Metal Film Resistors REPORT

[MarshallPlexi]

In an attempt to get less noise from this amp (Express) I used some Vishay/Dale 1/2W Metal Film resistors for all of the plate resistors in the amp, and they were horrible. They were quiet to me, but I can't compare directly to the Carbon Films that I used before since it took a while to swap them out.

Anyway after buttoning the amp back up I was playing it and I was thinking, "Man! It sounds like an output tube is dying!" There was this square wave sound on the high end that was horrible. The notes seemed to be dying way too quick and the sustain was just buzzy.

All I had done earlier was to swap to Metal Film resistors so I put the Carbon Films back in and INSTANTLY the tone was back. The notes had a natural bloom to them and sustain was smooth and without edge.

Lesson learned: Stay away from Metal Film resistors when good tone is the priority.

Having said this, I'm going to try some 1W carbon comps and see what that does.

[MarshallPlexi]

One more thing I forgot to mention. My Dad, who was in the driveway while I was testing this amp came in and said, "What's wrong with your amp? It sounds like a square wave distortion."

I say this to say that the sound of the metal film resistors is NOT subtle.

[Robert]

I recently swapped my carbon film plate resistors and slope resistor for carbon composition resistors. I can't really say they altered the tone noticeably, however I was surprised they did not increase the background noise as I expected. They were all higher , 104-108K range, rather than 100K as needed. That may have altered the tone in itself. For now, I'm satisfied to leave them in place since they did no harm to the tone.

[HiGain]

Thanks for the report!

I'm starting an express build, and am using Riken Ohm resistors for the plates (supposed to have the warmth of CC, without the noise).

However, on my current (modded) two rock EL84 express clone, I have metal film resistors for the 22k/82k split load arrangement on the second stage. I have to say, the distoriton is smooth, with good sustain.

I therefore wonder if metal films have different effects, depending on their placement in the circuit. For instance, maybe it's best to use them in earlier stages that are not overdriven, where they will produce (theoretically) less noise to be amplified downstream, without degrading overdrive characteristics.

Just a thought.

Jake

[Funkalicousgroove]

Are you sure you had correct values? 1% Resistors have a 4th band that's also a value place holder not a multiplier. I've used metal film's in alot of amps without problem, but I can say that there is a noticable musicality when using Carbon Film or Carbon Comp, Carbon Comp especially since they actually PRODUCE 2nd order harmonics.

[MarshallPlexi]

All I did was change the 100k plate resistors and nothing else. I'm very sure.

[dms]

I have a JTM 45 clone that I built with Vishay Dale metal films; I have also been trying to figure out why the tone has a weird fizz/sizzle/crackle to it after the initial attack. It's subtle, but bad, and it drives me nuts. Maybe this is the square-wave-ish distortion that MarshallPlexi mentioned in the post? Maybe this has been the problem I was having all along. Man! Now I'm really curious... could this have been my build's problem all along?

Where the hell's the soldering iron...

[Funkalicousgroove]

What does it look like on your scope?

[wiggle]

Metal film = BOo

Carbon Film = eeeh well naah

Carbon Comp = Yeah

[Funkalicousgroove]

Is that technical Jargon?

[dms]

Hi Funkaliciousgroove, all,

I'm eager to get going, have a scope, function generator, powerbrake, but no significant audio signal tracing experience... here are the newbie questions:

I assume I begin with a sinewave around 1kHz, right?

1. What should the input signal level be to not overdrive/overload anything... another post mentioned a guitar pu is around 0.1Vpp... should I use that?
2. What should the gain controls be set at (barely on or full blast, etc.)?
3. Where are the best points along the signal path? I used the metal film resistors for all but the B+ supplies (i.e. grid stoppers too).
4. Finally... what is the approximate visual characteristic of "bad distortion" (i.e. visibly flattened sinewave, jaggedness,imposed additional waveforms, etc..... as opposed to just gradual sqashing)?

Thanks for your help; hope the topic is of interest to more than just me!

[gnugear]

I found that the Vishay metal film resistors have a darker, richer tone in my Marshalls.

[EL34]

I have used metal films for years with no problems.

[Moose]

Hi Funkaliciousgroove, all,

4. Finally... what is the approximate visual characteristic of "bad distortion" (i.e. visibly flattened sinewave, jaggedness,imposed additional waveforms, etc..... as opposed to just gradual sqashing)?

Good lord, that's like asking what the description of a bad haircut is. Make a sound you hate and look at the scope -- that's bad distortion!

Seriously, there are some ways to see distortion that you may know is bad. Like jagged edges on one slope of the wave or radically different distorted waveforms on the output of the PI, but a lot of it is experience. And, more to the point, a lot of it is HEARING things and then scoping to see what you hear might look like. You just can't separate the two because, technically, most engineers would look at any distortion as bad.

Poke around here:

http://geofex.com/effxfaq/distn101.htm

and here:

http://members.aol.com/sbench102/thd.html

As for how what and where -- scope everywhere compared to everywhere. For example, scope the outputs of the PI to see how symetrical they are and which breaks up first. Scope before the tonestack and after the recovery stage. Scope the input and the poweramp. Etc.

Also, for frequency, remember that a lot of tuning is tuning frequency response. Try a variety of frequencies. Ideally, you should have a sweep function generator and sweep from a very low value to very high, with the scope trace in sync with the sweep This give you a picture of the total frequency response of the amp right there on the scope. With a non-sweep generator, you can vary the frequency and wait for the wavform to shrink to guess where your rolloff frequencies are, and adjust response accordingly.

If you have a DSO instead of an analog scope, you might also be able to perform a fourier transfer on the wavform, and that'll clearly show you the intensity of any given harmonic. This is a killer way to see how harmonics affect the wavform itself.

[dms]

Moose - thanks for the chuckle and the tips. I spent the weekend creating a better workspace and lining up my test equipment, so that I can work slowly and carefully, and of course, start looking for the source of the bad hair.

dms