I've had my tires balanced twice in the past two weeks, but I have shuddering over the speed range 115-120kph. As I continue to accelerate, it comes back as I approach 130kph. I don't intend to drive at 130kph (at least not consistently), but it was just a test to see whether I can repeatably cause the vibration at certain speeds. The fact that it is associated with certain speeds suggests that it is a balance problem, but the complicating factor is that even though it only occurs at those speeds, it doesn't always do so. It seems to be somewhat dependent on how much I'm stepping on the gas, and tends to show up when I'm stepping on the gas more, e.g., accelerating through the problematic speeds, or climbing a hill, in contrast to slowing down, holding steady, or travelling downhill.
Somewhat contrary to the above observation, I found that it seemed to occur more often when I was driving West (our main highway runs east-west), and the wind was coming from the East. So it seems more noticeable when I'm travelling with the wind, which sort of goes against the perception that the shuddering is more likely to occur when stepping more on the gas pedal.
Navel-gazing sense-making conceptual modelling
As someone with engineering schooling, I tried to dream up a mental model for the shuddering. If it was a balance issue, then the reason why shuddering occurs at certain speeds is due to resonance. So I pictured a linear system with resonance peaks across the frequency spectrum. Whatever the mechanical input disturbance is, the frequency components at the resonances would be most pronounced in the output. At certain speeds, the imperfections in the rotating tires yield input disturbances with a lot of power at those resonances, leading to the pronounced shuddering at the output as, felt by the driver at the steering wheel and where the feet touch the floor.
However, it's not just the right frequency spectrum of the input that yields pronounced shuddering. The amount of power (or in audio terms, volume) of the input also affects the power in the output (i.e., the strength of the shuddering), even if the input spectrum mismatches the system spectrum. I figure that this might explain why stepping on the gas more seems to yield greater shuddering. The tires are pushing harder, so any non-roundness and lateral shimmying might translate into more forceful bumpiness, and same with variability in the springiness of the sidewalls. As well, I seemed to notice that shuddering is less when traversing newly paved dark road surface, as opposed to more light-gray (more solid and concrete-looking) old pavement. Perhaps the newer pavement is smoother, leading to less power at the input due to random disturbances from the road, and hence, less shuddering. Random "noise" has a flat-ish spectrum, but it does contribute power at the resonance frequency ranges.
I have several questions about my characterization of the shuddering.
Does the fact that the shuddering is not just speed dependent, but also present only sometimes (depending on other conditions) indicate whether it is a balance problem, roundness problem, or gross variation in sidewall springiness problem?
Does the observation that it is at least partly depending on steppiness on the gas pedal provide any further clue?
Even with even pressure on the gas, on level ground, the vibration wavers in and out over a 5 second cycle. The tire changing outlet says that vibrations due to imbalance don't wax and wane like that, and I'm wondering if anyone can corroborate that statement?
Is my rationalization in terms of linear system model with resonances approximately reflective of what is happening at a gross level?
I haven't yet decided to go for road force balancing.
2017-12-13 New Info
There seems to be negligible shuddering this evening, and the only difference is that the wheel changing outlet tightened the nuts this morning (they recommend torquing it to a specific tightness after driving on the changed tires for 80km) and it's darn cold. -15 degrees Celsius. I've heard that coldness can make the resonance go away, and that makes sense since the rubber is of a different softness. I found data on the past week or so to see if there has been such cold temperatures. This following graph shows that the answer is no (during driving hours, 9-10am and 7-8pm), so I have more confidence in the coldness as a primary factor.
Note that on the very next day, 2017-12-14, 12 hours later, the same tire establishment was willing to make another attempt at re-balancing. The person who did this found that all tires were significantly out of balance, and could not explain why this was the case after two visits. I suspect that it may have been related to the crazy demand and schedule for tire changes in the two weekends that I visited previously. I haven't found that this 3rd balancing changed the vibration noticably. It is still present, and approximately as subdued as yesterday (but still noticable). But then, the temperature is just as cold tonight as it was last night. The acid test will be during warming temperatures, when the vibration was much worse in the past.
In fact, I'm not sure if very cold temperatures provide for a fair test of what's acceptable, since there will be many days that are not that cold, especially toward mid-day. What is the prevailing idea on this, both as an explanation for the presence and absence of shuddering, and as a fair condition of what is acceptable in a tire? Thanks.
On an tangentially related matter, is there anything I have to do to re-obtain attention to my question after adding such significant new info?