Merlin negative G cutout too quick?

[Viper2000]

Originally Posted by klem

Hi Viper,

well I think I followed that (amazed).

Just two points.

1. You estimate it takes 1/4 second for the small chamber to refill as it heads towards Rich cut but previously you said the entry holes were sized to permit the full power demand which you felt would contribute an inlet hole emptying under -G in 0.75 seconds (along with the emptying due to engine demand). So wouldn't it take 0.75 seconds to refill the small chamber through the inlet holes as it heads towards Rich cut?

The holes must be bigger than required for max engine demand otherwise the rich cut couldn't happen. The fuel pump is providing fuel at roughly 2.4 times maximum engine demand (2 pumps each sized for 120% of the requirement). I rounded to ¼ second because the 0.75 second figure was also an estimate, so there's not much point being over-precise; in the end the diagram is somewhat schematic, and only shows one of the 2 jets. There's also obviously the float adding depth to the float chamber, so the error bars on the calculation are quite large. But having said that, I think that the results are in pretty good agreement with the available data, so the chances are that it's not a million miles away from the truth.

Originally Posted by klem

2. I didn't understand the part under sudden -G where you said "as soon as the fuel hits the top of the float chamber, the float will instantly float downwards". Why would the float float downwards when it is being held to the top by the raised fuel surface?

Once the fuel hits the top of the float chamber, the situation looks like the "Negative G conditions" part of the figure I uploaded; remember that local acceleration is pointing upwards and so buoyancy pushes the float down.

To put it another way, if the float chamber was huge and you were sat in an inflatable boat inside it when negative g was applied, your experience would be:

1. weightlessness, with the ceiling appearing to rush towards you
2. headache from hitting the ceiling!
3. then you'd start floating "up" towards the surface of the liquid again; but upon reaching the surface you'd realise that the floor and ceiling had changed places...

Originally Posted by klem

I think your estimates fit in with the info I was given by the MkI Hurricane pilot at reduced G where he felt it did not cause a problem down to about 0.3G with around a 2 second delay before engine response. His rich cut recovery from that he estimates to take about 1 second which is pretty close to your 1.5.

The 1.5 second figure is from RR; it's approximate and it's predicated upon the assumption of negative g rather than reduced positive.

In case of reduced positive g the rich cut won't last as long because the equilibrium is just that the float will sit higher. So there's more fuel than the 1 g equilibrium, but it's not like the negative g case where the float chamber is literally filled to overflowing. As such, recovery would be quicker, because the rate at which the engine can suck away the excess fuel is fixed for any given rpm and OAT (since the supercharger is supersonic and therefore the non dimensional flow passing though its diffuser is fixed if you want to be technical about it).

So the figures line up quite nicely.

Originally Posted by klem

btw for interest and on a parallel topic, the Spitfire MkIa/Ib pilots notes say that it is quicker to make a turning dive onto a target passing below you in the opposite direction than to roll inverted and pull through, presumably because of the fairly slow roll rate. Its not the tail chase pushover case we are talking about but based on that and the earlier barrel roll comments I have found it very effective to barrel or corkscrew down into the dive in a tail chase with the slightest stick-back as it maintains +ve G and I think is quicker than the pushover.

In the end, this sort of thing turns into a parametric study because there are quite a lot of factors involved.

But if you're unconstrained by other threats and have the necessary energy then it's probably best of all to loop and roll off the top, because you'll exit the manoeuvre with at least as much energy as you came into it with.

But in the end this is just another way of restating the energy vs angles/ lead vs lag tradeoff, and the best option will always be a function of the geometry.

The bigger the height difference, the more attractive the idea of going straight into the vertical becomes.

[klem]

Got it

For the float I forgot the G had reversed to upside down.

its too sensitive definitely...even a very gentle push forward with the stick and engine cuts...

[IvanK]

The documented value in RAE documentation specifically investigating this problem is cutout onset at 0.1G ...."i.e. at accelerometer readings of less than 0.1g"





The evolution of the cutout and time taken for recovery is also well documented in AVIA 18/1281 Tests of RAE devices for the reduction of "Negative G" engine cutting on merlin engined fighter aircraft" Though specifically looking at various cutout reduction methods some good info on cut duration and recovery in there with various amounts of negative G application.

Both these documents are available at the UK National Archives. The devs have copies of both these documents.

[Crumpp]

http://forum.1cpublishing.eu/showpos...0&postcount=36

[Scavenger]

Hi! what about Tilly orifice? Do you have any data about minimal G with this improving?

[klem]

Originally Posted by IvanK

The documented value in RAE documentation specifically investigating this problem is cutout onset at 0.1G ...."i.e. at accelerometer readings of less than 0.1g"





The evolution of the cutout and time taken for recovery is also well documented in AVIA 18/1281 Tests of RAE devices for the reduction of "Negative G" engine cutting on merlin engined fighter aircraft" Though specifically looking at various cutout reduction methods some good info on cut duration and recovery in there with various amounts of negative G application.

Both these documents are available at the UK National Archives. The devs have copies of both these documents.

Excellent IvanK,

now perhaps we can all stop guessing.