Mark, I don't have any "tone loss" with my plain ol' pot pedal! Anyone who thinks that they do should read Carl Dixon's post in the following thread...
http://steelguitarforum.com/Forum11/HTML/001627.html

If we use good quality cable (Belden 8410) it has a capacitance of about 33 pF/foot. A 15 foot cable has a total capacitance of about 500 pF (5x10^-10). Most of these foot pedals use a 500 kΩ pot (for reasons stated below). At say the 50% position, because these are log taper pots, they add a series resistance of about 400 kΩ.

This sets a rolloff frequency (the frequency at which the output is down by one-half) at 1/(2πRC), or 800 Hz! Your typical "presence" control on a guitar amp is set at about 3 kHz and up. I am afraid that I find Dixon's posting a bit naive. He is probably right that the pot is not the issue. It is the cable.

BTW: this is one of the reasons that commercial micropohones are all low impedance devices. Guitar pickups, however, have higher impedances for historical reasons. The volume pedals must use high resistances to minimize the loading on the pickups. If they did not, the relatively high inductance of the pickups would cause even more sever high frequency attenuation.

If you don't believe all of this (and still belive that this is a result of the Fletcher-Munson curves (see my post below) for auditory sensitivity) try this:

1. Set your amp volume to its maximum and use your volume pedal to play at a low to moderate loudness.
   
2. Now, bring the volume pedal to its maximum and back off on the amp volume. Tell me if it sounds the same when the volume levels are about equal.
   

I've never taken an o-scope and traced the signal (or use a signal generator) to see what happens to the signal at various volume levels. Actually a freq run over the entire spectrum, at various volume levels, would need to be done to determine the differences at the various frequencies.

But, it's not a concern to me. If I had a concern, I'd go with a Hilton infrared pedal.

I also found that when I set the output of a powered pedal (I recently bought one of the more popular ones) so it matched the output of the guitar pickup (zero gain) that the sound is exactly the same as a passive (pot) pedal.

Amplifiers (whether in a volume pedal, or in the amp) can only replace what's lost by brute force. For example, say you're losing treble (for any reason). More amplification will boost the overall signal, and then all you have to do is back down bass and mids to make the signal sound the same. But, when some people hear phrases like "restores a signal", or "eliminates tone losses", they fail to realize that all an amplifier really does is amplify. No magic...no voodoo...no "restoration".

Also, your observation on the "amp volume low-pedal volume high" and "amp volume high-pedal volume low" dilemma is due to both hearing deficiencies, and the amplifier characteristics! Ever notice...no amp will put out as much bass as low volumes as it will at high volumes? The more signal a preamp has, (up to the point of clipping) the more effective the tone control circuits will be, since they are (most always) passive devices. Stated more simply, the more an amp (a preamp, or a power amp) has to "work with", the more it can do. (This is why I run my amp volumes "wide-open".)When you reverse the volume controls in the manner you describe, only the output stage "sees" the same amplitude. The preamp (having less signal to work with) just doesn't perform as well as it should.

Mark is definitely correct about the effect of putting 400Kohms in series with the audio line. Regular guitars have employed treble bypass capacitors for decades to offset this problem, but it never seems to match up with the system one is using at the time.

The Goodrich type preamp solves only one component of the problem...the pickup impedance. When a pickup is loaded with a volume pot, a cable capacitance, and an amplifier impedance, high frequency loss will be noticeable. This is probably why Jack Stoner seems to be OK with his setup because his pickups have such a low impedance. Many people have a preamp to drive the volume pot. This improves the response across volume control position. But it is not the whole picture.

Carl brings up a good point about the Fletcher-Munson curves. This is a fact of human hearing response, but in no way is the only contribution to high frequency loss in passive volume controls.

Every component is working against us on the high frequency loss issue.

1. Pickup impedance
2. Volume control resistance insertion
3. Cable capacitance (mostly when considered with high volume pot insertion resistance)
4. Amplifier impedance (mostly when considered with high volume pot insertion resistance)

If you are playing into a good high impedance amp (low capacitance) with low impedance pickups you may never see a problem. You may only be affected by the Fletcher-Munson criteria. I think Jack and Carl have very professional equipment and are probably only experiencing mostly Fletcher-Munson type effects. The rest of us out here with cheap stuff are the ones with problems.

For many other applications, any other variations will call out for the need of a preamp or an active volume pedal. Almost any instrument that has magnetic pickups with volume pots will exhibit this problem. The steel guitar is excerbated by the fact that we rarely run full volume (the sustain enhancer). Most other instruments play at 100% volume on the instrument.

Mark, I designed a remote volume control circa 1985 with a similar concept. I used a D-to-A converter with the signal going into the reference and the D-to-A output as the output. The pot controlled an A-to-D converter that addressed the log D-to-A. It had clicks, so I used two D-to-As out of phase and reconstructed the signal using a differential amplifier. Your modern day use of newer components is a better solution. I think you have a good idea.

For my own use, a Goodrich type preamp solves most of my high frequency loss problems. (2) 9V batteries last for years and I don't need extra power hookups. You might consider (2) 9 volt batteries in series. Go back through your design and reduce power dissipation as much as possible. I think one of the reasons why the Goodrich products are so popular is because no wall warts are required.

My 3 cents

Regards,
Terry

------------------
Terry Downs
http://nightshift.net
terry@nightshift.net

Maybe it's the opposite. I don't hear or percieve losses that many say or think are there.

Being an electronics tech I probably tend to keep my equipment in top notch shape, but other than that there's nothing special about my setup.

Donny: Your notes on amplification are certainly correct. In principle you could even set up a pedal that emphasized highs as the volume was backed off, but it would be nearly impossible to match to the cable effects which would differ with each combination. Even in the best of circumstances, it would come at the cost of extra hiss in the system.

I agree that if you have an amp that has power limitations, you could expect a flattening of the gain response - either through clipping or soft-gain limitations in the protection circuitry. For what it's worth, your super high powered amps like the 300W Peavey (Nashville 1000, Session 2000, RPC...) series are not likely to show this effect. The speakers, however, are the weak point at high volume levels. It is unlikely that a single 15" speaker can properly handle more than a few tens of watts of continuous power.

Terry: See the note above about trying to compensate the high frequency loss. By the way, the Fletcher-Munson curves for human hearing are remarkably flat above about 80 dB (see below) - not even loud enough to play with an unamplified trap set.

The power requirements of my circuit aren't likely to change. It is pretty easy, though, to use rechargeable batteries. If there is sufficient interest out there, I will create a battery charger drawing.

All: Another perfectly good solution is to use the MIDI master volume available on some amps (such as my Session 2000 with the RC2000 controller). My little fix is for the rest of the world, and is my choice in studio settings, where you can control your gain after the amplifier's preamp output.

--- Mark
This is the famous Fletcher-Munson Curve:

I've been using a Goodrich H10k for several years now, and have no complaints. What keeps this popular model from being "a good active pedal", in your opinion?

The graph below shows a simple model that includes only the effect of a volume pot set to about half the maximum volume (the frequency response would be completely flat at full volume). As you can see, the gain is reasonably flat to 1 kHz, then drops dramatically.

When you include the coil self-inductance (on the order of 3 H - though actually this is conservatively low...) The effects of the volume control on the brightness of the instrument are dramatic. The first plot shows the frequency response of the system with the volume high (about 95% of maximum). The most striking effect is that the large inductance of the coil adds a very sharp peak at about 4 kHz. This is no doubt well known to the manufacturers.

Finally, when you put this pickup in front of a volume pedal, at about half volume, this is the response that you get:

Notice that all of those highs are completely lost.

An impedance mataching circuit will only get you part way there, as the effects shown in the first plot will go uncorrected. You will still get the brightness peak, but it will be much attenuated.

I hope this helps.

Mark, what you are getting across is true, but to further support your conclusion, it is actually worse than you are showing. Your model only show a series inductor, resistive divider, and a shunt cap. The cable has a distributed capacitance in addition to the amplifer input RF bypass capacitor. This changes the circuit from a 12dB/octave network to an 18dB/octave network. If you modeled the 500pF as the total capacitance of the cable and the amp input, your breakpoint is probably about the same, but the rolloff rate would be even sharper than you modeled.

One thing this exercise has taught me is that the actual guitar cable is a big deal. Belden 8410 is unusually good stuff - better, I think, than you can expect in many commercial cables - I have seen 100 pF/foot often. Changing the length from 15 to 6 feet can mean the difference between having nice brilliant highs and having weak signal from the 24th fret.

The models that I used are clearly simplified, though. For example, the pickups themselves are not simple inductors but have distributed parallel capacitance and series resistance that will greatly alter the peakiness present in the plots above.

Anyone tried my circuit yet?

First, I find your statement following one of your graphs quite troubling...

quote:

---

Notice that all of those highs are completely lost.

---

My observations on your graph show that the highs at 10Khz (where we seldom venture) are not completely lost, but are down by less than 10db! Now, this graph represents a static model, and were we playing with the pedal always in one position, this would indeed be a loss---not a terribly significant one, but a loss nevertheless. However, when we actually play, and there is such a loss (which we see as an increased decay rate, and a decreased volume) we naturally add more pedal. This almost unconscious, instantaneous, "correction" that we make probably negates this loss to a considerable extent.

Also, by reading curves (frequency response measurements) on popular speakers, I find that many of them are down by 15-20db at only half that frequency (5Khz)! That's twice as much of a loss at only half the frequency!!! So, it can be said that, while the pedal and cable may be a problem, they are certainly not the most significant one in the tone department. Cumulative effects may cause more significant problems, but I feel the pedal, and judicious choice of cable lengths and types, really makes this a much smaller problem in actual practice than it appears on paper.

Also, I must add this last comment, which is based on over 35 years of playing, and many more years of "just listening". It would seem to me that today, there is no great emphasis on these high frequencies. Modern pedal steel sounds have moved away from pronounced highs, and towards a fuller, fatter, mid-range sound. One only has to listen to the recordings of today (and compare them to those of 35 years ago) to verify this. The significant highs which once characterized players like Emmons, Brumley, and Mooney, are no longer in demand. Therefore, I think it is safe to say that, while boosting highs might have been desirable years ago, it is now not very significant (for most players). In modern pickups, high impedance, and high "ohms measurments" are the rule. I would imagine that the Lawrence, and George-L companies get very few requests for the old high-dominated sounds. But, before I end this diatribe, perhaps someone can answer this one question. Knowing that Buddy used a plain old (nowadays often derised!) pot-pedal, and old technology cables and amp, just how did he produce all those delicious, clear, razor-sharp highs on his famous "Black Album"?

[This message was edited by Donny Hinson on 13 September 2001 at 04:18 PM.]

You are correct about the speaker response - esp. these 15" beasts are far from flat at the high end. That's one of the reasons that most folks, as far as I can tell, set their treble and presence knobs to boost positions - it tends to compensate the speaker rolloff (see the PDF files of "typical/recommended" settings for the Peavey amps, all of which show a treble and/or presence boost). The difference is that this portion of the equation doesn't change as you move the pedal.

Your comment about changes in taste is certainly relevant. As Bobby Lee & Jack Stoner have pointed out, they don't find a problem. Others, as you can see above, do. Part of this might be equipment differences, part of it might be taste and, because this happens to all of us as we get older, part of it might be related to differences in our ears (as we age, we become less able to hear high frequencies, and thus less sensitive to problems with them.)

All: Anyway, I am not trying to force this on anyone - I offer an analysis of the problem - which is pretty straightforward, a suggestion for an experiment you can do on your own rig (see the note posted on 9/10 about playing with different volume settings) and a solution that is well-tested. I will also try to work with the vendors (Mike Brown, are you listening?) to see if some variant of this might be cost-effective to put into an amp.

Mark Cohen

It rolls off fairly fast after 2Khz. At 3Khz its approx -10db from the 2Khz point. At 5Khz it's approx -20db from the 2Khz point.

If I remember correctly from my basic electronics, -3db is a half power point.