3
\$\begingroup\$

As a guitarist, very familiar with the various types of noise that can wreak havoc with electric guitars, I've been mystified at how infrequently guitar amplifier inputs have employed differential stages. Of course low impedance microphone inputs have done so for a long time, specifically to reject as much common mode noise as possible, especially given the typical use of long cables. Now I understand that most magnetic pickups used in guitars have comparatively higher outputs then microphones. But these guitar pickups generally perform best when connected to high impedance inputs, which always works against noise immunity, and their amplifiers need to cover a very wide dynamic range. Of course many guitars employ dual coil "humbucking" pickups, which help null out both electric and magnetic field induced noise. But all such approaches change the sound, and many players favor the tonal quality of simple single coil pickups.

So my question here is simply whether the failure to transition to balanced differential inputs for guitars is due to some disadvantage I'm not taking into account, or if its simply a matter of how difficult it is to change the "status quo" with anything related to audio and psycho-acoustics. I've made many home brewed amplifiers over time, and have re-wired enough guitars to consider all kinds of experimental changes. And of course a differnial input can still function with an old fashioned single conductor coax cable. But I'd still like to hear from others on this. If differential inputs have drawbacks I'm not considering, it wouldn't be the first time I thought I had a better way to do things, and was later proven dead wrong.

\$\endgroup\$
  • 1
    \$\begingroup\$ Asking for opinions? Ok, mine is that it's mainly because of status quo/inertia. Active preamps can combat this, but most guitar players seem to be a conservative bunch, who prefer the "genuine" "vintage" Fender/Gibson 1950s sound, or some such. It is much more common to find active preamps in bass guitars. \$\endgroup\$ – Dampmaskin Jun 16 '16 at 16:35
  • \$\begingroup\$ One thing to note is that balanced guitar pickups are uncommon, and the fully balanced pickup/tone/volume/selector switching in a guitar would make the internals much more complicated (dual pots, for one). In my opinion, yes, balanced guitars should be a thing, but influencing a market that is heavily based on historical (and sometimes mythological) factors is slow going. \$\endgroup\$ – uint128_t Jun 16 '16 at 16:50
  • \$\begingroup\$ @Dampmaskin - I actually wasn't asking about active pre-amps INSIDE guitars, but more about the first preamp stage in the guitar amplifier. \$\endgroup\$ – Randy Jun 16 '16 at 19:10
  • \$\begingroup\$ @uint128 I would counter that a guitar pickup is by nature "balanced", since its is essentially just a coil. Typically they are mounted on metal that serves as shielding, and the better ones also have conductive tape around the coils and additional metal shielding. That shielding is then connected to whichever side of the coil is decided to be ground (usually the outermost coil windings). But were the shield kept separate, its pretty near an ideal balanced source, and would require so little effort to take advantage of that, while retaining backward compatibility. \$\endgroup\$ – Randy Jun 16 '16 at 19:12
  • \$\begingroup\$ @Randy As Andy aka explained in his answer, guitar electronics do not have a fully differential signal path, as that would require a center-tapped pickups (you need a ground node to avoid common-mode excursion) and "dual" internal signal paths. As far as reducing noise in a guitar goes, well-designed active preamps (sending a low-impedance balanced signal out of the guitar) will get you a whole lot of improvement. \$\endgroup\$ – uint128_t Jun 17 '16 at 1:47
5
\$\begingroup\$

If the guitar pick-up were all that were in a guitar body then it would make sense to wire it differentially back to the differential input amplifier. That's where the story ends because to make the volume pot(s), tone pot(s) and pup-switching balanced needs much more complication.

So, what about a standard guitar feeding a balanced input? No big benefit. To be balanced you need: -

  • A balanced input
  • A balanced drive - nope a standard guitar does not do this
  • A balanced impedance to ground and a standard wired guitar just does not do this.

So, there is no benefit wiring an unbalanced signal and impedance (wrt ground) to a balanced input. In fact you are likely to cause common-mode problems on the input because of all the capacitamce to ground through the body when intermittently touching the strings.

\$\endgroup\$
  • \$\begingroup\$ That makes sense. Thanks. Were this a case of dropping a first stage pre-amp inside the guitar, there might be some noise immunity gained, but with all the other existing guitar wiring done with a single ended ground philosophy, I see now where there would be little gained, and possibly additional problems in that case. \$\endgroup\$ – Randy Jun 16 '16 at 22:26
1
\$\begingroup\$

The PRIMARY reason that "instrument" connections have remained high-impedance, unbalanced is simply TRADITION (or INERTIA if you like.) This is been the standard for many decades simply to preserve BACKWARDS-COMPATIBILITY.

To be sure, there are SOME instruments (synths, etc.) that have low-impedance, balanced, even XLR outputs. Especially where they are expected to be connected to house reinforcement, stage monitoring, or studio recording systems.

I presume that it took a great many windings on the pickup coils to produce enough voltage to survive ~1m of cable over to the guitar amp. Presumably in modern times with much better rare-earth magnets, it takes fewer turns which results in lower source impedance.

And because the connection is high-impedance, and VERY sensitive to parallel capacitance in the cable (which is why there are special "guitar cables"), making the signals balanced/differential would make the signal TWICE as vulnerable to high-frequency loss from cable capacitance.

Of course impedance changing transformers could be used to convert the high-impedance source to low impedance (and balanced). This is the method used in many early microphones which were high impedance. And there is an active equivalent that can be found in any condenser microphone. The impedance of the microphone capsule is even order of magnitude higher than any guitar pickup. But active circuits are used to buffer the signal, lower the source impedance, and typically balance the signal so it looks like a conventional low-impedance balanced mic output.

\$\endgroup\$
  • \$\begingroup\$ Well I'll grant the cable capacitance might change, but I'm not sure it would be for the worse. Most ordinary mic cables, with two conductors (often twisted) and a separate shield, are often much better quality then an awful lot of guitar cables I've used, and a twisted pair often has a better transmission characteristic. Its a good point though, further analysis needed! :-) The impedance of most guitar pickup coils is not really high, usually around 5K give or take. The need for high input impedance is due to the loading affect, which drops the peak resonance bump and robs high end. \$\endgroup\$ – Randy Jun 16 '16 at 19:07
  • \$\begingroup\$ But I'm not suggesting making the amplifier's input low impedance. I'm more suggesting a differential input, with both wires from the coil brought in as usual, but with the shield kept separate... as with those mic elements. \$\endgroup\$ – Randy Jun 16 '16 at 19:10
  • 1
    \$\begingroup\$ Traditional instrument impedance is rather HIGH. That makes it vulnerable to high-frequency attenuation from cable capacitance. That is why you typically see special instrument cable with low capacitance. Making the input differential would mean that BOTH of the internal signal wires would be subject to high-frequency attenuation quite possibly making the problem even worse than it is. You cannot compare instrument connections with microphone connections because mic signals are on the order of 50~100x lower impedance. \$\endgroup\$ – Richard Crowley Jun 16 '16 at 21:19
1
\$\begingroup\$

When should an audio pre-amp use a differential input ? Generally speaking when the difference signal is weak, compared to the common signal. The common signal is usually coming from the electromagnetic environment, usually electromagnetic power wiring, light, transformers, motors etc. Every weak magnetic field sensor (aka microphone, phone pickup and guitar pickup etc.) benefits from differential signal amplification. But that is not in contradiction in what is said in other answers. In all of the answers one or more practical aspect is answered. I'm a guitarist that is researching and designing (unfortunately, not often building) active guitar preamps, so let me focus on guitar pickups. In short: it makes a lot of sense to build in a differential pre-amp into the guitar, also for effects by mixing pickup signals (adding/in-phase or subtracting/out-of-phase), but once external, you better of with a normal single wire signal, mainly due to the connectivity to and portability with effects and amplifiers. What no one here really mentions is that the load by amplifier input impedance and cable capacitance has a acoustic dimming effect on the guitar strings. The current (caused by the load) in the pickup elements causes a reaction force in the strings, hence it destroys sustain and signal brightness. What it to be done to avoid this, is to make the pre amplifier impedance as high as possible. Not too high because that would pick up to much of that electromagnetic noise from the environment. However, if a differential input amplification is used, this common signal should be eliminated and the input impedance can be several times higher. Note that such build in differential pre amplifier is expected to be particular effective with single coil pickups since they do not cancel out the common signal like humbuckers do (although, those would also benefit). Having differential pre amplifiers, one for each pickup element before their signal is combined and providing a low output impedance to drive the external wiring, would to my opinion give the best results. "Best" is subjective - certainly to guitarists -, but be sure to have more sustain :-)

\$\endgroup\$
  • 5
    \$\begingroup\$ Welcome to Electronics.stackexchange! Comprehensive first answer, but it could really do with some more formatting - it's a big wall of text at the moment, which makes it hard to read and understand :) \$\endgroup\$ – Araho Sep 21 '17 at 14:11
  • \$\begingroup\$ Thanks for posting. I am in agreement with virtually all you've said. And yes, I too see where any pickup load, aside from changing the freq. peak of the pickup, will also dampen string vibration. I don't want to take the conversation into the weeds (my question is old now anyway), but feel free to PM. Sounds like we likely have a lot of projects in common. Or use the contact form on my own website (www.elfintechnologies.com) \$\endgroup\$ – Randy Sep 22 '17 at 18:45
0
\$\begingroup\$

Partly it is inertia, but also it is because in most situations the lions share of the noise pickup is NOT in the cable.... Also, guitar amps have to be seen as part of the instrument, which makes them hard to change much without causing all sorts of conservatism on the part of the musicians to appear (Not without some cause, the early transistor and early modelling amps were horrible).

For the most part noise pickup by electric guitars is either due to electrostic pickup due to unshielded pickup and electronics cavities in the instrument, which balancing after the fact will do nothing for (But for which appropriate application of copper tape is usually an almost total cure, balancing might also help but...), or magnetic pickup by the pickup coils themselves, which again is not something really fixable after the fact.

Now a pickup could reasonably be designed with a humbucking coil as a separate winding in the structure and active electronics then used to combine the two which may well reduce the hum without impacting the sound as much as a humbucker does, but I am having trouble seeing balancing helping much for either of the common noise sources.

Guitars are (when the on board controls are fully open) a few k of resistance in series with hundreds of mH of inductance, and it is the latter which makes guitar cables tricky (and actually one of the few places in audio that wire really matters, phono cartridges are the other one for much the same reason), and it is this inductance that can make input stage RFI filtering a pain when designing amps.

Unbalanced audio is all kinds of Ick, but in the case of a guitar it is probably not the biggest source of the problems.

Incidentally, when working on a guitar, placing a few nF of class Y cap in the bridge earth lead provides some useful harm reduction in the event of some idiot with a disconnected earth on the amp having an accident.

\$\endgroup\$
0
\$\begingroup\$

This a very interesting topic. My two bits is as follows,.. If you view the guitar pick up as a magnetic transducer.. with just an impedance in series with the tone/volume (resistance + capacitance) one could reduce it to a two port network of simple RLC. The frequency response of such a net work is limited by the coil inductance and the parallel capacitance.. assuming constant pot position. Hence the frequency plot will essentially be the same in a single ended or differential input stage. So a differential input stage would not make any difference to the frequency response. Noise figures also have no appreciable difference in either differential or single ended pre-amplifiers with respect to the input transducer. However the main issue (practically) is mains frequency (hum) that is picked up along with the guitar pick up.... the solution for this is may fold... Simplest one that comes to my mind is differential cancellation .. in short.. derive a mains frequency from the power supply and introduce an out of phase signal in the input itself. The level of this signal can be manually adjusted or if one has the technical verve..!! an elegant solution would be to have a DSP to track the mains frequency, analyze the preamp output, and then introduce a scaled out of phase cancellation signal. I would recommend a mosfet differential input pre-amplifier, not for the noise issue, but for the high impedance such a system provides, the frequency response of a high impedance input beats any thing else. Acoustic guitars or acoustic sounding electric guitar will sound more original.. however the contemporary rock fella may not appreciate all that.

\$\endgroup\$
  • 1
    \$\begingroup\$ Welcome to EE.SE! Please break up your answer into more than one paragraph for better readability. \$\endgroup\$ – Daniel Aug 15 '18 at 17:34
  • 1
    \$\begingroup\$ S/he tried. Use 2 x <Enter> for a paragraph break. To improve legibility avoid your odd '..' punctuation and capitalise properly. \$\endgroup\$ – Transistor Aug 15 '18 at 17:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.