Magnetic resistor end caps - do they really matter at audio frequencies?

lowredmoon

lowredmoon

Active Member
I'm building a moving coil head amp.

These days, it's hard to find non-magnetic 1/4 watt resistors. Vishay/Dale RN and CMF used to be non-magnetic, but they now use magnetic end caps, except in the CMF line that ends in "143," which no one seems to stock (Mouser has maybe 4 values). Takman REY are non-magnetic, nut their noise figures are 3x what Vishay publishes (0.3uV/V vs. 0.1 uV/V), and with a moving coil amp, small amounts of noise matter A LOT.
Vishay bulk foil resistors are perfect for this application, but they cost $15 each, so I'm not going to use them.

Any insight on how magnetic resistors effect a circuit and whether it really has an effect at audio frequencies?

Thank you for your time and input.

Adam
 
Register to hide this ad
I don't have a direct answer to your question but I do have some information for you.

My day job is making emissions measurements (and other Electromagnetic Compatibility tests) for electronic equipment for airplanes. I don't remember the levels exactly, I can check tomorrow at work, but one of the MIL-Std 461 tests we do is RE101, Radiated Emissions, Magnetic Field, 30 Hz to 100 kHz so definitely in the audio freq band.

There are some Conducted Emissions tests, CE101 and CE102 that both start at 30 Hz.
 
SteveJewels said:
I don't have a direct answer to your question but I do have some information for you.

My day job is making emissions measurements (and other Electromagnetic Compatibility tests) for electronic equipment for airplanes. I don't remember the levels exactly, I can check tomorrow at work, but one of the MIL-Std 461 tests we do is RE101, Radiated Emissions, Magnetic Field, 30 Hz to 100 kHz so definitely in the audio freq band.

There are some Conducted Emissions tests, CE101 and CE102 that both start at 30 Hz.
Click to expand...
Thank you for the value added reply.

I'm not sure what alloy they use for the end caps, but it has been reported that many values are attracted to a magnet. I haven't had luck in locating what the material is, or how much of it is present in the resistor.

Kind regards,

Adam
 
I checked both the Bob Cordell and Douglas Self audio design books that discuss component distortion mechanisms and neither talk about resistor cap magnetic distortion. So I’m guessing it is not a very dominant issue. I checked some of my Digikey 1/4 W metal films and the end caps are definitely attracted to a magnet but the leads themselves are not. I would be much more concerned about the leads since the current flows through them for more distance.
I checked some Vishal 1/2 Watt (CCF601) and they are not attracted to a magnet.
I checked some other components (small signal transistors in TO-92 and small capacitors) and found several with leads that were attracted to a magnet.
Interested if anyone knows more about how much this matters in component selection.
 
Given that just about every metal film resistor has magnetic end caps, it must be of little issue...at audio frequencies or beyond. Otherwise, there would be many non magnetic resistors to choose from.

I mean, it's not a high-technology process to use non-magnetic metals in a component.
I'm guessing that at some point, a company used the inclusion of non magnetic end caps as a marketing point, and it took off.

As of now, I know of only two companies that market non magnetic resistors as "audio grade" -- Vishay (CMF audio grade) and Takman. CMF non magnetic are difficult to find (Mouser and Digikey have only a few values) and Takman REY published noise figures are 3x RN or CMF.

Vishay PTF look promising (and pretty), but no empirical noise figures are published. I expect them to be about what the CMF are, but with much lower TCR and much better tolerance. Also, 2x or more the price.

Adam
 
I've got a pair of F5's built with Vishay/Dale and Panasonic thin films and some big Ohmite non-inductive wire wound resistors, and that thing goes WAY past 100KHz. Like 10 times higher. Usually the characteristics that are problematic are inductance and capacitance with capacitance being the dominant issue with 1/4 watt resistors. The capacitance is usually directly related to the size of the resistor and the Vishay/Dale I used are about half the size of a grain of white rice. Naked resistors have even less capacitance which is why some people are fanatical about them. I'm not familiar with magnetic end caps ever being an issue.
 
>100Khz response is admirable, but not really needed to get excellent audio, limited as our hearing is in comparison. I put together a 3-triode per side line stage modeled after an oscilloscope preamp, all direct coupled except for the output cap to the power amp, and I deliberately limited the reach of the upper corner. High bandwidth came with the design, want it or not. I did not want my power amp full of RF that the preamp could have passed through to it. In short, there is no real need for excessive bandwidth in an audio amp. I set my 3 dB down corner at 40kHz, down from the 1.5MHz I would have had otherwise.
 
egellings said:
>100Khz response is admirable, but not really needed to get excellent audio, limited as our hearing is in comparison. I put together a 3-triode per side line stage modeled after an oscilloscope preamp, all direct coupled except for the output cap to the power amp, and I deliberately limited the reach of the upper corner. High bandwidth came with the design, want it or not. I did not want my power amp full of RF that the preamp could have passed through to it. In short, there is no real need for excessive bandwidth in an audio amp. I set my 3 dB down corner at 40kHz, down from the 1.5MHz I would have had otherwise.
Click to expand...

It's a fairly easy problem to solve with wires.
 
Since you are going to be amplifying very low level signals why take a chance ? The Vishay (CMF audio grade) and Takman are great choices. You could also consider PRP Resistors.
 
I say build it with what's available and you're likely to love it. The CMF60 line from Dale has been quiet enough in my tbe phonostage that I'd say they are safe to use.
A head amp is going to find out real quick.
I dont know of any method of accurately predicting the noise contribution of a resistor. Theres too many variables for a model to be useful. Build it, and find out​
 
I tend to prefer measurements. If we're just looking at noise levels, thats relatively easy to quantify. if resistor type A has a higher noise level than resistor type B thats a hard data point. The rest gets into subjective stuff that doesn't actually have a real answer.
 
gadget73 said:
I tend to prefer measurements. If we're just looking at noise levels, thats relatively easy to quantify. if resistor type A has a higher noise level than resistor type B thats a hard data point. The rest gets into subjective stuff that doesn't actually have a real answer.
Click to expand...
That's my goal -- use as much objective data as possible to make decisions. I'm now at a point where I have a foothold in understanding parts data and material properties and how it is relevant in a circuit, and want to use what I've learned to build. It's a start, at least, and I'll learn more as I build more.
 
Noise caused by a noisy resistor may depend upon where in a vacuum tube circuit the resistor is employed. As an example, such a noisy resistor is more likely to audible when used as a plate resistor rather than when used as a cathode resistor in a cathode follower circuit.
 
You must log in or register to reply here.
Back
Top Bottom