ESE's works engine tuner

[TZ350]

Well tonight's effort was a total bust, the two injectors send out such a cloud that it wets out on the bellmouth and dribbles down into the plenum which totally negates the reason for having injectors in the first place. Ie to stop fuel pooling in the plenum.

I can see the advantage of having the injectors behind the throttle slide like they have on the BRC engine. Gave up for the night after the EFI filled the crankcase with raw fuel.

Neels sent me some very informative links.

http://www.sandia.gov/ecn/gasolineSprayCombustion.php

http://www.cmt.upv.es/ECN.aspx

Neels also suggested I take a closer look at what happens at an injector, things like the "liquid length" and atomization and finally evaporation. Also the effect of impinging on a wall at what stage of the plume development.

[Frits Overmars]

how about shrouds in the transfer roofs? If you lower them, you will lower the transfer timing, so wrongly-timed exhaust pulses will have less chance of shoving fresh cylinder charge back into the transfer ducts. Moreover, lowering the transfer timing while leaving the exhaust ports alone will increase the blowdown angle.area....you could even use the movable transfer ceiling instead of a throttle.

That's actually quite hard to do, there is not much room as the exhaust ports are almost right there! BUT I have a plan, as you could imagine. The CNC will come in very handy, are you sure you don't mind me bastardising your FOS system? How much are we allowed to angle the transfers up? To give room for the exhausts to angle down over them.

You can bastardize all you like, Neil. I have the transfers angled up 25°; any less would make the exhaust floor too thin locally; any steeper would not only limit the transfers' cross-sectional area, but also 'ease' the collision of the converging transfer streams to such an extent that too much of their initial flow velocity would survive, resulting in a central scavenging column with so much velocity that too much of the fresh charge would loop via the cylinder head into the exhaust ducts.
It also depends on the envisaged piston speed. A high flow velocity of the central scavenging column suits a high-revving engine. If you want to keep the revs moderate, you ought to reduce the axial transfer angle.

[AndreasL]

i just used the area values from engmod. i assume thats the correct way ? it said total area of all 3 ports was 1229.5mm sqaure. roof and floor angle is around 15* . im not sure how to figure area with cosine involved

heres a cosine calculator but i dont know what number values to use http://www.calculatorsoup.com/calcul...of-cosines.php

The EngMod numbers should be good to use as is. That's how I have used it anyway…
I think EngMod is set to a 95% value per default if letting the program do the calc for you. I'm sure the more experienced users can give us the correct answer.

[wobbly]

Yes, all the area/effective diameter numbers in EngMod are already calculated with the duct down angle taken into account.
Make the angle greater and you will see the area and then the STA drop proportionally.
The duct to effective window area ratio, is shown automatically in the display panel, so getting 90% or 75% numbers are easy to do.

[Flettner]

You can bastardize all you like, Neil. I have the transfers angled up 25°; any less would make the exhaust floor too thin locally; any steeper would not only limit the transfers' cross-sectional area, but also 'ease' the collision of the converging transfer streams to such an extent that too much of their initial flow velocity would survive, resulting in a central scavenging column with so much velocity that too much of the fresh charge would loop via the cylinder head into the exhaust ducts.
It also depends on the envisaged piston speed. A high flow velocity of the central scavenging column suits a high-revving engine. If you want to keep the revs moderate, you ought to reduce the axial transfer angle.

Thank you Frits, set at 25 then. Angle will reduce as the transfers and revs lower. As Freddie said, "I want it all".

[peewee]

Yes, all the area/effective diameter numbers in EngMod are already calculated with the duct down angle taken into account.
Make the angle greater and you will see the area and then the STA drop proportionally.
The duct to effective window area ratio, is shown automatically in the display panel, so getting 90% or 75% numbers are easy to do.

cool i just noticed it shows the passage to window area ratio. i never noticed that before. guess i didnt pay attention. so it seems the downward angle is already figured in and no manual calculations are needed

[TZ350]

.

Ok so I have starting to work it out.



It was a bit hard but finally ... Success .....

I have been struggling with detto and running out of injection time trying to squirt in enough fuel to stop it.

In the end, the issue was all about the piston, which was (only just) tapping the head, not a big problem and I was to lazy to fix it, well tonight I did and the deto has disappeared.



I shifted the injectors back to the transfer ports and now with a bit of decent squish clearance the Beast rev's to 12,500rpm plus with injection time to spare.



So transfer port injection is a realistic option.

Initially I thought 9,000rpm was going to be the limit for transfer port injection but the Beast is now running to well past 12,000 rpm with the original transfer port injection setup.



The injectors swap over at about 7k rpm and the two main injectors still have some injection time left at 12,800 rpm and so it looks like it could go to 13 easy enough.

Ok so now that it works, its time to tune it properly and see if we can better the old carb setup.

[Farmaken]

That is some cool news TZ, best bring it to T-nui so we can have a play....

[TZ350]

That is some cool news TZ, best bring it to T-nui so we can have a play....

The Beast is probably not ready for that but I will see if I can take Chambers wire wheel FZR/GP125.

[breezy]

Thank you Frits, set at 25 then. Angle will reduce as the transfers and revs lower. As Freddie said, "I want it all".

how about a crecy sleeve set up to completly close the exhaust ports... make Sir Harry Ricardo happy..

[Frits Overmars]

I have been struggling with deto and running out of injection time trying to squirt in enough fuel to stop it.
In the end, the issue was all about the piston, which was (only just) tapping the head, not a big problem and I was to lazy to fix it, well tonight I did and the deto has disappeared.

The piston to head tapping that caused the vibration frequency that fooled your deto sentor into thinking there was deto has disappeared.
You're not the first person something like this has happened to, TeeZee. Once upon a time Volvo went Touring car racing with their 850 Turbo.
But whatever they tried (less ignition advance, more fuel, even water injection) the engine would detonate at 8.500 rpm. Or so they thought.
It turned out that the bouncing of the closing inlet valves at 8.500 rpm generated exactly the frequency that the deto sensor was trained to recognize as deto.

[Frits Overmars]

Thank you Frits, set at 25 then. Angle will reduce as the transfers and revs lower. As Freddie said, "I want it all".

You got me curiouser and curiouser. Are you thinking sliding or hinged transfer roofs? Me, I've even considered reducing the piston stroke while the engine was running, raising the BDC position while maintaining the TDC position, so the transfers would only partly open (I'd probably had too much to drink).
I hope you will come up with something much, much simpler.

[seattle smitty]

Maybe a sleeve valve arrangement, not the usual kind, but the sleeve moving up and down (very short travel, just changing port timing) with throttle position or rpm-related control (or in a boat, your slider-pipe control) . . . . It would violate the KISS principle, however.

(Frits, I'm hoping you will say something about my 24/7 notions on the French site, please.) EDIT: Damn, I just looked, and somehow that post got almost completely deleted!! Nevermind, sorry.

[seattle smitty]

Okay, since the French cut me short (for once I don't think I did it, because it was all up for a while), here's the gist of my 24/7 speculation:


It draws from Frits' 24/7 intake concept, with which i'll assume you all are familiar; he uses a reedblock in which the reeds swing out of the way once the engine rpm has risen past the point at which the pipe comes in. It also relates to a recent remark by Frits that he had experimented with bypassing the crankcase, and mounting a pair of carbs over a pair of transfers; he said that the engine ran okay but failed due to insufficient lubrication at the bearings. Finally, it relates to Wobbly's story of a water injection experiment in which the bearings didn't like being lubricated partly with emulsified water.

My first notion for a 24/7 system was to use two carbs. The first carb, which would get the engine up to "pipe-speed" would be valved in one of the usual ways (piston-port, rotary valve, or reed valve), while the second, valveless carb would feed directly to the crankcase when it opened. I suggested this on Pit-Lane a while back; Frits or others said that the idea had already been proposed.

My current suggestion, yet to receive much feedback (possibly because it got deleted???), is to locate the second (24/7) carb on the side of the cylinder opposite the exhaust, directly above the lower portion of the boost-port. I would re-direct the bottom of the boost port to connect directly and smoothly with this second carburetor, and plug off its previous connection to the crankcase. The boost port would only "draw" when this second carb is open, which is only when the engine is on the pipe (obviously part of the success of a 24/7 scheme using two carbs depends on good sealing of the second carb when it is not in use).

Now rather than having a modern-style skinny boost-port, one could try an older-style wide boost-port, as used to be seen in some lower-power 2-strokes, Or, you could have dual boost-ports. Or, you could make your direct connection to the B transfer ports. Or to the "finger-ports" that we used to grind into old-fashioned 3-port cylinders (I have several such engines, mostly Konigs). Or any combination of these. You could end up having only the "A" pair of transfers connected to the crankcase, rather like very old loop-scavenged 2-strokes, and none of the other ports would come into play until the engine got up to "pipe-speed."

Plainly, such a development program would call for having a good supply of spare pistons!! But at least as a concept (until you tell me that it's unworkable or pointless for some reason) it would seem to me that at least it would solve both the problem Frits had with lack of lube to the bearings (and there must be a reason he tried that layout), and the problem Wobbly had with water in the bearings. The crankcase is always getting lubricating oil from the primary carburetor. If desired, water injection could go in to the secondary carb, which has no direct connection to the crankcase.

I'd be very interested in your reactions. I have boxes of obsolete engine parts I can play with and screw up, LOL.

FWIW, I just bought a machine shop (well, that's over-stated, but it feels like it). Where previously I have owned a number of small lathes, currently a 13X36", and have had a small Taiwanese knee mill, soon I'll have set up a heavy 16X40" lathe and a 2J head Bridgeport with all attachments, oh the glory!!

--Smitty

[Frits Overmars]

Maybe a sleeve valve arrangement, not the usual kind, but the sleeve moving up and down (very short travel, just changing port timing) with throttle position or rpm-related control (or in a boat, your slider-pipe control) . . . . It would violate the KISS principle, however.

I would like to vary the transfer port height so much that it can be used instead of an inlet throttle; that would mean closing the transfer ports almost completely.
Using a sleeve would mean varying the exhaust timing as well. I'm not sure I want that. Using a sleeve also means creating a thermal barrier between the piston and the surrounding cylinder material. And I'm quite sure I don't want that. So I would KISS the sleeve goodbye (although I remain curious about Flettners sleeve creation).