Well, I disagree. I have heard many TL072-driven guitar devices that I thought were NOISY AS HELL. You can't rely on figures drawn from a certain position on the volume pot----many times it may be all the way up, and your theory goes all to hell. An OPA1652 is only a coupla $$, and it's WAY quieter than your noisy old TL072!! In theory, it could be 12db quieter; while I don't think that translates to reality completely, it's damn sure gonna be NOTICEABLY quieter.
Rod Elliot recommends OPA2134 for guitar first and second stage--with significant gain, saying it's very quiet. From all the comments here, the myriad papers on noise, and data sheets, it appears to me the end result is determined by several factors, not just device noise on a laboratory test jig. For my feedback resistors, I chose values that minimized noise from those components, and also made sure the current was nowhere near max. Too low values will increase current through the devices raising heat and increasing thermal noise and noise current, which would swamp opamp noise. There is also the possibility of defective or damaged components. Sampling one rendition of a first stage, or grabbing the first opamp out of a box is not scientific at all. Right?
Last edited:
I would agree; it's just that the OPA2134s have gotten very expensive lately---almost ~$5 apiece! In any case, either the 2134 or 1652 is gonna be noticeably quieter than a TL07x.
Well, no! The nice thing about theory is that you can always re-calculate it with new numbers to try out those "what if?" questions. So, let's do that.
Firstly, keep in mind that the 34k grid stopper resistance (two 68ks in parallel) in any number of classic Fender guitar amps is already noisier than a TL072. Thermal noise from a 34k resistor at room temperature (20 degrees C) is about 24 uV/root Hz - appreciably more than the 18 nV/root Hz spec of a TL072.
We can't omit that input resistor entirely without inviting RF interference problems, but we can lower it. Suppose we wise up and use a 10k input resistor instead, as is widely recommended these days. Add that to the roughly 10k DC resistance of the pickup, and you're at roughly 20k. (We're being very kind - as PRR pointed out, the source impedance of the pickup rises to much more than 20k at around 3 - 4 kHz, where the pickup inductance resonates with cable capacitance.)
Anyway, let's go with the "kind" 20k source impedance. Look back at the graph I posted. Find 20k source impedance on the x-axis, go vertically up until you hit 18 nV/root Hz (the noise spec of the TL072). You'll find you are now more or less on the diagonal (noise figure = 3 dB) line.
So, if you are that rare guitarist who always leaves the volume control at maximum, a TL072 still gives you a NF of 3 dB. Not perfect, but neither is there a whole lot of room for improvement - maybe another 2 dB at best, to get you to a NF of 1 dB. A 2 dB improvement is just barely audible; it's certainly not significant.
No - if the OPA had unicorn-magic and came from the Roswell UFO and had ZERO noise of its own, and you left the guitar volume control always up, that would still only get you about a 2 dB improvement in noise figure - see my preceding paragraph!
It's quite possible that the OPA-whatsit might indeed be quieter in a different application, one where the source impedance was only a few tens of ohms instead of tens of kilo-ohms. A moving-coil record player pickup, say, or a low impedance dynamic microphone without a transformer in it. But connected to a conventional e-guitar pickup, the OPA-hoozy-whatsit is just a waste of money.
It's only a small waste of money, so it certainly won't hurt much to use it if you want to. It won't help much, either, though!
-Gnobuddy
Exactly! The key thing that's usually overlooked is that the resistances inside the guitar make the guitar signal inherently noisy all by itself. In other words, straight out of the guitar, there is only so much signal-to-noise ratio to be found. Not infinite.
All that the guitar preamp has to do is not worsen the guitar's inherent signal to noise ratio much. This doesn't require a zero-noise op-amp; it only requires an op-amp that makes less noise than the guitar itself. As we've seen, the venerable and inexpensive TL072 op-amp does the trick.
And I agree with you entirely that other external components (resistors) in the preamp matter a great deal. Leonidas Fender didn't realize it, but the 34k grid-stopper he used at the input of several of his amp designs contributes more noise than everything else in the entire amplifier!
-Gnobuddy
Thanks!
-Gnobuddy
"Fender commonly used two jack sockets each connected to 68k grid stoppers which appeard in parallel when using only the 'hi' input, making 34k." Blencowe.
I see 68k in many single input amps with 12AX7 first stage. It may be replaced with 10k along with the addition of a cap across grid to ground of the appropriate value pF range). Also Blencowe.
Better to buffer with silicon, or eliminate the first tube stage altogether since it usually only amplifies. Fender amps are quiet enough stock. Hi-gain is quite another story.
In other news, after fixing the opamp adapter board for the wrong wiring, it now has no output whatsoever. Checking the output of the opamp, it went over 5vrms with the guitar volume control at zero. I am making another from scratch. I may find either something burned out, or the pedal was a lemon from the get go, or both!!
I'm still alive by the way. It's just a matter of slowing the roll and taking many breaks.....
Not forgwtting anything. Nobody plays a metal distortion oedal with their guitar volume at halve. From a noise perspective, the noisefigure of the 10k input resistor is more than 3dB for typical Gibson humbuckers, and up to 6dB for singlecoils ( but who plays a metal pedal with a singlecoil?). The tl071' 18nV adds allmost 6dB to the noisefigure of a typical humbucker.
You want to lower the noise? Change these parts.
You want to lower the noise? Change these parts.
I've seen that, too. As you say, it isn't usually a problem with low-gain guitar amps.
I've read this from several sources. However, I found it to be incorrect. I could hear slight but very definite "tubeyness" (i.e., good-sounding harmonic distortion) using a single-coil guitar and a single half-12AX7 gain stage. With a humbucker, this was even more easy to hear.
With one of my (humbucker-equipped) guitars, an Ibanez AS73 semi-hollow, I could even drive that input stage all the way to audible clipping. Ibanez designers chose to put a rather high-output ceramic humbucker in that guitar.
It may be that a JFET input stage can be made to produce similar (small) amounts of good-sounding THD. I haven't experimented enough to be able to tell, one way or the other.
Sorry to hear you're dealing with a few headaches.
The enormous output suggests that either the feedback resistor from (-) input to ground is shorted, or the feedback resistor from output to (-) input is open-circuit. Either of those errors will unleash the full open-loop gain of the op-amp, which is typically of the order of 100,000 - 1000,000 for most audio op-amps.
-Gnobuddy
My error. All tubes have some distortion as we all know. There is some minimal clipping in guitar amps first stage, and obviously if over-driven. Some SS amps have one AX7 for either coloring or outright distortion. But take a fuzz box like mine. The buffer does not distort, and it could easily be made to have some clean gain, like my opamp buffer. The end result should be the focus, not so much how we get there. I have heard of fuzz boxes feeding high-gain amps, fuzz feeding fuzz in series, and for the life of me I don't see the usefulness of that kind of thing. More equipment, more expense, more complications (maybe).
The entire thing has been built anew. Testing soon. I hope nothing got smoked downstream....
In "proper" engineering, attenuating to half voltage IS 6dB Noise Figure (plus amplifier noise). If you wanted a clean signal, you would not turn-down before the first amplifier. Clearly this is not a high priority for the majority of guitarists. If it was, the on-guitar knob would work its deeds after the first amplifier stage (buffer inside the axe, remote control via extra wire or supersonic bleeps).
As nearly all other "frills" have been offered, but not this, I conclude that the S/N of guitar is "quite acceptable" to 99% of players, either at 10 or at 5.
As I have been greatly annoyed by the on-axis hiss of amps several times, I have other opinion.
Exactly. The signal is small enough to never actually clip, but it does get deformed, just a little, and I found this was audible. It still sounds "clean", but not solid-state clean; it's "tubey" clean.
I scratched my head over this one, because the concept didn't seem to make sense to me. Until, that is, I watched an episode of "That Pedal Show" on You Tube, and saw/heard what happened when one of the hosts (Dan) kicked in a second distortion pedal (a Boss Blues Driver) in series with his first one (an EHX Soul Food).
First, the distortion suddenly got much richer and much more amp-like. I think this is related to the discussion we had earlier about multiple clipping stages with dynamic DC bias shifts sounding better.
Then Dan tweaked a knob or two, and kicked in a Maxon delay as well, following the Blues Driver. Holy flutter-winged rabid Batman, where did David Gilmour's spirit suddenly come from? Where is that 50-lb violin I'm hearing?
I don't know if it will have the same effect on you, but the bit I'm describing happens starting at about 19:00 minutes into this clip: YouTube
I had a Blues Driver already, and an EHX Glove overdrive, so I tried stacking them, along with a delay and reverb. Sure enough, I too got some beautiful sounds reminiscent of David Gilmour. After a few hours of enjoying this, it dawned on me that beautiful as they are, these sounds don't really go with any of the music I play...
But, having tried it myself, now I do think there is something different and interesting that can happen when you stack one distortion pedal into another. It's probably the same reason why heavy distortion sounds better when spread out across several gain stages, rather than driving just one amp stage or diode clipper very hard.
-Gnobuddy
I think I was unclear. I was just looking at where the TL072's noise figure spec put it, in term of noise figure, against first a 10k, and then a 100k, source impedance. It didn't occur to me to consider the change in S/N ratio from the guitar itself, due to turning down by 6 dB.
Agreed. Listen to the amount of crud and crap in Jimi Hendrix' guitar sound (never mind his bloodstream)...I don't recall ever hearing about anyone asking for their money back because Jimi's guitar's S/N ratio wasn't good enough.
I agree, though I must qualify that I don't use lots of gain in my own guitar playing.
In fact, the change in S/N due to turning the volume knob has never been audible to me - I don't think I've ever had an amp with low enough input noise to make this audible, not to mention the sounds from the ventilation system in the room are usually louder and hissier than anything from my guitar amp.
I don't play at ear-shattering SPL levels either, which also contributes to having lower hiss levels from the amp.
I have never met an electric guitar player who intentionally sits on-axis with his / her amp, though I have met several acoustic-electric players who prefer exactly that position.
e-guitar speakers are bizarre things, beaming treble like a lighthouse. Acoustic-electric amps tend to have relatively flat frequency responses and include actual tweeters with good treble dispersion. So these two different preferences aren't too surprising.
-Gnobuddy
Help
The new opamp worked on the bench.
When installed in the box, the following describes the result:
1. There is zero output and no noise whatsoever, whether in bypass or stomp.
2. There is 9v on the rails.
4. There is 1mA of DC on the output
5. The output of the OUT terminal of the opamp is vanishingly low when it should be over 300 on a medium strum.
6. Checking for shorts after the OUT, my ohmmeter reads 6.9k--precisely what Rg and Rs are in series.
Looking at the schematic, since there is zero output in either mode, it appears it must be something after Q3 which is where the two signals would meet, if they could. Still mulling over what to do next.
Thanks @Gnobuddy for the clarification of series boxes. I also read Gilmour uses two different echo units set with different delays. For some reason, his sound seems too sterile to me. I would bet it sounds better live than on a consumer grade stereo..... I have not been to that many concerts, but Foreigner in their prime at the Capitol Theater in Passaic, NJ (a classic movie house) was perfect in my estimation; I could not imagine it able to be better. Everywhere else--meh.
The new opamp worked on the bench.
When installed in the box, the following describes the result:
1. There is zero output and no noise whatsoever, whether in bypass or stomp.
2. There is 9v on the rails.
4. There is 1mA of DC on the output
5. The output of the OUT terminal of the opamp is vanishingly low when it should be over 300 on a medium strum.
6. Checking for shorts after the OUT, my ohmmeter reads 6.9k--precisely what Rg and Rs are in series.
Looking at the schematic, since there is zero output in either mode, it appears it must be something after Q3 which is where the two signals would meet, if they could. Still mulling over what to do next.
Thanks @Gnobuddy for the clarification of series boxes. I also read Gilmour uses two different echo units set with different delays. For some reason, his sound seems too sterile to me. I would bet it sounds better live than on a consumer grade stereo..... I have not been to that many concerts, but Foreigner in their prime at the Capitol Theater in Passaic, NJ (a classic movie house) was perfect in my estimation; I could not imagine it able to be better. Everywhere else--meh.
Make sure that all the opamp inputs (inverting and noninverting) have the nominal +4.5 V half-supply bias on them. Same for the outputs. If you find any outliers, note down which ones. That one stage with the capacitor in the feedback loop's ground leg, make sure said cap is not shorted.
Troubleshooting opamp circuits isn't rocket science:
In order for an opamp circuit to stand any chance of working at all, the opamp must be able to reduce the voltage difference between its inputs to essentially nil. Input voltage must be within specified common-mode input voltage range, and must also be accessible to the output stage (but usually the former is the limiting factor). For traditional non-RRIO opamps like the M5218 (4558) used here, at least 1.2-2 V away from either positive or negative supply is usually a good rule of thumb, though up to 3 V on one side may be encountered with FET input types. In this circuit they are at a nominal +4.5 V, well away from either V+ = +9 V or V- = 0 V.
Let's say you had a noninverting gain-of-11 circuit. +in: DC +4.5 V + signal at AC Feedback network: 10k of feedback + 1k to ground, no capacitor in sight. The result? The opamp would be trying to get the -in to +4.5V. For that the output has to go through the 10k/1k feedback network, so any output voltage is attenuated by a factor of 11. So for +4.5 V at -in, you'd need 11 * 4.5 V = +49.5 V at the output. IOW, it's a gain-of-11 amplifier right down to DC. Since your V+ is only +9 V, the opamp is going to latch up and do nothing useful, since -in is going to be at 0.8 V or something, a whopping 3.7 V away from +in. A bipolar input part like the M5218 has a linear differential input range of maybe 65 mV (which clearly is far less than 3.7 V) and will run its output into the rails beyond that.
Now let's modify our feedback network with a capacitor in series with the 1k resistor to ground. The result? The capacitor is open at DC, so at DC the 1k resistor vanishes, leaving only the 10k feedback resistor - DC gain thus becomes unity. Without much current going in or out of the -in, the -in is essentially going to track the output. So what you end up with is +in = +4.5 V, out = +4.5 V, and -in = +4.5 V, and lo and behold, the thing actually starts doing what it's supposed to. At AC, the capacitor is essentially a short, so the whole shebang will be a gain-of-11 noninverting amplifier as desired.
Troubleshooting opamp circuits isn't rocket science:
In order for an opamp circuit to stand any chance of working at all, the opamp must be able to reduce the voltage difference between its inputs to essentially nil. Input voltage must be within specified common-mode input voltage range, and must also be accessible to the output stage (but usually the former is the limiting factor). For traditional non-RRIO opamps like the M5218 (4558) used here, at least 1.2-2 V away from either positive or negative supply is usually a good rule of thumb, though up to 3 V on one side may be encountered with FET input types. In this circuit they are at a nominal +4.5 V, well away from either V+ = +9 V or V- = 0 V.
Let's say you had a noninverting gain-of-11 circuit. +in: DC +4.5 V + signal at AC Feedback network: 10k of feedback + 1k to ground, no capacitor in sight. The result? The opamp would be trying to get the -in to +4.5V. For that the output has to go through the 10k/1k feedback network, so any output voltage is attenuated by a factor of 11. So for +4.5 V at -in, you'd need 11 * 4.5 V = +49.5 V at the output. IOW, it's a gain-of-11 amplifier right down to DC. Since your V+ is only +9 V, the opamp is going to latch up and do nothing useful, since -in is going to be at 0.8 V or something, a whopping 3.7 V away from +in. A bipolar input part like the M5218 has a linear differential input range of maybe 65 mV (which clearly is far less than 3.7 V) and will run its output into the rails beyond that.
Now let's modify our feedback network with a capacitor in series with the 1k resistor to ground. The result? The capacitor is open at DC, so at DC the 1k resistor vanishes, leaving only the 10k feedback resistor - DC gain thus becomes unity. Without much current going in or out of the -in, the -in is essentially going to track the output. So what you end up with is +in = +4.5 V, out = +4.5 V, and -in = +4.5 V, and lo and behold, the thing actually starts doing what it's supposed to. At AC, the capacitor is essentially a short, so the whole shebang will be a gain-of-11 noninverting amplifier as desired.
Last edited:
OK. You can keep your $1.29---the difference in cost between a TL072 and an OPA1652. My figuring, with a 25 KΩ source impedance puts the TL072 (18 nV√Hz Voltage Noise; 0.01 pA@ 1 Khz Current Noise) 12 db noisier than an OPA 1652 (4.5nV√Hz;0.003 pA). Seems money well spent to me.
Oh, that must be the one I eliminated. It works perfect now. Thanks a million, you saved me probably days of aggravation.
The box is super quiet, but much less icy. I suspect by reducing the 220k R44 down to 73k flattened the ramp slope a little. I did this to correct the overall gain of these two stages. This probably affected the harmonics significantly, since the square wave corners are not as sharp.
"I did my own studies and found that I rather preferred the humble NE5532 and biploar electrolytic caps in my mixer rather than buying insanely expensive super-duper audiophile quality opamps and esoteric caps, and/or specialized things like DC servos and such.
"Personally, I've always found the best way to upgrade cheap gear is usually to increase the local decoupling on the opamps and the bulk decoupling for the local opamp groups. Cheap gear tends to really lack in these areas. Most designs that use "cheap" opamps like the NE5532 also cheapen out on the power supplies and decoupling, so people who listen to these usually unfairly blame the active part instead of the supporting circuitry." Op Amps: Horrible Truth - Personal View Talks
Can this viewpoint apply to the humble TL072 as well?
Thanks for those. We should still know the nature of what we are dealing with before proceeding, as I just recently been reminded of.
An NE5532 is a fine BiPOLAR opamp, but guitar circuits (due to their high impedance) cry for FET opamps such as the OPA1652. It certainly is NOT "insanely expensive"---it's $1.97 @ Mouser, while an NE5532 is $1.01. Not much diff.
I agree completely. It's an inexpensive but VERY effective upgrade. Another VERY cheap upgrade (if you have LM317/337-regulated supplies) is Elvee's "DeNoiser" circuit, posted here on diyaudio. A coupla bucks to reduce power supply noise by ~30 db or so.
Component Quality
It was noticed the stock resistors in the MT-2 have lead/body junctions that look like industrial spot welds with no oxygen shielding whatsoever, whereas the Vishay thin film MBA series have junctions that appear to be spun on a lathe.
I have no idea if the cheap construction could affect noise, but how about a cracked junction after bending gone unnoticed? The lead/body attachment strength is worthy enough to be mentioned in the Vishay data sheet.
The OPA1656 is recommended for guitar effects. First time I ever saw that mentioned in a data sheet.
But again, in the end--the end product, that is--be it possible to chase or believe something that does not even exist? Is this not tragic?
In some cases, however, this might be desirable. For instance, the sensitivity and compression offered by fuzz boxes could not be achieved with skill alone. Another example is two echo units with different setting to sound unearthly. In fact, it is unearthly because how can that type of sound ever be reproduced in a natural setting?
You guys have emboldened me to split the diode clipping stage in the MT-2. Don't know if it will ever come to pass, though.
It was noticed the stock resistors in the MT-2 have lead/body junctions that look like industrial spot welds with no oxygen shielding whatsoever, whereas the Vishay thin film MBA series have junctions that appear to be spun on a lathe.
I have no idea if the cheap construction could affect noise, but how about a cracked junction after bending gone unnoticed? The lead/body attachment strength is worthy enough to be mentioned in the Vishay data sheet.
The OPA1656 is recommended for guitar effects. First time I ever saw that mentioned in a data sheet.
But again, in the end--the end product, that is--be it possible to chase or believe something that does not even exist? Is this not tragic?
In some cases, however, this might be desirable. For instance, the sensitivity and compression offered by fuzz boxes could not be achieved with skill alone. Another example is two echo units with different setting to sound unearthly. In fact, it is unearthly because how can that type of sound ever be reproduced in a natural setting?
You guys have emboldened me to split the diode clipping stage in the MT-2. Don't know if it will ever come to pass, though.
I should probably start by saying that, at least for me, this isn't about trying to prove who is right. For me, it's about trying to prove what is right. I hate to leave technical errors uncorrected on a forum like this, where thousands of other people might continue to get wrong information and be misled for years to come.
It would help if you showed us how you calculated that number - then we could spot the error.
However, the bottom line is that the claim that the TL072 is 12 dB noisier than the OPA1652 is incorrect.
Let's re-do that noise calculation, (for the third time in this thread I believe), to find out why that claim is wrong.
A 25k resistor (your chosen source impedance) has a Johnson noise of 20.113 nV/root Hz. You can verify this using this thermal noise calculator (enter 1 Hz for the bandwidth): Thermal Noise Calculator
(Since I can't see what's going on in that calculator, I also did the same calculation from scratch using a spreadsheet, and I got the same answer.)
From this number we can immediately see that the "noisy" TL072 with its 18 nV/root Hz noise spec is actually a bit quieter than the 25k resistor by itself, with its 20.113 nV.
Even before we do any further math, this is a big whopping clue: the TL072 isn't going to make the total noise much worse than the resistor alone. That means S/N will be largely set by the resistor, not the TL072. And therefore, no other opamp, no matter how quiet, is going to be able to beat the TL072 by a signficant amount.
But let's proceed with the math.
When we have two independent noise sources (like the 25k resistor and the TL072), they add as the square root of the sum of the squares: square each noise source, add all of these squared values, take the square root of the result.
Doing that with 20.113 and 18 nV/root Hz gives us a combined (resistor + op-amp) noise of 26.99 uV/root Hz for the TL072.
Doing the same thing with the OPA1652 (with its 4.5 nV/ root Hz spec) gives us a combined (resistor + op-amp) noise of 20.22 uV/ root Hz.
We can see that 26.99 isn't a whole lot of decibels bigger than 20.22 - in fact, its about a 2.5 dB difference. The superbly-quiet OPA1652 is unable to lower the total noise much compared to the older-generation, noisier, TL072, because most of the noise comes from the 25k source impedance, not the op-amp.
But we can formalize this a bit further by calculating the noise figure in each case. The TL072 has 20.11 uV / root Hz in, and 26.99 uV/root Hz out, for a noise figure of 2.55 dB.
The OPA1652 has 20.11 uV/ root Hz in, and 20.22 uV/ root Hz out, for an exemplary noise figure of 0.05 dB.
The improvement, going from TL072 to OPA 1652, is the difference between these two: 2.55 - 0.05, or 2.5 dB.
This is audible, but only just. It isn't what I would call a significant improvement, though this opinion is subjective, and you might feel differently.
Let's not forget that even this small 2.5 dB improvement evaporates the moment the guitarist turns her volume control knob anywhere below maximum. The source impedance shoots up to far above 25 kHz, the noise figure of the TL072 falls to essentially nothing, and the lovely OPA1652 can't beat nothing, so it won't do any better than the TL072 - the difference between the two op-amp types will be even less than the small 2.5 dB we have now.
(Don't take my word for it, do the math yourself, using a 100k source impedance instead of 25k.)
In other words, as I keep saying, you really can't audibly beat the noise performance of a TL072 when it comes to an electric guitar!
-Gnobuddy
If the cracked junction still has electrical contact, it introduces zero noise or distortion. If contact is entirely lost, the circuit stops working.
Technically illiterate audiophools have claimed that magic "micro-rectification" occurs at ugly-looking joints; science and engineering both say otherwise. The electrons inside a rusty length of barbed-wire fencing are exactly identical to the electrons inside a polished bar of Fort Knox gold; they behave identically where electricity is concerned.
I have heard distortion from a corroded RCA connector once - but it wasn't subtle "micro distortion", it was outright half-wave rectification, just as physics predicts; the rusty junction had created an accidental "crystal and cat's-whisker" semiconductor diode, as used in thousands of foxhole radios in WWII. Plugging and unplugging the cheap RCA connector fixed the problem, and it never came back.
And it might provide significant reliability improvements. If a fifty-cent resistor fails in a $100,000,000 space satellite, that is an enormous loss; all components in satellite electronics are engineered for extreme reliability to avoid financial catastrophes like this.
There is something we haven't talked about on this thread: most op-amps are not designed to operate with the output slamming from rail-to-rail, and some early op-amps behave very badly when operated in this way.
There are some fuzz-boxes that operate the op-amp in this way, and in those, the actual type of op-amp used will matter.
I think humanity in general spends much of its time chasing after, and believing in, things that don't exist. And I agree that this is tragic.
Sure, but there is nothing "earthly" about an electric guitar itself, either! We left "natural" behind a long time ago, for better or for worse, and now we live in an alien world irreparably stamped with the impact of humanity, from the sounds we hear to the food we eat to the floods, droughts, heat-waves, and tornadoes we endure.
By the way, you might like the book "Perfecting Sound Forever": https://www.amazon.com/Perfecting-Sound-Forever-History-Recorded/dp/0865479380/ref=sr_1_1?
This is usually the point at which one might start to consider buying a breadboard, a few pots, a couple of small spools of 22-gauge hookup wire, a few 1/4" jacks, and a 9V switching power supply. Then it's off to the races - you can tinker to your heart's content, in relative ease, until you come up with something worth building into more permanent form.You guys have emboldened me to split the diode clipping stage in the MT-2. Don't know if it will ever come to pass, though.
It can be a long slippery downhill road from there, though. One day you look up at the dungeon you've lived in for years, surrounded by wire-clippings and dusty electrical measurement equipment, and you wonder what happened to years of your life.
-Gnobuddy
Thanks for that - Google found the original full 'Web page for me. Wow, Mr. Murphy was preaching that gospel in 1994!John Murphy said:
That sounds like the Marshall recipe - and what amazing sounds it's created over the years!John Murphy said:
I've never built anything Marshally, mostly because their ridiculous naming scheme confused the heck out of me to the point where I couldn't make head or tail of which was what. But I think it might be time to take a second look.
-Gnobuddy
I would draw the analogy of a live concert to the satellite!! The MBAs are 14 cents ea. plus shipping in low quantities, and they fit anywhere.
Speaking of noise sources and focusing on first and second stages, the guitar is the first device of the first stage! I will try to get hold of a 2w conductive plastic pot and see if it's worth the bucks. It could be used for a gain pot in a high gain tube amp, someday. I will have to agree with using any component if there is the slightest advantage, as long as the cost is relatively insignificant.
I've blown many thousands over the years on "chasing the wind." A couple of bucks more for the first stage is meaningless for me.
What's your opinion on using oxygen free copper wire in the pickups?
That sounds good!
Indeed! And the first guitar pickups were designed by tinkerers who probably knew nothing about thermal noise, nor was it really an issue in those early days of low-gain amps and clean guitar tone.
Radio engineers were already very aware of noise, as it is a bread-and-butter thing for radio designs. But audio? People were happy if it actually made noises, never mind S/N ratios.
If S/N ratio of the guitar is the focus, an on-guitar preamp can help a bit, making the signal bigger, and lowering its impedance, before sending it out into the hostile world via the guitar cable. (But the preamp brings headaches too, including the need to power it, and the fact that the guitar may no longer sound like it did without the preamp.)
Shortly after I began playing electric guitar, I built a 2-stage JFET/BJT preamp into a guitar I'd built from scratch. In those pre-Internet days, it was hard to get any information on how electric guitars were made, so I got just about everything wrong - the guitar had a lightweight pine body, the pickup ended up at about 4k DC resistance, and my preamp used negative feedback, taking out any "colour" the JFET might otherwise have added.
It worked, and I played it for years, but the guitar had very little sustain, and the sound was more like today's plugged-in acoustic-electric guitars, than like a proper electric guitar.
Commercially manufactured onboard preamps have been popular briefly - for instance, in the case of those EMG active pickups in the shred era.
After the culturally bankrupt 1980s came to a close, most electric guitarists woke up to the fact that the most musically attractive electric guitar sounds tended to come out of primitive guitars plugged into primitive guitar amps. And so solid-state MIDI-controlled rack mounted guitar preamps and sequencers and drum machines mostly disappeared from music stages. Thank goodness.
More recently, Fishman has made low-impedance active guitar pickups (their Fluence (TM) line). For the most part, Fishman carefully refrain from calling them "active", and I don't think they ever mention the dreaded term "solid state", though that is exactly what they are.
The Fluence line has heavy EQ built into the onboard solid-state active preamp, in the attempt to make a very-low-impedance printed-circuit coil sound like a many-thousand-turn high impedance copper wire coil interacting with a guitar cable.
The pickup wire might be in better shape a couple of centuries from now, when your great-great-great-great grandkids inherit your stuff.
More seriously: oxygen free means better thermal conductivity (completely irrelevant for a guitar pickup coil.) It also means greater current handling capacity (also completely irrelevant in a guitar pickup coil, where currents are so small as to be virtually nonexistent - not even microamps.)
There will be no sonic benefit whatsoever. If the extra bucks are burning a hole in your pocket, I suggest buying a hungry homeless person a sandwich - that will make both of you happier than any amount of oxygen-free copper wire ever could!
-Gnobuddy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.