. It's really best for things that are mass produced by yourself or a larger entity. The alternative is if you design a PCB with some modularity involved in the setup so that different products can fit inside it. You may often see PCB with what appear to be missing spots for extra components for different versions of the same thing. You could also work out different effects and try to make the parts layout as nearly identical as possible, with some additions or subtractions here and there, and then get PCBs printed that have extra traces and holes where necessary.

Performance wise, there's nothing to worry about as long as you're not using inductors or high frequency parts, and then that's just a layout issue. There's nothing really different about PCB compared to perfboard electrically speaking.
If I can solve the problem for someone else, I've learned valuable skill and information that pays me back for helping someone else.