The severe distortion sounds absolutely horrible. Even the clean sound (that sounded much better) had obvious harmonic and intermodulation distortion. I hate the sounds of the CRUNCH.

 

You buy it and then stuck. But if you build it you can change it make the sounds that are not de-sensitizing.

rythm guitar the sound of overdrive simple circuit.

 

How about the best of both worlds - JFET + OPAMP : https://www.ti.com/lit/ds/symlink/tl072.pdf
I have used these for years (usually the TL072) as a low noise, low distortion amplifier.
The manufacturer says "ideally suited for high-fidelity and audio pre-amplifier applications", and they are cheap and readily available. Other variations include TL06x and TL08x. Using the dual IC allows for huge amounts of amplification.

 

Just checking - your input source is high impedance, yes? If not, BJT's might be a better route.

If you are concerned about a low power design (e.g. battery driven, portable) then I would recommend a JFET-input op amp approach.

ac

 

I also made many circuits with the TL07x audio opamps in the '70s, '80's and '90s. They have Jfet high impedance inputs and low distortion. Newer audio opamp designs produce less noise.

 

There is one opamp with reasonable specs and it's the LM4562 (dual), I bought an off-the-shelf phono preamp, yanked and improved on the compensation parts, then dropped this into the existing socket and it worked pretty good, flat as a board on the RIAA test record. It's 2.7nv/root hertz for noise, remember even the discrete JE-990 was about 1.2 (if I recall), I mean it's not as good but it's no slouch either. Also 20v/us and 55 MHz GBW and the distortion figures are hard to beat (0.00003%), my vintage THD meter doesn't even read that low. If you need to make an LV supply from B+ consider the LT8316 if you have decent enough filtering skills that you can tolerate a switcher (yeah I know just ruined my cred, sorry but just don't be all that easily offended, I just get tired of using 10 watt dropping resistors to heat the entire neighborhood).

 

Interesting discussion going on – I have one question for Dynaman though. Are you building a preamplifier designed as part of a musical instrument amplifier or are you building a preamplifier as part of a high-fidelity reproduction amplifier? This is not a “dumb question” as they each have very different design requirements beyond simply passing a reasonable facsimile of the input signal to the output.

If the need is for a high-fidelity preamp, the goal is to EXACTLY reproduce the input signal except only larger in amplitude and add absolutely nothing else. This has often been described as the proverbial “wire with gain”. By ‘nothing else’, I mean zero harmonic distortion, zero intermodulation distortion, zero hum and noise and zero extraneous additional signals within the full audio band of 20 Hz to 20,000 Hz. (These goals are of course theoretical and have nothing to do with how it sounds to a person’s ear.)

Generally, to achieve these goals the opamp approach (as opposed to a one or two transistor circuit) is generally best, and certainly has lower distortion if a precision opamp is used. Bipolar input stages potentially have the lowest noise figure, followed by JFET input stages. MOSFET input stages are usually the worst-performing in terms of noise figure. If high input impedance is a primary concern, then the MOSFET wins, followed by the JFET and finally the bipolar input stage. These are not hard-fast rules because there is huge overlap in specification and how each device performs in reality also depends on the exact circuit used.
If the need for the preamp is part of a musical instrument amplifier – well, then anything goes because that circuit now becomes part of the musical instrument. If the musician (an artist) is looking for a certain sound from the instrument, who is to say they are wrong? Both odd and even harmonic distortions and the various intermodulation components all add to the richness of the note played. Hum, noise and other extraneous signals also add to the musical sound produced. You may have a different personal preference in sound than what the artist prefers, but they are not wrong in wanting that particular sound.

Generally speaking, musical instrument amplifiers are not overly concerned with the lowest possible noise because of where and how they are typically used. For that reason, a preamp with (I assume minimal gain because it is driving an input stage) does not have to have extremely low noise figures; the following amplifier stages probably don’t have more than a 60dB S+N ratio. The need for low noise is further reduced because the frequency response of most guitar amplifiers does not extend much beyond 12 kHz or so before rolling off significantly. A 741 type opamp operating with a gain of less than 4 would be overkill in this respect. (No, I’m not advocating a 741.)

If the purpose of the preamp is to add “richness” to the sound, your primary design goal would be the harmonic content generated by the preamp. The harmonic content generated depends not only on the device, but also on the exact circuit configuration and values used. Even a different manufacture of the same device type will affect the precise balance of harmonics produced. There will also be an interaction between the preamp and the existing input stage that will change the sound with the circuit configuration of the preamp and its component values. Even if the gain is more like 10 or so, noise generated by any modern opamp or a single discrete device would still be pretty negligible for this application. When you get into higher impedances, the dominant noise source is usually the high-value resistors, followed by a wide bandwidth.

Without knowing the purpose of or why you are building this preamp and what you are hoping to achieve with the preamp, it is kind of hard to make any further recommendations.

 

Check out a LT1115. I've used it in ultra low noise instrumentation lab applications.

 

Maybe he still plays old records. My Shure V15 player cartridges required that the preamp had an input resistance of 47k which was done with a discrete low noise transistor. I had a Yamaha stereo receiver that used a Japanese audio IC for its phono input.

 

Maybe he still plays old records. My Shure V15 player cartridges required that the preamp had an input resistance of 47k which was done with a discrete low noise transistor. I had a Yamaha stereo receiver that used a Japanese audio IC for its phono input.

In my book, the Shure V15 was one of the best cartridges ever produced, although many would not agree. The biggest complaint was the excessive high frequency response. I found the "excessive" high frequency response was only in comparison to the sort of dull response of the other "high end" cartridges. When loaded properly the response was actually quite flat, but the loading was somewhat critical. As I recall, it liked a slightly higher resistive loading ( ~51K for less ringing) and a bit higher capacitive loading to roll off the 'building' high frequency response. Using "cheap" 4' cables usually provided just about the right amount of added capacitance. The best thing about the V15 is its compliance. The spec was 1.25 grams maximum, but if you tracked it at 2 grams it would follow ANYTHING. Ortophons (considered the best), would follow about 95% of the records, but the other 5% it would destroy when the compliance was taxed. If you tried to increase the tracking force they would bottom out. BTW - Shure cartridges were made by Pickering with the better Pickerings being identical to the Shure cartridges except for the shape of the plastic needle grip.

The typical phono preamp of that era used 2SC458 transistors. They had a noise figure of 4 typical and a maximum of 10. You can get away with that level of noise figure because of the RIAA rolloff characteristic that very effectively reduces noise. As a comparison a 2N3904 transistor has a noise figure of 5 maximum. Because of this rolloff, a mic preamp is actually much more demanding in terms of noise than a phono preamp, despite the higher gain requirements of the phono preamp.

 

When I was a teenager, a few times my friend in a wealthy family bought the latest Shure V-15 cartridge then gave me his old one that was only 1 year old. I never wore out the needles.
Then later a compact tape recorder/player with Dolby B and C.
Then later again a portable recorder/player using a hard drive.

 

When I was a teenager, a few times my friend in a wealthy family bought the latest Shure V-15 cartridge then gave me his old one that was only 1 year old. I never wore out the needles.
Then later a compact tape recorder/player with Dolby B and C.
Then later again a portable recorder/player using a hard drive.

You will never wear out a diamond stylus playing vinyl at 2 grams tracking force. Used to play 78's at a radio station; they never showed any signs of stylus wear either and 78's have an abrasive built in to purposefully wear down the steel needles so they would match the groove shape. (That is why you need to change them after every play and ABSOLUTELY NEVER reuse a steel needle.)

 

Just out of interest...
Kicking around here somewhere, I have a 30+ year old copy of Wireless World mag that had an article on low noise mic preamp design, and that used matched power transistors as the front end.
I found an archive of the mag.
https://worldradiohistory.com/Wireless_World_Magazine.htm
And I think this is the article remembered...
https://worldradiohistory.com/UK/Wireless-World/90s/Wireless-World-1992-05.pdf page 402.

 

Just out of interest...
Kicking around here somewhere, I have a 30+ year old copy of Wireless World mag that had an article on low noise mic preamp design, and that used matched power transistors as the front end.
I found an archive of the mag.
https://worldradiohistory.com/Wireless_World_Magazine.htm
And I think this is the article remembered...
https://worldradiohistory.com/UK/Wireless-World/90s/Wireless-World-1992-05.pdf page 402.

Thanks for the links. It is always interesting how one goes about finding an "optimum" circuit design. 10 active devices seems like a lot of silicon for the task though. The use of a quality input transformer in a properly designed circuit can achieve equal or better noise performance with a lot less silicon. The key is three-fold.
1) - A transformer is able to provide exact impedance matching for greatest transfer of signal energy.
2) - A transformer is able to provide "gain" through a turns ratio.
3) - A transformer is able to achieve 1 and 2 with a lower circuit resistance resulting in lower circuit noise.

I would think that a circuit designed to take advantage of an input transformer's strengths along with state-of-the-art pro audio opamps would be the best possible way to design a low noise mic preamp today. It is too bad that modern-day engineers and technicians are taught that magnetic components are the least perfect components so should be avoided at all costs where possible. Their minds become closed to the possibilities of a transformer.

 

I'm sorry, "matched impedances" has nothing to do with optimum signal-to-noise, besides when you match impedances you lose half the power you generated in the source, why is that "optimal"? That's actually the reason professionals DON'T for example design mic input transformers to "terminate" a mic with a 200 ohm source impedance, the correct solution is a "bridging" impedance. Now optimum input transformer design is one of the hardest things to do correctly, but in a properly designed input stage with a properly designed transformer, the transformer itself can endow the circuit with 20 or more extra decibels of what's called "noise-free gain" over a direct connection, now that may be less critical nowadays but back when the best IC parts available were the NE5532 or the RC4136 they were essential. Oh and don't expect to find an accurate description of this technique in a circuit design textbook, the academics have never got this right, not in any book I'm aware of anyway. (The other problem with cheap mic transformers is instead of providing a proper magnetic shield to block external fields like hum they deliberately eliminate the air gap and use a "soft" magnetic material for the core, so at high levels and low frequencies the core saturates and sounds horrible.)

 

I'm sorry, "matched impedances" has nothing to do with optimum signal-to-noise, besides when you match impedances you lose half the power you generated in the source, why is that "optimal"? That's actually the reason professionals DON'T for example design mic input transformers to "terminate" a mic with a 200 ohm source impedance, the correct solution is a "bridging" impedance. Now optimum input transformer design is one of the hardest things to do correctly, but in a properly designed input stage with a properly designed transformer, the transformer itself can endow the circuit with 20 or more extra decibels of what's called "noise-free gain" over a direct connection, now that may be less critical nowadays but back when the best IC parts available were the NE5532 or the RC4136 they were essential. Oh and don't expect to find an accurate description of this technique in a circuit design textbook, the academics have never got this right, not in any book I'm aware of anyway. (The other problem with cheap mic transformers is instead of providing a proper magnetic shield to block external fields like hum they deliberately eliminate the air gap and use a "soft" magnetic material for the core, so at high levels and low frequencies the core saturates and sounds horrible.)

I don’t think we should hijack Dynaman’s thread, so I will keep this brief. I actually agree to everything you have said except for “you lose half the power you generated in the source, why is that optimal?” Because with any other termination impedance you lose even more POWER.

In the world of voltage amplifying devices (i.e. vacuum tubes); generally a bridging circuit is the best way to go because you transfer the most VOLTAGE into the load with minimal current requirements. In the world of current amplifying devices (i.e. bipolar transistors) you want to have a circuit that maximizes the CURRENT transferred and minimizes voltage requirements of the devices. But the telephone, radio, motion picture and recording industry standards evolved during the age of batteries and vacuum tubes where achieving high current was a premium, but high voltages were relatively easy to attain. In the RF world maximizing POWER transfer is of the paramount concern along with minimizing reflections that can degrade (distort) the signal, so (vacuum tube) RF circuitry evolved accordingly.

As with any design, there are several ways it can be attacked and the optimum design is the one that come the closest to finding the balance which satisfies all requirements and priorities. If the overriding priority was to maximize energy efficiency, I’m sure things would have evolved differently.

There was one magazine article I recall reading that explained the technique of optimizing with an input transformer. Unfortunately, I don’t remember exactly where I read it. It was about 30 years ago, and probably EDN magazine.

 

I mostly agree but the reason you optimize voltage to best preserve signal-to-noise isn't because of how anything "evolved". It's because that's how you allow the transformer to provide you with noise-free gain, because you're taking optimum advantage of the properties of the transformer. I suppose you could postulate there "ought to be" a dual circuit where you could have the transformer drive an ultra-low impedance output winding into the inverting input configuration and get super noise performance but it wouldn't work with an actual transformer, it would be the dual equivalent of a transformer which doesn't really exist. I used to have a subscription to the Journal of the Audio Engineering Society and I don't recall the topic ever being discussed there, I believe it has been described in some publications, the problem is the topic isn't based on "pure theory" which means you can't summarize the solution to the problem with an equation so the academics aren't interested in publishing it. All I meant is that "optimizing" by losing half the power DOESN'T get you best SNR but that's what the academics insist on having you learn, because they've never been shown that it's a total fallacy! Also do you know why mic preamps need such a low noise floor? It's because many of your favorite artists (I could name names but I won't) naturally have a pretty "nasally" quality to their voices when close-mic'ed, so the producer will have them stand as much as TEN FEET BACK from a large-format condensor mic to keep that tone from getting picked up when they do their vocals, that drops the sound level but you still have to pick up that track with an acceptably high SNR to make a decent record out of it. You know, necessity being the mother of invention...

 

I concede a transformer does not give better S/N due to exact impedance matching. I went through the math for a couple of examples based on Johnson noise and amplifier gain, assuming perfect op-amps for simplicity The example with a 20X bridging impedance provided about 5.7dB improved S/N over the matched load. Both impedances were reflected loads on the transformer secondary. I also tried a current amplifier configuration, but it was far worse than either voltage amplifier configuration. Perhaps if the microphone was designed as a current source instead of a voltage source there might be some advantage, but that is not how they are designed. (They would probably be more susceptible to noise pickup in the studio.) It has been about 40 years since I calculated noise in a circuit so it was a nice exercise – thanks for the refresher.

Interesting you mentioning keeping the artist 10’ away because of their nasal quality. Listen to the 1937 recording of Marie by Tommy Dorsey (Victor 25523-B). About 2/3 the way through during the chorus, Bunny Berigan plays the famous solo trumpet that completely changes the mood of the song from an almost waltz into hot jazz with 2 notes. He was placed 30’ from the microphone because he was such a strong player. Back in those days the mixing console was changing the physical placement of the artists around a single microphone; and of course direct to disk.

 

An artist singing 10' from the mic has a longer distance reflection off the floor to cancel the nasal frequencies that have a shorter distance to the mic (if the floor is hard).

 

In the same way, the microphone itself in relation to the amplifier.
Defining what a moving coil or a moving magnet in a coil will produce that is what flux pattern over time.
How well an amplifier can fill a room that is conducive to the sound waves is in part the volume and shape.

Taking away the sound pedal and doing surgery on the guitar's circuit itself in the same way as the microphone
can lend itself to understanding particular effects. Probably this guitar is not a good victim for surgery into it's circuit
there might be a damaged guitar or microphone in salvage somewhere that could be explored.

In many ways the entertainers came from an era where amplification was less reliant than stage performance
without. The size of the gestures used on stage without cameras as well as the volume needed took a stronger voice that carried
was in some ways an advantage in delivering wider variety of sounds naturally. Going back further the round pancake shaped microphones
used before the magnetic coil influenced sound reproduction. It was all of the equipment together developed when sound fidelity
made rapid advancement.