https://youtu.be/3fdkUvnmI9w
I thought this was rather Brilliant.
Single ended only.
Hum neutralizing circuit
Moderators: pompeiisneaks, Colossal
Re: Hum neutralizing circuit
Masco, nice... have you tried this on any of your SE builds?
I did on this circuit and it did not work out. I did elevate the virtual tap to the Cathode of the power tube and that did further lesson the very low hum
A
I did on this circuit and it did not work out. I did elevate the virtual tap to the Cathode of the power tube and that did further lesson the very low hum
A
You do not have the required permissions to view the files attached to this post.
Re: Hum neutralizing circuit
I remember a clever hum neutralizer article that can get rid of most hum issues. I'll dig up a link as soon as I can find it again.
This thing took a divided -down line voltage signal, maybe a couple of volts, then half wave rectified it. The half wave rectified signal was fed to the inputs of bandpass filters at 2X, 3X, 4X line frequency, and then each of the filter outputs as well as the line signal was fed to a variable phase shift stage and a final volume control for each before a mixing stage.
This sounds complicated, but only took a few quad opamp chips.
In use, you fed the output of the filter setup to an input on your amp with all the filter volumes at zero. Then you turn up the volume and adjust the phase control for each frequency while listening to the amp's hum. This let you dial in a canceling anti-hum by ear for the line frequency and the next few harmonics. Again, sounds complicated, but in practice it's a quick run-through as you tune it by ear.
N.B.: found one version of it, although not the article I remember. It's the second section of this: https://ethanwiner.com/filters.html
I also notice that I got the functional order a little wrong, as the volume controls are at the front end, not after the filters.
This thing took a divided -down line voltage signal, maybe a couple of volts, then half wave rectified it. The half wave rectified signal was fed to the inputs of bandpass filters at 2X, 3X, 4X line frequency, and then each of the filter outputs as well as the line signal was fed to a variable phase shift stage and a final volume control for each before a mixing stage.
This sounds complicated, but only took a few quad opamp chips.
In use, you fed the output of the filter setup to an input on your amp with all the filter volumes at zero. Then you turn up the volume and adjust the phase control for each frequency while listening to the amp's hum. This let you dial in a canceling anti-hum by ear for the line frequency and the next few harmonics. Again, sounds complicated, but in practice it's a quick run-through as you tune it by ear.
N.B.: found one version of it, although not the article I remember. It's the second section of this: https://ethanwiner.com/filters.html
I also notice that I got the functional order a little wrong, as the volume controls are at the front end, not after the filters.
Re: Hum neutralizing circuit
with proper engineering there is no need for any hum canceling systems. also for S.E. no need to mention PP.
-
dorrisant
- Posts: 2628
- Joined: Tue Sep 21, 2010 1:27 pm
- Location: Somewhere between a river and a cornfield
- Contact:
Re: Hum neutralizing circuit
"Education is what you're left with after you have forgotten what you have learned" - Enzo
Re: Hum neutralizing circuit
I'm cursed with one of those inquiring minds.
I went off and looked at the link, downloaded and read the user manual. Hmmm...
It pretty much sticks with "just plug this in and all audio hum is removed." It's not clear exactly how that would happen. The manual is very clear that you must use the included 7.5Vac power adapter for the thing to work, so that's a vote for it looking for something from voltage-lowered AC mains, in addition to whatever power it needs.
There's no comment on how the thing works inside, although the note that it's a Class B computing device at the end of the manual seems to imply that there is some kind of switching logic inside. Maybe a baby DSP doing ... something??
It's tempting too speculate that with a DSP you could sense the AC line voltage from the 7.5Vac, derive the several harmonics as in the Winer hum canceller and then somehow use that to cancel the line frequency and harmonics. I did some speculative design on that approach, and ran into issues with - how does it know when it's cancelled enough, and where?? The manual is very clear about putting this thing right after the guitar and before going into pedals. It sure seems to me that it could not know what hum the amp adds, if any, because it doesn't get to look at the amplifier's output. Would it work equally well with all amps, including a carefully de-hummed masterpiece and an un-isolated widowmaker?
Occam's Razor says that the simplest explanation is nearly always the correct one. Does that mean that "advertising enhancement" is the primary operating method?
I'm nearly tempted to buy one and look inside. Ummm... nearly.
Re: Hum neutralizing circuit
Ah. Thanks.
That's a SHARC DSP, which is relatively higher performance than I was speculating. Given that, the speculation on freestompboxes is close to what I'd guess. It likely looks at the incoming AC line, then tries to remove power line harmonics with super sharp notch filters.
That does work, as long as it can avoid notching out the desired signal too, and can track the actual incoming AC line's actual frequency. With a higher performance DSP it might be able to make good guesses about the incoming AC signal and stop notching if there's enough signal at a power line harmonic frequency to cover it up. That might even account for the "normal/strong" switch, turning predictive notching on and off. Maybe.
Unless there is more magic in there than I can see or guess at, it has the problem I was thinking of - it can't detect or measure any hum added later in the chain after the device. I'm guessing that it can't affect amplifier-added hum at all. Just a guess.
That's a SHARC DSP, which is relatively higher performance than I was speculating. Given that, the speculation on freestompboxes is close to what I'd guess. It likely looks at the incoming AC line, then tries to remove power line harmonics with super sharp notch filters.
That does work, as long as it can avoid notching out the desired signal too, and can track the actual incoming AC line's actual frequency. With a higher performance DSP it might be able to make good guesses about the incoming AC signal and stop notching if there's enough signal at a power line harmonic frequency to cover it up. That might even account for the "normal/strong" switch, turning predictive notching on and off. Maybe.
Unless there is more magic in there than I can see or guess at, it has the problem I was thinking of - it can't detect or measure any hum added later in the chain after the device. I'm guessing that it can't affect amplifier-added hum at all. Just a guess.
-
dorrisant
- Posts: 2628
- Joined: Tue Sep 21, 2010 1:27 pm
- Location: Somewhere between a river and a cornfield
- Contact:
Re: Hum neutralizing circuit
A customer of mine brought this device to my attention just a few weeks back. He says that it cuts out certain notes of the guitar, which we figured was around 60hz. It is mostly noticable in the 2nd mode, where you probably can't tell it by ear in the the lesser mode. Has me interested anyway. I will probably end up with one sooner or later.
Bill, had the problem RG mentioned. It only works on what is coming into the unit. I suggest putting it in the FX loop. He says that he will use it more knowing that and has several amps he want to have me install passive loops in.
Bill, had the problem RG mentioned. It only works on what is coming into the unit. I suggest putting it in the FX loop. He says that he will use it more knowing that and has several amps he want to have me install passive loops in.
"Education is what you're left with after you have forgotten what you have learned" - Enzo
Re: Hum neutralizing circuit
Yep, that mostly confirms that its using notch filters, given that you lose certain notes. A normal guitar can't get down to 60 hz in normal tuning, as low E is 82Hz, but some slack tunings could get to that. Probably the losses are near 120Hz, 180Hz, 240Hz, etc. and it would vary by how close your guitar was to modern A = 440 Hz.
Using a convenient note-to-frequency chart from https://mixbutton.com/mixing-articles/m ... ncy-chart/ the likely interferences for an accurately tuned guitar are likely at low A#/Bb (116.54 Hz) or B (123.47 Hz) ; F..F#Gb ( 174.61 Hz to 185.00 Hz) ; A#/Bb and B again ( 233.08 Hz, 246.94 Hz ) and possibly D (293.66 Hz) to D#/Eb(311.13 Hz).
The necessary filter Qs are modestly insane. To stop 120Hz with only a half-power loss at Bb and B, you need a filter with a Q of about 120Hz/7Hz or a bit higher than 17. This is crazy difficult to do in analog, but not too difficult in a DSP; although you do have to pay attention to some of the finer points in getting very high Q filters.
CBS (IIRC) tried to use a series of notch filters as a copy protection scheme back when the fight over copying music was copying to tape. It sounded horrible with music notching issues and big phase shifts near the filter centers. Maybe the DSP programming in this one can do better. But the issue will still be that variable guitar tuning will wander notes over to the notch frequencies, as we don't use the power line as a reference frequency for tuning our guitars. Well, I don't ...