Amp Feel, Speaker Bobbing, Tube, Solid State and Digital

[loudthud]

The first pro quality amp I owned around 1970 had JBL speakers with the aluminum dome on each speaker. Right away I noticed that when the amp was cranked up, there was a low frequency bobbing of the cone that accompanied each picked note. The cone slowly moves in and out over a distance much greater than the vibration required to reproduce the note or chord being played at the volume the amp is telling the speaker to reproduce.

Clearly, no DC can come through an output transformer, but as Steve Conner pointed out, a non-symetrical squarewave has a DC component. The DC dissipates slowly and the baseline of the non-symetrical squarewave resets off center to the tops and bottoms of the non-symetrical squarewave. The non-symetrical squarewave comes from baseline shift within the amp when grid current flows in various stages in the amp. Most of the time you can't see the dynamic behavior with an oscilloscope unless you are using digital storage scope capable of recording long records. Below I have attached some single shot photos of a 50W Marshall driven by a burst function generator.

MVC-319F shows the non-symetrical wave and how the bottom makes a shark fin when drive is increased.
MVC-320edit shows how the wave changes at the beginning of the burst and starts to shark fin.
MVC-321F shows a longer view of MVC-320edit.

My theory at this point is that speaker bobbing is one component of tube amp feel and touch sensitivity. Gain compression and power supply sag also are components that may be part of a complex interaction.

Question for Daz: Does the Mustang III amp exhibit speaker bobbing?

[Enzo]

Cool.

[daz]

Question for Daz: Does the Mustang III amp exhibit speaker bobbing?

Me? You're asking me? If you are asking me if the mustangs PA exhibits it.....you know i'm a electronic cripple so i dunno ! I think looking at it logically my answer would be as follows. What you are describing is something that happens in a tube PA right? It causes a certain sound and feel. If you ask me whether my mustang's PA does that i couldn't tell you, but i will take your word that it can't. But see, heres the thing, and i think you realize this but i'll say it anyways. The people who develop these digital circuits today know exactly what they have to accomplish unlike in the past. So today it's a matter of figuring out how to accomplish whats needed, not what to accomplish. And once they know that i'm sure they also look at things like you mentioned and consider it from the point of view that if the PA can't be made to do that, then make the PA clean and accurate and figure out a way to make the amp model include that feel or whatever it is so the tone and feel is already there and the PA only had to accurately amplify it. Remember van halen in the early days? He would run his marshal into a attenuator and create the tone he wanted then into SS power amps and finally marshall cabs. Everything was there....the preamp tone, the distorted PA tone, and the cabs. But the power amps were in between those doing nothing more than amplifying the already fully created guitar tone aside from the speakes which came later.

So the same kind of thing is going on here but it's created digitally, the same as the preamp of a tube amp is. So i'd say probably no, i doubt the mustang's PA does what you describe. But the point of digital modeling in the R&D of it is to find a way to reproduce the same feel, dynamics and tone of a tube amps in the digital domain. It really has nothing to do with a preamp and a pA that need to do the job the same as a tube amp.....it's just a digital picture of the entire path from input to speaker. And it has about a dozen different speaker sims. I'm not trying to convince you of what the mustang is capable of because you won't believe me till you try one thats been tweaked well. And thats not likely unless YOU yourself tweak it by spending hours with it. I'm just starting to really nail the feel and tone to my expectations now, and i know i'll get it even better. But unless you do the same you will remain a non believer. I was, and big time. I'm still not convinced it's AS good in all ways. I feel it is BETTER in MOST ways and possibly in all, tho i will remain skeptical till i've had more time. After all, i've only had it a week. And as i just posted in the other thread, i found the recipe for clean tones and just nailed it big time. I'm getting clean tones now that just floor me. And all this said remember this one thing that i've been saying forever here....by far the most important thing to me when it comes to guitar tone isn't the tone, it's the feel and dynamics. That should tell you something about this amp. Its what till now they could never do worth cr@p. Now i thing they do it better than most tube amps, at least the mustang does. (as i keep saying, WHEN TWEAKED RIGHT)

[J M Fahey]

Your scope is wrong.


EVERYBODY knows that transistors make squarewaves and tubes make smoothly rounded tops and bottoms.

I have SEEN all the nice waveforms drawn and posted by respected Sound Gurus such as Randall Smith and tons of others who should know !!!!!!!!!!!!!!!!!!!!!!!!!!



THANKS Loudthud for investigating and sharing what you find

[exclamationmark]

Modelling has come pretty far since the days of the humble waveshaper... though just how far is somewhat hard to judge since manufacturers are typically very secretive about methods used. I can't speak for the Fender modelling amplifier, but I do know that modellers such as the Axe-FX, Kemper profiling amp and the Eleven Rack all exhibit non-linear behaviour with some form of memory/dynamics. If you input a static waveform, you can observe a bias shift and watch it converge to a steady state response. I'm not sure if these systems use non-linear memory modelling or if they use linear memory (usually just an impulse response - doesn't model things like blocking distortion/transient clipping effects etc). So even though the model may show signs of memory, it may not necessarily be that accurate. It depends on the situation - for example, the microphone -> speaker interaction can be modelled with just a simple impulse response that doesn't even take into account non-linearities, because the original system is quite linear. A surprising number of albums have been silently recorded this way!

As an interesting side note, probably around 60-70% of the bands I've seen over the last couple of years have been spotted with an Axe-FX or Kemper profiling amp.

[teemuk]

It's sort of a myth that early modeling amplifiers employed only static waveshaping functions. If you study at patents from likes of Line 6 or Yamaha they do outline "piecewise" transfer functions but they also present methods to make waveshaping dynamic. Loading coefficients for transfer functions from several lookup tables based on input amplitude, or simply using feedback to alter virtual DC offsets of waveshaper stages are few examples of techniques employed already in the earliest modeling amplifiers. They've been pretty dynamic from the beginning. The effects of dynamic signal symmetry/asymmetry have been known and studied at least since early 1980's, whether to make more "tubey" solid-state amplifiers, or simply more "tubey" tube amplifiers. It would be ignorant to think the people designing amps for big companies would be unaware of such issue.

How much does asymmetry affect the touch sensitivity and feel then? It is an interesting concept to study because you can A/B a push-pull amp to a single-ended one, and the latter portrays signal asymmetry much more prominently while also lacking other dynamic behaviour like sagging or dynamic crossover distortion. So, which one is more touch sensitive and "tubey" feeling; PP or SE? ....Or are they just different.

Another interesting part of this topic to explore would be interaction with speaker's behaviour. Clearly DC offset will also offset the voice coil, leading to greater asymmetry in cone excursions. What makes this even more interesting is that cone excursion tends to be pretty asymmetric from the beginning so we arrive to issues like what would reversing the polarity of the speaker connections do... I think this is a topic which can't be explored with a mere oscilloscope. And which really hasn't been studied in great depth.

[Leo_Gnardo]

Loudthud , I first noticed that "cone bobbing" with JBL's in 1975, while watching Bob Weir & Kingfish from the stage wing. At first I thought I was just "seeing things" due to wacky light show effects, but have seen it many times since, so it's real. In another thread Juan mentioned the slow settling of aluminum voice coil formers into their magnetic gaps due to eddy currents. With a working SS amp attached, that effect might be minimized due to high damping factor, but with a tube amp, the settling-in would be slower. As the JBL coil is pushed well out of its rest position by large power pulses delivered from a tube amp - pick attack and loud low notes for instance - you might notice an overall pulsing of the cone position. Maybe this is what's going on. A/B between Al-former JBL's and some non Al-former, watch those cones & see what they do.

[loudthud]

Me? You're asking me? If you are asking me if the mustangs PA exhibits it.....you know i'm a electronic cripple so i dunno !

You have the amp and I assume you are not blind. It's a simple observational experiment. It would be nice to play the amp through a cabinet without grill cloth so you can see the front of the cone, but it should be obvious even looking at the back of the speaker.

I think looking at it logically my answer would be as follows. What you are describing is something that happens in a tube PA right? It causes a certain sound and feel. If you ask me whether my mustang's PA does that i couldn't tell you, but i will take your word that it can't. But see, heres the thing, and i think you realize this but i'll say it anyways.

It's no problem to make a solid state power amp do it. Any power amp will do it if it is fed a non-symetrical squarewave through a capacitor.

The people who develop these digital circuits today know exactly what they have to accomplish unlike in the past. So today it's a matter of figuring out how to accomplish whats needed, not what to accomplish. And once they know that i'm sure they also look at things like you mentioned and consider it from the point of view that if the PA can't be made to do that, then make the PA clean and accurate and figure out a way to make the amp model include that feel or whatever it is so the tone and feel is already there and the PA only had to accurately amplify it. Remember van halen in the early days? He would run his marshal into a attenuator and create the tone he wanted then into SS power amps and finally marshall cabs. Everything was there....the preamp tone, the distorted PA tone, and the cabs. But the power amps were in between those doing nothing more than amplifying the already fully created guitar tone aside from the speakes which came later.

Bottom line is the feel has to come through the speaker. The recording process to put a track on a CD or video will almost certainly filter it out, but the guitar player in the same room with the amp will feel it. This is why you can't tell squat on a You-Tube demo.

So the same kind of thing is going on here but it's created digitally, the same as the preamp of a tube amp is. So i'd say probably no, i doubt the mustang's PA does what you describe. But the point of digital modeling in the R&D of it is to find a way to reproduce the same feel, dynamics and tone of a tube amps in the digital domain. It really has nothing to do with a preamp and a pA that need to do the job the same as a tube amp.....it's just a digital picture of the entire path from input to speaker. And it has about a dozen different speaker sims. I'm not trying to convince you of what the mustang is capable of because you won't believe me till you try one thats been tweaked well. And thats not likely unless YOU yourself tweak it by spending hours with it. I'm just starting to really nail the feel and tone to my expectations now, and i know i'll get it even better. But unless you do the same you will remain a non believer. I was, and big time. I'm still not convinced it's AS good in all ways. I feel it is BETTER in MOST ways and possibly in all, tho i will remain skeptical till i've had more time. After all, i've only had it a week. And as i just posted in the other thread, i found the recipe for clean tones and just nailed it big time. I'm getting clean tones now that just floor me. And all this said remember this one thing that i've been saying forever here....by far the most important thing to me when it comes to guitar tone isn't the tone, it's the feel and dynamics. That should tell you something about this amp. Its what till now they could never do worth cr@p. Now i thing they do it better than most tube amps, at least the mustang does. (as i keep saying, WHEN TWEAKED RIGHT)

I theorize that one of the byproducts of cone bobbing is some slight compression. The cone moves past Xmax which means not all of the voice coil is in the magnetic gap. This lowers the speakers efficiency. Many guitar speakers have very small Xmax to begin with, but the farther away from center the cone moves, the lower it's incremental efficiency.

Have you hooked up the USB interface and looked at any patches on the Fender website?

[daz]

Yes, but thier patches sucked. I was surprised that the ones with the most positive reviews by far weren't good at all to say the least. Real surprised.
I don't understand what you guys do and i don't think about all that tech stuff. Tonite was my sat nite jam test where I play a load of mp3's and jam along. Mustang, then tube amp. Honestly, neither was 100% satisfying but the modeler easily took the prize. I think more tweaking will get a lot closer too. I feel like junking my tube amp at this point. It does however do one thing better....high gain. I'm not sure but i don't think the modeler can do it the way i like. l like the sound and feel of a clean boost but the modeler isn't good with that.