#124 NFB Resistor

[steve_p]

Hi All,
I am getting ready to get started on a 100W #124 build.
I am gathering parts and have had trouble figuring out what to use for a 4.7k NFB resistor.
I have seen mention of a 2W or 3W resistor used in the original. however I am having trouble finding a carbon film resistor that meets these specs.

Does this resistor need to be a higher wattage, or will 1W suffice?
If 1W is not enough, what are people using in this location, metal film or metal oxide?

Thanks in advance,
Steve

[martin manning]

I don't know why that resistor would have to have such a high power rating. Even with a 16-ohm speaker load and 100W average output power it will only have to dissipate about 0.29W; a 1/2-Watt would be big enough.

[Structo]

I had some trouble finding that as well locally and ended up using a 1 or 2 watt 4K7 metal oxide, not really ideal but it works.

I like to use 1 watt film resistors because of their larger size and voltage handling.

You can also play with that value with regards to the NFB loop.

But Mouser or similar places have them in carbon and metal film.

KOA Speer or Vishay/Dale are nice resistors.

http://www.mouser.com/Passive-Component ... _/N-7gz3w/

[odourboy]

I've always used 1 W resistors in that location myself and it's probably overkill.

[steve_p]

Thanks guys, I will probably start with a 4.7k 1 watt carbon film.


Steve

[thorens]

Hi Steve!
1 Watt is enough in a electrically manner. But Mr. Dumble is aware of choosing his materials as a sound designer. The 2 watts Carbon Comps are famous for their overtones on big voltage jump positions(and their noise and heat drift too). Nowadays 2 Watts CC are rare and expensive. However the neg feedback resistor has a big tonal influence. Please try out and let us know about your experiences.
Tom

[ChrisM]

Hi Steve!
1 Watt is enough in a electrically manner. But Mr. Dumble is aware of choosing his materials as a sound designer. The 2 watts Carbon Comps are famous for their overtones on big voltage jump positions(and their noise and heat drift too). Nowadays 2 Watts CC are rare and expensive. However the neg feedback resistor has a big tonal influence. Please try out and let us know about your experiences.
Tom

[thorens]

Don't want sombody confusing. Sorry, english is not my home language. 1 Watt is enough and stable at this position. Mr. Dumble often choose the bigger 2 Watts because they sound warmer. Resistors: Carbon Compress resistors sound different than carbon film and metal oxide. From warmer to more compact, but also from more hiss to few hiss. 2 Watts ones sound different to 1/4 Watters (the ability to dissipate heat). The effect of the warmer sound the carbon compress have is the biggest at positions where big voltage differences are (at both ends of the resistor). I mean especially all anode resistors. Ok we are talking about the NFB resistor. You add at this position an amplified signal to the original. So the influence of the used material is notable.
Tom

[martin manning]

My understanding is that for resistors to exhibit this behavior there must be around 100V across them. The NFB resistor does not get near that. See: http://www.geofex.com/article_folders/c ... oncomp.htm

[crbowman]

My understanding is that for resistors to exhibit this behavior there must be around 100V across them. The NFB resistor does not get near that. See: http://www.geofex.com/article_folders/c ... oncomp.htm

Hadn't seen that before Martin. Pretty interesting stuff! I have to admit, the only noticeable difference I've heard with carbon comps was noise level, but I never knew exactly where to use them for max benefit.

[talbany]

What about signal path?
IMO..I don't think Dumble was concerned with the power rating of the NFB resistor (although there is a rather large amount of current there) more so than frequency response...
Ideally, resistors should act as pure resistors, without any of the characteristics of other types of components and when they are used in DC circuits they do IMO. In AC circuits (like NFB resistor) however some resistors may have characteristics that make them unsuitable for a particular purpose. At high frequencies, some resistors also have characteristics of capacitance and/or inductance.(Depending on power rating value and type) Because of this they will have have a property called reactance, similar to resistance but dependant on the frequency of AC signals passing through the component. The frequency response of a resistor tells us at what frequencies the resistor still acts as a pure resistor... For this reason this parameter is of interest to people working with high frequency AC circuits like NFB which has a direct relationship on harmonic content..Anyone else have a opinion on this?

Thanks!!

Tony

[odourboy]

My understanding is that for resistors to exhibit this behavior there must be around 100V across them. The NFB resistor does not get near that. See: http://www.geofex.com/article_folders/c ... oncomp.htm

Interesting article! I'm a believer in engineering over mojo. If someone had the inclination and the equipment, it would be pretty easy to measure the 2nd harmonic distortion generated by a CC versus another resistor technology in the feedback loop. However, sounds like it's not worth the trouble. Even in plate resistor position, a 0.008%/V resistor distortion (as quoted in the article for modern CC resistors) would be hard to hear. Although, the cumulative effect through all the stages might be significant - IF you could hear it over the hiss!

[vibratoking]

There is an effective parallel capacitance and effective series inductance associated with resistors. For CC, CF, MF types used in most guitar amps, the effective reactances are neglible IMO. This is a can of worms open for a lot of debate such as "you can't measure it, but I can hear it". Here is a link to some data that shows the effect of these reactances occurs in the MHz and above regions. I doubt that this would be important with regard to the NFB loop in a guitar amp.

http://ieeeb.okstate.edu/lecturenotes/E ... ematic.pdf

[talbany]

This is an interesting topic for sure!!
Most of the articles out there (like the one Martin posted) talk about the properties of resistors when DC voltage 100v is applied..What about in the signal path..Grid resistors in pre amp or 100k/150k/ Drive, level feeders or OD entrance where no DC exists..I can definitely hear the difference between say a MF and CC in these spots..

Tony

[thorens]

My understanding is that for resistors to exhibit this behavior there must be around 100V across them. The NFB resistor does not get near that. See: http://www.geofex.com/article_folders/c ... oncomp.htm

The geofex article is talking about a sinus wave 100V AC over a 50V DC resulting in a resistor distorted signal. Please think of position where the grid stopper are. This is another example for a great influence of cc's at bias voltage (DC).
very interesting discussion
Tom