[wrecked_porsche]

Hey guys, quick question about the im716.

Why do you guys always recommend that the metal film resistors be used with the iM716? Why is it not good to use the standard carbon ones?


Standard carbon film:

Metal film:

If its just a case of better tolerance on the metal film, then lets say an exact matching pair of carbon resistors are obtained, then it should not make any difference?

Reason I ask is because I have a pair of of exactly matching 48ohm metal film resistors (Tokyo Koon Denpa (TKD) High Precision Metal Film Resistors) which I've soldered to my im716. While they do sound great, I have itchy fingers because I have a fistful of 30~40++ ohm carbon ones that I wanna mess around with...

Is it ok to use standard carbon ones? Will the sound be deteriorated in any way?

If it does affect the sound, then I'll leave my master piece alone then...

Thanks in advance for the input guys.

Cheers!

 

[TheMarchingMule]

Maybe this would be better suited for the DIY forum, but maybe you'll get the answer here anyway.

 

[Uncle Erik]

Eh, I don't like carbon comp resistors. They drift a lot depending on temperature and age. I poke at and restore old radios, as well, and I usually like to put what I remove on the DMM to see where it's at. Old electrolytic caps are usually toast, but some are OK. If the paper/wax caps haven't opened up from heat (common) they'll usually hit within 20% of spec. Mica caps and wirewound resistors are awesome - they rarely need replacing.

Carbon comp resistors? Never, ever, ever, ever have I found one within spec. We're talking about maybe a thousand I've removed over the past ten years. Carbon comp resistors are always several thousand percent out of spec. Some radio guys only recap sets. I think they're insane. I replace every last carbon comp in anything I work on.

If you want a counterpoint, a close friend of mine is a guitar nut and plays in at least one band all the time. He loves tube amps, and he loves old carbon comp resistors. Why? Because they get hot and distort all over the place. He loves that sound.

 

[wrecked_porsche]

Quote:

Originally Posted by TheMarchingMule /img/forum/go_quote.gif Maybe this would be better suited for the DIY forum, but maybe you'll get the answer here anyway.

Yeah, I kinda though about that.... but I already hit the "post" button...

Hopefully I'll get some good answers here...

Quote:

Originally Posted by Uncle Erik /img/forum/go_quote.gif If you want a counterpoint, a close friend of mine is a guitar nut and plays in at least one band all the time. He loves tube amps, and he loves old carbon comp resistors. Why? Because they get hot and distort all over the place. He loves that sound.

Oh noes... that's it then. I'm not gonna touch the carbons then if that's the case.

 

[craiglester]

I must admit, that it seems like a given that lower tolerance resistors should perform better, but I've yet to find a resistor that's outside of it's spec by 1000% , when used within normal operating temperatures, voltage and current guidelines.
Now it's possible that these degenerate over time, but by such a large margin? Unlikely I'd imagine.
I think maybe you're getting carbon composition, and carbon film types mixed up. Carbon film are a lot closer to spec than carbon composition types, especially old ones.

 

[ericj]

Quote:

Originally Posted by Uncle Erik /img/forum/go_quote.gif Eh, I don't like carbon comp resistors.

Yeah but carbon comp resistors are little tubes full of compacted carbon dust. These experience weird things, especially at high voltages where tiny little arcs of electricity zap between particles of carbon.

The question was about carbon films. Whole different ball game.

Thermal stability still isn't as good as metal films, but thermal stability isn't an issue with the im716. They'll never ever get hot in that application.

Carbon film resistors are usually only sorted into 10% tolerances. A 10% difference is too big when you want two channels to be matched. However, if you have a dmm, you can take a bag of 10 resistors and find two that are matched closer than 1%. It's just more work.

You can get 1% carbon film resistors, but they don't cost substantially less than 1% metal films. Not at DIY quantities anyway.

There is more contact noise and schott noise in carbon films than in metal films, but these noise sources are directly related to the amount of current pulled through the resistor.

In this case, it's not going to be very much. In guitar amps, they intentionally overload the current ratings of their resistors to enhance the distortion. We're not going anywhere near those power levels.

So, the final question, why do we use metal film resistors in low-current parts of DIY audio circuits?

Because they're A Little Tiny Bit Better, and when you add it all up, the difference between buying metal films vs. carbon films adds up to far less than what your time is worth.

If you're ordering a bunch of stuff from Mouser or Digikey you can get KOA-Speer or Yageo resistors for less than 10 cents each, sometimes as little as 6 cents qty1. You can get 10 of them for 3 cents each sometimes. So there's absolutely no financial reason, no more than a dollar or so, to try and go cheaper than that.

It's unfortunate that in a lot of locations, there's few places locally to buy metal film resistors in specific values.

 

[ericj]

Quote:

Originally Posted by craiglester /img/forum/go_quote.gif I must admit, that it seems like a given that lower tolerance resistors should perform better

Not really. That 1% tolerance only means that the machine at the factory sorted them to within 1% of their nominal value.

The other rating - the ppm rating - that's their thermal stability. But thermal statbility is a moot point in circuits where the resistors run cool.

 

[craiglester]

Exactly, I just matched my resistors to as close as my Multimeter would read, and then popped them in. I very much doubt the current going through them will be a factor, nor should heat (and they should be at roughly the same temperatures seeing as they're right next to each other)

 

[wrecked_porsche]

Quote:

Originally Posted by ericj /img/forum/go_quote.gif Yeah but carbon comp resistors are little tubes full of compacted carbon dust. These experience weird things, especially at high voltages where tiny little arcs of electricity zap between particles of carbon. The question was about carbon films. Whole different ball game. Thermal stability still isn't as good as metal films, but thermal stability isn't an issue with the im716. They'll never ever get hot in that application. Carbon film resistors are usually only sorted into 10% tolerances. A 10% difference is too big when you want two channels to be matched. However, if you have a dmm, you can take a bag of 10 resistors and find two that are matched closer than 1%. It's just more work. You can get 1% carbon film resistors, but they don't cost substantially less than 1% metal films. Not at DIY quantities anyway. There is more contact noise and schott noise in carbon films than in metal films, but these noise sources are directly related to the amount of current pulled through the resistor. In this case, it's not going to be very much. In guitar amps, they intentionally overload the current ratings of their resistors to enhance the distortion. We're not going anywhere near those power levels. So, the final question, why do we use metal film resistors in low-current parts of DIY audio circuits? Because they're A Little Tiny Bit Better, and when you add it all up, the difference between buying metal films vs. carbon films adds up to far less than what your time is worth. If you're ordering a bunch of stuff from Mouser or Digikey you can get KOA-Speer or Yageo resistors for less than 10 cents each, sometimes as little as 6 cents qty1. You can get 10 of them for 3 cents each sometimes. So there's absolutely no financial reason, no more than a dollar or so, to try and go cheaper than that. It's unfortunate that in a lot of locations, there's few places locally to buy metal film resistors in specific values.

Wow, ericj, thank you sooo much. Since I'm studying engineering, I understood every word you said there and it completely makes sense to me. That's the kind of info I was looking for. Thanks.

Reason I was asking is because I have easy access to tons of 5% tolerance carbon film resistors in my Uni engineering lab. So, getting a matching pair ain't gonna be a problem. I already went into town during the weekend and got my 48ohm exact matching pair of 1% TKD metal film resistors and soldered it.

I know, I know, I should have bought other ohm values too, but I figured 48 would be fine. Now I find myself curious about how the other resistor values would sound, so I thought I could mess around with resistors from my eng lab. Oh well, I guess its ok to use them temporarily to mess around then.

The answer to my original question is obvious then:
Head-fiers will do everything they can to squeeze even 0.0000001% more performance out of their audio gear if they can; there is no such thing as overkill in the audiophile world!

Woooooooooo! I love being a head-fier!

A big thanks to everyone for their input.