I'm "bettering" my Fender blues Jr. So far I've changed my Filter Capacitors, Coupling Caps, Tone Stack values, and Plate resistors with better quality components.
The difference is massive! The sound is full and rich with a lot more pleasent low end.
The next thing to tackle is some high end noise floor hiss. It's nothing particularly strong but it is definitely when I play very lightly.
I'm considering changing V1 from a 12AX7 (100% Gain) to a TAD 5751 (70% Gain).
Rat Valves also removed the *R1-1M resistor (*R2 is a 10k and then it goes to *V1). Would this benfit the circuit in any way? Lower hiss?
If not should i replace them with some better low noise resistors?
My circuit board is the third image.
The difference is massive! The sound is full and rich with a lot more pleasent low end.
The next thing to tackle is some high end noise floor hiss. It's nothing particularly strong but it is definitely when I play very lightly.
I'm considering changing V1 from a 12AX7 (100% Gain) to a TAD 5751 (70% Gain).
Rat Valves also removed the *R1-1M resistor (*R2 is a 10k and then it goes to *V1). Would this benfit the circuit in any way? Lower hiss?
If not should i replace them with some better low noise resistors?
My circuit board is the third image.
Attachments
Moved to Instruments and Amplifiers where the musical instrument and amplifier folks hang out. 12AX7 / ECC83 is pretty good regarding noise. It has a relatively low transconductance but its 1/f noise is remarkably low according to measurements of Merlin Blencowe ( @Merlinb ). You might have a bad one, though; some valves are functional, but much more noisy than others of the same type.
Removing R1 eliminates the thermal noise current it injects into the guitar, but it also makes the bias point with no guitar connected unpredictable. 1 Mohm is not an unusually low value, so it seems unlikely that its noise would be unacceptable. I wouldn't remove it.
Disclaimer: I have some practical experience with valve circuits, but none with guitar amplifiers.
I have built 22 guitar amps, repaired a bunch, helped new builders with their builds. Adding to what MarcelvdG noted, the 12AX7 has a pretty good noise level in general, but gain between tubes varies a lot from their quoted gain. You should try a couple of different tubes and see if you notice a change in hiss. The overall gain in a circuit, as well as the effects of the tone stack and any other tone shaping also affects hiss. Many tube makers (mostly in the pre-Ukraine war/Russian embargo years) sold 12AX7 tubes graded for gain.
If you are playing at very low, "bedroom" levels, and you have any sort of high-gain circuit, or are using the higher-gain channel in a 2 channel amp, you will notice the hiss more. Most of my amps are higher-gain amps (Trainwreck style, Marshall style, Komet style), so hiss is a given. When I can't turn up the volume and don't want to hear the hiss, I will use my RedPlate or one of my lower-gain amps.
The change to a lower-gain tube will generally help with hiss, as will changing out any carbon comp or carbon film resistors to metal film, but some players don't like how much parts changes like these can clean up the amp. My preference, for example, is for carbon film, with selected carbon comp in a few positions, and polyester caps rather than polypropylene, as I like the distortion characteristics of the former compared to the latter, with exceptions for certain types of builds (e.g. D-style). It sounds like you may prefer the clean, quiet sounds of a lower-gain amplifier, so after you try a different 12AX7 tube, try the 5751 and see what you think.
I highly recommend against removing the input grid-ground 1 meg resistor, for the reasons cited by MarcelvdG. It is a high value and could contribute thermal, shot or contact noise, so you could try a metal film in that position, and in the plate load positions if you haven't already, but try tube rolling first.
Oh - one more thing. Are you using single coils or humbuckers?
If you are playing at very low, "bedroom" levels, and you have any sort of high-gain circuit, or are using the higher-gain channel in a 2 channel amp, you will notice the hiss more. Most of my amps are higher-gain amps (Trainwreck style, Marshall style, Komet style), so hiss is a given. When I can't turn up the volume and don't want to hear the hiss, I will use my RedPlate or one of my lower-gain amps.
The change to a lower-gain tube will generally help with hiss, as will changing out any carbon comp or carbon film resistors to metal film, but some players don't like how much parts changes like these can clean up the amp. My preference, for example, is for carbon film, with selected carbon comp in a few positions, and polyester caps rather than polypropylene, as I like the distortion characteristics of the former compared to the latter, with exceptions for certain types of builds (e.g. D-style). It sounds like you may prefer the clean, quiet sounds of a lower-gain amplifier, so after you try a different 12AX7 tube, try the 5751 and see what you think.
I highly recommend against removing the input grid-ground 1 meg resistor, for the reasons cited by MarcelvdG. It is a high value and could contribute thermal, shot or contact noise, so you could try a metal film in that position, and in the plate load positions if you haven't already, but try tube rolling first.
Oh - one more thing. Are you using single coils or humbuckers?
Interesting, I will try some different AX7's to see if the tubes I have are the problem.
I play a thinline 72 telecaster which has some reissue wide range humbuckers. The amp is usually a clean pedal platform.
Thank you for the reply. 🙂
I play a thinline 72 telecaster which has some reissue wide range humbuckers. The amp is usually a clean pedal platform.
Thank you for the reply. 🙂
Good place to start, I think. A quick online search for this problem resulted in several confirmations of this suggestion, plus; loose sockets (I usually find this causes squealing, not hiss), and a bad batch of caps (would expect hum, not hiss).
R1 is in parallel with the guitar pickup so its current noise is shorted through the pickup and negligible.
R2 (10k in series) is a direct source of voltage noise however and might be worth reducing in value.
R2 (10k in series) is a direct source of voltage noise however and might be worth reducing in value.
A guitar pick-up is hardly a short circuit.
@PRR once posted actual impedance values for various types of guitar pick-ups, but I can't find it anymore. As far as I recall, the inductance was of the order of 2 H, or even 8 H for humbuckers. I haven't the faintest idea what a humbucker is, but apparently it's quite inductive.
@PRR once posted actual impedance values for various types of guitar pick-ups, but I can't find it anymore. As far as I recall, the inductance was of the order of 2 H, or even 8 H for humbuckers. I haven't the faintest idea what a humbucker is, but apparently it's quite inductive.
Last edited:
Well at higher frequencies yes, you're right that the inductance makes the pickup high impedance, but 1M is a lot of impedance to compete with and in practice the pickup is significantly attenuating the effect of R1's current noise. The reason most guitar amps have a switch to ground un-plugged inputs is the large increase of hiss that would be present without the pickup to reduce it. A metal film 1M resistor alone generates 130nV/√Hz, or about 18µV rms in the audio band, carbon ones can be a lot worse.
A pickup with 10H of inductance has an impedance of 62k or so at 1kHz, which is a lot less than 1M.
A pickup with 10H of inductance has an impedance of 62k or so at 1kHz, which is a lot less than 1M.
Well your not going to get much better than a 12ax7 maybe a 5751 as said above. That said there are good and bad 12ax7's so if you have a few try them. Its a pretty standard input. The guitar does need to work into a high impedance input to make the pickup resonate with the self capacitance and not damp the strings. The 1M pot of the guitar does not help unless its set to max volume. If you want quieter maybe a jfet cascode with the tube at the top.
At 100 kohm of inductive reactance, the 1 Mohm parallel resistor produces about as much noise as the 10 kohm series resistor (just calculate the real part of the reciprocal of 1 uS - 10 j uS).
Agreed, but - with the parallel capacitance of several 100pF the resulting resonant peak may reach 1MOhm, limited to ca 250k by passive potentiometer wiring. Thus an input impedance of 1MOhm is appropriate.
Do you mean that there is a 250 kohm potentiometer between the guitar pick-up and the amplifier? They really do their utmost to make the noise level as high as possible!
Actually it will still inject noise current into the signal path, although its noise contribution will be far less than when it is not set to its maximum.
I had to laugh at that. 😀
I've had the same thought, and certainly the design is very far from optimal when you focus on thermal noise. But there are some practical reasons for this design, and most crucially, it works well enough to get the job done.
Believe it or not, the onboard volume pot can have even more resistance than 250k. Most guitars with humbucker pickups (which have much more series inductance) tend to use 500k or 470k volume pots to avoid loading down the guitar pickup too heavily. That means the pot itself can create up to 125k Thevenin source impedance when set to half-resistance. Factor in the additional winding resistance of the pickups, and the total source impedance can rise to around 140k, worst case.
So why is this very imperfect design used?
1) For starters, the first commercially manufactured electric guitars with a practical pickup design date back to 1932 ( https://www.vintageguitar.com/3588/ro-pat-ins-first-electric-spanish/ ).
In that early era of electronics, I don't think thermal noise was on anyone's mind, least of all the tinkerers trying to electrify guitars. If you could actually get a noise out of the speaker, that was already success enough!
Keep in mind that few, if any, of these tinkerers were actually college-educated in electronics theory. Most knew just enough to heat up a soldering iron and build or repair a (tube) radio. Early guitar amps were pretty well identical to the audio sections of radios or record players.
2) Having a volume control on the guitar itself allows the guitarist to quickly and unobtrusively make repeated changes to the signal level. This has many uses, including switching from rhythm guitar (strumming six-string chords) to single-note lead guitar (picking only one string). To keep the chords from being far too loud, and the solo far too quiet, something has to be done to adjust the signal level.
3) For the first couple of decades (1930s - 1950s) most electric guitars were played "clean", meaning amplifiers didn't have all that much gain. Hiss wasn't really a problem. (Maybe people were also used to hearing hiss from records and tape recorders, so a little hiss from a guitar wouldn't be objectionable.)
4) I've never owned a guitar amplifier with a noise floor low enough to actually hear thermal hiss from the guitar volume pot. In principle the hiss should increase when you turn the volume pot to about half-resistance (raising the Thevenin source resistance to its maximum). In practice, I've never heard this.
For me, then, thermal noise from the guitar's volume pot has never been an issue. The much bigger issue is that the combination of guitar cable capacitance and volume pot source impedance create a low-pass filter. Turn the (guitar) volume down from maximum, and a lot of the high treble disappears from your guitar sound, particularly if you're using a long guitar cable.
Those first Ro-Pat-In e-guitars were sold nearly a century ago, and the crudity of the electronics makes sense in context. But why hasn't anyone updated this antiquated design? Why not just put a modern low-noise preamp into the guitar, eliminating the issue of volume potentiometer source impedance?
I think the reasons for this are much subtler. Many bass guitars do contain onboard preamps nowadays. Bass guitarists have been relatively quick to embrace new and improved technology, starting with pioneering companies like Alembic in the 1970s ( https://en.wikipedia.org/wiki/Alembic_Inc )
But very few regular (tenor) e-guitars have active electronics onboard. Part of this is definitely cultural - most e-guitar players remain much more conservative than bassists, for reasons I've never entirely figured out.
There have been various attempts to build tenor e-guitars with onboard preamps, including EMG's "active pickups" from the 1980s, which integrate a preamp into the pickup itself, and more contemporary Fishman Fluence pickups (a low-output, low-impedance printed-circuit "coil" combined with onboard active filtering and pre-amplification).
Those EMG pickups did became popular, but only for use in a couple of fairly niche heavily distorted styles of music. The pickups didn't sound like traditional e-guitar pickups, and had zero chance of becoming popular with, say, guitarists in country music.
The newer (2013 - present) Fluence pickups are an attempt to create pickups that sound entirely traditional, but with less noise, lower output impedance, and fewer loading effects. They're still available ten years after being introduced, so somebody must be buying enough of them to keep the product line alive. But I've never seen a single one in use by any musician I've ever met, so I guess they're a pretty niche product, too.
-Gnobuddy
If you ever listened to condenser microphones from the early 1930's (not old recordings of them, but how the ones that still work sound now), you know that there must have been people back then who managed to design reasonably low noise circuitry somehow, even though Johnson and Nyquist's famous articles about thermal noise had only been published a few years earlier, in 1928. (Geertruida de Haas-Lorentz already wrote something about thermal noise in her 1912 PhD thesis, by the way.) It could well be that those people didn't get involved in electric guitars, though.
Regarding your statement: "4) I've never owned a guitar amplifier with a noise floor low enough to actually hear thermal hiss from the guitar volume pot. In principle the hiss should increase when you turn the volume pot to about half-resistance (raising the Thevenin source resistance to its maximum). In practice, I've never heard this." Is that because you only use poorly designed guitar amplifiers or because you play at limited volumes? I guess the first, as you write about the amplifier's noise floor not being low enough. It should be fairly easy to design an amplifier with a lower noise floor.
In any case, if the thermal noise of 62.5 kohm...125 kohm straight in series with the input is small enough not to matter, the noise of just about any healthy valve is bound to be negligible.
Regarding your statement: "4) I've never owned a guitar amplifier with a noise floor low enough to actually hear thermal hiss from the guitar volume pot. In principle the hiss should increase when you turn the volume pot to about half-resistance (raising the Thevenin source resistance to its maximum). In practice, I've never heard this." Is that because you only use poorly designed guitar amplifiers or because you play at limited volumes? I guess the first, as you write about the amplifier's noise floor not being low enough. It should be fairly easy to design an amplifier with a lower noise floor.
In any case, if the thermal noise of 62.5 kohm...125 kohm straight in series with the input is small enough not to matter, the noise of just about any healthy valve is bound to be negligible.
Last edited:
Greg Koch, guitarist, podcaster, etc., is a Fluence advertiser/user. I have a set for one of my Strats, and I've talked with a few guitarists who swear by them, so I think they have a pretty strong market.
Guitar pickups range in impedance between about 8k and 15k depending on type and desired output/sound. Volume controls on guitars are essential, so as gnobuddy stated, 250k pots are generally used for single-coil pickups that tend to be wound on the lower end, and 500k are used for double-coil (humbucking) pickups. Guitarists also use volume foot pedals with essentially similar characteristics.
The electric guitar was invented around the mid-1930's and was initially only intended to amplify without distortion that is so prized today. Early amps made plenty of undesired noise, but were low gain.
I agree with gnobuddy's observations, having used guitar amps for 55 years, through the many different eras of design, including gain and distortion characteristics. With single coil pickups, the dominant noise is typically hum. With any guitar and any higher-gain amplifier, there will inevitably be hiss, almost always caused by what is happening in the tubes, and compounded by choice of resistors and tone circuit design. Old amps used CC, then CF, and amps built for cleaner tones or more controlled preamp-generated distortion typically use quieter metal film.
Guitar players needing a quiet amp for small venues and lighter music generally select lower-gain amps and don't add noise-inducing pedals to their signal chain, such as boost pedals that add lots of hiss. However, those who play loud, heavy, more distorted rock music types use their volume knobs to control noise between songs or in quiet passages, and when the music is cooking, no one hears hiss. I can turn up my amps and get the typical gain-induced hiss and as soon as I start playing the noise is swamped. Add other musicians, and the hiss becomes noticeable only when everyone stops playing.