[time-nuts] Harmonics suppression in ring oscillators

[Florian Teply]

Am Thu, 19 Mar 2015 14:17:58 +0100
schrieb Attila Kinali <attila at kinali.ch>:

> On Wed, 18 Mar 2015 21:19:55 +0100
> Florian Teply <usenet at teply.info> wrote:
> 
> > > Good, I am not alone.. I felt stupid not being able to find
> > > something this basic.
> > >
> > Maybe we are stupid not being able to find something this basic, but
> > then we're stupid together, so at least we have company ;-)
> 
> :-)
>    
> > > Hm.. so only odd harmonics? What prevents the even harmonics?
> > 
> > Now you got me thinking...
> > 
> > Based on my train of thought of yesterday, the prevention of even
> > harmonics is caused by the need of an odd number of stages. Now as I
> > rethink about it I'm no longer sure that there couldn't possibly any
> > even harmonic. From what it seems to me now, it doesn't even need to
> > have an odd number of stages, it just happens to need to have an odd
> > number of INVERTING stages for it to self-start oscillation
> > reliably.
> 
> Yes, I have seen reports of ring oscillators with even number of
> inverters. But all of them said that they needed to kick the
> oscillator to reliably start it. 
> And yes, if you look at an inverter as an analog amplifier with
> a negative amplification that is non-linearly dependent on its
> input, then it is not clear at all, why there aren't any harmonics,
> both odd or even. 
> 
Well, possibly a circuit design guy can shed some light onto that. I
myself am more a technology guy. I simply use ring oscillators to
assess device degradation as they are pretty simple to measure and
provide easy numbers. After all, they are much easier to continuously
monitor than, say, I-V-curves of a single transistor...

> My only guess is, that the R_DS_on of the transistors, together
> with the C_GS of the next stage form a first order low-pass filter
> that dampens the higher modes, such that the Barkhausen Criteria
> is violated.
>
Well, indeed the R_DS_on and C_GS of the next stage do form a simple
RC low pass filter. But the time constant of that is essentially the
gate delay, and thus would allow for the higher harmonics. So, at these
harmonics there is still sufficient gain for oscillation. After all,
the output load for each stage does not change depending on the number
of stages in the chain, so why should some cell that can oscillate at
high frequencies in a 3 stage ring oscillator suddenly be unable to do
so just because now there might be 303 stages.

My guess would be slightly different: the fundamental mode of
oscillation could be considered the lowest energy state of all
oscillation modes. Assuming that the system wants to minimize energy,
this would be the mode to choose if it can't get into a steady state.
But here we are back in wild guess land, and I'm not even sure that the
concept of minimum energy states has any meaning in this context.

> I would like to do a simulation of this to get some understanding, 
> but i'm lacking the right simulation software and data of real
> transistors. (the disadvantage of being in an theoretical
> computer science group)
>
> 
> > > Ok, so you are saying, that if you start the ring oscillator
> > > in the right way, you get only the fundamental mode. What prevents
> > > higher modes from apearing during runtime? What happens if a
> > > particle passes trough the oscillator and switches one of the
> > > transistors?
> > > 
> > Well, assuming that we are talking about sane environments (which
> > your mentioning of particle strikes basically renders null and void,
> > pointing to either high energy physics or space applications which
> > can not be considered sane in this context due to their
> > posssibility to switch logic states in circuits), all possible
> > causes of introduction of higher order oscillation are excluded by
> > definition ;-) Joking aside, the case that one cell is switched
> > should be covered above.
> 
> Well.. SEUs are common enough on earth that, if you are talking
> about high reliability, you need to take them into account. Of course
> if you are going into space, then it's a rather common event.
> The goal of what I am doing here is to have a clocking system that
> can withstand arbitrary faults and self-stabilize again after a
> number of nodes (not necessarily all) regain their composure and
> start working correctly again.
> 
On the other hand, SEU are rare enough that most people couldn't spot
them. After all, a crashing computer usually is blamed on Windows or
some obscure software, but not on a bit flip caused by a rotten ion
wandering around in the wrong place. It's only a very few people that
actually are aware of Single Event Effects, and of these as far as I
can tell, 99% work either in space business or in high energy physics.
In commercial electronics, it has been a topic around the year 2000
when a great part of the industry had problems with trace impurities in
their packaging materials, but quickly after that it reduced again to
insignificance outside of military and space applications.

> Part of that is asking all those "what if..." questions that usually
> get discarded because they have low probability or because people
> think that the system will stabilize again in that case... maybe.
> 
Well, I'd say most of these questions usually do not get discarded but
do not arise in the first place.

> > Hmm, I'm already mentally sorting the list of past and potential
> > project partners to see where this might lead. In any case, should
> > you come close to Frankfurt (or Berlin for that matter), notify me
> > so we can have a beer together. If it's on my boss, even better ;-)
> 
> I definitly will :-)
> 
Perfect.

Best regards,
Florian