Honda Power Unit Hardware & Software

[gruntguru]

Its a long straight.

[_cerber1]

I understand that this is a straight line, but why does the turnover schedule have such a shape? Mercedes and Ferrari do not have this, problems with vibrations?

[kasio]

What is it?

https://radikal.ru

It is not actual telemetry. its good to look, but its somehow not right. if Ver Speed curve at no moment is in front of botas how can he be in front. so Its a good overview, but do not take it so serious.
As about the part You have highlighted - it can be an error of "telemetrizing" (when adding telemetry to video) or "detelemetrizing" (when extracting telemetry from the Youtube video ) or it can be as well the honda mguk-to-mguh-to-mguk or just simple air resistance effect.

[_cerber1]

It is not actual telemetry. its good to look, but its somehow not right. if Ver Speed curve at no moment is in front of botas how can he be in front. so Its a good overview, but do not take it so serious.
As about the part You have highlighted - it can be an error of "telemetrizing" (when adding telemetry to video) or "detelemetrizing" (when extracting telemetry from the Youtube video ) or it can be as well the honda mguk-to-mguh-to-mguk or just simple air resistance effect.

On the Chinese line, similar fluctuations of revolutions are visible, and only at Honda.

[restless]

Could it be due to reaching fuel-flow cap?
Oscilating RPMs at upper fuel-flow limit?

[henry]

If the engine speed varies then the road speed must vary, unless there is clutch slip or wheel slip, which is very unlikely. So my take would be that this is telemetry transcription errors.

[Laserguru]

What is it?

https://radikal.ru

Is it resonance or mgu-h/turbo spooling?

[gruntguru]

Sorry _cerber1 I ignored the noisy trace - I am used to data that looks like that.

@kasio. The displacement x axis can be misleading. You cannot compare laptime or lap progress with this chart. It allows you to compare what each car is doing at any position on the track. You could plot a Model T Ford on the same track and it wouldn't be immediately obvious who was in front. (The speed curve is the only giveaway)

[Laserguru]

Sorry _cerber1 I ignored the noisy trace - I am used to data that looks like that.

@kasio. The displacement x axis can be misleading. You cannot compare laptime or lap progress with this chart. It allows you to compare what each car is doing at any position on the track. You could plot a Model T Ford on the same track and it wouldn't be immediately obvious who was in front. (The speed curve is the only giveaway)

Sensors tend to produce noisy data, but the Honda rpm trace demonstrates more oscillations and with higher amplitude?
Also, when listening to the Verstappen Baku on board soundtrack the oscillations seem real?
Is this the famous Honda mguh alternating between motor and generator?
Could these then be the same oscillations mclaren suffered from?
Could this have played a role in braking Gasly’s driveshaft?

A lot of questionmarks, so obviously I am not a gruntguru .

[henry]

FWIW, the speed fluctuation at the end of the straight is about 100rpm or roughly 1% variation in engine speed. The frequency is about 1hz. I measured off the screen so not the highest of accuracy but enough to further the discussion I think.

Honda’s extra harvest, switching between motoring & generating happens at between 20 and 40 hz.

I don’t know what 1% slip would do to the rear tyres. I’m guessing it’s not beneficial.

[gruntguru]

The average slip at that speed would be in the vicinity of 1% or 2% so a 1% oscillation would be significant.

It would be good if someone more knowledgeable on tyres can chip in here. The tractive force would be about 6.7 kN (assuming 330 km/hr and 600kW). Normal force on the rear tyres would be about 4.5 kN plus aero - say 15 kN total?

[Craigy]

The average slip at that speed would be in the vicinity of 1% or 2% so a 1% oscillation would be significant.

It would be good if someone more knowledgeable on tyres can chip in here. The tractive force would be about 6.7 kN (assuming 330 km/hr and 600kW). Normal force on the rear tyres would be about 4.5 kN plus aero - say 15 kN total?

We're making the assumption that the gearbox is built to transfer torque to the wheels in a progressive way. If there are a pair of non-circular (slightly elliptical) gears somewhere in the transmission (such as you might find inside a fluid pump, in ordinary use), that's not necessarily going to be true.

Having 100% linear torque delivery to the tyres would be optimal in some sort of closed loop system where a low-latency sensor can detect and vary throttle application to ensure the optimal slip for traction and type temperature, but it'd be hard for the driver to feel exactly where the limit is if the driver was the sensor.
Given that the driver is the sensor and is necessarily laggy and error-prone, it may make sense to have an element in the transmission that varies the torque delivery in a nonlinear (eg. sine) pattern. The driver can then feel where the limit of adhesion is because every few feet the tyres will breach and retreat from that limit. Such a system might explain this phenomenon: https://www.motor1.com/news/39418/photo ... rol-video/

This sort of thing is used in karting, where electronics aren't permitted in things like braking systems (non-circular rotors and non-circular venting patterns on brake discs are common because they permit feel for where the limit actually is).

That said, personally I think the RPM up/down seesaw on the graph is just noise in the data and nothing to do with the above hypothetical system, but everyone likes theories, so there you go.

The only other reason I can think of why it'd do that would be to deploy energy in the ES via the H. You would pulse the H up, suck in more air, shove it through the engine like in "Q3 free" mode, just for a short period of time to let the ICE climb up into the next few tens of revs each time without getting into horrific det issues as you would if you deployed it all at the start of the straight in one lump.

It would also explain why Honda were having so many problems with instability, if this is their strategy for long straights.

I still think this is just noise on a graph. A real Honda data graph would be very helpful; obviously that isn't going to be forthcoming so soon after a race, but Honda have released this sort of thing in the past for old races.

[restless]

Maybe they do extra harvest?
1Hz oscillation for harvesting and accumulating ... somewhere, while with 20Hz they transfer via H to K ?

[HPD]

Here is a good video if you want to analyze more.
The Mercedes RPM obviously looks more stable than the Honda RPM. But I still think that it is within the normal range.

Max q3 https://streamable.com/r2tbc
Ham q3 https://streamable.com/nld6t

[roon]

That sawtooth is also apparent on the left side of the graph. Between T2 and T3 there are some noticeable differences. Looks like one gearshift where HAM stayed flat. The shape of the Honda gearshifts seem slow. Error in graph or was the gear change really so slow? Don't hear delayed shifts in ths video.

How was the graph made? Manual data entry from video reference? Sorry if this has been explained already.