What about 0805 smd resistors?

[2N1305]

Originally Posted by JAY X

Hi Geeks!

I'm planning to get into smd and start using 1% thin film smd resistors for small signal circuits. 1206 and/or 0805 resistors. The 0805 is quite easy to layout in a pcb, but i have read that at high values 10k, 22k..100k they behave non linear...¿is that true? ¿is this really a problem?.

I've never heard of this and I've been using SMD components for over 10 years. Maybe their tolerance becomes worse than at lower ohmic values, I don't know. You'd have to check the datasheet of a particular type of resistor to see if the linearity is what you want.[/B][/B]The circuits i usually build (Differential amplifiers, line outputs, etc..)use dual supply +/- 18v or +/-15v.

Originally Posted by JAY X

I also read that to improve resistor voltage handling 2or 3 resistors in series are placed...but i think this is only useful for smaller footprints like 0602 or 0402.

I would like to have your opinions on that matter,
Thank you very much,

JAY X

It can be useful for whichever type of resistor, be it surface mount, axial, power resistor.. It just depends on the voltage you're working at. Again, verify the datasheet.
It is good that you are concerned about these kinds of things, and the datasheet should tell you what you need to know.

cheers,
2N

[JohnRoberts]

For an important audio path you probably need to avoid small SMD resistors.. stick to larger footprint (say 1206 ) and high quality parts.

JR

[mellow]

Originally Posted by JohnRoberts

For an important audio path you probably need to avoid small SMD resistors.. stick to larger footprint (say 1206 ) and high quality parts.

JR

Are you implying that the small SMD resistors have distortion problems? If so, is there a scientific reason that you know of?

[johnnyc]

In general higher wattage (bigger size) equals less noise.

[dcollins]

Originally Posted by mellow

Are you implying that the small SMD resistors have distortion problems? If so, is there a scientific reason that you know of?

The value changes with the applied voltage. It's primarily a problem with thick-film devices, and if you go with MELF construction (or Susumu) it's fine.

The first time you see a resistor that has more distortion than the whole entire circuit, it's a surprise.........

http://www.davehilldesigns.com/smt_r...rtion_rev1.pdf


DC

[mellow]

Originally Posted by dcollins

The value changes with the applied voltage. It's primarily a problem with thick-film devices, and if you go with MELF construction (or Susumu) it's fine.

The first time you see a resistor that has more distortion than the whole entire circuit, it's a surprise.........

http://www.davehilldesigns.com/smt_r...rtion_rev1.pdf


DC

Hmm, interesting. Thanks for that.

[Ike Zimbel]

Originally Posted by dcollins

The value changes with the applied voltage. It's primarily a problem with thick-film devices, and if you go with MELF construction (or Susumu) it's fine.

The first time you see a resistor that has more distortion than the whole entire circuit, it's a surprise.........

http://www.davehilldesigns.com/smt_r...rtion_rev1.pdf


DC

Yes, that is interesting. Some of what is going on is explained here:
Analog Devices : Rarely Asked Questions (RAQs) : Resistors in Analog Circuitry

Best,
Ike

[johnnyc]

Originally Posted by Ike Zimbel

Yes, that is interesting. Some of what is going on is explained here:
Analog Devices : Rarely Asked Questions (RAQs) : Resistors in Analog Circuitry

Best,
Ike

Nice link. Thanks for posting.

[JohnRoberts]

Originally Posted by mellow

Are you implying that the small SMD resistors have distortion problems? If so, is there a scientific reason that you know of?

I haven't looked into this personally, but a friend I trust, using SMD resistors in premium audio paths, tells me it can be measurable using a decent test bench. Perhaps related to the dimensions involved for making large resistance in such small packages. I only recall one issue of measurable distortion from a through hole resistor, not counting faulty parts.

While I am generally inclined to dismiss wild claims trying to justify expensive exotic components, I believe the reports that SMD resistors can be too small and too cheap.

The distortion is probably not very audible, but if it is measurable and relatively easy to avoid, just avoid it... .

JR

[Jim Williams]

MELF metal film resistors are about the only low cost linear resistors you will find in that size. Most of the metal oxide resistors have THD and tempco problems.

The Vishay bulk foils in the 0805 or 1206 size run about $18 each, but the 0805 size is .2 watt, much better than the metal oxides. They are .01% tolerance and are the lowest noise available.

There were some Mytek converters made with the MELF's for general audio and the Vishay bulk foils for current to voltage conversion stages.

[JAY X]

Susumu resistors are 0,1% not available in 1% ...and expensive at 0,47 Euros/unit... Vishay are also pricey..besides, Pick and place machines don't like MELF so what kind of smd resistors do use Neve, SSL, Moog, etc?

I think the only way out is to stay with thru hole, or learn to live with smd resistor behaviour...

JAY X

[JohnRoberts]

While I don't recall the brand my friend is using (he did tell me but I forgot). It is possible to get respectable path linearity, using SMD without trading your first born for each resistor. That said you can't just use the smallest and cheapest SMD parts. Look for a voltage coefficient spec.

For one-off DIY projects maybe stick to through hole for now. The serious manufacturers do their own bench testing to qualify parts. Which takes time and money and time (and a test bench).

JR

[JAY X]

Although this thread is a bit old,


RT0805FRE071KL - YAGEO (PHYCOMP) - RESISTOR, 0805, 1% 50PPM 1K | Farnell España

I found these resistors, but not in bigger size. They don't say anything about voltage coefficient, only temperature coefficient.

¿Maybe these SMD resistors are ok?

JAY X

P.S. The opening of the games was impressive!!

[JohnRoberts]

Sorry link didn't work.

Voltage coefficient i the important parameter for distortion.

You can build and test to find out... that is often what manufacturers do to answer such questions.

JR

[Jim Williams]

If you want good and small, the Dale CMF50 series 1/4 watters are very good. They are small in the 1/8 watt size. Mount them vertically and they take no more space than an 0805 resistor. They are not cheap, about a quarter each.

[dcollins]

Originally Posted by Jim Williams

If you want good and small, the Dale CMF50 series 1/4 watters are very good. They are small in the 1/8 watt size. Mount them vertically and they take no more space than an 0805 resistor. They are not cheap, about a quarter each.

While you do see this done, it's not really recommended to mount vertically especially for precision circuits.

DC

[Jim Williams]

Good to know if you are building time based circuits of very high precision. Encapsulating also helps spread the thermals around evenly. The CMF50's are rather thick, almost oval looking next to 1/4 watt designs. Since they are also so short, there's little surface area for thermals to build up on.

Not a problem for audio though.

[Dan Kennedy]

I've run across the vertical mounting issue in thermocouple inputs, but it was one of the lesser sources of error in the whole thing. Board gradients, light leakage into protection diodes, excess junctions all bit me.

Like Jim says, not much of an audio issue with RN50's.