Unnatural sounding 'fizz' on sustained notes

Rex said:
And there's one of my pet peeves. Guitarists use the word gain as if it means distortion
Click to expand...
Probably at least partly because many of us don't have much chance of understanding (without some deep study) that description of what gain actually is.
 
  • Like
Reactions: Rex
Tubes (and transistor) are active component, depending on circuit values have a GAIN (20 to 60:1 in a typical tube amp). Let's 20:1 the gain of each stage of a 12AX7. A signal of 1 volt get out of stage 1 at 20v, into stage2, out at 400v. The power supply is, say, 200v. So the second stage CLIP AND SATURATE.
Pots are passive component, a voltage divider, than can varies output v from input value down to 0. If we put the a pot between 2 stages, label it gain, put if halfway, the second stage will have 10v input 200v out, at the verge of breakup.

Well, it's not that simple, but it gives you an idea! Tubes are complex, component are not ideal: after about 100 years of amp tech knowledge evolution, tube amps nowadays use each imperfection to reach a desired behavior (compression, sag, clipping, saturation, brightness, ...). Digital tech is a lot younger, and can copy those imperfections in lots of different ways. I hope we will be able, in the digital domain, to mix those imperfection to ours artistic needs (like fizz).
 
Rex said:
That's my point. It doesn't apply here. Open- and closed-loop gain apply to any amplifier, but @Cuiken is approaching this as a gain-bandwidth product issue, which is pretty much an op-amp concern.
Click to expand...
For what it is worth, open loop gain applies to any amplifier that uses negative feedback to improve linearity (not just op-amps). The fact that open loop gain will always fall with frequency means higher frequency signals will show higher distortion and intermod distortion than lower frequency signals. Hence my question about choice of frequency of test tone.

I was, and remain, curious as to whether the gain bandwidth of a classic tube amp is sufficiently large to demonstrate similar distortion characteristics for a 1kHz tone when compared to, for example, an 80Hz tone (where, as a guitar player, I am most interested). The answer may be an emphatic 'yes' or the choice of tone may simply have been driven by the available FFT bin width of the True RTA analyser.

I wasn't trying to kick up a fuss here, just trying to make sense of my experience of using my Fractal.
 
Reducing master volume on my dirty amp (currently ODS 100 HRM) drastically reduced this issue for me. The resulting sound is slightly different, of course, but I was able to adjust elsewhere and get back to the feel I like. Incidentally, it didn’t take much - I reduced it from 7.5 to 6.
 
Yes, the ODS 100 HRM is particularly notable for this characteristic. I’ve done exactly the same and rolled back the master volume. I have also increased the grid bias of the power stage, reduced the hardness of the preamp tubes, dialled back the output EQ of the amp just an smidge and dug around for a less bright IR.

After all this, I guess I have probably changed the original character of the model a lot but I have something that is still bright with that hint of fuzz distortion on top of the overdrive that I associate with these amps.

Took a lot of tweaking to get there but I do like the end tone I arrived at.
 
Last edited:
Cuiken said:
Yes, the ODS 100 HRM is particularly notable for this characteristic. I’ve done exactly the same and rolled back the master volume. I have also increased the grid bias of the power stage, the hardness of the preamp tubes, dialled back the output EQ of the amp just an smidge and dug around for a less bright IR.

After all this, I guess I have probably changed the original character of the model a lot but I have something that is still bright with that hint of fuzz distortion on top of the overdrive that I associate with these amps.

Took a lot of tweaking to get there but I do like the end tone I arrived at.
Click to expand...
Sounds interesting.
Post the preset?
 
Cuiken said:
Yes, the ODS 100 HRM is particularly notable for this characteristic. I’ve done exactly the same and rolled back the master volume. I have also increased the grid bias of the power stage, reduced the hardness of the preamp tubes, dialled back the output EQ of the amp just an smidge and dug around for a less bright IR.

After all this, I guess I have probably changed the original character of the model a lot but I have something that is still bright with that hint of fuzz distortion on top of the overdrive that I associate with these amps.

Took a lot of tweaking to get there but I do like the end tone I arrived at.
Click to expand...
Rolling back the master volume will move from power amp to preamp distortion. That could have lots of impact on tone.
 
Jim Amsden said:
Rolling back the master volume will move from power amp to preamp distortion. That could have lots of impact on tone.
Click to expand...

Indeed. Power and pre clipping give very different distortion profiles. That said, it's an amp model so it's no different to how I might use an amp in the real world. The original amp has a master volume, I doubt I'd be running that at full tilt if I owned the real amp. Or, at least I'd roll it back if I wasn't enjoying the power amp distortion.

Personally, I find the power amp distortion profile to be, for want of a better term, overly 'fizzy' so I am inclined to roll the Master back in most models (maybe this would be true with the real amps too, I don't know). For some models, I guess this is not truly authentic (where the real amp would not have a master volume for example) but that's surely where a modeller shines. You can do stuff that wouldn't be possible with the actual amp.

Like I say, I've found careful use of Master Volume, Grid bias, preamp Tube Hardness and Saturation (to compensate in some cases for lack of power tube clipping) gets me tones I'm happy with. Authentic tones? Probably not but, for me, that's where the Fractal shines.

Naturally all this is subjective. YMMV and all that.
 
For what it's worth, attached is my ODS_100 drive preset. You can AB against the stock preset (073) to see the reasult of the changes I've made.

I'm not going to claim for a second that it's 'better'. But it's more to my taste and sounds very convincing through headphones and my HR108s..

I'd forgotten that I'd moved over to the Dyna-Cab for this preset (which can often be a bit brighter) but I tamed that a bit with mic placement and choice of the 'Rumble EV12L' IR.

Feel free to have a compare.
 

Attachments

FractalAudio said:
You'd probably be happier with a Quad Cortex. Our modeling stresses accuracy. The "fizz", as you call it, is caused by power tube grid clipping and has a somewhat disassociated quality.
Click to expand...
Well, I played and played with it over the holiday break but, ultimately, just didn't get on with my FM9. So, I sort of took Cliff's advice above and purchased a QC.

Not going to claim it is 'better' but I definitely find it easier to dial in tones that don't exhibit the 'fizz' that I am hearing on my FM9. So, FM9 is sold and I'll see how I get on with the QC.

Thanks for all the advice in this thread and a happy new year!
 
You must log in or register to reply here.
Back
Top Bottom